University of Southern California Dissertations and Theses

Towards groundbreaking green energetic materials

(USC Thesis Other)

Towards groundbreaking green energetic materials

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content TOWARDS GROUNDBREAKING GREEN ENERGETIC MATERIALS by Guillaume Bélanger-Chabot A Dissertation Presented to the FACULTY OF THE USC GRADUATE SCHOOL UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment of the Requirements for the Degree DOCTOR OF PHILOSOPHY (CHEMISTRY) August 2015 Copyright 2015 Guillaume Bélanger-Chabot ii To my lovely wife Kathleen, who keeps making my life and my work enjoyable... iii Acknowledgements I would first like to thank my supervisor Karl Christe. Besides the far-reaching aesthetics and newsworthiness of his contribution to chemistry, he truly was an ideal boss, giving me all the freedom I wanted while being available all the time. He has helped me many times professionally as well as personally and I will be forever grateful for that. I would also like to thank Prof. Ralf Haiges who, unfortunately for him, had to answer most of my questions. He and Karl taught me most of what I know now as far as inorganic chemistry goes, but they also taught me how to think critically of everything and to conduct nice (in many meanings of the term) research. I would like to thank my colleague and good friend Martin Rahm. He was a constant source of encouragement during the challenging times at the beginning of my PhD and his unwavering optimism was a nice complement to my trademark cynicism. He and I closely collaborated on the nitryl cyanide project (and then some) and his chemical intuition really changed the way I think about chemistry (and about theoreticians). I would like to thank my friends and colleagues Piyush Deokar, Amanda Baxter, Juri Skotnitzki, Thomas Saal, Norbert Szimhardt, Dominik Leitz, Christian Mende, Andy Clough, Max Kaplan, Igor Federov, Ross Wagner and Bill Wilson. They were a joy to work with and made the whole graduate experience pleasant. Max Kaplan in particular has conducted very nice work for me and was instrumental in quickly completing a few of the projects presented herein. I would like to thank my friends and colleagues from the department of chemistry Matt Greaney, Socrates Muñoz, Fang Wang, Arjun Narayanan, Xing Ping Wu, Zhiyao Lu and Andrey Rudenko. I really enjoyed hanging out and discussing with them and it was a real privilege to share some "grad school moments" with such good people and chemists. I am thankful for the input from my PhD guidance committee, Profs. Karl Christe (he deserves to be mentioned twice), Surya Prakash, Travis Williams, Richard Brutchey, and Priya Vashishta. These people have worked on everything, it seems, and it was a real honor to have them evaluate my work and my qualifying exam proposals. I would like to thank Prof. Curt Wittig and his wife Michele who made the moving to Los Angeles much easier for us. Kathleen and I have very fond memories of their beautiful home in Santa Monica, and of the bunnies. I would like to thank the South L.A. Christian Life church community for their interest in my work and for their support. John Gregorchuk, in particular, showed a much appreciated enthusiasm for the research conducted in the Christe-Haiges group. I would like to thank my M.Sc advisor Fred Fontaine and my colleagues from back then for being a source of inspiration and encouragement and for their much needed support. I would like to acknowledge the Fonds de recherche sur la nature et les technologies (Québec) and the National Science and Engineering Research Council of Canada, the Loker Hydrocarbon Research Institute and USC's Department of Chemistry for their financial support. I also acknowledge the Office of Naval Research and the National Science Foundation for financing the work presented herein. I iv would like to thank my friends and family, who have always backed me up in what I was undertaking. I would like to thank my professors from my undergrad days. My early mentors from Université Laval, Prof. Turcotte for being a gentleman, Prof. Barbeau for teaching me good old-fashioned professionalism and Mario Boucher, M.Sc for his unwavering support; my high school professors, Annie Tanguay and Lina Carrier, for being very good people and for nurturing my budding inclination towards intellectual concerns. Almost lastly but first, really, I would like to thank my beautiful wife Kathleen for her constant support. She truly is a perfect wife and gave meaning to all these efforts. Lastly, I would like to thank God for making a universe with chemistry in it. Studying it has been a blast (pun intended). v Publications and Presentations The work presented herein had resulted in the publication of five peer-reviewed articles in high-impact journals and one conference proceedings paper at the time this section was written. "Preparation and characterization of 3,5-dinitro-1H-1,2,4-triazole" Haiges, R.; Bélanger-Chabot, G.; Kaplan, S. M.; Christe, K. O. (2015): Dalton Trans. 2015, 44, 7586. "Synthesis and Structural Characterization of 3,5-dinitro-1,2,4-triazolates" Haiges, R.; Bélanger-Chabot, G.; Kaplan, S. M.; Christe, K.O. Dalton Trans., 2015, 44, 2978. "Synthesis and Characterization of Fluorodinitramide, FN(NO 2 ) 2 " Christe, K. O.; Wilson, W. W.; Bélanger-Chabot, G.; Haiges, R.; Boatz, J. A.; Rahm, M.; Prakash, G. K. S.; Saal, T.; Hopfinger, M. Angew. Chem. Int. Ed., 2015, 54, 1316. "Nitryl Cyanide, NCNO 2 " Rahm, M.†; Bélanger-Chabot, G.†; Haiges, R.; Christe, K. O. Angew. Chem. Int. Ed. 2014, 53, 6893(†Have equally contributed). "[BH 3 C(NO 2 ) 3 ] − : The First Room-Temperature Stable (Trinitromethyl)borate" Bélanger-Chabot, G.; Rahm, M.; Haiges, R.; Christe, K. O. Angew. Chem. Int. Ed. 2013 52, 11002. "The Quest for the Missing Energetic Building Block Nitryl Cyanide, NCNO 2 " Rahm, M.; Bélanger- Chabot, G.; Haiges, R.; Christe, K. O. Proceedings of the 39th International Pyrotechnics Seminar, 2013. In addition, four oral presentations at Conferences have been given on the work presented herein. (Presenter in bold) "Reaction of NO 2 BF 4 with Trialkylsilyl Cyanides; Synthesis and Characterization of the Novel Nitryl Cyanide, NCNO 2 ." Bélanger-Chabot, G.; Rahm, M.; Haiges, R.; Christe, K. O. 248th ACS National Meeting, San Francisco, California, August 12, 2014. "Synthesis and Characterization of Fluorodinitramine, FN(NO 2 ) 2 " Christe, K. O.; Wilson, W. W.; Bélanger-Chabot, G.; Haiges, R.; Boatz, J. A.; Rahm, M.; Prakash, G. K. S.; Hopfinger, M. 248th ACS National Meeting, San Francisco, California, August 12, 2014. vi "The Quest for the Missing Energetic Building Block Nitryl Cyanide, NCNO 2 " Rahm, M.; Bélanger- Chabot, G.; Haiges, R.; Christe, K. O. 39th International Pyrotechnics Seminar, Valencia, Spain, May 2013. "High Oxygen Borates as Potential Green Replacements for Ammonium Perchlorate-based Solid Propellants" Bélanger-Chabot, G.; Rahm, M.; Haiges, R.; Christe, K. O. 243rd ACS National Meeting, San Diego, California, March 27, 2012. vii Contributions The work presented herein was for the most part published in peer-reviewed journals or was written as manuscripts intended for publication. Each chapter of this dissertation corresponds to one published article or a manuscript, with a respective appendix containing the Supporting Information. The content of these articles or manuscripts is presented herein with minor alterations to fit the format of this dissertation. Many people contributed to the work. In order to give proper credit to everyone, the following describes the individual contributions to each chapter. All of the computational work throughout was expertly performed by Dr. Martin Rahm (CHAPTER 5, CHAPTER 6, CHAPTER 8 and CHAPTER 9). Profs. Ralf Haiges and Karl Christe were extensively involved in co-writing and proof-reading all of the written material. Dr. Martin Rahm wrote most of the computational-related material. Prof. Haiges was extensively consulted in matters involving X-ray crystallography. The experimental work on the synthesis of HDNT (CHAPTER 2) was performed mostly by Prof. Ralf Haiges. I contributed several crystal structures of impurities (including one in collaboration with S. Max Kaplan) and performed two syntheses of the title compound, one of which demonstrated the formation of 5-azido-3-nitrotriazole as a dangerous side-product. The data were processed and analyzed by Prof. Haiges and me. I was heavily involved in the writing of the manuscript, which was published Dalton Transactions (Haiges, R.; Bélanger-Chabot, G.; Kaplan, S. M.; Christe, K. O. Dalton Trans. 2015, 44, 7586). The work on the synthesis of DNT salts (CHAPTER 3) was initially performed by Prof. Ralf Haiges, who contributed most of the crystal structures. S. Max Kaplan, under my supervision, reproduced most of the work. Prof. Haiges and I processed and interpreted the data. I was heavily involved in the redaction of the manuscript, which was published in Dalton Transactions (Haiges, R.; Bélanger-Chabot, G.; Kaplan, S. M.; Christe, K. O. Dalton Trans. 2015, 44, 2978). The work on (boranyl)nitroazolates (CHAPTER 4) was initiated by me and performed by S. Max Kaplan (under my supervision) and me. Piyush Deokar, Prof. Ralf Haiges and Norbert Szimhardt provided four of the five azole precursors used in the work. Norbert Szimhardt contributed one crystal structure, which I was also able to duplicate under different conditions. viii The work on (trinitromethyl)borates (CHAPTER 5) was initiated and performed by me. The article was published in Angewandte Chemie International Edition (Bélanger-Chabot, G.; Rahm, M.; Haiges, R.; Christe, K. O. Angew. Chem. Int. Ed. 2013, 52, 11002). The work on dinitramide-substituted ammonia borane (CHAPTER 6) and dinitramidoborates (CHAPTER 7) was initiated and performed by me. The work on nitryl cyanide (CHAPTER 8) was performed by both Dr. Martin Rahm, who initiated the project, and me. I performed most of the characterizations in all of the experiments. I was heavily involved in the design of the experiments conducted by Dr Rahm. He performed several very critical experiments and was instrumental in predicting the vibrational spectra and the stability of the compound, which ultimately allowed to confirm its identity beyond doubt. I detected the compound for the first time by NMR spectroscopy, developed an efficient synthetic procedure allowing its isolation in high yields and purity, fully characterized it and explored its reactivity. The article was co-written by Dr. Rahm and me and was published in Angewandte Chemie International Edition (Rahm, M.; Bélanger-Chabot, G.; Haiges, R.; Christe, K. O. Angew. Chem. Int. Ed. 2014, 53, 6893). The work on fluorodinitramide (CHAPTER 9) was mostly performed by Prof. Karl Christe and Dr. William Wilson over the course of almost twenty years. Many other people contributed to the project. I performed the last critical experiments before the publication and reproduced most of the data under several different conditions and in different solvents. I succeeded in concentrating the compound to an extent to which its most intense vibrational bands could be observed by Raman spectroscopy, which established the link between the NMR and Raman spectra, which had been obtained separately for the compound. I contributed to writing some parts of the manuscript, which was published in Angewandte Chemie International Edition (Christe, K. O.; Wilson, W. W.; Bélanger-Chabot, G.; Haiges, R.; Boatz, J. A.; Rahm, M.; Prakash, G. K. S.; Saal, T.; Hopfinger, M. Angew. Chem. Int. Ed. 2015, 54, 1316). ix Table of contents Acknowledgements ................................................................................................................................ iii Publications and Presentations ................................................................................................................. v Contributions......................................................................................................................................... vii List of isolated or observed chemical species ....................................................................................... xvii List of tables ....................................................................................................................................... xxiv List of figures ..................................................................................................................................... xxix List of abbreviations ............................................................................................................................ xliii Abstract ................................................................................................................................................ xlv CHAPTER 1 : INTRODUCTION ........................................................................................................... 1 1.1 Energetic Materials 1-4 ..................................................................................................................... 1 1.1.1 Deflagration and Detonation .................................................................................................... 1 1.1.2 Explosophores ......................................................................................................................... 1 1.1.3 Properties ................................................................................................................................ 2 1.1.4 Classification .......................................................................................................................... 6 1.1.5 Brief Historical Overview ..................................................................................................... 10 1.1.6 Research in Energetic Materials 7 ........................................................................................... 11 1.2 Environmental and Technological Challenges 1,2 ........................................................................... 12 1.2.1 Toxicity and Environmental Impact ....................................................................................... 12 1.2.2 Performance and Cost Reduction ........................................................................................... 12 1.2.3 Need for Energetic Material Breakthroughs ........................................................................... 13 1.3 Objectives .................................................................................................................................... 13 1.4 Experimental Methods ................................................................................................................. 13 1.4.1 Synthesis ............................................................................................................................... 13 1.4.2 Safety.................................................................................................................................... 15 1.5 References ................................................................................................................................... 16 x CHAPTER 2 : IMPROVED SYNTHESIS AND STRUCTURAL CHARACTERIZATION OF 3,5- DINITRO-1H-1,2,4-TRIAZOLE AND IDENTIFICATION OF IMPORTANT IMPURITIES ............... 18 2.1 Introduction ................................................................................................................................. 18 2.2 Synthesis of HDNT ...................................................................................................................... 18 2.3 Impurities and Decomposition Products of HDNT........................................................................ 21 2.4 Structural Characterization ........................................................................................................... 23 2.5 Conclusion................................................................................................................................... 32 2.6 References ................................................................................................................................... 32 CHAPTER 3 : SYNTHESIS AND CHARACTERIZATION OF SALTS OF THE 3,5-DINITRO- 1,2,4-TRIAZOLATE ANION ................................................................................................................ 34 3.1 Introduction ................................................................................................................................. 34 3.2 Synthesis ..................................................................................................................................... 35 3.3 Structural Characterization ........................................................................................................... 36 3.4 Thermal Stability ......................................................................................................................... 51 3.5 Spectroscopy ............................................................................................................................... 52 3.6 Conclusion................................................................................................................................... 53 3.7 Reference..................................................................................................................................... 53 CHAPTER 4 : SYNTHESIS OF BORANYL-BASED COMPLEX ANIONS OF NITRO- SUBSTITUTED AZOLATES ............................................................................................................... 55 4.1 Introduction ................................................................................................................................. 55 4.2 Detection of Boranyl-azolate Adducts by NMR............................................................................ 55 4.3 Isolation ....................................................................................................................................... 58 4.4 Stability ....................................................................................................................................... 60 4.5 Structural Characterization ........................................................................................................... 62 4.6 Vibrational Spectroscopy ............................................................................................................. 71 4.7 NMR Spectroscopy ...................................................................................................................... 72 4.8 Conclusion................................................................................................................................... 73 4.9 References ................................................................................................................................... 74 xi CHAPTER 5 : (TRINITROMETHYL)BORATE COMPLEX ANIONS ................................................ 76 5.1 Introduction ................................................................................................................................. 76 5.2 Detection of (trinitromethyl)borate Anions ................................................................................... 76 5.3 Isolation and Properties ................................................................................................................ 78 5.3.1 Structural Characterization .................................................................................................... 79 5.3.2 NMR Spectroscopy ............................................................................................................... 81 5.3.3 Vibrational Spectroscopy ...................................................................................................... 81 5.3.4 UV-Vis Spectroscopy ............................................................................................................ 82 5.3.5 Sensitivity and Stability......................................................................................................... 82 5.4 Theoretical Insight ....................................................................................................................... 83 5.5 Conclusion................................................................................................................................... 86 5.6 References ................................................................................................................................... 87 CHAPTER 6 : REACTION OF DINITROAMINE WITH AMMONIA-BORANE ................................ 90 6.1 Introduction ................................................................................................................................. 90 6.2 Detection of Dinitramide-substituted Ammonia-borane Species ................................................... 90 6.3 Synthesis and Isolation................................................................................................................. 91 6.4 Theoretical Insight ....................................................................................................................... 92 6.5 Structural Characterization ........................................................................................................... 93 6.6 Vibrational Spectroscopy ............................................................................................................. 95 6.7 Stability ....................................................................................................................................... 96 6.8 B-O vs B-N Isomers .................................................................................................................... 96 6.9 Conclusion................................................................................................................................... 96 6.10 References ................................................................................................................................. 97 CHAPTER 7 : ON THE EXISTENCE OF DINITRAMIDO-BORATES ............................................... 98 7.1 Introduction ................................................................................................................................. 98 7.2 NMR Observations ...................................................................................................................... 98 7.3 Conclusion................................................................................................................................. 101 xii 7.4 References ................................................................................................................................. 102 CHAPTER 8 : SYNTHESIS OF NITRYL CYANIDE, NCNO 2 ........................................................... 103 8.1 Introduction ............................................................................................................................... 103 8.2 Synthesis and Properties ............................................................................................................ 103 8.3 Potential Applications ................................................................................................................ 109 8.4 Conclusion................................................................................................................................. 110 8.5 References ................................................................................................................................. 110 CHAPTER 9 : SYNTHESIS OF FLUORODINITRAMINE ................................................................ 112 9.1 Introduction ............................................................................................................................... 112 9.2 Synthesis and Properties ............................................................................................................ 113 9.3 Conclusion................................................................................................................................. 119 9.4 References ................................................................................................................................. 120 CHAPTER 10 : SUMMARY AND OUTLOOK .................................................................................. 122 10.1 Summary, Relevance of Results, and Research Outlook ........................................................... 122 10.2 Conclusion ............................................................................................................................... 126 10.3 References ............................................................................................................................... 126 BIBLIOGRAPHY ............................................................................................................................... 128 APPENDIX 1 : ADDITIONAL INFORMATION ON IMPROVED SYNTHESIS AND STRUCTURAL CHARACTERIZATION OF 3,5-DINITRO-1H-1,2,4-TRIAZOLE AND IDENTIFICATION OF IMPORTANT IMPURITIES (CHAPTER 2) .................................................. 138 A1.1 Experimental Details ............................................................................................................... 138 A1.2 Crystallographic Information................................................................................................... 142 A1.2.1 Crystal structure report for the monoclinic modification HDNT-1 ..................................... 142 A1.2.2 Crystal structure report for the triclinic modification HDNT-2 .......................................... 144 A1.2.3 Crystal structure report for (HDNT) 3 ·4 H 2 O...................................................................... 146 A1.2.4 Crystal Structure Report for 5-ethoxy-1-methyl-3-nitro-1H-1,2,4-triazole (1) .................... 148 A1.2.5 Crystal Structure Report for 1-acetyl-3,5-diamino-1H-1,2,4-triazole (2) ............................ 150 xiii A1.2.6 Crystal Structure Report for 1-(i-propyl)-3,5-dinitro-1H-1,2,4-triazole (3) ........................ 152 A1.2.7 Crystal Structure Report for the Co-crystal of 1-(i-propyl)-3,5-dinitro-1H-1,2,4-triazole and 3-nitro-1H-1,2,4-triazole (3·4). .............................................................................................. 154 A1.2.8 Crystal Structure Report for Sodium 3-nitro-1,2,4-triazol-5-olate monohydrate (5·H 2 O) ... 156 A1.2.9 Crystal Structure Report for the Co-crystal of 5-azido-3-nitro-1,2,4-triazole and PPN + 5- azido-3-nitro-1,2,4-triazolate (6) .................................................................................................. 158 A1.2.10 Crystal Structure Report for 5-azido-3-nitro-1,2,4-triazolate-containing PPN + 3,5- dinitro-1,2,4-triazolate ................................................................................................................. 160 A1.3 References .............................................................................................................................. 162 APPENDIX 2 : ADDITIONAL INFORMATION ON SYNTHESIS AND CHARACTERIZATION OF SALTS OF THE 3,5-DINITRO-1,2,4-TRIAZOLATE ANION (CHAPTER 3) .............................. 163 A2.1 Experimental details ................................................................................................................ 163 A2.2 Crystallographic details ........................................................................................................... 172 A2.2.1 Crystal Structure Report for LiDNT·2H 2 O (2 . 2H 2 O) ......................................................... 172 A2.2.2 Crystal Structure Report for NaDNT·2H 2 O (3·2H 2 O) ....................................................... 174 A2.2.3 Crystal Structure Report for KDNT (3) ............................................................................. 176 A2.2.4 Crystal Structure Report for KDNT·2H 2 O (3 . 2H 2 O) .......................................................... 178 A2.2.5 Crystal Structure Report for RbDNT (4) ........................................................................... 180 A2.2.6 Crystal Structure Report for CsDNT (5)............................................................................ 182 A2.2.7 Crystal Structure Report for Sr(DNT) 2 ·6H 2 O (6 . 6H 2 O) ..................................................... 184 A2.2.8 Crystal Structure Report for Ba(DNT) 2 ·11H 2 O (7·11H 2 O) ................................................ 185 A2.2.9 Crystal structure report for [Ag(NH 3 )][DNT] (9) .............................................................. 188 A2.2.10 Crystal Structure Report for NH 4 [DNT] . 2H 2 O (10 . 2H 2 O) ................................................ 189 A2.2.11 Crystal Structure Report for [HNEt 3 ][DNT] (11)............................................................. 190 A2.2.12 Crystal Structure Report for Monoclinic [H 2 NEt 2 ][DNT] (12a) ....................................... 193 A2.2.13 Crystal Structure Report for Triclinic [H 2 NEt 2 ][DNT] (12b) ........................................... 195 A2.2.14 Crystal Structure Report for Guanidinium DNT (13) ....................................................... 197 A2.2.15 Crystal Structure Report for Aminoguanidinium DNT (14) ............................................. 199 xiv A2.2.16 Crystal Structure Report for Pyridinium DNT·H 2 O (15·H 2 O).......................................... 200 A2.2.17 Crystal Structure Report for Aminotetrazolium DNT·H 2 O (16·H 2 O) ............................... 201 A2.2.18 Crystal Structure Report for PPh 4 [DNT] (17) .................................................................. 204 A2.2.19 Crystal Structure Report for PPN[DNT] (18) .................................................................. 207 A2.2.20 Crystal Structure Report for TMA[DNT] (19) ................................................................. 210 A2.2.21 Crystal Structure Report for TMA[DNT]·HDNT ............................................................ 212 A2.3 References .............................................................................................................................. 214 APPENDIX 3 : ADDITIONAL INFORMATION ON SYNTHESIS OF BORANYL-BASED COMPLEX ANIONS OF NITRO-SUBSTITUTED AZOLATES (CHAPTER 4) ................................ 215 A3.1 Experimental Details ............................................................................................................... 215 A3.2 Crystallographic Information................................................................................................... 224 A3.2.1 Crystal Structure Report for PPh 4 [DNT-BH 3 ] (1a) ............................................................ 224 A3.2.2 Crystal Structure Report for PPN[DNT-BH 3 ] (1b) ............................................................ 227 A3.2.3 Crystal Structure Report for PPh 4 [TNMeNTrz-BH 3 ] (2a) .................................................. 230 A3.2.4 Crystal Structure Report for PPh 4 [FDNMeNTrz-BH 3 ] (4a) ............................................... 232 A3.2.5 Crystal Structure Report for PPh 4 [TNTz-BH 3 ] (6a) ........................................................... 234 A3.2.6 Crystal Structure Report for PPh 4 [FDNTz-BH 3 ] (8a) ........................................................ 236 A3.2.7 Crystal Structure Report for Tetraphenylphosphonium dinitro(1-H-tetrazol-5- yl)methanide monohydrate (12) ................................................................................................... 238 A3.3 References .............................................................................................................................. 240 APPENDIX 4 : ADDITIONAL INFORMATION ON (TRINITROMETHYL)BORATE COMPLEX ANIONS (CHAPTER 5) ..................................................................................................................... 242 A4.1 Experimental Details ............................................................................................................... 242 A4.2 Additional Remarks ................................................................................................................ 247 A4.3 Further Analysis of Experimental Data .................................................................................... 248 A4.4 Theoretical Data ...................................................................................................................... 249 A4.5 Predicted Vibrational Data ...................................................................................................... 253 A4.6 Crystallographic Data for PPN[BH 3 C(NO 2 ) 3 ] .......................................................................... 254 xv A4.7 References .............................................................................................................................. 255 APPENDIX 5 : ADDITIONAL INFORMATION ON REACTION OF DINITROAMINE WITH AMMONIA-BORANE (CHAPTER 6) ................................................................................................ 257 A5.1 Experimental Details ............................................................................................................... 257 A5.2 Additional Remarks ................................................................................................................ 260 A5.3 NMR Spectra .......................................................................................................................... 261 A5.4 Computational Details ............................................................................................................. 264 A5.5 Crystal Structure Report for NH 3 BH 2 N 3 O 4 (1) ......................................................................... 266 A5.6 References .............................................................................................................................. 270 APPENDIX 6 : FURTHER INFORMATION ON THE EXISTENCE OF DINITRAMIDOBORATES (CHAPTER 7)......................................................................................... 271 A6.1 Experimental Details ............................................................................................................... 271 A6.2 Additional Remarks ................................................................................................................ 273 A6.3 References .............................................................................................................................. 275 APPENDIX 7 : ADDITIONAL INFORMATION ON SYNTHESIS OF NITRYL CYANIDE, NCNO 2 (CHAPTER 8) ........................................................................................................................ 276 A7.1 Experimental Details ............................................................................................................... 276 A7.1.1 Measurement of the Physical Properties of NCNO 2 (1)...................................................... 278 A7.1.2 Computational Details ...................................................................................................... 280 A7.2 Further Computational Results and Vibrational Analysis ......................................................... 282 A7.3 Other Synthetic Approaches to Prepare NCNO 2 (1) ................................................................. 291 A7.3.1 TMSCN + NO 2 BF 4 ........................................................................................................... 291 A7.3.2 t-BuMe 2 SiCN + NO 2 BF 4 in Various Solvents. .................................................................. 292 A7.3.3 Metathesis of CN - and NO 2 - .............................................................................................. 294 A7.3.4 Nitration of HCN .............................................................................................................. 296 A7.3.5 Reactions with FNO 2 ........................................................................................................ 298 A7.3.6 Reactions with ClNO 2 ...................................................................................................... 300 A7.3.7 Nitration of Me 3 SnCN ...................................................................................................... 301 xvi A7.3.8 Nitration of Cyanamide .................................................................................................... 302 A7.3.9 Reactions of BrCN and NO 2 - ............................................................................................ 302 A7.3.10 Attempts at Chemical Purification of NCNO 2 and Reactivity of NCNO 2 ......................... 304 A7.4 References .............................................................................................................................. 306 APPENDIX 8 : ADDITIONAL INFORMATION ON SYNTHESIS OF FLUORODINITRAMINE (CHAPTER 9) ..................................................................................................................................... 308 A8.1 Experimental Details ............................................................................................................... 308 A8.2 Computational Method Description ......................................................................................... 309 A8.3 Bond Dissociation Energies ..................................................................................................... 309 A8.4 Vibrational Analysis................................................................................................................ 311 A8.5 Bonding, Structure and Electron Density Analysis ................................................................... 311 A8.6 References .............................................................................................................................. 312 xvii List of isolated or observed chemical species CHAPTER 2 xviii CHAPTER 3 xix CHAPTER 4 xx xxi CHAPTER 5 CHAPTER 6 xxii CHAPTER 7 xxiii CHAPTER 8 CHAPTER 9 xxiv List of tables Table 2.1: Crystallographic data for the three HDNT crystal structures. ................................................. 24 Table 2.2: Selected bond lengths [Å] and angles [°] for HDNT in the crystal structures. ........................ 25 Table 2.3: Crystallographic data for the crystal structures of the identified HDNT impurities and decomposition products. ........................................................................................................................ 28 Table 3.1: Selected bond lengths [Å] and angles [°] for alkali metal DNT salts 1-5. ............................... 37 Table 3.2: Crystallographic data for the DNT metal salts 1-9. ................................................................ 38 Table 3.3: Crystallographic data for salts 10-16. .................................................................................... 44 Table 3.4: Crystallographic data for the [PPh 4 ] + , [(Ph 3 P) 2 N] + , and [NMe 4 ] + DNT salts 17-19. ................ 50 Table 3.5: Sensitivity and stability data for the DNT salts studied. ........................................................ 51 Table 4.1: NMR chemical shifts (ppm) of the observed BH 3 -azole species in deuterated acetonitrile solution. ................................................................................................................................................ 57 Table 4.2: Crystallographic data for the salts 1a, 1b, 2a, 4a, 6a, 8a and 12 obtained at 100(2) K. .......... 62 Table 4.3: Selected bond lengths (Å) and angles (°) for the [Trz-BH 3 ] - salts 1a, 1b, 2a and 4a. .............. 63 Table 4.4: Selected bond lengths (Å) and angles (°) for the [Tz-BH 3 ] - salts 6a and 8a ............................ 64 Table 5.1: Comparison of selected averaged structural parameters of PPN[BH 3 C(NO 2 ) 3 ] and several covalent and ionic trinitromethanide derivatives. ................................................................................... 80 Table 6.1: Predicted NMR chemical shifts for the two linkage isomers of 1. 1 H and 11 B chemical shifts were referenced to the experimentally observed values for NH 3 . BH 3 . 14 N chemical shifts were referenced to MeNO 2 . ............................................................................................................................ 93 Table 6.2: Crystallographic details for 1. ............................................................................................... 95 Table 8.1: Fundamental vibrational modes of nitryl cyanide ................................................................ 108 Table 9.1 :Observed and calculated Raman spectra of FN(NO 2 ) 2 ......................................................... 118 Table A1.1: Sample and crystal data for the monoclinic modification HDNT-1. .................................. 143 Table A1.2: Data collection and structure refinement for the monoclinic modification HDNT-1. ........ 143 Table A1.3: Sample and crystal data for the triclinic modification HDNT-2. ....................................... 145 Table A1.4: Data collection and structure refinement for the triclinic modification HDNT-2. .............. 145 Table A1.5: Sample and crystal data for (HDNT) 3 ·4H 2 O. .................................................................... 147 Table A1.6: Data collection and structure refinement for (HDNT) 3 ·4H 2 O. ........................................... 147 xxv Table A1.7: Data collection and structure refinement for 5-ethoxy-1-methyl-3-nitro-1H-1,2,4- triazole (1). .......................................................................................................................................... 149 Table A1.8: Sample and crystal data for 1-acetyl-3,5-diamino-1H-1,2,4-triazole (2). ........................... 151 Table A1.9: Data collection and structure refinement for 2. ................................................................. 151 Table A1.10: Sample and crystal data for 1-(i-propyl)-3,5-dinitro-1H-1,2,4-triazole (3). ...................... 153 Table A1.11: Data collection and structure refinement for 3. ............................................................... 153 Table A1.12: Sample and crystal data for the co-crystal of 1-(i-propyl)-3,5-dinitro-1H-1,2,4-triazole and 3-nitro-1H-1,2,4-triazole (3·4)....................................................................................................... 155 Table A1.13: Data collection and structure refinement for 3·4. ............................................................ 155 Table A1.14: Sample and crystal data for sodium 3-nitro-1,2,4-triazol-5-olate monohydrate (5·H 2 O). .............................................................................................................................................. 157 Table A1.15: Data collection and structure refinement for 5·H 2 O. ....................................................... 157 Table A1.16: Sample and crystal data for co-crystal of 5-azido-3-nitro-1,2,4-triazole and PPN + 3- azido-5-nitro-1,2,4-triazolate (6). ......................................................................................................... 159 Table A1.17: Data collection and structure refinement for 6. ............................................................... 159 Table A1.18: Sample and crystal data for 5-azido-3-nitro-1,2,4-triazolate-containing PPN + 3,5- dinitro-1,2,4-triazolate. ........................................................................................................................ 161 Table A1.19: Data collection and structure refinement for 5-azido-3-nitro-1,2,4-triazolate-containing PPN + 3,5-dinitro-1,2,4-triazolate. ......................................................................................................... 162 Table A2.1: Sample and crystal data for LiDNT·2H 2 O (2·2H 2 O). ....................................................... 173 Table A2.2: Data collection and structure refinement for 2·2H 2 O. ....................................................... 173 Table A2.3: Sample and crystal data for NaDNT·2H 2 O (3·2H 2 O). ...................................................... 174 Table A2.4: Data collection and structure refinement for (3·2H 2 O). .................................................... 175 Table A2.5: Sample and crystal data for KDNT (3). ............................................................................ 176 Table A2.6: Data collection and structure refinement for 3. ................................................................. 176 Table A2.7: Sample and crystal data for KDNT·2H 2 O (3·2H 2 O). ........................................................ 178 Table A2.8: Data collection and structure refinement for 3·2H 2 O. ....................................................... 179 Table A2.9: Sample and crystal data for RbDNT (4). .......................................................................... 180 Table A2.10: Data collection and structure refinement for 4. ............................................................... 181 Table A2.11: Sample and crystal data for CsDNT (5). ......................................................................... 182 Table A2.12: Data collection and structure refinement for 5. ............................................................... 183 Table A2.13: Sample and crystal data for Sr(DNT) 2 . 6H 2 O (6·6H 2 O). .................................................. 185 Table A2.14: Data collection and structure refinement for 6·6H 2 O. ..................................................... 185 Table A2.15: Sample and crystal data for Ba(DNT) 2 ·11H 2 O (7·11H 2 O). ............................................. 186 xxvi Table A2.16: Data collection and structure refinement for 7·11H 2 O. ................................................... 187 Table A2.17: Sample and crystal data for [Ag(NH 3 )][DNT] (9). .......................................................... 188 Table A2.18: Data collection and structure refinement for 9. ............................................................... 189 Table A2.19: Sample and crystal data for NH 4 [DNT] ·2H 2 O (10·2H 2 O) ............................................. 190 Table A2.20: Data collection and structure refinement for 10·2H 2 O .................................................... 190 Table A2.21: Sample and crystal data for [HNEt 3 ][DNT] (11). ............................................................ 191 Table A2.22: Data collection and structure refinement for 11. ............................................................. 191 Table A2.23: Sample and crystal data for [H 2 NEt 2 ][DNT] (12a). ......................................................... 194 Table A2.24: Data collection and structure refinement for 12a. ........................................................... 194 Table A2.25: Sample and crystal data for triclinic [H 2 NEt 2 ][DNT] (12b)............................................. 196 Table A2.26: Data collection and structure refinement for 12b. ........................................................... 197 Table A2.27: Sample and crystal data for guanidinium DNT (13) ........................................................ 198 Table A2.28: Data collection and structure refinement for 13. ............................................................. 198 Table A2.29: Sample and crystal data for aminoguanidinium DNT (14). ............................................. 199 Table A2.30: Data collection and structure refinement for 14. ............................................................. 200 Table A2.31: Sample and crystal data for pyridinium DNT . H 2 O (15·H 2 O). ......................................... 201 Table A2.32: Data collection and structure refinement for 15·H 2 O. ..................................................... 201 Table A2.33: Sample and crystal data for aminotetrazolium DNT·H 2 O (16·H 2 O). ............................... 203 Table A2.34: Data collection and structure refinement for 16·H 2 O. ..................................................... 203 Table A2.35: Sample and crystal data for PPh 4 [DNT] (17). ................................................................. 205 Table A2.36: Data collection and structure refinement for 17. ............................................................. 206 Table A2.37: Sample and crystal data for PPN[DNT] (18). ................................................................. 208 Table A2.38: Data collection and structure refinement for 18. ............................................................. 209 Table A2.39: Sample and crystal data for TMA[DNT] (19). ................................................................ 210 Table A2.40: Data collection and structure refinement for 19. ............................................................. 211 Table A2.41: Sample and crystal data for TMA[DNT]·HDNT ............................................................ 213 Table A2.42: Data collection and structure refinement for TMA[DNT]·HDNT. .................................. 213 Table A3.1: Sample and crystal data for PPh 4 [DNT-BH 3 ] (1a) ............................................................ 225 Table A 3.2: Data collection and structure refinement for (1a)............................................................. 226 Table A 3.3: Sample and crystal data for PPN[DNT-BH 3 ] (1b) ........................................................... 228 Table A3.4: Data collection and structure refinement for 1b ................................................................ 229 Table A3.5: Sample and crystal data for PPh 4 [TNMeNTrz-BH 3 ] (2a) .................................................. 230 Table A 3.6: Data collection and structure refinement for 2a. .............................................................. 231 Table A3.7: Sample and crystal data for PPh 4 [FDNMeNTrz-BH 3 ] (4a) ............................................... 233 xxvii Table A3.8: Data collection and structure refinement for 4a ................................................................ 234 Table A3.9: Sample and crystal data for PPh 4 [TNTz-BH 3 ] (6a) ........................................................... 235 Table A3.10: Data collection and structure refinement for 6a. ............................................................. 235 Table A3.11: Sample and crystal data for PPh 4 [FDNTz-BH 3 ] (8a) ...................................................... 237 Table A3.12: Data collection and structure refinement for 8a .............................................................. 237 Table A3.13: Sample and crystal data for tetraphenylphosphonium dinitro(1-H-tetrazol-5- yl)methanide monohydrate (12) ........................................................................................................... 239 Table A3.14: Data collection and structure refinement for 12 .............................................................. 239 Table A4.1: Selected vibrational data for [BH 3 C(NO 2 ) 3 ] - derivatives and for NaC(NO 2 ) 3 . .................... 249 Table A4.2: Energetics of considered decomposition pathways. All energies are relative 2 unless otherwise stated. Free energies in acetonitrile (1 M and 298 K) and given within parentheses. ............. 252 Table A4.3: Calculated vibrational spectra of the gas phase C 3 symmetric [BH 3 C(NO 2 ) 3 ] - anion (B3LYP/aug-cc-pVTZ level, unscaled frequencies). ............................................................................ 253 Table A4.4: Sample and crystal data for PPN[BH 3 C(NO 2 ) 3 ]. ............................................................... 255 Table A4.5: Data collection and structure refinement for PPN[BH 3 C(NO 2 ) 3 ]. ...................................... 255 Table A5.1: Calculated and experimental vibrational frequencies (in cm -1 ) for 1. Harmonic frequencies were calculated at the B2PLYP 11 /Def2-TZVPP 12 level of theory using the ORCA 3.1 code, 13 with implicit consideration of CH 3 CN solution, as treated by the COSMO method. Frequencies were scaled by 0.98 for best fit with experimental values observed for 1 its B-O isomer. .. 264 Table A5.2: Calculated vibrational frequencies (in cm -1 ) for the B-O isomer of 1. ............................... 265 Table A5.3: Sample and crystal data for 1. .......................................................................................... 268 Table A5.4: Data collection and structure refinement for 1. ................................................................. 269 Table A7.1: Vapor pressures of NCNO 2 measured at various temperatures .......................................... 279 Table A7.2: Unscaled vibrational frequencies (cm -1 ) for C 2v nitryl cyanide (1) at four levels of theory, together with experimentally observed values. a ........................................................................ 283 Table A7.3: Unscaled vibrational frequencies (cm -1 ) for C 2v nitryl isocyanide (2) at two levels of theory showing a poor match with the observed spectra attributed to 1. a ............................................... 283 Table A7.4: Complete assignment of the observed vibrational frequencies of C 2v nitryl cyanide (1) ..... 284 Table A7.5: Experimental and theoretical NMR data used to construct fitting equation in Figure A7.5. ................................................................................................................................................... 288 Table A8.1: N-N bond distances (M06-2X/aug-cc-pVTZ) and bond dissociation enthalpies and free energies (CBS-QB3) for N(NO 2 ) 3 , FN(NO 2 ) 2 and F 2 NNO 2 ................................................................... 310 Table A8.2: Summary of calculated frequencies and Raman intensities for FN(NO 2 ) 2 , together with experimental solid-state results. ........................................................................................................... 311 xxviii Table A8.3: Summary of calculated properties for N(NO 2 ) 3 , FN(NO 2 ) 2 , F 2 NNO 2 and NF 3 ................... 312 xxix List of figures Figure 1.1: Common explosophores. ....................................................................................................... 2 Figure 1.2: Differential thermal analysis thermogram of tetraphenylphosphonium dinitrotriazolate (CHAPTER 3) showing an endotherm at 185 °C (melting) and a decomposition exotherm with an onset at 360 °C. ....................................................................................................................................... 3 Figure 1.3: Drophammer setup for the determination of impact sensitivity (OZM Research) used in the work presented herein. ....................................................................................................................... 4 Figure 1.4: Friction sensitivity apparatus (OZM research) used in the work presented herein. .................. 4 Figure 1.5: Lead azide and styphnate, two typical primary explosives. .................................................... 7 Figure 1.6: Common military secondary explosives. ............................................................................... 7 Figure 1.7: Common civil secondary explosives...................................................................................... 8 Figure 1.8: 1-phenylazonaphth-2-ol, or Sudan I, an orange dye part of several additives used for colored-smoke generation. ..................................................................................................................... 10 Figure 1.9: Pentaerythritol, a new high explosive used in World War II ................................................ 10 Figure 1.10: Prominent research explosives. ......................................................................................... 11 Figure 1.11: High-vacuum line used throughout the work of this dissertation, with the U-trap train in the center and the Heise pressure gauge on the right hand side. .............................................................. 14 Figure 1.12: Stainless steel high-vacuum line. Notice the Teflon-FEP U-trap train on the right hand side. ...................................................................................................................................................... 15 Figure 1.13: Teflon-FEP reactor after an explosion, showing the "peeled " bottom end.......................... 16 Figure 2.1: Synthesis of HDNT. ............................................................................................................ 19 Figure 2.2: Possible pathways for the formation of the HDNT impurities 1-5. ....................................... 21 Figure 2.3: Formation of azidotriazoles ................................................................................................. 23 Figure 2.4: Part of a chain made through hydrogen bonding in the monoclinic crystal structure HDNT-1. Thermal ellipsoids are shown at the 50% probability level. Hydrogen atom positions were determined from the electron density map and are depicted as spheres of arbitrary radius. The N1-N3 distance is 2.837(1) Å. ........................................................................................................................... 25 Figure 2.5: Hydrogen bonding in the crystal structure of triclinic HDNT-2. Thermal ellipsoids are shown at the 50% probability level. Some nitro groups have been omitted for clarity. Hydrogen atoms are depicted as spheres of arbitrary radius. The hydrogen atoms H1/H1a and H3/H3a show a 1:1 positional disorder, only one of the disordered atoms is shown per molecule. Selected distances (Å): N1-N12 2.959(2), N2-N2 2.975(3), N6-N13 2.858(3), N11-N11 2.971(3). ..................................... 26 xxx Figure 2.6: Hydrogen bonding in the crystal structure of (HDNT) 3 ·4H 2 O. Thermal ellipsoids are shown at the 50% probability level. Hydrogen atom positions were determined from the electron density map and are depicted as spheres of arbitrary radius. Selected distances (Å): N1-O13 2.849(4), N7-O13 2.587(4), N8-O16 2.882(3), N13-O14 2.984(5), O3-O16 2.927(4). ............................ 27 Figure 2.7: Crystal structure of 5-ethoxy-1-methyl-3-nitro-1H-1,2,4-triazole (1). Thermal ellipsoids are shown at the 50% probability level. Hydrogen atoms have been omitted for clarity. Selected distances (Å) and angles (°): C1-N1 1.351(2), C2-N1 1.326(2), N2-C2 1.346(2), N3-C1 1.311(2), C1-N4 1.455(2), N2-N3 1.364(2), N2-C5 1.461(2), N4-O2 1.223(2), N4-O3 1.227(2), C2-O1 1.321(2), C3-O1 1.467(2), C1-N1-C2 100.1(1), N1-C2-N2 117.7(1), C2-N2-N3 109.1(1), N2-N3-C1 100.9(1), N3-C1-N1 118.2(1), O2-N4-O3 125.4(1). ............................................................................... 27 Figure 2.8: Intermolecular hydrogen bonding in the crystal structure of 1-acetyl 3,5-diamino-1H- 1,2,4-triazole (2). Thermal ellipsoids are shown at the 50% probability level. Some hydrogen atoms have been omitted for clarity. Selected distances (Å): N2-N4’ 3.0771(5), N3-N5’ 3.0160(5), N5-O1’ 3.0174(6). .............................................................................................................................................. 29 Figure 2.9: Crystal structure of 1-(i-propyl)-3,5-dinitro-1H-1,2,4-triazole (3). Thermal ellipsoids are shown at the 50% probability level. Hydrogen atoms have been omitted for clarity. Selected distances (Å) and angles (°): C2-N3 1.311(2), C1-N3 1.334(2), C2-N1 1.345(2), C1-N2 1.318(2), N1-N2 1.349(2), C2-N5 1.456(2), C1-N4 1.451(2), N4-O1 1.219(2), N4-O2 1.218 (2), N5-O3 1.2147(19), N5-O4 1.221 (2), N1-C3 1.491(2), N1-N2-C1 101.75(13), N2-N1-C2 107.67(12), C2- N3-C1 99.91(13), N3-C2-N1 113.05(14), N3-C1-N2 117.62(14), O1-N4-O2 124.94(14), O3-N5-O4 125.38(15). ............................................................................................................................................ 29 Figure 2.10: Asymmetric unit of the co-crystal of 1-(i-propyl)-3,5-dinitro-1H-1,2,4-triazole and 3- nitro-1H-1,2,4-triazole (3 . 4). Thermal ellipsoids are shown at the 50% probability level. Some hydrogen atoms have been omitted for clarity. ....................................................................................... 30 Figure 2.11: Crystal structure of sodium 3-nitro-1,2,4-triazol-5-olate (5·H 2 O). Thermal ellipsoids are shown at the 50% probability level. Hydrogen atoms are depicted as spheres of arbitrary radius. Selected distances (Å) and angles (°):N4-C2 1.446(3), C1-N2 1.370(3), C2-N1 1.308(3), C2-N3 1.342(3), C1-N3 1.353(3), N1-N2 1.359(2), C1-O3 1.273(2), N4-O1 1.226(2), N4-O2 1.229(2), Na1-O1 2.5203(17), Na1-O3'' 2.3377(16), Na1-O3' 2.5658(17), Na1-N3 2.4282(19), Na1-O4 2.3365(19), Na1-O4' 2.454(2). Bond angles (°):O1-N4-O2 124.29(18), C2-N4-O1 117.94(18), C2- N4-O2 117.76(18), N3-C1-O3 127.43(19), N2-C1-O3 124.33(19), O2-N4-C2-N1 -2.3(3). .................... 30 Figure 2.12: The anion part of the crystal structure 6. The hydrogen atom is depicted as a sphere of arbitrary radius and shows a 1:1 positional disorder located at N3 and N10. Only one of the disordered positions is shown. Selected distances (Å) and angles (°): C1-N1 1.445(3), C3-N8 xxxi 1.451(3), C2-N4 1.395(3), C4-N11 1.400(3), N3-N10 2.655(3), N4-N5 1.247(3), N5-N6 1.123(3), N11-N12 1.245(3), N12-N13 1.127(3), N4-N5-N6 171.4(3), N11-N12-N13 171.5(3). ........................... 31 Figure 3.1: Synthesis of the metal salts of DNT 1-9. ............................................................................. 34 Figure 3.2: Synthesis of DNT salts 10-19. ............................................................................................. 35 Figure 3.3: The solid-state structure of NaDNT∙2H 2 O (2∙2H 2 O). Thermal ellipsoids are shown at the 50% probability level. Hydrogen atom positions were determined from the electron density map and are depicted as spheres of arbitrary radius. Selected bond distances (Å): Na1-N2 2.380(1), Na1-O1 3.033(1), Na1-O3 2.340(1), Na1-O3’ 2.469(1). ...................................................................................... 39 Figure 3.4: The solid-state structure of 3∙2H 2 O. Thermal ellipsoids are shown at the 50% probability level. Hydrogen atom positions were determined from the electron density map and are depicted as spheres of arbitrary radius. Selected bond distances (Å): K1-O1 3.144(3), K1-O1’’ 3.143 (3), K1-O2 2.906(3), K1-O4’’’ 2.909(3), K1-O5 2.770(3), K1-O6 2.703(3), K1-O6’ 2.832(3). ................................ 39 Figure 3.5: The solid-state structure of 3. Thermal ellipsoids are shown at the 50% probability level. Some nitro groups have been omitted for clarity. Selected bond distances (Å): K1-N2’ 2.871(4), K1- N5 2.714(4), K1-N8’’ 3.017(4), K1-N10 2.829(4), K1-O1 2.967(4), K1-O2’ 2.775(3), K1-O7’ 2.826(4), K2-O1 2.866(4), K2-O2 3.191(3), K2-05 3.146(5). ................................................................. 40 Figure 3.6: The solid-state structure of 4. Thermal ellipsoids are shown at the 50% probability level. Selected bond distances (Å): Rb1-N1 3.340(5), Rb1-N2 3.421(5), Rb1-N3 3.019(4), Rb1-O1 3.379(5), Rb1-O3 3.676(4), Rb1-O3’ 3.076(4), Rb1-O4’ 3.197(4). ........................................................ 41 Figure 3.7: The solid-state structure of the 5. Thermal ellipsoids are shown at the 50% probability. Selected bond distances (Å): Cs1-N1 3.340(3), Cs-N1’ 3.591(4), Cs1-N2 3.236(3), Cs1-N3 3.207(4), Cs1-O1- 3.288(3), Cs1-O2 3.392(3), Cs1-O4 3.233(3). .......................................................................... 41 Figure 3.8: One of the two independent Sr(H 2 O) 5 (DNT) 2 units in the solid-state structure of 6·6H 2 O. Thermal ellipsoids are shown at the 50% probability level. Hydrogen atom positions were determined from the electron density map and are depicted as spheres of arbitrary radius. Selected bond distances (Å): Sr1-N1 2.773(2), Sr1-N6 2.741(1), Sr1-O3 2.720(1), Sr1-O7 2.798(1), Sr1-O17 2.578(1), Sr1-O18 2.601(1), Sr1-O19 2.548(1), Sr1-O20 2.618(1), Sr1-O21 2.709(2). ........................... 42 Figure 3.9: A Ba(H 2 O) 6 (DNT) 2 unit in the solid-state structure of 7·11H 2 O. Thermal ellipsoids are shown at the 50% probability level. Hydrogen atoms were omitted due to difficulties in determining their positions. Selected bond distances (Å): Ba1-N1 2.987(5), Ba1-O2 3.143(5), Ba1-O5 2.865(5), Ba1-O6 2.854(5), Ba1-O7 2.680(5), Ba1-O8 2.754(5), Ba1-O9 2.761(5). .............................................. 42 Figure 3.10: The solid-state structure of 9. Thermal ellipsoids are shown at the 50% probability level. Hydrogen atom positions were determined from the electron density map and are depicted as xxxii spheres of arbitrary radius. Selected bond distances (Å): Ag1-N1 2.393(2), Ag1-N2 2.242(2), Ag1- N6 2.198(2). .......................................................................................................................................... 43 Figure 3.11: Part of a helical chain in the solid-state structure of 9. The nitro groups and ammonia ligands have been omitted for clarity. .................................................................................................... 44 Figure 3.12: Hydrogen bonding in the solid-state structure of 10∙2H 2 O. Thermal ellipsoids are shown at the 50% probability level. Hydrogen atom positions were determined from the electron density map and are depicted as spheres of arbitrary radius. Selected distances (Å): N3-O6’ 2.852(3), N6-O1 3.453(3), N6-O4 3.050(3), N6-O5 2.862(3), N6-O6 2.885(3), N6-O6’ 2.828(4). ......................... 45 Figure 3.13: Hydrogen bonding in 11. Thermal ellipsoids are shown at the 50% probability level. The hydrogen atoms of the alkyl chains have been omitted for clarity. Depicted hydrogen atom positions were determined from the electron density map and are depicted as spheres of arbitrary radius. The N2-N6 distance is 2.947(1) Å. ............................................................................................. 45 Figure 3.14: Hydrogen bonding in the diethylammonium salts 12a and 12b. Thermal ellipsoids are shown at the 50% probability level. The hydrogen atoms of the alkyl chains have been omitted for clarity. Depicted hydrogen atom positions were determined from the electron density map and are depicted as spheres of arbitrary radius. Selected atomic distances in the monoclinic [triclinic] polymorph (Å): N2-N6 2.929(1) [2.946(1)], N3-N6 2.962(1) [3.016(1)]. ............................................... 46 Figure 3.15: Asymmetric unit in the crystal structure of 13 with hydrogen bonding. Thermal ellipsoids are shown at the 50% probability level. Hydrogen atom positions were determined from the electron density map and are depicted as spheres of arbitrary radius. For clarity, only some of the observed hydrogen bonds are included. Selected atomic distances (Å): N1-N6 3.044(2), N2-N8 2.897(2)................................................................................................................................................. 46 Figure 3.16: Asymmetric unit in the crystal structure of 14 with hydrogen bonding. Thermal ellipsoids are shown at the 50% probability level. Hydrogen atom positions were determined from the electron density map and are depicted as spheres of arbitrary radius. For clarity, only the major component of the two disordered anion orientations and only some of the observed hydrogen bonds is included. Selected atomic distances (Å): N3-N6 3.354(7), N3-N8 3.374(6), N6-O2 2.901(6), N8- O4 2.916(5). .......................................................................................................................................... 47 Figure 3.17: Hydrogen bonding in 15∙H 2 O. Thermal ellipsoids are shown at the 50% probability level. Some hydrogen atoms have been omitted for clarity. The positions of shown hydrogen atoms were determined from the electron density map and are depicted as spheres of arbitrary radius. Selected atomic distances (Å): N2-O1w 2.861(1), N3-O2w 2.900(1), N6-O1w 2.624(1), N7-O2w 2.666(1)................................................................................................................................................. 48 xxxiii Figure 3.18: The hydrogen bonding in 16 . H 2 O bridges two cation-anion chains. Thermal ellipsoids are shown at the 50% probability level. Hydrogen atom positions were determined from the electron density map and are depicted as spheres of arbitrary radius. For clarity, only some of the observed hydrogen bonds are included. Selected atomic distances (Å): N1-O9 2.895(2), N2-N16 2.737(2), N3- N14 2.890(2), N6-O10 2.870(2), N7-N19 2.748(2), N8-N11 2.835(2), N14-O1 3.281(2), O4-O10 2.922(2), O8-O9 2.921(2). ..................................................................................................................... 49 Figure 3.19: Molecular structure of 19. Thermal ellipsoids are shown at the 50% probability level. Hydrogen atoms were omitted for clarity. .............................................................................................. 49 Figure 4.1: Coordination chemistry of the azolate ligands of this study with BH 3 . ................................. 56 Figure 4.2: Two possible isomers for the bis(boranyl)tetrazolate species observed based on the two signals observed in 11 B NMR spectroscopy. ........................................................................................... 58 Figure 4.3: ORTEP plot of the asymmetric unit of PPh 4 [DNT-BH 3 ] (1a) showing only one part of the two-part positional disorder along the C1-N4 axis. Thermal ellipsoids are drawn at the 50% probability level. Hydrogen atom positions were idealized. Hydrogen atoms are depicted as spheres of arbitrary radius. Selected structural parameters: Bond distances (Å) : N1-N2 1.342(12), N2-C1 1.321(4), C1-N3 1.331(4), N3-C2 1.361(12), C2-N1 1.341(8), C1-N4 1.458(4), N4-O1 1.219(5), N4- O2 1.222(5), N1-B1 1.585(12), C2-N5 1.460(13), N5-O3 1.208(7), N5-O4 1.219(7). Bond angles (°): N1-N2-C1 103.2(5), N2-C1-N3 117.8(3), C1-N3-C2 98.7(6), N3-C2-N1 113.0(11), C2-N1-N2 107.2(10), O1-N4-O2 125.9(3), C1-N4-O1 117.4(3), C1-N4-O2 116.6(3), O3-N5-O4 126.1(5), C2- N5-O3 116.1(8), C2-N5-O4 117.8(7), O2-N4-C1-N2 0.2(5), O4-N5-C2-N1 33(2). ................................ 65 Figure 4.4: ORTEP plot of the asymmetric unit of PPN[DNT-BH 3 ] (1b) showing only one part of the disordered DNT anion. Thermal ellipsoids are drawn at the 50% probability level. Hydrogen atom positions were idealized. Hydrogen atoms are depicted as spheres of arbitrary radius. Selected structural parameters: Bond distances (Å): N2-N1 1.355(4), N1-C2 1.330(7), C2-N3 1.314(7), N3- C1 1.322(5), C1-N2 1.340(4), C2-N5 1.441(5), N5-O3 1.214(2), N5-O4 1.225(2), N2-B1 1.608(4), C1-N4 1.441(5), N4-O1 1.228(3), N4-O2 1.210(3). Bond angles (°): N2-N1-C2 102.1(3), N1-C2-N3 118.0(3), C2-N3-C1 99.3(3), N3-C1-N2 114.3(4), C1-N2-N1 106.3(3), O3-N5-O4 125.06(18), C2- N5-O3 117.78(17), C2-N5-O4 117.17(16), C1-N4-O1 117.1(2), C1-N4-O2 117.9(3), O2-N4-C1-N2 29.6(6), O4-N5-C2-N1 15.7(4). ............................................................................................................. 66 Figure 4.5: ORTEP plot of the asymetric unit of PPh 4 [TNMeNTrz-BH 3 ] (2a). Thermal ellipsoids are drawn at the 50% probability level. Hydrogen atom positions were idealized. Hydrogen atoms are depicted as spheres of arbitrary radius. Selected structural parameters: Bond distances (Å): N1-N2 1.354(4), N2-C1 1.332(5), C1-N3 1.324(4), N3-C2 1.334(4), C2-N1 1.348(4), C1-N4 1.451(5), N4- O1 1.226(4), N4-O2 1.231(4), N1-B1 1.591(5), C2-C30 1.483(5), C30-N5 1.521(4), C30-N6 xxxiv 1.526(4), C30-N7 1.540(5), N5-O3 1.213(4), N5-O4 1.221(4), N6-O5 1.217(5), N6-O6 1.208(4), N7-O7 1.218(4), N7-O8 1.220(4). N1-N2-C1 102.5(3), N2-C1-N3 117.8(3), C1-N3-C2 99.3(3), N3- C2-N1 113.8(3), C2-N1-N2 106.5(3), O1-N4-O2 125.4(3), C1-N4-O1 116.2(3), C1-N4-O2 118.4(3). Bond angles (°): O3-N5-O4 126.9(3), O5-N6-O6 128.1(3), O7-N7-O8 126.3(3), O3-N5-C30 114.3(3), O5-N6-C30 117.2(3), O8-N7-C30 116.7(3), N5-C30-N6 106.5(3), N6-C30-N7 104.5(3), N7-C30-N5 109.7(3), O2-N4-C1-N2 -6.5(5). ......................................................................................... 67 Figure 4.6: ORTEP plot the asymmetric unit of PPh 4 [FDNMeNTrz-BH 3 ] (4a) showing only one part of the disordered FDNMeNTrz anion. Thermal ellipsoids are drawn at the 50% probability level. Hydrogen atom positions were idealized. Hydrogen atoms are depicted as spheres of arbitrary radius. Selected structural parameters: Bond distances (Å):N1-N2 1.356(12), N2-C1 1.423(17), C1- N3 1.299(18), N3-C2 1.336(12), C2-N1 1.353(13), C1-N4 1.425(9), N4-O2 1.204(7), N4-O1 1.228(7), N1-B1 1.596(14), C2-C3 1.470(14), C3-F1 1.367(15), C3-N6 1.524(12), C3-N5 1.549(13), N5-O3 1.277(15), N5-O4 1.250(16), N6-O5 1.208(17), N6-O6 1.208(16). Bond angles (°): N1-N2-C1 102.5(9), N2-C1-N3 113.8(9), C1-N3-C2 103.2(10), N3-C2-N1 113.3(11), C2-N1- N2 107.1(10), O2-N4-O1 124.5(6), C1-N4-O2 116.8(6), C1-N4-O1 118.6(5), O3-N5-O4 128.4(11), O5-N6-O6 126.0(13), O3-N5-C3 110.9(11), O4-N5-C3 120.6(11), O5-N6-C3 119.2(12), O6-N6-C3 114.8(12), F1-C3-N5 107.4(12), F1-C3-N6 107.0(11), N5-C3-N6 104.3(8), O1-N4-C1-N2 7(1). ........... 68 Figure 4.7: ORTEP plot of the asymmetric unit of PPh 4 [TNTz-BH 3 ] (6a). Thermal ellipsoids are drawn at the 50% probability level. Hydrogen atom positions were idealized. Hydrogen atoms are depicted as spheres of arbitrary radius. Selected structural parameters: Bond distances (Å): C1- N4(1.325(2), N4-N3(1.3272(19), N3-N2(1.324(2), N2-N1(1.324(2), N1-C1(1.335(2), N3- B1(1.583(2), C1-C2(1.480(2), C2-N5(1.535(3), C2-N6(1.521(3), O1-N5(1.198(2), O2-N5(1.205(2), O3-N6(1.210(2), O4-N6(1.212(2), C2-N7(1.527(3), O5-N7(1.210(3), O6-N7(1.206(3). Shortest intermolecular distance : O6-O6 2.860(2) Å. Bond angles (°): C1-N4-N3(101.84(14), N4-N3- N2(112.25(13), N3-N2-N1(107.66(13), N2-N1-C1(104.41(14), N1-C1-N4(113.83(15), O1-N5- O2(127.23(19), O3-N6-O4(129.92(19), O2-N5-C2(113.68(18), O3-N6-C2(115.50(16), N5-C2- N6(108.38(15), O1-N5-C2(119.10(17), O4-N6-C2(117.53(19), N5-C2-C1(113.89(15), N6-C2- C1(109.30(15), O6-N7-C2(116.59(17), O5-N7-O6(127.5(2), N5-C2-N7(104.91(16), N6-C2- N7(107.32(16), O5-N7-C2(115.94(19), N7-C2-C1(112.73(15). ............................................................. 69 Figure 4.8: ORTEP plot of the asymmetric unit of PPh 4 [FDNTz-BH 3 ] (8a) showing only one part of the disordered FDNTz anion. Thermal ellipsoids are drawn at the 50% probability level. The positions of the B-H hydrogen atoms were determined from the difference electron density map. All other hydrogen positions were idealized. The hydrogen atoms are depicted as spheres of arbitrary radius. Selected structural parameters: Bond distances (Å): C1-N4 1.329(4), N4-N3 1.318(4), N3-N2 xxxv 1.333(4), N2-N1 1.324(5), N1-C1 1.330(5), B1-H1a 1.1586(fixed), B1-H1b 1.1547(fixed), B1-H1c 1.1709(fixed), N3-B1 1.579(5), C1-C2 1.470(5), C2-N6 1.537(9), C2-N5 1.594(8), O3-N6 1.212(9), O4-N6 1.262(10), O2-N5 1.231(8), O1-N5 1.225(9), C2-F1 1.374(6). Bond angles (°): C1-N4-N3 102.5(3), N4-N3-N2 111.8(3), N3-N2-N1 107.6(3), N2-N1-C1 104.7(3), N1-C1-N4 113.4(3), O3- N6-O4 128.5(8), O1-N5-O2 124.4(8), O3-N6-C2 114.9(8), O4-N6-C2 115.7(7), O2-N5-C2 115.7(5), O1-N5-C2 119.6(7), N6-C2-N5 100.1(5), N6-C2-C1 114.0(5), N5-C2-C1 117.4(4), F1-C2- C1 116.1(4), N6-C2-F1 106.3(5), N5-C2-F1 100.9(4). ........................................................................... 70 Figure 4.9: ORTEP plot of the asymmetric unit of tetraphenylphosphonium dinitro(1-H-tetrazol-5- yl)methanide monohydrate (12), showing only one part of the disordered dinitromethanide moiety. Thermal ellipsoids are drawn at the 50% probability level. Hydrogen atom positions were idealized. Hydrogen atoms are depicted as spheres of arbitrary radius. Selected structural parameters: Bond distances (Å): C1-N1 1.322(3), N1-N2 1.364(3), N2-N3 1.301(4), N3-N4 1.334(3), N4-C1 1.337(3), C1-C2 1.465(3), C2-N6 1.477(10), C2-N5 1.291(12), O1-N5 1.238(11), O2-N5 1.246(9), O3-N6 1.247(10), O4-N6 1.252(7), N4-H5N 0.9926(fixed). Bond angles (°): C1-N1-N2 104.9(2), N1-N2- N3 110.7(2), N2-N3-N4 106.9(2), N3-N4-C1 108.2(2), N4-C1-N1 109.3(2), O1-N5-O2 122.7(11), O1-N5-C2 125.2(7), O2-N5-C2 112.2(11), O3-N6-O4 119.3(9), O3-N6-C2 121.8(6), O4-N6-C2 118.9(9), N5-C2-N6 121.7(6), N5-C2-C1 125.7(5), N6-C2-C1 112.5(4), O2-N5-C2-C1 1(2), O4-N6- C2-C1 -5(1). .......................................................................................................................................... 71 Figure 5.1: Synthesis of [BH 3 C(NO 2 ) 3 ] - salts (PPN = [Ph 3 P=N=PPh 3 ] + ). ............................................... 77 Figure 5.2: Multinuclear NMR spectra of [BH 3 C(NO 2 ) 3 ]- measured in CD 3 CN for 1 H and 11 B and in glyme for 13 C. ........................................................................................................................................ 78 Figure 5.3: ORTEP plot for one of the two symmetry-independent anions in the asymmetric unit of PPN[BH 3 C(NO 2 ) 3 ]. Thermal ellipsoids are shown at the 50% probability level (hydrogen atom positions are idealized). Selected structural parameters with M06-2X/aug-cc-pVTZ calculated values in parentheses: Bond distances (Å): B1a-C1a 1.629(3) (B-C 1.645); C1a-N3a 1.513(2); C1a-N1a 1.517(2); C1a-N2a 1.527(2) (C-N 1.523); N1a-O1a 1.215(3); N1a-O2a 1.220(3); N2a-O3a 1.223(2); N2a-O4a 1.211(2); N3a-O5a 1.225(2); N3a-O6a 1.215(2) (N-O 1.201; 1.210). Bond angles (°). O1a- N1a-O2a 125.1(2); O6a-N3a-O5a 125.82(18); O4a-N2a-O3a 124.98(19) (O-N-O 125.75). .................... 79 Figure 5.4: (Top) Relative energies of C(NO 2 ) 3 - , [BH 3 C(NO 2 ) 3 ] - , [H 3 BO-N(O)-C(NO 2 )] - and [BCl 3 C(NO 2 ) 3 ] - in the gas phase (black) and in acetonitrile solution (1M, red, within parenthesis) shown in kcal/mol. (Bottom) Electrostatic surface potentials (ESPs) plotted on constant 0.001 e/bohr 3 electron density isosurfaces, derived from M06-2X/aug-cc-pVTZ wave functions in vacuum. .... 84 Figure 5.5: The intramolecular isomerization of [BH 3 C(NO 2 ) 3 ] - to [H 3 BO-N(O)-C(NO 2 ) 2 ] - offers an approximate free energy estimate of the dissociation barrier toward BH 3 elimination. Relative xxxvi energies are shown in kcal/mol and bond lengths in Ångström (Å). Values for the process in acetonitrile solution (implicitly treated) are shown in parentheses. ......................................................... 86 Figure 6.1: Reaction of ammonia borane with dinitroamine in glyme or acetonitrile at room temperature. The last step might also involve some hydrolysis. .............................................................. 91 Figure 6.2: Relative energies of the most stable B- N and B-O isomers of 1 in the gas phase and in acetonitrile solution (red), obtained using solvent corrected coupled-cluster calculations (CBS- QB3+SMD-M06-2X/cc-pVTZ). ............................................................................................................ 92 Figure 6.3: One of the two molecules of the asymmetric unit of 1. N-H hydrogen atoms positions were idealized. B-H hydrogen atoms were determined from the electron density map and then restrained. Selected structural parameters: Bond distances (Å): N1-N2 1.3926(15), N1-N3 1.4046(15), O1-N2 1.2157(15), O2-N2 1.2214(16), O3-N3 1.2160(15), O4-N3 1.2197(15), N1-B1 1.5798(19), B1-N4 1.5895(18), B1-H1 1.100(14), B1-H2 1.080(14), Bond angles (°): O1-N2-O2 125.20(12), O3-N3-O4 126.04(12), O4-N3-N1 113.26(11), O3-N3-N1 120.61(11), O1-N2-N1 115.57(11), O2-N2-N1 119.16(11), N2-N1-N3 115.64(11), N2-N1-B1 125.69(11), N3-N1-B1 118.47(11), N1-B1-N4 107.23(11), H1-B1-H2 115.7(13), H1-B1-N4 107.8(9), H1-B1-N1 107.7(9), H2-B1-N4 110.7(9), H2-B1-N1 107.4(9), O1-N2-N1-B1 -9.4(2), O4-N3-N1-B1 -33.4(2). ..................... 94 Figure 7.1: 11 B NMR spectrum of NaBH 4 + HDN in glyme at 0°C. ....................................................... 99 Figure 7.2: 11 B NMR spectrum of TMABH 4 + excess HDN at room temperature, showing 2 and probably the glyme solvated B 2 H 5 + cation. ........................................................................................... 100 Figure 7.3: 11 B NMR spectrum showing species tentatively assigned as 3 and 4 from TMABH 4 + excess HDN in glyme which had reacted at room temperature for several days. ................................... 100 Figure 7.4: Reactions of HDN with BH 4 - as observed by multinuclear NMR spectroscopy. ................. 101 Figure 8.1: 14 N (230 K) and 13 C (238 K) NMR spectra of nitryl cyanide in SO 2 solution. ..................... 107 Figure 8.2: Calculated and observed infrared spectra of NCNO 2 in the gas-phase. ............................... 107 Figure 8.3: Observed Raman spectra of NCNO 2 in the solid state and liquid phases, together with the calculated spectrum........................................................................................................................ 108 Figure 8.4: Nitryl cyanide (1), Nitryl isocyanide (2) and 2,4,6-trinitro-1,3,5-triazine (3). Calculated relative enthalpies and Gibbs free energies are calculated at the M06-2X/cc-pVTZ level (kcal/mol, 1 atm, 298 K). Geometries (Å) are from M06-2X/cc-pVTZ and RI-B2PLYP/def2-TZVPP (*) optimizations. ...................................................................................................................................... 110 Figure 9.1: The simple fluoro- and nitro-substituted amines ................................................................ 112 Figure 9.2: 19 F NMR spectra of the KN(NO 2 ) 2 + NF 4 SbF 6 system in SO 2 solution recorded as a function of temperature........................................................................................................................ 114 xxxvii Figure 9.3: Raman spectrum of solid FN(NO 2 ) 2 recorded at -130 °C with the 4880-Å exciting line of an Ar-ion laser at two different attenuations; the bands marked by an asterisk are due to a small amount of N 2 O 4 . .................................................................................................................................. 116 Figure 9.4: 14 N NMR spectrum of FN(NO 2 ) 2 in SO 2 solution recorded at -43 °C; the weak signal at - 19.7 ppm is due to a trace of N 2 O 4 . 19 .................................................................................................... 117 Figure 9.5: Minimum energy structures of NF 3 , F 2 N(NO 2 ), FN(NO 2 ) 2 , and N(NO 2 ) 3 predicted at the M06-2X/aug-cc-pVTZ level (bond lengths in Å) viewed along the sterically active free valence electron pair on the central nitrogen atom. ........................................................................................... 119 Figure 10.1: HDNT along other useful nitroazole energetic building blocks ........................................ 122 Figure 10.2: Representative examples of [BH 2 (Azolyl) 2 ] - (Top: Haiges et al.; Bottom: Klapötke et al.). ..................................................................................................................................................... 123 Figure 10.3: NCNO 2 was shown to behave as a cyanogen radical source. ............................................ 125 Figure 10.4: Possible cyano nitration reaction of NCNO 2 at low temperature ...................................... 125 Figure 10.5: (From left to right) fluorodinitroamine and the known trinitroamine, fluoronitramide, trifluoramine, and the still unknown difluoronitroamine. ...................................................................... 125 Figure 10.6: Preliminary results indicate that significantly better yields of trinitroamine are obtained by this method than the published method, which involves the reaction of NO 2 BF 4 with KDN. ............ 126 Figure A1.1: Projection of the packing in the monoclinic modification HDNT-1 perpendicular to the 001 plane. ............................................................................................................................................ 142 Figure A1.2: Projection of the packing in the triclinic modification HDNT-2 perpendicular to the 100 plane. ............................................................................................................................................ 144 Figure A1.3: Projection of the packing in (HDNT) 3 ·4H 2 O perpendicular to the 100 plane. .................. 146 Figure A1.4: Projection of the packing in 5-ethoxy-1-methyl-3-nitro-1H-1,2,4-triazole (1) perpendicular to the 001 plane. ............................................................................................................ 148 Figure A1.5: Projection of the packing in 1-acetyl-3,5-diamino-1H-1,2,4-triazole (2) perpendicular to the 100 plane. .................................................................................................................................. 150 Figure A1.6: Projection of the packing in 1-(i-propyl)-3,5-dinitro-1H-1,2,4-triazole (3) perpendicular to the 100 plane. ............................................................................................................ 152 Figure A1.7: Projection of the packing in the co-crystal of 1-(i-propyl)-3,5-dinitro-1H-1,2,4-triazole and 3-nitro-1H-1,2,4-triazole (3·4).perpendicular to the 100 plane. ...................................................... 154 Figure A1.8: Projection of the packing in sodium 3-nitro-1,2,4-triazol-5-olate monohydrate (5·H 2 O) down the b-axis (blue= nitrogen, grey= carbon, red= oxygen, purple= sodium). ................................... 156 xxxviii Figure A1.9: Projection of the packing in co-crystal of 5-azido-3-nitro-1,2,4-triazole and PPN + 3- azido-5-nitro-1,2,4-triazolate (6) perpendicular to the 010 plane. Hydrogen atoms have been omitted for clarity. ........................................................................................................................................... 158 Figure A1.10: Disordered anion in the crystal structure of 5-azido-3-nitro-1,2,4-triazolate-containing PPN + 3,5-dinitro-1H-1,2,4-triazolate with refined ratios of 1:3 (blue= nitrogen, grey= carbon, red= oxygen, orange= phosphorus). ............................................................................................................. 160 Figure A2.1: Projection of the packing in LiDNT·2H 2 O (2·2H 2 O) along the b-axis (blue=nitrogen, red=oxygen, black=carbon, gold=lithium). .......................................................................................... 172 Figure A2.2: Projection of the packing in NaDNT·2H 2 O (3·2H 2 O) down the b-axis............................ 174 Figure A2.3: Projection of the packing in KDNT (3) down the a-axis (blue= nitrogen, grey= carbon, red= oxygen, purple= potassium). ........................................................................................................ 176 Figure A2.4: Projection of the packing of RbDNT (4) down the a-axis. ............................................... 180 Figure A2.5: Projection of the packing in CsDNT (5) down the c-axis (blue= nitrogen, grey= carbon, red= oxygen, purple= cesium). ................................................................................................ 182 Figure A2.6: Solid state structure of Sr(DNT) 2 . 6H 2 O (6·6H 2 O). Thermal ellipsoids are shown at the 50% probability level. Hydrogen atom positions were determined from the electron density map and are depicted as spheres of arbitrary radius. ........................................................................................... 184 Figure A2.7: Projection of the packing of (6·6H 2 O) down the a-axis................................................... 184 Figure A2.8: Solid state structure of Ba(DNT) 2 ·11H 2 O (7·11H 2 O). Thermal ellipsoids are shown at the 50% probability level. Hydrogen atom positions were determined from the electron density map and are depicted as spheres of arbitrary radius. No satisfactory hydrogen atoms for water molecules O5 through O9 could be found. These hydrogen atoms are therefore not depicted. ............................... 186 Figure A2.9: Projection of the packing of (7·11H 2 O) down the a-axis blue= nitrogen, grey= carbon, red= oxygen, green= barium. ............................................................................................................... 186 Figure A2.10: Projection of the packing of [Ag(NH 3 )][DNT] (9) perpendicular to the 010 plane. ........ 188 Figure A2.11: Projection of the packing of NH 4 [DNT] ·2H 2 O (10·2H 2 O) perpendicular to the 100 plane. .................................................................................................................................................. 189 Figure A2.12: Projection of the packing in [HNEt 3 ][DNT] (11) perpendicular to the 010 plane. .......... 191 Figure A2.13: Projection of the packing of monoclinic [H 2 NEt 2 ][DNT] (12a) perpendicular to the 100 plane. ............................................................................................................................................ 193 Figure A2.14: Projection of the packing of triclinic [H 2 NEt 2 ][DNT] (12b).perpendicular to the 001 plane. .................................................................................................................................................. 195 Figure A2.15: Projection of the packing of 12b perpendicular to the 010 plane. .................................. 195 Figure A2.16: Projection of the packing of 12b perpendicular to the 100 plane. .................................. 196 xxxix Figure A2.17: Projection of the packing of guanidinium DNT (13) perpendicular to the 001 plane. ..... 197 Figure A2.18: Projection of the packing of aminoguanidinium DNT (14) perpendicular to the 100 plane. .................................................................................................................................................. 199 Figure A2.19:Projection of the packing of pyridinium DNT . H 2 O (15·H 2 O) perpendicular to the 100 plane. .................................................................................................................................................. 200 Figure A2.20: Projection of the packing of aminotetrazolium DNT·H 2 O (16·H 2 O) perpendicular to the 001 plane. ...................................................................................................................................... 202 Figure A2.21: Molecular structure of PPh 4 [DNT] (17). Thermal ellipsoids are shown at the 50% probability level. Hydrogen atoms were omitted for clarity. Bond distances (Å): N1-O3 1.2261(17), N1-O4 1.2258(17), N4-O1 1.2294(17), N4-O2 1.2306(17), C1-N1 1.4516(19), C2-N4 1.4494(19), C1-N2 1.3310(19), C2-N3 1.3330(19), C2-N5 1.3347(19), C1-N5 1.3348(19), N2-N3 1.3637(18). Bond angles (°): O1-N4-O2 124.61(14), O3-N1-O4 124.43(14), C1-N1-O3 117.63(13), C1-N1-O4 117.94(13), C2-N4-O1 117.88(13), C2-N4-O2 117.51(13), O1-N4-C2-N3 -8.1(2), O3-N1-C1-N2 0.5(2). ................................................................................................................................................. 204 Figure A2.22: Projection of the packing of 17 perpendicular to the 100 plane. Hydrogen atoms were omitted for clarity. ............................................................................................................................... 205 Figure A2.23: Molecular structure of PPN[DNT] (18). Thermal ellipsoids are shown at the 50% probability level. Hydrogen atoms were omitted for clarity. Selected structural parameters: Bond lengths (Å): N4-O1 1.221(2), N4-O2 1.230(2), N5-O4 1.220(2), N5-O3 1.223(2), C1-N5 1.450(3), C2-N4 1.444(3), C1-N2 1.329(3), C2-N1 1.332(3), C1-N3 1.330(3), C2-N3 1.329(3), N1-N2 1.364(2); Bond angles (°): O1-N4-O2 124.03(19), O3-N5-O4 123.9(2), C1-N5-O3 118.20(18), C1- N5-O4 117.84(18), C2-N4-O1 117.92(19), C2-N4-O2 118.05(18), O2-N4-C2-N1 13.5(3), O4-N5- C1-N2 1.9(3). ...................................................................................................................................... 207 Figure A2.24: Projection of the packing of 18 perpendicular to the 001 plane. Hydrogen atoms were omitted for clarity. ............................................................................................................................... 208 Figure A2.25: Molecular structure of TMA[DNT]·HDNT. Thermal ellipsoids are shown at the 50% probability level. Hydrogen atom H1 has a site occupancy factor of 50%. Selected structural parameters: Bond distances (Å): N4-O1 1.2155(17), N4-O2 1.2168(17), N5-O3 1.2214(16), N5-O4 1.2220(16), C1-N4 1.4524(17), C2-N5 1.4524(16), C1-N2 1.3241(17), C2-N1 1.3199(17), C1-N3 1.3145(16), C2-N3 1.3334(16), N1-N2 1.3503(15). Bond angles (°):O1-N4-O2 126.11(13), O3-N5- O4 125.11(12), C1-N4-O1 116.88(12), C1-N4-O2 117.01(12), C2-N5-O3 117.71(11), C2-N5-O4 117.18(12), O2-N4-C1-N2 -3.2(2), O4-N5-C2-N1 5.3(2). .................................................................... 212 Figure A2.26: Projection of the packing of TMA[DNT]·HDNT perpendicular to the 100 plane. .......... 213 xl Figure A3.1:ORTEP plot of the asymmetric unit of PPh 4 [DNT-BH 3 ] 1a showing the two part disorder. Hydrogen atoms were omitted for clarity. Thermal ellipsoids are drawn at the 50% probability level. ................................................................................................................................. 224 Figure A3.2: Projection of the packing showing one part of the disordered anion 1a perpendicular to the 100 plane. ...................................................................................................................................... 225 Figure A3.3:ORTEP plot of the asymmetric unit of PPN[DNT-BH 3 ] (1b) showing the disordered [DNT-BH 3 ] - ring. Hydrogen atoms were omitted for clarity. ................................................................ 227 Figure A3.4: Projection of the packing of one part of the disordered asymmetric unit of 1b perpendicular to the 001 plane. Hydrogen atoms were omitted for clarity. ............................................ 228 Figure A3.5: Projection of the packing of PPh 4 [TNMeNTrz-BH 3 ] (2a) perpendicular to the 001 plane. .................................................................................................................................................. 230 Figure A 3.6: Asymmetric unit of PPh 4 [FDNMeNTrz-BH 3 ] (4a) showing the two-part disorder in the [FDNMeNTrz-BH 3 ] - anion. ............................................................................................................ 232 Figure A3.7: Projection of the packing of one part of the disordered asymmetric unit of 4a perpendicular to the 100 plane. ............................................................................................................ 233 Figure A3.8: Projection of the packing of PPh 4 [TNTz-BH 3 ] (6a) perpendicular to the 100 plane. ........ 234 Figure A3.9: ORTEP plot of the disordered asymmetric unit of PPh 4 [FDNTz-BH 3 ] (8a). Hydrogen atoms were omitted for clarity. ............................................................................................................ 236 Figure A3.10: Projection of the packing of one part of the disordered asymmetric unit of 8a perpendicular to the 100 plane. Hydrogen atoms were omitted for clarity. ............................................ 236 Figure A3.11: ORTEP plot of the asymmetric unit of tetraphenylphosphonium dinitro(1-H-tetrazol- 5-yl)methanide monohydrate (12) showing the two-part disordered anion. ........................................... 238 Figure A3.12: Projection of the packing of one part of the disordered asymmetric unit of 12 perpendicular to the 100 plane. Hydrogen atoms were omitted for clarity. ............................................ 239 Figure A4.1: 14 N NMR spectrum in glyme showing trace amounts of [BH 2 (C(NO 2 ) 3 ) 2 ] - at ca -15 ppm. .................................................................................................................................................... 244 Figure A4.2: 11 B NMR spectrum in glyme showing trace amounts of [BH 2 (C(NO 2 ) 3 ) 2 ] - at ca -17.5 ppm. .................................................................................................................................................... 245 Figure A4.3: Considered pathways for intramolecular decomposition of [C(NO 2 ) 3 ] - , [BH 3 C(NO 2 ) 3 ] - (2) and [BH 3 ON(O)C(NO 2 ) 2 ] - (3). Selected bond lengths are shown in Ångström (Å). Values for acetonitrile are shown within parentheses. ........................................................................................... 251 Figure A4.4: ORTEP plot of the two symmetrically independent [BH 3 C(NO 2 ) 3 ] - anions in the asymmetric unit of PPN[BH 3 C(NO 2 ) 3 ]. Thermal ellipsoids are drawn at the 50% probability level. Hydrogen atom positions were idealized. ............................................................................................. 254 xli Figure A4.5: Projection of the packing of PPN[BH 3 C(NO 2 ) 3 ] perpendicular to the 010 plane. ............. 254 Figure A5.1: Possible auto-ionisation of 1. .......................................................................................... 260 Figure A5.2: Possible formation of ammonium dinitramide from the decomposition of 1 .................... 261 Figure A5.3: 11 B{ 1 H} NMR spectrum of a DCM extract containing 1 as the major product. Signals at ca -4 ppm are associated to the major unassigned by-products. ......................................................... 261 Figure A5.4: 14 N NMR spectrum of a dichloromethane extract containing 1 as the major species. The signal at ca -33 ppm is associated to one of the major unassigned by-products. ............................. 262 Figure A5.5: 11 B NMR spectrum of isolated 1 in CD 3 CN, showing the triplet B-H coupling pattern. Signals at ca -4 ppm are associated to the major unassigned by-products, while the triplet at ca -15 is likely (NH 3 ) 2 BH 2 + ................................................................................................................................ 262 Figure A5.6: 11 B NMR spectrum of a reaction mixture of NH 3 BH 3 + excess HDN after two days at room temperature, which shows significant amounts of a species assigned to 2. The doublets at ca +2 and ca -3 ppm belong to unassigned species. The very intense signal at ca +2 ppm only has the HDN/DN - signal as a possible match (with respect to intensities) in the 14 N NMR spectrum, suggesting it does not belong to a dinitramide-substituted species. ....................................................... 263 Figure A5.7: 14 N NMR spectrum of a reaction mixture of NH 3 BH 3 + excess HDN after two days of reaction at room temperature, showing significant amounts of a species assigned to 2. The intense signal at ca -29 ppm moves towards lower field as the concentration of DN - increases. ........................ 263 Figure A5.8: ORTEP plot of the complete asymmetric unit of 1. Hydrogen atoms were found in the Fourrier difference map. N-H hydrogen atoms positions were idealized, while the B-H distances were restrained. Selected structural parameters: Bond lengths (Å): N1-N2 1.3926(15), N1-N3 1.4046(15), O1-N2 1.2157(15), O2-N2 1.2214(16), O3-N3 1.2160(15), O4-N3 1.2197(15), N1-B1 1.5798(19), B1-N4 1.5895(18), B1-H1 1.100(14), B1-H2 1.080(14), N6-N5 1.4137(16), N7-N5 1.3992(15), O5-N6 1.2199(15), O6-N6 1.2091(15), O7-N7 1.2135(15), O8-N7 1.2261(15), N5-B2 1.5825(18), N8-B2 1.5809(18), B2-H6 1.097(13), B2-H7 1.106(13); Bond angles (°): O1-N2-O2 125.20(12), O3-N3-O4 126.04(12), O4-N3-N1 113.26(11), O3-N3-N1 120.61(11), O1-N2-N1 115.57(11), O2-N2-N1 119.16(11), N2-N1-N3 115.64(11), N2-N1-B1 125.69(11), N3-N1-B1 118.47(11), N1-B1-N4 107.23(11), H1-B1-H2 115.7(13), H1-B1-N4 107.8(9), H1-B1-N1 107.7(9), H2-B1-N4 110.7(9), H2-B1-N1 107.4(9), O1-N2-N1-B1 -9.4(2), O4-N3-N1-B1 -33.4(2), O5-N6-O6 125.79(12), O7-N7-O8 126.03(11), O5-N6-N5 114.52(11), O6-N6-N5 119.65(11), O7-N7-N5 120.63(11), O8-N7-N5 113.22(10), N6-N5-N7 115.89(10), N6-N5-B2 123.38(10), N7-N5-B2 120.61(11), N5-B2-N8 109.01(11), H6-B2-H7 115.5(13), H6-B2-N8 108.6(9), H6-B2-N5 105.1(9), H7-B2-N8 109.6(9), H7-B2-N5 108.8(9), O5-N6-N5-B2 -30.4(2), O8-N7-N5-B2 -12.2(2). ................. 266 Figure A5.9: Projection of the packing of 1 perpendicular to the 001 plane. ........................................ 267 xlii Figure A5.10: Projection of the packing of 1 perpendicular to the 010 plane. ...................................... 267 Figure A6.1: Attempts at isolating and observing dinitramidoborates through halide elimination from boron trihalides. .......................................................................................................................... 274 Figure A6.2: 11 B NMR TMABH 4 + excess HDN in glyme heated for two days at 50 °C, showing an unusual signal at ca 23 ppm which could correspond to 5. ................................................................... 275 Figure A7.1: Selected tetrameric cyclo-oligomerization products of 1. Calculated geometries (Å) and relative enthalpies and Gibbs free energies are calculated at the M06-2X/cc-pVTZ level (kcal/mol, 1 atm, 298 K). ..................................................................................................................... 282 Figure A7.2: Relaxed broken symmetry scans at the PCM-UM06-2X/6-31+G(d) level hint at the reaction dynamics in acetonitrile solution. The ambivalence of CN - is illustrated by a carbon- dominated HOMO orbital and a negative electrostatic potential that favors nitrogen. MOs and ESPs are plotted on constant 0.01 and 0.001 a.u. isosurfaces, respectively. ................................................... 285 Figure A7.3: Isomerization transition states between 1 and 9, 2 and 8, 9 and 8, calculated at the PCM-M06-2X/6-31+G(d) level of theory. All energies in kcal/mol, and are shown relative to 1........... 286 Figure A7.4: Transition state corresponding to the reaction of NO 2 with NCNO 2 . Energies (relative NCNO 2 + NO 2 ) and geometries are at the CBS-QB3//M06-2X/cc-pVTZ level of theory (kcal/mol, 1 atm, 298K). Bond lengths are in Å. ..................................................................................................... 287 Figure A7.5: Theoretical GIAO-PCM-B3LYP 14 N NMR shifts versus experimental values in acetonitrile. ......................................................................................................................................... 288 Figure A7.6: Raman spectrum of a partially purified reaction mixture at -60 °C. Spectral features of 1 are seen together with HCN and one CN-stretching mode of an unidentified compound. The spectrum of TMSF has been subtracted. ............................................................................................... 292 Figure A7.7: 14 N-NMR (25 °C) of the reaction between HCN (-117.7 ppm) and NO 2 BF 4 in CFCl 3 . NCNO 2 is observed at -64.2 ppm and NCNO 2 at -171.9 ppm. The minor peak at -64.7 ppm is likely ClNO 2 (see section below on reaction with ClNO 2 ). ............................................................................. 297 Figure A7.8: Calculated and observed infrared spectra of 1 in the gas-phase; CFCl 3 background subtracted. ........................................................................................................................................... 297 xliii List of abbreviations ADN: Ammonium dinitramide ANTA: Amino-nitrotriazole CL-20: Hexanitrohexaazaisowurtzitane DCM: Dichloromethane DMS: Dimethyl sulfide DN: Dinitramide DNT: Dinitrotriazolate DSC: Differential scanning calorimetry DTA: Differential thermal analysis ESD: Electrostatic discharge sensitivity ESP: Electrostatic surface potential FDNMeNTrz: (Fluorodinitromethyl)nitrotriazolate FDNTz: (Fluorodinitromethyl)tetrazolate FOX-7: 1,1-diamino-2,2,-dinitroethene FOX-12: Guanylurea dinitramide FS: Friction sensitivity GEM: Green energetic material HE: High explosive HEDM: High energy-density material HMX: octahydro-1,3,5,7-tetranitro-1,3,5,7-tetrazocine IS: Impact sensitivity xliv NMR : Nuclear magnetic resonance (spectroscopy) PPh 4 + : Tetraphenylphosphonium PPN + : Bis (triphenylphosphine) ammonium RDX: Cyclotrimethylenetrinitramine (from Research Department Explosive) TGA: Thermogravimetric Analysis THF: Tetrahydrofuran TMA: Tetramethylammonium TNMeNTrz: (Trinitromethyl)nitrotriazolate TNT: Trinitrotoluene TNTz: (Trinitromethyl)tetrazolate xlv Abstract There is an increasingly pressing need for new, high-performance, environmentally benign energetic materials. Indeed, the state-of-the-art explosives and propellants are causing growing concern over their detrimental impact on the environment and human health. It is extremely challenging to find alternatives to the compounds currently in use, which are high performing, relatively insensitive and inexpensive to synthesize. This challenge requires the synthesis of new types of compounds. This dissertation summarizes my contribution to this effort. The approach of the Christe-Haiges research group to the synthesis of new energetic materials involves the targeting of groundbreaking, novel, highly challenging species, the development of efficient synthetic pathways and the characterization of these species. Chapters 2-3 deal with derivatives of 3,5-dinitro-1,2,4-triazole, an energetic building block with suitable properties for potential applications. In CHAPTER 2 the improved synthesis of 3,5-dinitro-1H- 1,2,4-triazole (HDNT) and the first structural characterizations of that compound, which had been used in situ for decades, are described. This chapter also discusses the identifcation and structural characterization of common impurities formed when literature synthetic methods are followed. Some of these impurities were found to be dangerously more sensitive than the targetted HDNT. CHAPTER 3 describes the synthesis and structural characterization of over twenty 3,5-dinitro-1,2,4-triazolate (DNT) salts and polymorphs thereof, most of which are highly thermally stable and insensitive. CHAPTER 4-CHAPTER 7 describe the syntheses of hydro-borane and -borate compounds with energetic ligands. Halogen-free boron-based energetic materials are promising compounds because of their environmentally benign combustion products and because of the high heat of formation of boron oxide. CHAPTER 4 describes the synthesis of salts of nitroazolate-trihydroborate complex anions using, among several nitroazolates, some of the precursors described in CHAPTER 2 and CHAPTER 3. CHAPTER 5 describes the synthesis and full characterization of the first room temperature-stable solid salts of trinitromethyl-trihydroborate. The compounds demonstrate that boron hydrides can form stable derivatives even with highly oxidizing moieties. CHAPTER 6 describes the synthesis of the dinitramide- substituted ammonia-borane NH 3 . BH 2 [N(NO 2 ) 2 ], another example of a surprisingly stable energetic boron compound and the first structurally characterized Group 13-dinitramide compound. CHAPTER 7 describes the observation of several dinitramide-hydroborates, which demonstrates the potential use of dinitramide as a weakly coordinating anion. CHAPTER 8 describes the synthesis and characterization of nitryl cyanide, a small molecule which had long eluded synthesis. This species is another spectacular example of the combination of a strong oxidizer xlvi and a reducing group. The energy density of this molecule is extremely high and the compound could eventually serve as a replacement for hydrazine as a monopropellant. CHAPTER 9 describes the synthesis of fluorodinitramide, a particularly reactive and marginally stable compound which was characterized by low-temperature NMR and vibrational spectroscopy. The study of the compound provided insight into the stability of the intriguing fluoronitroamine family. 1 CHAPTER 1: INTRODUCTION 1.1 Energetic Materials 1-4 The American Society for Testing and Materials (ASTM) defines an energetic material as a compound or mixture that contains both fuel and oxidizer which can readily react, releasing energy and gases. Such compounds are typically endothermic species (with respect to their decomposition products) in a metastable state, the integrity of which is maintained solely by their kinetic barrier towards decomposition. Their thermodynamically favorable decomposition (or self-combustion) can be triggered by heat, shock, friction or other stimuli. Because the term "energetic material" is quite vague and therefore can be abused, the term high energy density material (HEDM), which is more appropriate for propellants, explosives and pyrotechnics, is preferred. 1.1.1 Deflagration and Detonation A deflagration is the exothermic oxidation of a fuel the propagation of which proceeds through heat transfer through the reacting material at subsonic speed. In a detonation, a shockwave propagates at a supersonic speed throughout the burning energetic material. Under the influence of such a shockwave, the unreacted energetic material is compressed, resulting in a temperature increase which initiates its combustion with further heat release. A detonation is therefore a self-sustaining process. It is noteworthy that the rate at which the combution front propagates through a burning material is strongly dependent upon confinement. Indeed, when a deflagrating material is significantly confined, the reaction front may travel faster than the respective speed of sound in the material at hand, resulting in a detonation. 1.1.2 Explosophores Explosophores are chemical functional groups which confer the positive heat of formation (endothermicity) or the high reactivity which characterize energetic materials. The highly favorable formation of a N-N triple bond (with a bond dissociation energy of ca 226 kcal/mol) 5 makes nitrogen-rich groups containing N-N single or N=N double bonds, such as azido, useful explosophores. Oxygen-rich groups, such as nitro or nitrato, are commonly used as the oxidizer. Even in molecules where the decomposition of nitrogen-containing fragments to dinitrogen is the major source of energy and gas, high- 2 oxygen content groups are often used to allow the complete oxidation of the other fuel fragments, such as carbon or hydrogen atoms. Figure 1.1 shows some common explosophores. Figure 1.1: Common explosophores. 1.1.3 Properties 1.1.3.1 Stability As stated earlier, energetic materials are typically in a thermodynamically metastable state. This obviously means that the thermodynamically favored decomposition reaction will inevitably happen beyond a certain temperature. The temperature at which decomposition occurs at a significant rate is therefore a practical way of quantifying the stability of an energetic material. This is typically accomplished by identifying major exotherms on a differential scanning calorimetry (DSC), a thermogravimetric analysis (TGA) or a differential thermal analysis (DTA) thermogram of the compound (Figure 1.2). Energetic materials with practical applications generally have decomposition onsets well above 150 °C. 3 Figure 1.2: Differential thermal analysis thermogram of tetraphenylphosphonium dinitrotriazolate (CHAPTER 3) showing an endotherm at 185 °C (melting) and a decomposition exotherm with an onset at 360 °C. 1.1.3.2 Sensitivity The sensitivity of an explosive is a measure of the ease with which a specific stimulus will initiate the deflagration or detonation of an energetic material. The impact sensitivity (IS), expressed in Joules, relates to the amount of energy transferred from the impact of a falling weight required to initiate the explosion of a material (Figure 1.3). A material initiated with 39 J or less is considered impact-sensitive and hazardous. Friction sensitivity (FS) is a similar property which relates to the ease with which friction will initiate the explosion of a material. It is typically determined by varying the friction, expressed in Newtons, required for a material to explode. This is typically accomplished by placing a sample of the material between two ceramic surfaces to which a horizontal movement is applied. The weight on the top ceramic surface (and hence the friction applied to the sample) can be increased, thus increasing the friction, until an explosion is obtained (Figure 1.4). A material which can be initiated with 353 N or less of friction energy is considered friction-sensitive and hazardous. The electrostatic discharge sensitivity (ESD), expressed in Joules, relates to the amount of energy, in the form of a static discharge spark, required to set off an energetic material. ESD sensitivities are in the order of 1-20 mJ for primary explosives to a few hundred mJ for secondary explosives. Given the fact that a single person can supply up to 20 mJ of electrical discharge, 6 the ESD of a material becomes especially important for large scale operations. 4 Figure 1.3: Drophammer setup for the determination of impact sensitivity (OZM Research) used in the work presented herein. Figure 1.4: Friction sensitivity apparatus (OZM research) used in the work presented herein. 5 1.1.3.3 Performance The performance of an explosive is usually defined by its brisance, B, that is, its destructive fragmentation effect on its surrounding, expressed in kg/s 3 . This is in turn influenced by several parameters. The heat of explosion, Q (in kJ/kg), the detonation velocity D (in m/s) and the detonation pressure P (in kbar) are factors of great importance, and high values for these parameters are always desirable. The explosion temperature, T (in K) and the volume of gas V released per kg of explosive (in l/kg) are also, to a lesser extent, relevant parameters. The density, ρ, of the material is also of critical importance, since it directly affects Q, D and P. Equations 1.1 and 1.2 relate these parameters. (1.1) where F is the specific energy (in J/kg), or force of an explosive, which can be described by: (1.2) where is the maximum pressure throughout the explosion (not the detonation pressure), V is the volume of gases produced by the explosion, n is the number of moles of gas per kg of explosive decomposed, R is the gas constant and T is the temperature of explosion. For rocket propellants, a useful performance measurement is the specific impulse, I sp (in m/s), which represents the impulse delivered per mass unit of a given propellant. The specific impulse generally includes the weight of the propellant and not its mass and is then expressed in seconds. Because the specific impulse is related to the pressure difference between the interior and exterior of the nosel, it varies significantly depending on altitude and is the highest in vacuum. Specific impulses are therefore often reported at sea level (sl) or in vacuum (vac). From a chemical point of view, it is noteworthy that (1.3) which means that the specific impulse is related to the square root of the ratio of the combustion chamber temperature (T c ) and the average molecular weight of the combustion products (M). It follows that highly exothermic combustion reactions and the formation of low molecular weight combustion products (such as H 2 O, CO and N 2 ) favor the performance of propellant formulations. Typical values for propellants are ca 100-200 s for monopropellants, such as hydrazine or nitromethane and ca 250-460 s (vac) for bipropellants, such as liquid oxygen/liquid hydrogen. 7 6 1.1.3.4 Oxygen Balance An important factor affecting the performance, especially of rocket propellants, is the amount of oxygen available for the combustion of fuel fragments. This measurement is called the oxygen balance, Ω, and is expressed in %. For a given compound of general formula C a H b N c O d the oxygen balance is defined as (1.4) where M is the molecular mass of the compound. A compound which possesses enough oxygen to completely combust itself to N 2 , H 2 O and CO 2 will have a of 0 whereas a compound which only contains enough oxygen to combust to N 2 , H 2 O and CO will have a of 0. Similarly, a compound containing more oxygen than needed to reach a complete combustion will have a positive oxygen balance. Generally, a good oxygen balance, even for an explosive, yields greater heats of explosions and therefore better performances. It should be noted that, even if is 0 for a given compound, it does not indicate that CO 2 will be the only carbon product in the combustion product. Indeed, since the detonation of energetic compounds can generate temperatures of ca 3000 K, the Boudouard equilibrium (Equation 1.5) becomes important and CO will form as a major product, which will affect the expected heat of explosion and the performance of the compound as a propellant. 2CO C + CO 2 (1.5) 1.1.4 Classification 1.1.4.1 Primary Explosives Primary explosives are able to rapidly transition from combustion or deflagration to detonation. They can typically be easily initiated by relatively low intensity stimuli. Since they either release large amounts of heat and/or produce a shockwave, they are typically used to transfer detonation to larger amounts of much more insensitive explosives (secondary explosives). Typical primary explosives in use include lead azide and lead styphnate (Figure 1.5). Primary explosives usually have IS and FS values below 4 J and 10 N, respectively. Detonation velocities for primary explosives usually range from 3500 to 5500 m/s, while their heats of explosion range from 1000 to 2000 kJ/kg. 7 Figure 1.5: Lead azide and styphnate, two typical primary explosives. 1.1.4.2 Secondary Explosives Secondary explosives, or high explosives (HE), are typically much more insensitive than their primary explosive counterparts. They usually form the bulk of the energetic material used in explosive formulations and require a shockwave, usually from the detonation of a more sensitive, primary explosive, to be initiated. Although the much higher sensitivity of primary explosives and the loudness of their explosions would intuitively indicate that their detonation velocities are significantly higher than for secondary explosives, the opposite is actually the case. Indeed, secondary explosives usually have detonation velocities of 6500-9000 m/s and heats of explosion of 5000-6000 kJ/kg, both parameters being roughly twice as large for secondary explosives than for primary explosives. Military secondary explosives commonly in use today include the very well known trinitrotoluene (TNT), cyclotrimethylenetrinitramine (RDX, for Research Department Explosive), octahydro-1,3,5,7-tetranitro- 1,3,5,7-tetrazocine (HMX), nitroguanidine (NQ) and the extremely insensitive triaminotrinitrobenzene (TATB) (Figure 1.6). Common secondary explosives for civil applications include hexanitrostilbene (HNS) and the famous nitroglycerin (NG), which is used in its insensitive dynamite form (Figure 1.7). Figure 1.6: Common military secondary explosives. 8 Figure 1.7: Common civil secondary explosives 1.1.4.3 Propellants Propellants are compounds or mixtures (formulations) which, upon decomposing, generate gases used to provide forward momentum to a projectile or vehicle. Rocket propellants, in particular, are classified in solid and liquid categories. Solid propellants are typically used for the boosting stage of rocket launches, because they are easier to manufacture. They are typically used as formulations containing an oxidizer (most often ammonium perchlorate), a fuel (typically a metal such as aluminum) and a polymer binder (most often hydroxy-terminated polybutadiene), which is also a fuel. The major drawback of solid propellants is the near impossibility of stopping and controlling their combustion once initiated. Moreover, since the exhaust velocities tend to be lower for solid propellants, their specific impulse also tends to be smaller than for liquid propellants. Therefore, solid propellants are often used in a first stage in conjunction with liquid propellants. Liquid propellants are further separated into two categories, monopropellants and bipropellants. Monopropellants are single compounds which decompose exothermically in the absence of oxygen, forming hot gases which provide thrust. Perhaps the most prominent example of monopropellant is hydrazine, the catalytic decomposition of which generates hot dinitrogen, dihydrogen and ammonia gases. Other monopropellants include hydrogen peroxide and nitromethane. Because of their relatively lower energy content and specific impulse, monopropellants are typically used in small satellites for orbit adjustments and attitude control. Larger vehicles typically require the use of bipropellants, which are systems including two separate liquids, an oxidizer and a fuel, which are initiated upon contact in the combustion chamber of a rocket. Some bipropellants, such as liquid H 2 /O 2 (I sp =455 s in vacuum), 7 are cryogenic and require prohibitively low temperatures of storage for most military applications and are therefore mostly used for civil rocket and shuttle launches. Earth- storable bipropellants make use of relatively low vapor-pressure liquids. Mostly HNO 3 and N 2 O 4 (oxidant) and hydrazine derivatives (fuel), with specific impulses of ca 340 s in vacuum, 7 have found actual applications, despite the fact that some systems, such as F 2 /H 2 have much higher specific impulses 9 (above 400 s). 7 Such high performance bipropellants are however usually severely limited by the dangers associated to their handling and by the toxicity of their combustion byproducts, such as HF. 1.1.4.4 Pyrotechnics Pyrotechnic compositions are designed to produce one or several of the light, heat, sound or smoke effects in conjunction with the highly exothermic reaction associated with the burning of an energetic composition. Whereas the desired combustion rates associated with explosives are very high and those for propellants should be lower, those associated with pyrotechnics should be in an intermediate range. Pyrotechnic compositions are typically mixtures containing an oxidizer, a fuel, binders and additives intended to generate the associated effect, such as colored-light emitting compounds, smoke generators, etc. Light generating pyrotechnic compositions usually involve the use of suitable metals which, at the high temperatures occurring during the combustion of the formulation, undergo electronic excitations, the relaxation of which generates light of defined wavelengths. For example, salts of sodium are used for yellow, salts of strontium for red and boron or salts of barium for green light emission. Metals such as magnesium are often used to increase the burning temperature, which also allows the emission of white light. For these applications, ammonium perchlorate and organic propellants, such as nitrocellulose, are used as the oxidizer. Smoke generating mixtures, usually of military use, involve the production of aerosol clouds intended for signaling or as a cover. In such compositions, red phosphorus is generally the smoke generating additive. Upon combustion, either in air or more often with the aid of oxidants and other additives, the phosphorus generates a cloud of P 4 O 10 (alternatively described as P 2 O 5 ), which, upon reaction with atmospheric moisture, generates droplets of phosphoric acid. In applications where such acidic aerosols must be avoided, formulations containing zinc oxide, aluminum and hexachloroethane have been used. For colored smokes, dyes (such as in Figure 1.8) are added, which sublime upon combustion of the formulation and become part of the aerosol, thus imparting their characteristic color to it. 10 Figure 1.8: 1-phenylazonaphth-2-ol, or Sudan I, an orange dye part of several additives used for colored- smoke generation. 1.1.5 Brief Historical Overview Second-century BC China is credited for the discovery of blackpowder, a mixture containing potassium nitrate, charcoal and sulfur. Its military use started much later, in Europe, in the 13th century. In 1846, nitroglycerine was synthesized by the Italian chemist Ascanio Sobrero. This important but dangerously sensitive compound was made tremendously safer to handle by the Swedish chemist Alfred Nobel some 20 years later. Nobel patented dynamite, based on nitroglycerin absorbed on diatomaceous earth (an absorbent clay) in 1867. Trinitrotoluene was prepared and characterized towards the end of the 19th century. This compound became the most commonly used explosive in World War I. Although TNT is nowadays outperformed by modern explosives, such as HMX and RDX, the possibility of melt-casting TNT makes it a still useful energetic material. World War II saw the wide use of two new explosives, RDX and pentaerythritol tetranitrate (PETN, Figure 1.9). The use of HMX began shortly after World War II. In the 60s and 70s, HNS and TATB were produced industrially for the first time. Figure 1.9: Pentaerythritol, a new high explosive used in World War II 11 1.1.6 Research in Energetic Materials 8 The last century has seen the appearance of several very promising energetic materials, some of which are shown in Figure 1.10. Perhaps the most prominent is hexanitroheaxaazaisowurtzitane (CL-20), discovered over 20 years ago. This structurally complex, high-performance secondary explosive has not yet seen any significant application because of issues related to its sensitivity, high cost and the stability of its polymorph. The Swedish defense's Totalförsvarets forkningsinstitut (FOI) developed two promising, insensitive, widely known energetic materials, called FOX-7 (1,1-diamino-2,2-dinitroethene) and FOX-12 (guanylurea dinitramide). While their performance is comparable or inferior, respectively, to that of RDX, their high insensitivity (>20 J IS and >350 N FS, respectively) with respect to RDX (7.5 J IS and 120 N FS) could allow them to be used in insensitive munitions. In the field of primary explosives, copper salts of tetrazole derivatives, such as copper(I) 5-nitrotetrazolate (DBX-1), show promise as environmentally friendly replacements for lead-containing compounds. Lastly, ammonium dinitramide (ADN) (Eurenco, Sweden), hydrazinium nitroformate (HNF) (APP, Netherlands) and triaminoguanidium nitroformate (TAGNF) (Germany) are being commercialized as ecologically friendly, chlorine-free replacements for ammonium perchlorate. These are promising materials, but factors such as stability and compatibility with binders and fuels used in propellant formulations are hampering their widespread application. Figure 1.10: Prominent research explosives. 12 1.2 Environmental and Technological Challenges 1,2 The energetic materials in use today are the fruit of centuries of intense research efforts. The incremental improvements that can be brought to the existing materials have most likely reached or are approaching their limit. While most materials used today are perfectly acceptable as far as manufacturing costs and performance are concerned, they nevertheless remain plagued with increasingly important drawbacks. 1.2.1 Toxicity and Environmental Impact The main drawback of the state-of-the-art energetic materials in use today is the acute and/or chronic toxicity of several of their components and their combustion products. For example, compounds of heavy metals, such as lead, cadmium and barium, are used extensively as primary explosives and in pyrotechnic formulations. Lead causes adverse gastrointestinal, neuromuscular and neurological effects, which are associated with a condition called plumbism. 9 Cadmium interferes with bone metabolism, damages the kidneys and the lungs and is a known carcinogen. 10 Water-soluble barium compounds are acutely toxic (the lethal dose for BaCl 2 is ca 900 mg) and affects the nervous system, while the inhalation of insoluble barium-containing salts causes lung symptoms. 11 Commonly used secondary explosives, such as RDX and TNT, also have adverse effects on the environment. 12 Ammonium perchlorate, used in ammunitions and as a major component of solid rocket propellants, is associated with chronic thyroid problems. 13 In addition, its combustion generates hydrogen chloride. It is estimated that 270 tons of HCl are produced during the launch of an Ariane 5 vehicle, which severely impacts the immediate surroundings. The incomplete combustion of formulations containing an excess of binder with respect to the oxidizer is also liable to generate soot-like exhaust products, as well as potentially contributing to lower performances. Hydrazine and its derivatives, extensively used as propellants and bipropellants, are known carcinogens. 14 1.2.2 Performance and Cost Reduction For every 20 s increase in specific impulse, the maximum payload carried by a given amount of propellant roughly doubles. This illustrates the tremendous potential of even small improvements in the existing state-of-the-art propellants. Similarly, new explosives or pyrotechnics with improved performances would reduce the amounts of material required to accomplish similar effects and as such reduce transportation and storage costs. 13 1.2.3 Need for Energetic Material Breakthroughs As seen previously, the environmental effects of the energetic materials currently in use alone warrant extensive research efforts for new energetic materials. While the new materials investigated should be non-toxic and environmentally benign, they should also meet or even surpass the favorable properties of the state-of-the-art materials, i.e., their low sensitivity, their high stability, their low manufacturing costs and their high performances. It can easily be seen that such a goal requires significant time and efforts, as can be seen by the fact that very few new compounds have seen a significant industrial use in the past fifty years. It is also obvious that entirely new types of materials and a better fundamental understanding of the factors influencing their stability, sensitivity and performance are needed. Throughout this dissertation, my contribution to this endeavor will be described. 1.3 Objectives The work presented herein intends to explore the synthesis of several challenging compounds which could either themselves have a potential as green replacements for the state-of-the art energetic materials or serve as proof-of-concept for new promising, practical or academically important compound types. This work focuses mainly on small inorganic molecules and boron-based energetic materials. In addition, further insight into known compounds was sought where relevant, in an attempt to contribute to the general, fundamental understanding of energetic material chemistry. 1.4 Experimental Methods 1.4.1 Synthesis Most of the manipulations were performed using high vacuum-line, glovebox and Schlenk techniques. 15 While Schlenk lines are very common and their design is relatively consistent throughout research groups, the high vacuum lines used in this work are less widespread. Their main features include an integrated U-trap series allowing the separation of volatile materials by fractional condensation. Each trap is individually connected to the rest of the vacuum line and to a Heise pressure gauge, which allows the determination of the amounts of each fraction collected by pressure measurements in calibrated volumes (Figure 1.11). Each fraction collected is analyzed by IR spectroscopy by expanding the contents of each trap into a glass IR cell attached to the line. For high pressure, high oxidizers and hydrogen fluoride handling, a stainless steel high-vacuum line equipped with Teflon-FEP U-traps of roughly the same 14 design as for the glass line was used (Figure 1.12). Such lines are typically passivated by exposing them to the highly reactive chlorine trifluoride (ClF 3 ), which eliminates traces of moisture and converts the metal surfaces to inert metal fluorides. Figure 1.11: High-vacuum line used throughout the work of this dissertation, with the U-trap train in the center and the Heise pressure gauge on the right hand side. 15 Figure 1.12: Stainless steel high-vacuum line. Notice the Teflon-FEP U-trap train on the right hand side. 1.4.2 Safety The safe distance from a detonating material is proportional to the cubic root of the weight of an explosive. In a practical setting, this means that the mass of explosives used should be minimized. Typically, scales exceeding a few milimoles of energetic materials should be avoided. Furthermore, the use of special protective equipment, such as heavy leather gloves and jackets, earplugs and face shields is essential in order to minimize the risks associated with energetic material handling. In the event of a detonation, high-velocity glass or metal fragments from the container are often the highest risk. When energetic materials are isolated, Teflon-FEP reactors are used whenever possible, since they usually peel instead of breaking into fragments (Figure 1.13). 16 Figure 1.13: Teflon-FEP reactor after an explosion, showing the "peeled " bottom end. 1.5 References (1) Klapötke, T. M. Chemistry of High-energy Materials; de Gruyter: Berlin, 2012. (2) Rahm, M. PhD dissertation, Royal Institute of Technology, 2010. (3) Stierstorfer, J. PhD dissertation, Ludwig-Maximilian University, 2009. (4) Brinck, T. Green Energetic Materials; Wiley: Chichester, 2014. (5) Darwent, B. deB. Bond Dissociation Energies in Simple Molecules, National Standard Reference Data System, 1970. (6) Talawar, M. B.; Agrawal, A. P.; Anniyappan, M.; Wani, D. S.; Bansode, M. K.; Gore, G. M. J. Hazard. Mater. 2006, 137, 1074. (7) Stratton, H. "Theoretical Performance of Rocket Propellant Combinations", Advanced Programs, Rocketdyne Division, Rockwell International, 537A-2 rev. 2-88. (8) Klapötke, T. M. High Energy Density Materials; Springer: Berlin, 2007; Vol. 125. (9) Pearce, J. M. S. Eur. Neurol. 2007, 57, 118. (10) Hartwig, A. In Cadmium: From Toxicity to Essentiality; Sigel, A., Sigel, H., Sigel, R. K. O., Eds.; Springer Netherlands: 2013; Vol. 11, p 491. (11) Doig, A. T. Thorax 1976, 31, 30. (12) Panz, K.; Miksch, K. J. Environ. Manage. 2012, 113, 85. (13) Braverman, L. E.; He, X.; Pino, S.; Cross, M.; Magnani, B.; Lamm, S. H.; Kruse, M. B.; Engel, A.; Crump, K. S.; Gibbs, J. P. J. Clin. Endocrinol. Metab. 2005, 90, 700. 17 (14) Vernot, E. H.; MacEwen, J. D.; Bruner, R. H.; Haun, C. C.; Kinkead, E. R.; Prentice, D. E.; Hall III, A.; Schmidt, R. E.; Eason, R. L.; Hubbard, G. B.; Young, J. T. Fundam. Appl. Toxicol. 1985, 5, 1050. (15) Shriver, D. F.; Drezdzon, M. A. The Manipulation of Air-Sensitive Compounds; J. Wiley and Sons: New York, 1986. 18 CHAPTER 2: IMPROVED SYNTHESIS AND STRUCTURAL CHARACTERIZATION OF 3,5-DINITRO-1H-1,2,4-TRIAZOLE AND IDENTIFICATION OF IMPORTANT IMPURITIES 2.1 Introduction In recent years, much effort has been devoted to the development of environmentally friendly, green energetic materials (GEM). 1 Important building blocks for energetic materials are nitrogen-rich cyclic compounds, such as triazoles and tetrazoles, which can be further functionalized with explosophore groups, such as nitro, N-nitro, azo, or azido. 3-nitro-1H-1,2,4-triazole 2-5 and especially 3,5-dinitro-1H- 1,2,4-triazole (HDNT) 6,7 have attracted considerable interest for energetic material and high oxygen carrier applications. A significant number of compounds containing the 3,5-dinitro-1,2,4-triazolate anion have been prepared and characterized. 5,8-16 The synthesis of HDNT was first described in the 1960s 8,9 but the neat compound has not been isolated. Instead, the compound has been handled in solution only and subsequently been converted into salts containing the 3,5-dinitro-1,2,4-triazolate anion. Although related compounds, such as 3-amino-5-nitro-1H-1,2,4-triazole, 17 5-azido-3-nitro-1,2,4-1H-triazole 18 and 3-nitro- 1-(2H-tetrazol-5-yl)-1H-1,2,4-triazol-5-amine 19 have been structurally characterized, the crystal structure of HDNT has not been reported so far. As part of the Christe-Haiges group's ongoing research on energetic materials, the preparation of neat HDNT was investigated. Herein the synthesis and purification of HDNT together with the structural characterization of two of its polymorphs, as well as of a HDNT hydrate are described. In addition, several impurities found in HDNT that was prepared according to literature methods were identified and structurally characterized. 2.2 Synthesis of HDNT The synthetic route for the preparation of HDNT is shown in Figure 2.1. Potassium 3,5-dinitro-1,2,4- triazolate (KDNT) was prepared from 3,5-diamino-1,2,4-triazole (DAT, guanazole) through a Sandmeyer reaction 20 according to a modified literature procedure. 21 While the DAT starting material required for the synthesis of HDNT is commercially available, it can be prepared in high yield from 2-cyanoguanidine (dicyandiamide) and hydrazine hydrate. 22 The addition of a sulfuric acid solution of DAT to a vigorously 19 stirred aqueous solution of excess sodium nitrite at 50-60 °C results in the nitration of the triazole. It was found that the yield of the synthesis and the purity of the resulting KDNT were highly dependent on the addition rate of the DAT solution as well as the speed at which the solution was stirred. Low addition rates coupled with vigorous stirring with a mechanical stirrer instead of a magnetic stirrer resulted in higher yields and higher purity of the resulting KDNT and, subsequently, HDNT. Figure 2.1: Synthesis of HDNT. Extraction of the reaction mixture with ethyl acetate, followed by immediate evaporation of the solvent using a rotary evaporator at ambient temperature resulted in the isolation of crude HDNT as a yellow to dark orange oil or paste. This crude product contained various acidic impurities, which rendered it prone to decomposition. On several occasions, samples of crude HDNT started to decompose exothermically 20 within minutes once the solvent had been removed. These samples started to heat up very fast, reaching temperatures in excess of 60 °C and released large quantities of brown NO 2 gas. It was possible to quench this potentially dangerous decomposition reaction through the addition of large amounts of water. The resulting aqueous HDNT solutions contained various amounts of decomposition products and were discarded. The decomposition of the crude HDNT was avoided by immediately re-dissolving the dark orange evaporation residue in acetone. When the resulting yellow acetone solution was treated with an excess of potassium carbonate, carbon dioxide was evolved and the mixture turned orange. After filtration and evaporation of the solvent a yellow to orange solid was obtained. Recrystallization of the solid from water resulted in the isolation of the dihydrate KDNT·2H 2 O as a crystalline solid which could be dried at 50 °C in vacuo. 16 Potassium 3,5-dinitro-1,2,4-triazolate was dissolved in 5M sulfuric acid and the solution extracted with ethyl acetate. The solvent was removed using a rotary evaporator, leaving behind yellow, very hygroscopic and deliquescent HDNT. According to its 1 H and 13 C NMR spectra, the HDNT prepared in this way contains various amounts of impurities. As already described before, it was possible to minimize the amount of impurities through careful control of the reaction condition in the KDNT synthesis. In addition, it was also noted that the amount of impurities increases the longer HDNT remains dissolved in ethyl acetate. HDNT is strongly acidic and catalyses the cleavage of ethyl acetate, resulting in the formation of acetic acid which in turn appears to promote the decomposition of HDNT. It is therefore recommended to remove the ethyl acetate solvent immediately after each extraction step. The crude yellow HDNT was further purified by sublimation in vacuo at 100-110 °C. Very pure 3,5-dinitro-1H-1,2,4-triazole was obtained as an off-white amorphous or colorless crystalline solid that is hygroscopic and deliquescent. The sublimation of HDNT should only be attempted on a small scale and behind blast shields. In addition, it is necessary to carefully monitor and control the temperature during the sublimation in order to avoid a possible explosion as pure HDNT decomposes explosively upon heating to 170 °C (DTA onset). The friction and impact sensitivity of HDNT was determined as 144 N and 35 J, respectively. However, it was found that certain impurities might increase the sensitivity of the material and also lower its explosion temperature. For example, one sample of crude HDNT showed an explosion temperature of 150 °C with friction and impact sensitivities of 130 N and 30 J, respectively. The isolation of DAT prepared from 2-cyanoguanidine and hydrazine hydrate is complicated and labor-intensive. 22 Chernyshev et al. developed a procedure for the synthesis of HDNT from 2- cyanoguanidine and hydrazine hydrate without isolation of the DAT intermediate. 21 Following this procedure, it was difficult to obtain pure HDNT. Even after repeated sublimations, the resulting HDNT 21 always had a greenish to light yellow-orange color and contained various unidentified impurities. A major concern is that HDNT prepared according to this procedure showed significantly lower explosion temperatures and higher impact sensitivities than HDNT that was prepared from commercial DAT. The 14 N NMR and vibrational spectra of these samples indicated the presence of azido compounds and single crystals of 3-azido-5-nitro-1H-1,2,4-triazolate were obtained from such an HDNT sample. Figure 2.2: Possible pathways for the formation of the HDNT impurities 1-5. 2.3 Impurities and Decomposition Products of HDNT During the course of this work, several HDNT impurities and/or decomposition products have been identified by their X-ray crystal structures. However, no efforts were made to purify and further characterize these compounds other than by their X-ray crystal structure. From various batches of HDNT that have been prepared by nitration of DAT, crystals of 5-ethoxy-1-methyl-3-nitro-1H-1,2,4-triazole (1), 1-acetyl 3,5-diamino-1H-1,2,4-triazole (2), 1-(i-propyl)-3,5-dinitro-1H-1,2,4-triazole (3) and a co-crystal of 3 and 3-nitro-1H-1,2,4-triazole (4) were obtained. The crystals of these compounds were generally obtained through recrystallization of crude HDNT from acetonitrile, ethanol, ethyl acetate, or acetone. A small amount of crystals of sodium 3-nitro-1,2,4-triazol-5-olate 5 were obtained when a batch of crude HDNT was treated with NaBH 4 , 10 and the resulting product was recrystallized from acetonitrile. The formation of compounds 1-5 can be rationalized through a radical decomposition of HDNT under the 22 formation of NO 2 , followed by the reaction of the resulting radical with the solvents ethyl acetate, water or acetone (Figure 2.2). This is consistent with the observation that the exothermic decomposition of crude HDNT involves the release of large quantities of brown NO 2 /N 2 O 4 gas. Batches of HDNT that had been prepared directly from 2-cyanoguanidine and hydrazine hydrate generally contained more impurities than HDNT that had been prepared by nitration of commercial DAT. It is troublesome that not even through repeated sublimation it was possible to obtain pure HDNT from these batches. It is important to note that HDNT that had been prepared from 2-cyanoguanidine and hydrazine hydrate consistently showed significantly lower explosion temperatures and higher impact sensitivities than the one prepared from commercial DAT. The 14 N NMR spectra of HDNT prepared from 2-cyanoguanidine showed weak resonances at -146 ppm and -232 ppm and the vibrational spectra of this HDNT showed bands at around 2100 cm -1 . These spectroscopic data indicate the presence of azido compounds, which would explain the increased sensitivity of the material. Conclusive evidence for the presence of an azido impurity was found when an aqueous solution of HDNT, which had been prepared from 2-cyanoguanidine was treated with PPN + Cl - , 10 and the resulting precipitate was recrystallized from acetone (PPN + = bis(triphenylphosphine)ammonium, ((Ph 3 P) 2 N + ). The X-ray structure determination of a resulting single crystal resulted in the structure of PPN + 3,5-dinitro-1H-1,2,4-triazolate in which the anion showed a 25 % substitution disorder with 5-azido-3-nitro-1,2,4-triazolate (AzNT). In another instance, the structure of a crystal containing equal amounts of 5-azido-3-nitro-1H-1,2,4-triazole (HAzNT) and PPN + 3-azido-5-nitro-1,2,4-triazolate (PPNAzNT) was obtained. The formation of the azido compounds can be rationalized according to Figure 2.3. 23 Figure 2.3: Formation of azidotriazoles DAT, prepared through the reaction of 2-cyanoguanidine and hydrazine hydrate (Figure 2.1), which has not been isolated and purified, contains small amounts of unreacted hydrazine hydrate. The reaction of nitrous acid, formed from sodium nitrite and sulfuric acid, with hydrazine hydrate results in the formation of hydrazoic acid, 23 which in turn will form azidotriazoles in the Sandmeyer reaction. 2.4 Structural Characterization Single crystals suitable for X-ray crystal structure determination were obtained for two polymorphs of neat HDNT and the hydrate (HDNT)·4 H 2 O. In addition, crystal structures were obtained of the HDNT impurities or decomposition products 1, 2, 3, the co-crystal 3 . 4, 5 . H 2 O as well as PPN[H(AzNT) 2 ] - (6). The relevant data and parameters for the X-ray structure determinations and refinements of the investigated compounds are summarized in Table 2.1-Table 2.3. Further crystallographic data and representations of the unit cells for all crystal structures can be found in the APPENDIX 1. 24 Table 2.1: Crystallographic data for the three HDNT crystal structures. HDNT-1 HDNT-2 (HDNT) 3 . 4 H 2O formula C 2HN 5O 4 C 2HN 5O 4 C 6H 11N 15O 16 mol wt [g/mol] 159.08 159.08 549.30 temp [K] 140(2) 140(2) 100(2) [Å] 0.71073 0.71073 0.71073 crystal system Monoclinic Triclinic Triclinic space group P2 1/c P P1 a [Å] 6.1585(14) 8.7465(15) 6.1906(2) b [Å] 9.083(2) 8.9684(16) 9.5492(3) c [Å] 9.858(2) 11.942(2) 9.5656(3) α [°] 90 111.927(2) 111.377(2) β [°] 93.892(3) 96.726(3) 93.467(2) γ [°] 90 93.853(2) 90.765(3) V [Å 3 ] 550.2(2) 856.7(3) 525.22(4) Z 4 6 1 ρ calc [g/cm 3 ] 1.920 1.850 1.737 μ [mm -1 ] 0.183 0.176 0.169 F(000) 320 480 280 reflns collected 11484 5462 16510 ind reflns 1470 3826 5371 R int 0.0308 0.0196 0.0241 no. of parameters 103 302 343 R 1 [I > 2σ(I)] 0.0327 0.0499 0.0251 wR 2 [I > 2σ(I)] 0.0831 0.1170 0.0554 GOF 1.050 1.036 1.015 Colorless crystals of 3,5-dinitro-1H-1,2,4-triazole that were obtained by sublimation in vacuo belong to space group P2 1 /c (HDNT-1). The unit cell of this monoclinic modification contains four molecules per unit cell (Z = 4). A triclinic modification with six HDNT molecules per unit cell (space group P ) was obtained by recrystallization from anhydrous acetonitrile (HDNT-2). The density of the triclinic modification (1.850 g/cm 3 ) is lower than the one of the monoclinic modification (1.920 g/cm 3 ). Crystals of the hydrate (HDNT) 3 ·4 H 2 O were obtained by recrystallization of neat HDNT from moist acetone. The hydrate crystallizes in the triclinic space group P1 with one formula unit per unit cell. Its density of 1.737 g/cm 3 is lower than the one of the two polymorphs of pure HDNT. The geometry of the HDNT molecule remains virtually unchanged between the three different crystal structures. Selected bond lengths and bond angles of the molecule are listed in Table 2.2. The observed N-N bond distances range from 1.343(4) to 1.3471(13) Å and are shorter than the ones found for the 3,5-dinitro-1,2,4-triazolate (DNT) anion (1.350(2) to 1.368(5) Å). 16 The five-membered ring contains two shorter C-N distances (C1-N2 and C2-N3) of 1.309(5) to 1.325(5) and two longer C-N distances (C1- N3 and C2-N1) of 1.3267(13) to 1.3479(13) Å. This is consistent with the common description of the 1H- 1,2,4-triazole ring having double bonds between the 2-3 and 4-5 positions. 25 Table 2.2: Selected bond lengths [Å] and angles [°] for HDNT in the crystal structures. HDNT-1 HDNT-2 a (HDNT) 3 . 4H 2O a N1-N2 1.3471(13) 1.346(3) 1.343(4) N1-C2 1.3267(13) 1.330(3) 1.329(5) N2-C1 1.3169(13) 1.313(3) 1.325(5) N3-C1 1.3479(13) 1.346(3) 1.336(5) N3-C2 1.3118(13) 1.312(3) 1.309(5) C1-N4 1.4544(14) 1.451(3) 1.451(5) C2-N5 1.4520(14) 1.445(3) 1.452(5) N4-O1 1.2191(13) 1.218(3) 1.225(4) N4-O2 1.2260(12) 1.221(3) 1.221(4) N5-O3 1.2221(12) 1.213(3) 1.216(4) N5-O4 1.2204(13) 1.223(3) 1.228(4) N1-N2-C1 100.97(8) 101.1(2) 101.6(3) N2-C1-N3 117.43(9) 118.0(2) 117.3(3) C1-N3-C2 99.66(8) 99.0(2) 99.3(3) N3-C2-N1 112.82(9) 113.3(2) 117.3(3) C2-N1-N2 109.11(8) 108.6(2) 107.9(3) O1-N4-O2 125.87(10) 125.7(2) 125.8(3) O3-N5-O4 126.99(10) 126.2(2) 125.9(3) a Values given for one of the independent molecules in the asymmetric unit. Figure 2.4: Part of a chain made through hydrogen bonding in the monoclinic crystal structure HDNT-1. Thermal ellipsoids are shown at the 50% probability level. Hydrogen atom positions were determined from the electron density map and are depicted as spheres of arbitrary radius. The N1-N3 distance is 2.837(1) Å. In the solid-state structure of the monoclinic modification HDNT-1, individual HDNT molecules are linked through single N1-H1 ⋯N3 hydrogen bonds, resulting in chains in which the individual HDNT molecules are rotated by 61° from each other (Figure 2.4). The chains are oriented along the c-axis of the crystal. 26 The hydrogen bonding of the triclinic polymorph HDNT-2 is more complex than the one of monoclinic HDNT-1. The molecules form chains along the (011) direction of the crystal in which units of two coplanar HDNT molecules are linked through hydrogen bonds to other units that are rotated by 75°. In addition, every HDNT molecule of a chain is linked to a single HDNT molecule through a hydrogen bond at the nitrogen atom in the 4-position of the triazole ring (N3 and N13). A part of the hydrogen bonding in the triclinic HDNT polymorph is depicted in Figure 2.5. Figure 2.5: Hydrogen bonding in the crystal structure of triclinic HDNT-2. Thermal ellipsoids are shown at the 50% probability level. Some nitro groups have been omitted for clarity. Hydrogen atoms are depicted as spheres of arbitrary radius. The hydrogen atoms H1/H1a and H3/H3a show a 1:1 positional disorder, only one of the disordered atoms is shown per molecule. Selected distances (Å): N1-N12 2.959(2), N2-N2 2.975(3), N6-N13 2.858(3), N11-N11 2.971(3). As expected, the solid-state structure of the hydrate (HDNT) 3 ·4H 2 O is dominated by hydrogen bonding. It is interesting to note that all hydrogen bonds involve water molecules and that no direct bonds between HDNT molecules can be observed. A part of the hydrogen bonding in (HDNT) 3 ·4H 2 O is shown in Figure 2.6. 27 Figure 2.6: Hydrogen bonding in the crystal structure of (HDNT) 3 ·4H 2 O. Thermal ellipsoids are shown at the 50% probability level. Hydrogen atom positions were determined from the electron density map and are depicted as spheres of arbitrary radius. Selected distances (Å): N1-O13 2.849(4), N7-O13 2.587(4), N8-O16 2.882(3), N13-O14 2.984(5), O3-O16 2.927(4). The crystallographic data for 1-3, 5 . H 2 O, 6 and the co-crystal 3 . 4 are listed in Table 2.3. The crystallographic data for the structure of PPN + 3,5-dinitro-1H-1,2,4-triazolate that is disordered with 5- azido-3-nitro-1,2,4-triazolate is given in APPENDIX 1. Figure 2.7: Crystal structure of 5-ethoxy-1-methyl-3-nitro-1H-1,2,4-triazole (1). Thermal ellipsoids are shown at the 50% probability level. Hydrogen atoms have been omitted for clarity. Selected distances (Å) and angles (°): C1-N1 1.351(2), C2-N1 1.326(2), N2-C2 1.346(2), N3-C1 1.311(2), C1-N4 1.455(2), N2- N3 1.364(2), N2-C5 1.461(2), N4-O2 1.223(2), N4-O3 1.227(2), C2-O1 1.321(2), C3-O1 1.467(2), C1- N1-C2 100.1(1), N1-C2-N2 117.7(1), C2-N2-N3 109.1(1), N2-N3-C1 100.9(1), N3-C1-N1 118.2(1), O2- N4-O3 125.4(1). 28 1 crystallizes in the triclinic space group P with two molecules per unit cell (Figure 2.7). In the solid- state structure, the molecules are aligned along parallel planes. The shortest distances between molecules of neighboring planes are 3.007(2) Å (O2-O2’) and 3.299(2) Å (C1-C2’). Table 2.3: Crystallographic data for the crystal structures of the identified HDNT impurities and decomposition products. 1 2 3 3·4 5·H 2 O 6 formula C 5 H 8 N 4 O 3 C 4 H 7 N 5 O C 5 H 7 N 5 O 4 C 7 H 8 N 9 O 6 C 2 H 3 N 4 NaO 4 C 40 H 31 N 15 O 4 P 2 mol wt [g/mol] 172.15 141.15 201.16 314.22 170.07 847.74 temp [K] 100(2) 100(2) 130(2) 130(2) 140(2) 100(2) [Å] 1.54178 0.71073 0.71073 0.71073 0.71073 0.71073 crystal system Triclinic Triclinic Orthorhombic Orthorhombic Monoclinic Triclinic space group P P Pbca Pbca P2 1 /c P a [Å] 6.6336(8) 5.25150(10) 9.4402(19) 9.467(2) 10.7152(18) 11.1121(11) b [Å] 7.6993(8) 7.69060(10) 10.611(2) 11.270(3) 8.3768(14) 12.5247(12) c [Å] 8.5129(9) 8.49150(10) 17.296(4) 23.865(6) 6.7473(11) 16.0476(16) α [°] 83.636(5) 67.2290(10) 90 90 90 70.424(2) β [°] 73.035(5) 85.4460(10) 90 90 97.084(2) 72.600(2) γ [°] 65.784(7) 70.1300(10) 90 90 90 83.653(2) V [Å 3 ] 379.25(7) 296.854(8) 1732.5(6) 2546.2(10) 601.01(17) 2007.9(3) Z 2 2 8 8 4 2 ρ calc [g/cm 3 ] 1.508 1.579 1.542 1.639 1.880 1.402 μ [mm -1 ] 1.086 0.122 0.134 0.144 0.232 0.172 F(000) 180 148 832 1288 344 876 reflns collected 8069 35219 13391 15158 6577 48323 ind reflns 1315 2865 2124 3102 1444 12028 R int 0.0241 0.0256 0.0393 0.0323 0.0417 0.0605 no. of parameters 141 109 129 201 112 550 R 1 [I > 2σ(I)] 0.0300 0.0288 0.0435 0.0551 0.0433 0.0692 wR 2 [I > 2σ(I)] 0.0823 0.0807 0.0943 0.1588 0.1008 0.1441 GOF 1.055 1.123 1.024 1.052 1.055 1.026 In the solid-state structure of 2, the molecules are associated by N-H ⋯N and N-H ⋯O hydrogen bonds, forming planar layers (Figure 2.8). The shortest distances between molecules of neighboring planes are 3.1592(8) Å (C3-N4’) and 3.3351(2) Å (N3-N4’). 29 Figure 2.8: Intermolecular hydrogen bonding in the crystal structure of 1-acetyl 3,5-diamino-1H-1,2,4- triazole (2). Thermal ellipsoids are shown at the 50% probability level. Some hydrogen atoms have been omitted for clarity. Selected distances (Å): N2-N4’ 3.0771(5), N3-N5’ 3.0160(5), N5-O1’ 3.0174(6). Figure 2.9: Crystal structure of 1-(i-propyl)-3,5-dinitro-1H-1,2,4-triazole (3). Thermal ellipsoids are shown at the 50% probability level. Hydrogen atoms have been omitted for clarity. Selected distances (Å) and angles (°): C2-N3 1.311(2), C1-N3 1.334(2), C2-N1 1.345(2), C1-N2 1.318(2), N1-N2 1.349(2), C2- N5 1.456(2), C1-N4 1.451(2), N4-O1 1.219(2), N4-O2 1.218 (2), N5-O3 1.2147(19), N5-O4 1.221 (2), N1-C3 1.491(2), N1-N2-C1 101.75(13), N2-N1-C2 107.67(12), C2-N3-C1 99.91(13), N3-C2-N1 113.05(14), N3-C1-N2 117.62(14), O1-N4-O2 124.94(14), O3-N5-O4 125.38(15). 3 crystallizes in the orthorhombic space group Pbca with eight symmetry related molecules in the unit cell (Z = 8). The molecule is depicted in Figure 2.9. The five-membered ring is almost co-planar with both nitro-groups (dihedral angles of 5.7° and 14.0°) and is perpendicular to the plane of the isopropyl group. 30 Figure 2.10: Asymmetric unit of the co-crystal of 1-(i-propyl)-3,5-dinitro-1H-1,2,4-triazole and 3-nitro- 1H-1,2,4-triazole (3 . 4). Thermal ellipsoids are shown at the 50% probability level. Some hydrogen atoms have been omitted for clarity. Similar to pure 1-(i-propyl)-3,5-dinitro-1H-1,2,4-triazole (3), the compounds 1-(i-propyl)-3,5-dinitro-1H- 1,2,4-triazole (3) and 3-nitro-1H-1,2,4-triazole (4) co-crystallize in the orthorhombic space group Pbca with eight symmetry related formula units per unit cell (Z = 8). However, the unit cell of the co-crystal 3·4 (V = 2546.2(10) Å 3 ) is almost 50% larger than that of the neat 3 (V = 1732.5(6) Å 3 ). The asymmetric unit of the crystal structure 3·4 is depicted in Figure 2.10. The closest distances between molecules 3 and 4 in the crystal structure are 2.820(3) and 2.854(3) Å. The geometry of compound 3 remains virtually unchanged going from the structure of the neat compound 3 to the co-crystal 3·4. Figure 2.11: Crystal structure of sodium 3-nitro-1,2,4-triazol-5-olate (5·H 2 O). Thermal ellipsoids are shown at the 50% probability level. Hydrogen atoms are depicted as spheres of arbitrary radius. Selected distances (Å) and angles (°):N4-C2 1.446(3), C1-N2 1.370(3), C2-N1 1.308(3), C2-N3 1.342(3), C1-N3 1.353(3), N1-N2 1.359(2), C1-O3 1.273(2), N4-O1 1.226(2), N4-O2 1.229(2), Na1-O1 2.5203(17), Na1- O3'' 2.3377(16), Na1-O3' 2.5658(17), Na1-N3 2.4282(19), Na1-O4 2.3365(19), Na1-O4' 2.454(2). Bond 31 angles (°):O1-N4-O2 124.29(18), C2-N4-O1 117.94(18), C2-N4-O2 117.76(18), N3-C1-O3 127.43(19), N2-C1-O3 124.33(19), O2-N4-C2-N1 -2.3(3). 5 crystallizes as a monohydrate in space group P2 1 /c with four symmetry related formula units in the unit cell (Z = 4). Not surprisingly, the crystal structure does not consist of isolated ions but is dominated by interactions between the sodium cation and the triazol-5-olate anion as well as the water molecule. The solid-state structure of 5·H 2 O is depicted in Figure 2.11. Figure 2.12: The anion part of the crystal structure 6. The hydrogen atom is depicted as a sphere of arbitrary radius and shows a 1:1 positional disorder located at N3 and N10. Only one of the disordered positions is shown. Selected distances (Å) and angles (°): C1-N1 1.445(3), C3-N8 1.451(3), C2-N4 1.395(3), C4-N11 1.400(3), N3-N10 2.655(3), N4-N5 1.247(3), N5-N6 1.123(3), N11-N12 1.245(3), N12-N13 1.127(3), N4-N5-N6 171.4(3), N11-N12-N13 171.5(3). When a sample of HDNT, which had been prepared from hydrazine hydrate, was reacted with PPN + Cl - and the resulting PPN + DNT - was recrystallized from acetone, single crystals of PPN[H(AzNT) 2 ] (AzNT = 5-azido-3-nitro-1,2,4-triazolate) were obtained. The solid-state structure of 6 contains isolated PPN + cations and anions in which two AzNT - parts that are associated through a hydrogen bond between the N atoms in the 1-position of both triazole moieties (N-N distance: 2.655(3) Å). Both 5-azido-3-nitro-1,2,4- triazolate moieties have essentially identical geometries and are depicted in Figure 2.12. The H-atom shows in a 1:1 positional disorder between both triazolate moieties, occupying positions near N3 and N10. 32 2.5 Conclusion The important energetic building block 3,5-dinitro-1H-1,2,4-triazole (HDNT) was structurally characterized for the first time. Neat HDNT was obtained in quantitative yield from potassium 3,5- dinitro-1,2,4-triazolate and sulfuric acid, followed by extraction with ethyl acetate. The compound was isolated as a pale yellow solid, which can be further purified by sublimation to give colorless crystals. Pure HDNT is a hygroscopic white solid that decomposes explosively upon heating to 170 °C. However, the presence of impurities might lower the decomposition temperature and increase the sensitivity of the material. Potassium 3,5-dinitro-1,2,4-triazolate was prepared from commercially available 3,5-diamino- 1,2,4-triazole through a Sandmeyer reaction with sodium nitrite and sulfuric acid. The synthesis of HDNT from 2-cyanoguanidine and hydrazine hydrate in one step without isolation and purification of the 3,5- diamino-1,2,4-triazole intermediate is not recommended as it might lead to the formation of very sensitive azidotriazole impurities. 3,5-dinitro-1H-1,2,4-triazole was characterized by its multinuclear NMR and vibrational spectra, as well as its X-ray crystal structure. The crystal structures of several HDNT impurities and decomposition products were obtained. Further experimental details, crystal packing diagrams and crystallographic information can be found in APPENDIX 1. 2.6 References (1) Talawar, M. B.; Sivabalan, R.; Mukundan, T.; Muthurajan, H.; Sikder, A. K.; Gandhe, B. R.; Rao, A. S. J. Hazardous Mat. 2009, 161, 589. (2) Langlet, A.; Oestmark, H. WO9402434A1 1994. (3) Lee, K. Y.; Storm, C. B.; Hiskey, M. A.; Coburn, M. D. J. Energ. Mater. 1991, 9, 415. (4) Simpson, R. L.; Pagoria, P. F.; Mitchell, A. R.; Coon, C. L. Propellants, Explos., Pyrotech. 1994, 19, 174. (5) Zhang, Y.; Parrish, D. A.; Shreeve, J. n. M. J. Mater. Chem. A 2013, 1, 585. (6) Huynh, M. H. V. WO2008US01904 2008. (7) Miller, C. G.; Williams, G. K. US2009199937 2009. (8) Burchfield, H. P.; Gullstrom, D. K. US3054800 1962. 33 (9) Wiley, R. H.; Smith, N. R. US3111524 1963. (10) Haiges, R.; Jones, C. B.; Christe, K. O. Inorg. Chem. 2013, 52, 5551. (11) Klapötke, T. M.; Penger, A.; Pflüger, C.; Stierstorfer, J.; Sućeska, M. Eur. J. Inorg. Chem. 2013, 2013, 4667. (12) Klapötke, T. M.; Piercey, D. G.; Stierstorfer, J. Dalton Trans. 2012, 41, 9451. (13) Kofman, T. P. Russ. J. Org. Chem. 2001, 37, 1158. (14) Pevzner, M. S.; Kofman, T. P.; Kibasova, E. N.; Sushchenko, L. F.; Uspenskaya, T. L. Chem Heterocycl Compd 1980, 16, 194. (15) Stinecipher, M. M. Investigation of the Physical and Explosives Properties of the Eutectic Explosive Ammonium Nitrate/Ammonium 3,5-Dinitro-1,2,4-Triazolate; 1978 Annual Report, Los Alamos National Laboratory, 1978. (16) Haiges, R.; Bélanger-Chabot, G.; Kaplan, S. M.; Christe, K. O. Dalton Trans. 2015, 44, 2978. (17) Garcia, E.; Lee, K.-Y. Acta Crystallogr. Sect. C: Cryst. Struct. Commun. 1992, 48, 1682. (18) Izsák, D.; Klapötke, T. M. Crystals 2012, 2, 294. (19) Bian, C.; Zhang, M.; Li, C.; Zhou, Z. J. Mater. Chem. A 2015, 3, 163. (20) Larina, L.; Lopyrev, V. Nitroazoles: Synthesis, Structure and Applications; Springer: Dordrecht ; New York, 2009. (21) Chernyshev, V. M.; Zemlyakov, N. D.; Il'in, V. B.; Taranushich, V. A. Russ. J. Appl. Chem. 2000, 73, 839. (22) Roemer, J. J.; Kaiser, D. W. US2648671 1953. (23) Thiele, J. Chem. Ber. 1908, 41, 2681. 34 CHAPTER 3: SYNTHESIS AND CHARACTERIZATION OF SALTS OF THE 3,5-DINITRO-1,2,4-TRIAZOLATE ANION 3.1 Introduction In recent years, the development of new, environmentally friendly energetic materials has sparked considerable attention in the synthetic community. 1 Nitrogen-rich cyclic compounds, such as triazoles and tetrazoles, are important and versatile motifs for energetic materials. These heterocycles can be further functionalized with explosophore groups, such as nitro, N-nitro, azo, or azido. Among these, 3-nitro-1H- 1,2,4-triazole 2-5 and 3,5-dinitro-1H-1,2,4-triazole (HDNT) 6,7 have attracted considerable interest for energetic material and high oxygen carrier applications. With five nitrogen and four oxygen atoms per molecule, the latter combines the advantages of a high nitrogen and high oxygen content in a single compound. Since the first preparation of HDNT, 8,9 a significant number of derivatives have been reported, 5,10-14 including complex borate anions. 15 During the course of the ongoing research on environmentally-benign energetic materials and high-oxygen in the Christe-Haiges group, 15-18 characterization data and crystal structures for several 3,5-dinitro-1,2,4-triazolates were obtained, which should be useful to the broader scientific community. Although several simple salts containing the 3,5- dinitro-1,2,4-triazolate (DNT) anion have been described in the literature, 19-29 they have not been fully characterized. In this work nineteen different salts of the DNT anions have been prepared and fully characterized, including their crystal structures, thermal stability, and friction and impact sensitivities. Figure 3.1: Synthesis of the metal salts of DNT 1-9. 35 3.2 Synthesis For the synthesis of LiDNT (1), NaDNT (2), KDNT (3), RbDNT (4), CsDNT (5), Sr(DNT) 2 (6), and Ba(DNT) 2 (7), the parent triazole HDNT was dissolved in acetone that contained a few drops of water and stirred over a suspension containing an excess of the corresponding metal carbonate (Figure 3.1). 8 With the exception of barium and strontium carbonate that remained white, the insoluble carbonate turned orange to reddish brown and the solution became light yellow. The reaction mixtures were filtered and crystals were grown from the filtrates by slow evaporation of the solvent. The alkali and alkali earth metal salts were obtained as light yellow to orange crystalline solids. Crystals of the water-free potassium salt 3 were obtained by recrystallization of the reaction product from acetone solution. The compound is hygroscopic and forms the dihydrate upon prolonged exposure to air. The DNT silver salt (8) 8,30 was obtained as a white amorphous precipitate when a silver nitrate solution was added to an aqueous solution of HDNT. Orange single crystals of Ag(NH 3 )DNT (9) were obtained when compound 8 was dissolved in aqueous ammonia and the solvent was allowed to evaporate. The ammonium salt NH 4 DNT 25 was prepared by reacting HDNT with NH 3 (Figure 3.2). Yellow crystals of the dihydrate NH 4 DNT∙2H 2 O (10 . 2H 2 O) were obtained by slow evaporation of an aqueous solution of NH 4 DNT (10). Figure 3.2: Synthesis of DNT salts 10-19. 36 For the synthesis of the triethylammonium (11), diethylammonium (12) and pyridinium (15) salts, an excess of the corresponding amine was added to an aqueous solution of HDNT. Yellow to orange crystals were obtained by slow evaporation of the solvent. Guanidinium DNT (13) 8 was obtained by stirring an acetone solution of HDNT over guanidine carbonate followed by filtration and slow evaporation of the solvent, yielding a crystalline orange material (Figure 3.2). Similarly, aminoguanidium DNT (14) was obtained as a yellow-orange crystalline solid from aminoguanidine bicarbonate and HDNT in ethanol. Colorless aminotetrazolium DNT (16) 7 was obtained from stoichiometric amounts of 5-aminotetrazole and HDNT in aqueous solution. The tetraphenylphosphonium (PPh 4 + ) 27 and bis(triphenylphosphoranylidene)ammonium (PPN) salts (17 and 18, respectively) were precipitated from an aqueous solution of HDNT by addition of aqueous solutions of PPh 4 Cl and PPNCl, respectively. The colorless to off-white precipitates were thoroughly washed with water. Crystalline compounds were obtained by recrystallization from acetone. The properties of PPh 4 DNT are in good agreement with those previously reported. 27 The tetramethylammonium (TMA) salt (19) 23 was obtained from stoichiometric amounts of tetramethylammonium hydroxide and HDNT in aqueous solution. Single crystals were grown by recrystallization of the crude product from ethanol solution. The obtained materials were characterized by multinuclear NMR and vibrational spectroscopy, as well as their X-ray crystal structure. 3.3 Structural Characterization Single crystals suitable for X-ray crystal structure determination were obtained for the metal salts LiDNT∙2H 2 O (1∙2H 2 O), NaDNT∙2H 2 O (2∙2H 2 O), KDNT (3), KDNT∙2H 2 O (3∙2H 2 O), RbDNT (4), CsDNT (5), Sr(DNT) 2 ∙6H 2 O (6∙6H 2 O), Ba(DNT) 2 ∙11H 2 O (7∙11H 2 O), and Ag(NH 3 )DNT (9). Structures were also obtained for the ammonium salts NH 4 DNT∙2H 2 O (10∙2H 2 O), [HNEt 3 ][DNT] (11), [H 2 NEt 2 ][DNT] (12), the guanidinium salt (13), the aminoguanidinium salt (14), monohydrates of the pyridinium salt (15∙H 2 O) and aminotetrazolium salt (16∙H 2 O), as well as PPh 4 [DNT] (17), PPN[DNT] (18), and NMe 4 [DNT] (19). The geometry of the five-membered ring in the dinitrotriazole anion remains virtually unchanged throughout the investigated structures. The N-N bond distances range from 1.350(2) to 1.368(5) Å, the C- N distances from 1.315(2) to 1.347(7) Å and the C-NN distances from 1.316(6) to 1.351(8) Å. Selected bond lengths and bond angles of the 3,5-dinitro-1,2,4-triazolate anion in the different alkali metal salts are summarized in Table 3.1. In the investigated salts, the DNT anion usually adopts a geometry in which 37 both NO 2 groups are almost coplanar with the plane of the five-membered triazole ring. The torsion angles for the NO 2 groups with respect to the triazole ring plane were found to be less than 6° for most salts. The major exception is the rubidium salt (4) for which torsion angles of 4.6° and 19.1° were found. The relevant data and parameters for the X-ray measurements and structure refinements of the 3,5-dinitro-1,2,4-triazolates are summarized in Table 3.2-Table 3.4. Further crystallographic data and representations of the unit cells for all crystal structures can be found in APPENDIX 2. Table 3.1: Selected bond lengths [Å] and angles [°] for alkali metal DNT salts 1-5. LiDNT·2H 2 O 1∙2H 2 O NaDNT·2H 2 O 2∙2H 2 O KDNT 3 a KDNT·2H 2 O 3∙2H 2 O RbDNT 4 CsDNT 5 N1-N2 1.366(1) 1.366(1) 1.368(5) 1.359(4) 1.353(7) 1.364(4) C1-N2 1.336(1) 1.336(1) 1.331(5) 1.322(4) 1.338(7) 1.339(4) C1-N3 1.336(1) 1.331(1) 1.322(5) 1.329(4) 1.343(7) 1.326(4) C1-N4 1.449(1) 1.451(1) 1.445(5) 1.445(5) 1.431(7) 1.445(5) C2-N1 1.336(1) 1.336(1) 1.327(6) 1.336(4) 1.341(7) 1.336(5) C2-N3 1.336(1) 1.331(1) 1.325(5) 1.333(5) 1.319(7) 1.331(5) C2-N5 1.449(1) 1.451(1) 1.448(6) 1.446(4) 1.433(7) 1.440(4) N1-N2-C1 104.12(5) 104.07(4) 103.2(3) 104.3(3) 104.3(5) 103.7(3) N2-C1-N3 116.84(8) 116.73(7) 117.6(4) 117.2(3) 116.4(5) 117.1(3) C1-N3-C2 98.09(10) 98.39(9) 97.9(3) 98.1(3) 98.0(4) 97.9(3) N3-C2-N1 116.84(8) 116.73(7) 117.1(4) 116.3(3) 117.2(5) 117.3(3) C2-N1-N2 104.12(5) 104.07(4) 104.1(3) 104.2(3) 104.1(5) 104.0(3) a Values given for one of the two independent anions of the asymmetric unit. 38 Table 3.2: Crystallographic data for the DNT metal salts 1-9. 1∙2H 2O 2∙2H 2O 3 3∙2H 2O 4 5 6·6H 2O 7·11H 2O 9 formula C 2H 4LiN 5O 6 C 2H 4N 5NaO 6 C 2 KN 5O 4 C 2H 4 KN 5O 6 C 2N 5O 4Rb C 2CsN 5O 4 C 4H 12N 10O 14 Sr C 4H 22BaN 10O 19 C 2H 3AgN 6O 4 mol wt [g/mol] 201.04 217.09 197.17 233.20 243.54 290.98 511.86 651.65 282.97 temp [K] 100(2) 100(2) 100(2) 130(2) 140(2) 133(2) 100(2) 100(2) 140(2) crystal system Monoclinic Monoclinic Monoclinic Triclinic Orthorhombic Monoclinic Triclinic Monoclinic Monoclinic space group C2/c C2/c P2 1/n P1 P2 12 12 1 P2 1/n P P2 1/m P2 1/n a [Å] 15.342(2) 15.3444(7) 6.741(2) 4.5220(15) 7.6561(12) 7.974(3) 9.2895(3) 6.9239(6) 7.8061(19) b [Å] 8.3048(11) 8.7270(4) 12.997(4) 6.355(2) 7.8172(12) 10.088(3) 12.8551(4) 18.7743(15) 5.5756(14) c [Å] 6.3573(8) 6.6379(3) 14.749(5) 7.770(3) 11.0946(17) 9.267(3) 16.5552(7) 9.2088(7) 17.034(4) α [°] 90 90 90 96.093(4) 90 90 108.2150(10) 90 90 β [°] 113.079(2) 114.9340(6) 97.697(5) 92.533(4) 90 109.941(4) 90.8870(10) 111.6350(10) 101.686(3) γ [°] 90 90 90 108.569(4) 90 90 111 90 90 V [Å 3 ] 745.17(17) 806.04(6) 1280.6(7) 209.75(12) 664.00(18) 700.8(4) 1736.03(11) 1112.73(16) 726.0(3) Z 4 4 8 1 4 4 4 2 4 [Å]  0.71073 0.71073 0.71073 0.71073 0.71073 0.71073 0.71073 0.71073 0.71073 ρ calc [g/cm 3 ] 1.792 1.789 2.045 1.846 2.436 2.758 1.958 1.945 2.589 μ [mm -1 ] 0.172 0.216 0.813 0.652 7.440 5.269 3.204 1.891 2.773 F(000) 408 440 784 118 464 536 1024 648 544 reflns collected 8446 9609 14946 2303 4210 3944 42154 26945 4279 ind reflns 1124 1231 3713 1445 1583 1589 10328 3476 1688 R int 0.0269 0.0207 0.0395 0.0228 0.0400 0.1014 0.0331 0.0424 0.0176 no. of parameters 71 72 217 127 109 109 619 178 130 R 1 [I > 2σ(I)] 0.0311 0.0237 0.0764 0.0255 0.0335 0.0332 0.0240 0.0370 0.0185 wR 2 [I > 2σ(I)] 0.0850 0.0662 0.1578 0.0714 0.0703 0.0848 0.0491 0.0831 0.0476 GOF 1.077 1.095 1.277 1.138 0.998 1.040 1.032 1.204 1.084 The crystallographic data for the metal salts of DNT are compiled in Table 3.2. Crystallized from acetone solutions and exposed to ambient air, the lithium, sodium and potassium DNT salt crystals contain two molecules of water per formula unit, whereas the rubidium and cesium salts crystallize free of water. Water-free single crystals of the potassium salt of DNT were also obtained. The alkali metal salts LiDNT (1) and NaDNT (2) crystallize as dihydrates in the monoclinic space group C2/c with four formula units in the unit cell (Z = 4). The solid-state structures of MDNT∙2H 2 O (M = Li 1∙2H 2 O, Na 2∙2H 2 O) do not consist of isolated ions but are dominated by interactions between the metal cation and a nitrogen atom of the triazole ring as well as two oxygen atoms from individual nitro groups of the anion. The structures of the lithium and sodium salts are quite similar. The solid-state structure of 2∙2H 2 O is depicted in Figure 3.3. 39 Figure 3.3: The solid-state structure of NaDNT∙2H 2 O (2∙2H 2 O). Thermal ellipsoids are shown at the 50% probability level. Hydrogen atom positions were determined from the electron density map and are depicted as spheres of arbitrary radius. Selected bond distances (Å): Na1-N2 2.380(1), Na1-O1 3.033(1), Na1-O3 2.340(1), Na1-O3’ 2.469(1). Figure 3.4: The solid-state structure of 3∙2H 2 O. Thermal ellipsoids are shown at the 50% probability level. Hydrogen atom positions were determined from the electron density map and are depicted as spheres of arbitrary radius. Selected bond distances (Å): K1-O1 3.144(3), K1-O1’’ 3.143 (3), K1-O2 2.906(3), K1-O4’’’ 2.909(3), K1-O5 2.770(3), K1-O6 2.703(3), K1-O6’ 2.832(3). The dihydrate of the potassium DNT salt (3∙2H 2 O) crystallizes in space group P1, with one formula unit per unit cell (Z = 1). The solid-state structure is dominated by cation-anion interactions (Figure 3.4). Four oxygen atoms and a nitrogen atom from three anions as well as three water molecules coordinate the potassium cations. Two of the water molecules are bridging two individual cations, while the third water molecule is terminal. Selected bond lengths and angles are listed in Table 3.1 40 Figure 3.5: The solid-state structure of 3. Thermal ellipsoids are shown at the 50% probability level. Some nitro groups have been omitted for clarity. Selected bond distances (Å): K1-N2’ 2.871(4), K1-N5 2.714(4), K1-N8’’ 3.017(4), K1-N10 2.829(4), K1-O1 2.967(4), K1-O2’ 2.775(3), K1-O7’ 2.826(4), K2- O1 2.866(4), K2-O2 3.191(3), K2-05 3.146(5). The water-free potassium salt (3) crystallizes in the monoclinic space group P2 1 /n with eight molecules units per unit cell (Z = 8). The asymmetric unit of the crystal structure consists of two KDNT units. Selected bond lengths and angles are listed in Table 3.1. The solid-state structure is dominated by interactions between the metal cations and the nitrogen and oxygen atoms of the DNT anions (Figure 3.5). The closest K-N and K-O distances are 2.714(3) and 2.888(4) Å, respectively. 41 Figure 3.6: The solid-state structure of 4. Thermal ellipsoids are shown at the 50% probability level. Selected bond distances (Å): Rb1-N1 3.340(5), Rb1-N2 3.421(5), Rb1-N3 3.019(4), Rb1-O1 3.379(5), Rb1-O3 3.676(4), Rb1-O3’ 3.076(4), Rb1-O4’ 3.197(4). Figure 3.7: The solid-state structure of the 5. Thermal ellipsoids are shown at the 50% probability. Selected bond distances (Å): Cs1-N1 3.340(3), Cs-N1’ 3.591(4), Cs1-N2 3.236(3), Cs1-N3 3.207(4), Cs1-O1- 3.288(3), Cs1-O2 3.392(3), Cs1-O4 3.233(3). The rubidium salt (4) and the cesium salt (5) of DNT crystallize in space group P2 1 2 1 2 1 and P2 1 /n, respectively, with four formula units per unit cell. Both solid-state structures do not consist of isolated ions but show interactions between the metal cations and the nitrogen atoms of the triazole ring as well as oxygen atoms of the nitro groups (Figure 3.6 and Figure 3.7). The closest Rb-N and Cs-N distances are 3.019(4) and 3.207(4) Å, respectively, and the closest Rb-O and Cs-O are 3.004(4) and 3.212(3) Å, respectively. 42 Figure 3.8: One of the two independent Sr(H 2 O) 5 (DNT) 2 units in the solid-state structure of 6·6H 2 O. Thermal ellipsoids are shown at the 50% probability level. Hydrogen atom positions were determined from the electron density map and are depicted as spheres of arbitrary radius. Selected bond distances (Å): Sr1-N1 2.773(2), Sr1-N6 2.741(1), Sr1-O3 2.720(1), Sr1-O7 2.798(1), Sr1-O17 2.578(1), Sr1-O18 2.601(1), Sr1-O19 2.548(1), Sr1-O20 2.618(1), Sr1-O21 2.709(2). The strontium salt Sr(DNT) 2 (6) crystallizes in the triclinic space group P with six molecules of water per Sr atom. The structure consists of two symmetry independent Sr(H 2 O) 5 (DNT) 2 units in which the Sr atom is coordinated by five water molecules and by two nitrogen and two oxygen atoms from the two DNT anions (Figure 3.8). The last water molecule is not coordinated to the metal atom so that the structure can be described as Sr(H 2 O) 5 (DNT) 2 ·H 2 O. Figure 3.9: A Ba(H 2 O) 6 (DNT) 2 unit in the solid-state structure of 7·11H 2 O. Thermal ellipsoids are shown at the 50% probability level. Hydrogen atoms were omitted due to difficulties in determining their positions. Selected bond distances (Å): Ba1-N1 2.987(5), Ba1-O2 3.143(5), Ba1-O5 2.865(5), Ba1-O6 2.854(5), Ba1-O7 2.680(5), Ba1-O8 2.754(5), Ba1-O9 2.761(5). 43 The barium salt Ba(DNT) 2 (7) crystallizes in the monoclinic space group P2 1 /m with a total of eleven molecules of water per metal atom. The solid-state structure consists of two symmetry-related units. Each unit consists of a central barium atom that is coordinated by two nitrogen and two oxygen atoms from two DNT anions. Due to the larger ionic radius compared to strontium, the barium atom is also coordinated by six water molecules instead of only five for the strontium salt (Figure 3.9). The remaining five water molecule are not coordinated to the metal atom so that the structure can be described as Ba(H 2 O) 6 (DNT) 2 ·5 H 2 O. Unfortunately, the quality of the obtained single crystals of the Ba compound was rather poor and it was not possible to determine the positions of all hydrogen atoms in the structure. All attempts to grow single crystals of AgDNT (8) suitable for X-ray structure determination were unsuccessful. Recrystallization of an amorphous sample of 8 from an aqueous ammonia solution resulted in crystals of the ammonia adduct [Ag(NH 3 )][DNT] (9) instead. Figure 3.10: The solid-state structure of 9. Thermal ellipsoids are shown at the 50% probability level. Hydrogen atom positions were determined from the electron density map and are depicted as spheres of arbitrary radius. Selected bond distances (Å): Ag1-N1 2.393(2), Ag1-N2 2.242(2), Ag1-N6 2.198(2). The silver ammonia compound 9 crystallizes in the monoclinic space group P2 1 /n with four formula units in the unit cell (Z = 4). The solid-state structure of 9 contains silver atoms coordinated by one NH 3 molecule that are linked in a 1,2-fashion to bridging DNT anions (Figure 3.10). The resulting helical chains of AgDNT units are oriented parallel along the b-axis of the crystal. Part of a helical chain is depicted in Figure 3.11. 44 Figure 3.11: Part of a helical chain in the solid-state structure of 9. The nitro groups and ammonia ligands have been omitted for clarity. Table 3.3: Crystallographic data for salts 10-16. 10∙2H 2O 11 12a 12b 13 14 15∙H 2O 16∙H 2O formula C 2H 8N 6O 6 C 8H 16N 6O 4 C 6H 12N 6O 4 C 6H 12N 6O 4 C 3H 6N 8O 4 C 3H 7N 9O 4 C 7H 8N 6O 5 C 3H 6N 10O 5 mol wt [g/mol] 212.14 260.27 232.22 232.20 218.16 233.18 256.19 262.18 temp [K] 100(2) 100(2) 100(2) 100(2) 100(2) 100(2) 100(2) 100(2) crystal system Triclinic Monoclinic Monoclinic Triclinic Monoclinic Monoclinic Monoclinic Orthorhombic space group P1 P2 1/n P2 1/n P C2/c P2 1/c C2/c P2 12 12 1 a [Å] 4.653(3) 7.0842(5) 6.4659(4) 7.3101(9) 8.3235(7) 9.0616(3) 13.9051(6) 6.0692(3) b [Å] 6.461(4) 9.3195(6) 8.9964(6) 8.1876(10) 16.7996(14) 16.2452(5) 26.2532(12) 17.6181(8) c [Å] 7.619(4) 19.3391(13) 18.8357(12) 9.9470(12) 12.3068(10) 6.5935(2) 6.3689(3) 19.4400(9) α [°] 93.105(7) 90 90 73.347(3) 90 90 90 90 β [°] 93.948(7) 100.3110(11) 99.5420(10) 73.673(2) 102.3250(13) 110.189(3) 111.9480(7) 90 γ [°] 108.360(7) 90 90 77.130(3) 90 90 90 90 V [Å 3 ] 216.2(2) 1256.17(15) 1080.51(12) 540.79(11) 1681.2(2) 910.98(5) 2156.48(17) 2078.68(17) Z 1 4 4 2 8 4 8 8 [Å] 0.71073 0.71073 0.71073 1.54178 0.71073 1.54178 0.71073 0.71073 ρ calc [g/cm 3 ] 1.630 1.376 1.427 1.426 1.724 1.700 1.578 1.676 μ [mm -1 ] 0.158 0.111 0.120 1.036 0.155 1.338 0.136 0.153 F(000) 110 552 488 244 896 480 1056 1072 reflns collected 2547 29946 24237 8580 20271 12104 26448 35316 ind reflns 1599 3841 2903 1606 2562 1692 3279 6290 R int 0.0122 0.0325 0.0293 0.0255 0.0333 0.0474 0.0274 0.0344 no. of parameters 159 169 153 153 160 201 174 373 R 1 [I > 2σ(I)] 0.0264 0.0397 0.0318 0.0267 0.0420 0.0790 0.0345 0.0372 wR 2 [I > 2σ(I)] 0.0610 0.0959 0.0842 0.0671 0.1147 0.2765 0.0974 0.0797 GOF 1.075 1.042 1.034 1.029 1.075 1.220 1.064 1.060 The crystallographic parameters for salts 10-16 are compiled in Table 3.3. The dihydrate of the ammonium salt (10 . 2H 2 O) crystallizes in the triclinic space group P1 with one formula unit per unit cell (Figure 3.12). The solid-state structure consists of ammonium cations and DNT anions that are associated through hydrogen bonds. Hydrogen bonding is observed with water molecules as donor and DNT - as acceptor sites and also between water molecules and the NH 4 + cation as acceptor. 45 Figure 3.12: Hydrogen bonding in the solid-state structure of 10∙2H 2 O. Thermal ellipsoids are shown at the 50% probability level. Hydrogen atom positions were determined from the electron density map and are depicted as spheres of arbitrary radius. Selected distances (Å): N3-O6’ 2.852(3), N6-O1 3.453(3), N6- O4 3.050(3), N6-O5 2.862(3), N6-O6 2.885(3), N6-O6’ 2.828(4). Figure 3.13: Hydrogen bonding in 11. Thermal ellipsoids are shown at the 50% probability level. The hydrogen atoms of the alkyl chains have been omitted for clarity. Depicted hydrogen atom positions were determined from the electron density map and are depicted as spheres of arbitrary radius. The N2-N6 distance is 2.947(1) Å. The triethylammonium salt of 3,5-dinitro-1H-1,2,4-triazole (11) crystallizes in the monoclinic space group P2 1 /n with four formula units per unit cell. As expected, the solid-state structure is dominated by hydrogen bonding between the cation and the anion (Figure 3.13). 46 Figure 3.14: Hydrogen bonding in the diethylammonium salts 12a and 12b. Thermal ellipsoids are shown at the 50% probability level. The hydrogen atoms of the alkyl chains have been omitted for clarity. Depicted hydrogen atom positions were determined from the electron density map and are depicted as spheres of arbitrary radius. Selected atomic distances in the monoclinic [triclinic] polymorph (Å): N2-N6 2.929(1) [2.946(1)], N3-N6 2.962(1) [3.016(1)]. Two polymorphs were found for diethylammonium 3,5-dinitro-1,2,4-triazolate (12). When the compound was recrystallized from aqueous solution, the monoclinic polymorph (12a) was obtained, while recrystallization from diethylamine resulted in crystals of the triclinic polymorph (12b). Both polymorphs consist of [(H 2 NEt 2 ][DNT]) 2 units which are formed through four hydrogen bonds between two diethylammonium cations and two DNT anions. One such unit is depicted in Figure 3.14. Figure 3.15: Asymmetric unit in the crystal structure of 13 with hydrogen bonding. Thermal ellipsoids are shown at the 50% probability level. Hydrogen atom positions were determined from the electron density map and are depicted as spheres of arbitrary radius. For clarity, only some of the observed hydrogen bonds are included. Selected atomic distances (Å): N1-N6 3.044(2), N2-N8 2.897(2). 47 Guanidinium 3,5-dinitro-1,2,4-triazolate (13) crystallizes in the monoclinic space group C2/c with eight formula units per unit cell. Selected crystallographic parameters of the structure are listed in Table 3.3. As has already been observed for 10∙2H 2 O, the solid-state packing in 13 is defined by strong hydrogen bonding between the guanidinium cations and the DNT anions (Figure 3.15). Figure 3.16: Asymmetric unit in the crystal structure of 14 with hydrogen bonding. Thermal ellipsoids are shown at the 50% probability level. Hydrogen atom positions were determined from the electron density map and are depicted as spheres of arbitrary radius. For clarity, only the major component of the two disordered anion orientations and only some of the observed hydrogen bonds is included. Selected atomic distances (Å): N3-N6 3.354(7), N3-N8 3.374(6), N6-O2 2.901(6), N8-O4 2.916(5). The aminoguanidinium salt (14) crystallizes in the monoclinic space group P2 1 /c with four formula units per unit cell. Similarly to 12, the solid-state structure is dominated by strong hydrogen bonding between the aminoguanidinium cations and the anions (Figure 3.16). In the structure, a positional disorder was observed in which the anion is split (ratio 9:1) between two orientations related by a two-fold rotation. 48 Figure 3.17: Hydrogen bonding in 15∙H 2 O. Thermal ellipsoids are shown at the 50% probability level. Some hydrogen atoms have been omitted for clarity. The positions of shown hydrogen atoms were determined from the electron density map and are depicted as spheres of arbitrary radius. Selected atomic distances (Å): N2-O1w 2.861(1), N3-O2w 2.900(1), N6-O1w 2.624(1), N7-O2w 2.666(1). Pyridinium 3,5-dinitro-1H-1,2,4-triazolate crystallizes with one molecule of water per formula unit in the monoclinic space group C2/c. The solid-state structure (15 . H 2 O) consist of [(HPy][DNT]·H 2 O units that are arranged in sheets that are coplanar to the ab plane of the crystal. One such unit is depicted in Figure 3.17. In each unit, two water molecules bridge two DNT anions through hydrogen bonds. For these bonds, the oxygen atom of the water molecules act as hydrogen donors with a nitrogen atom of a triazole ring as the acceptor. In addition, the two water oxygen atoms act as hydrogen acceptors in additional hydrogen bonds to pyridinium cations. It is interesting to note that the pyridinium cations of such a unit are not coplanar. The angle between the two planes containing the two six-membered rings is 28.7(1)°. The solid-state structure of the 5-aminotetrazolium salt [CH 2 N 5 ][DNT]·H 2 O (16 . H 2 O) is dominated by hydrogen bonds. The compound crystallizes in the orthorhombic space group P2 1 2 1 2 1 with eight formula units in the unit cell. The structure contains chains of alternating cation and anion units that are bridged by hydrogen bonds. The chains are oriented along the c-direction of the crystal. While the triazolyl rings of the anions within a chain are co-planar, the planes containing the tetrazolyl rings of the cations are rotated from each other by 73.5(1)°. The angles between the planes of the five-membered 49 rings of the cations and anions are 52.0(1) and 54.8(1)°. The chains are bridged through hydrogen bonds by water molecules. Parts of two such interconnected chains are depicted in Figure 3.18. Figure 3.18: The hydrogen bonding in 16 . H 2 O bridges two cation-anion chains. Thermal ellipsoids are shown at the 50% probability level. Hydrogen atom positions were determined from the electron density map and are depicted as spheres of arbitrary radius. For clarity, only some of the observed hydrogen bonds are included. Selected atomic distances (Å): N1-O9 2.895(2), N2-N16 2.737(2), N3-N14 2.890(2), N6-O10 2.870(2), N7-N19 2.748(2), N8-N11 2.835(2), N14-O1 3.281(2), O4-O10 2.922(2), O8-O9 2.921(2). The salts PPh 4 [DNT] (17), PPN[DNT] (18), and NMe 4 [DNT] (19) form crystals that do not contain incorporated water molecules. Selected crystallographic data for compounds 17-19 are listed in Table 3.4. The molecular structures of compounds 17 and 18 are given in APPENDIX 2. Figure 3.19 gives the molecular structure of 19. Figure 3.19: Molecular structure of 19. Thermal ellipsoids are shown at the 50% probability level. Hydrogen atoms were omitted for clarity. 50 The solid-state structures of the PPh 4 + , PPN + and NMe 4 + salts consist of well-separated cations and DNT - anions. The tetraphenylphosphonium compound 17 crystallizes with four formula units in space group P2 1 /c. The shortest cation-anion distance in the PPh 4 + salt was found at 3.188(2) Å (C7-O3). The bis(triphenylphosphine)iminium salt 18 crystallizes in the triclinic space group P with two formula units per unit cell. The shortest distances between the [(PPh 3 ) 2 N] + cations and DNT - anions are 3.163(3) Å (C8- O4). The closest anion-anion distance is 5.694(3) Å (N1-N2). Table 3.4: Crystallographic data for the [PPh 4 ] + , [(Ph 3 P) 2 N] + , and [NMe 4 ] + DNT salts 17-19. 17 18 19 formula C 26H 20N 5O 4P C 38H 30N 6O 4P 2 C 6H 12N 6O 4 mol wt [g/mol] 497.44 696.62 232.22 temp [K] 100(2) 100(2) 100(2) crystal system Monoclinic Triclinic Monoclinic space group P2 1/c P P2 1/c a [Å] 11.2261(5) 10.9099(7) 6.1515(3) b [Å] 15.0086(7) 11.9031(8) 17.1906(10) c [Å] 13.8078(6) 14.4603(10) 9.8599(5) α [°] 90 112.6130(8) 90 β [°] 95.0410(10) 90.0610(9) 92.6300(10) γ [°] 90 103.5780(9) 90 V [Å 3 ] 2317.45(18) 1676.28(19) 1041.57(10) Z 4 2 4  [Å] 0.71073 0.71073 0.71073 ρ calc [g/cm 3 ] 1.426 1.380 1.481 μ [mm -1 ] 0.164 0.182 0.124 F(000) 1032 724 488 reflns collected 50180 10859 3159 ind reflns 5872 7529 3159 R int 0.0558 0.0198 0.0146 no. of parameters 325 451 184 R 1 [I > 2σ(I)] 0.0373 0.0463 0.0406 [I > 2σ(I)] 0.0837 0.1150 0.1086 GOF 1.010 1.025 1.096 Tetramethylammonium 3,5-dinitro-1,2,4-triazolate (19) crystallizes in space group P2 1 /c. While the solid- state structure of the NMe 4 + salt is still composed of individual NMe 4 + and DNT - ions, the ions are not as well separated as in the PPh 4 + or PPN + salts. The shortest cation-anion distance is 2.935(2) Å (C3-O1) and the shortest anion-anion distance is 3.083(2) Å (C1-O3). The smaller distances between the ions is due to the smaller size of the NMe 4 + cation in comparison to PPh 4 + or PPN + . On one occasion, a crystalline sample of the composition NMe 4 [DNT]·HDNT was obtained from an impure sample of 19 that contained 51 unreacted HDNT. The molecular structure of NMe 4 [DNT]·HDNT and the full related crystallographic data are given in APPENDIX 2. 3.4 Thermal Stability The impact and friction sensitivities of the tetrazolates of this study were determined using a BAM Fall Hammer and a BAM Friction Tester, respectively. The obtained sensitivities are summarized in Table 3.5. Most compounds prepared in this study can be considered insensitive. Only the metal salts 4-8 have measurable friction sensitivities ranging from 160 for CsDNT to 324 N for the Sr 2+ salt. The metal salts 4- 9, the aminotetrazolium salt 16 and the tetramethylammonium salt 19 had measurable impact sensitivities ranging from the low value of 4.5 J for AgDNT to 85 J for the 5-aminotetrazolium salt. The two silver salts 8 and 9 can be considered highly impact sensitive with determined sensitivities of 4.5 and 10 J, respectively. With impact sensitivities below 39 J, the metal salts 4, 5, 8 and 9 must be considered hazardous explosive materials. The thermal stabilities of all compounds of this study were determined using Differential Thermal Analysis (DTA) scans with a heating rate of 5 °C min -1 . The results of these scans are summarized in Table 3.5. Compounds 11, 12, 15, and 17-18 melt without decomposition. Melting points ranged from 100 °C for the [HNEt 3 ] + salt to 225 °C for the TMA salt. Table 3.5: Sensitivity and stability data for the DNT salts studied. Compound m.p. [°C] a T decomp [°C] a FS [N] IS [J] LiDNT·2H 2 O 1·2H 2 O - 280 >360 >100 NaDNT·2H 2 O 2·2H 2 O - 260 >360 >100 KDNT·2H 2 O 3·2H 2 O - 265 >360 >100 RbDNT 4 - 250 192 20 CsDNT 5 - 250 160 25 Sr(DNT) 2 ·6H 2 O 6·6H 2 O - 290 c 324 55 Ba(DNT) 2 ·11H 2 O 7·11H 2 O - 245 192 55 AgDNT 8 - 240 192 4.5 Ag(NH 3 )DNT 9 - 245 d >360 10 NH 4 DNT·2H 2 O 10·2H 2 O - 190 >360 >100 [HNEt 3 ]DNT 11 100 190 >360 >100 [H 2 NEt 2 ]DNT b 12 150 190 >360 >100 [C(NH 2 ) 3 ]DNT 13 - 200 >360 >100 [CH 7 N 3 ]DNT 14 - 200 >360 >100 [C 5 H 6 N]DNT·H 2 O 15 120 160 >360 >100 [CH 4 N 5 ]DNT·H 2 O 16 - 160 >360 85 [PPh 4 ]DNT 17 180 350 >360 >100 [PPN]DNT 18 150 340 >360 >100 [TMA]DNT 19 225 235 >360 75 a DTA onset; b possible mixture of polymorphs; c explosion; d endotherm at 155 °C (loss of NH 3) 52 Decomposition onset temperatures, as determined by DTA, range from 245 °C to 290 °C for the metal salts. The three ammonium salts 11-13 showed decomposition temperatures of 190 °C while the two guanidinium salts 14 and 15 have slightly higher decomposition temperatures of 200 °C. With decomposition temperatures of 160 °C, the pyridinium salt 15 and the 5-aminotetrazolium salt 16 are the least thermally stable among the investigated DNT salts. Not surprisingly, the most stable compounds towards thermal decomposition are the PPh 4 + and PPN + salts 17 and 18. The decomposition temperatures of the two salts are 340 °C and 350 °C, respectively. It is interesting to note that, with a decomposition temperature of 235 °C, the tetramethylammonium salt 19 was found to be thermally less stable than any investigated metal DNT salt. While most investigated compounds showed smooth decompositions in the DTA scans, the aminoguanidinium salt 14 and the 5-aminotetrazolium salt 16·H 2 O exhibited sharp exotherms in the DTA. Only in the case of the strontium salt 6·6H 2 O was an explosions observed upon heating. In addition to DTA scans, the decomposition of the metal DNT salts was also investigated in a qualitative way by heating of the neat salts in the flame of a Bunsen burner. In all cases, the salts deflagrated, producing a colored flames characteristic for the respective metal cation. These salts could be useful for pyrotechnical applications. 3.5 Spectroscopy The nineteen investigated 3,5-dinitro-1,2,4-triazolates have been characterized by vibrational spectroscopy (IR and Raman) and multinuclear NMR spectroscopy ( 1 H, 13 C, 14 N, and 31 P). The observed frequencies and intensities of the observed vibrational bands as well as the recorded chemical shifts are listed in APPENDIX 2. Besides resonances due to the respective cations, the 13 C NMR and 14 N NMR spectra show the resonances expected for a dinitro-substituted symmetric 1,2,4-triazolyl anion. The 13 C NMR spectra of all investigated DNT salts exhibit one single resonance at 162 to 164 ppm in (CD 3 ) 2 CO, CD 3 CN, or DMSO-d 6 due to the DNT anion. The most noticeable feature in the 14 N NMR spectra of the DNT anion in the various salts is the resonance of the NO 2 -groups at -19 to -22 ppm. This resonance is relatively sharp with a line width of about 50 Hz. Two resonances at about -50 ppm (τ 1/2 ≈ 400 Hz) and - 145 ppm (τ 1/2 ≈ 300 Hz) are observed for the three nitrogen atoms of the triazolyl ring. Besides vibrational bands due to the cation or water, the vibrational spectra of the DNT salts display strong characteristic bands characteristic for the NO 2 groups at around 1550-1530 cm -1 and 1520-1480 cm -1 (ν as NO 2 ), as well as 1400-1370 cm -1 and 1360-1290 cm -1 (ν s NO 2 ). In addition, a strong band for the deformation of the C 2 N 3 ring is observed in the range 850-820 cm -1 . These bands are in good agreement with vibrational data 53 published previously for potassium 3,5-dinitro-1,2,4-triazolate and 1-methyl-3,5-dinitro-1,2,4-triazole. 31,32 Due to the negative charge, the observed vibrational bands of the DNT anion are shifted by 10-40 cm -1 to lower wavenumbers in comparison to the neutral parent compound 3,5-dinitro-1H-1,2,4-triazole (HDNT, APPENDIX 1). The effect is most pronounced for the two asymmetric stretch modes (in-phase and out- of-phase) of the NO 2 groups bands at 1550-1530 cm -1 and 1520-1480 cm -1 (HDNT: 1563 and 1530 cm -1 ). 3.6 Conclusion Numerous salts of 3,5-dinitro-1H-1,2,4-triazole have been prepared. Most of the studied 3,5-dinitro-1,2,4- triazolates have relatively high thermal stability and show only very low shock and friction sensitivity. The salts containing alkali metal, alkali earth metal, and silver cations exhibit colored emissions upon combustion and might be of interest as components in pyrotechnics formulations. Most compounds in this study have been fully characterized by their X-ray crystal structure, vibrational and multinuclear NMR spectra, their decomposition temperature, as well as friction and impact sensitivities. Further experimental details, crystal packing diagrams and crystallographic information can be found in APPENDIX 2. 3.7 Reference (1) Talawar, M. B.; Sivabalan, R.; Mukundan, T.; Muthurajan, H.; Sikder, A. K.; Gandhe, B. R.; Rao, A. S. J. Hazardous Mat. 2009, 161, 589. (2) Langlet, A.; Oestmark, H. WO9402434A1 1994. (3) Lee, K. Y.; Storm, C. B.; Hiskey, M. A.; Coburn, M. D. J. Energ. Mater. 1991, 9, 415. (4) Simpson, R. L.; Pagoria, P. F.; Mitchell, A. R.; Coon, C. L. Propellants Explos., Pyrotech. 1994, 19, 174. (5) Zhang, Y.; Parrish, D. A.; Shreeve, J. M. J. Mater. Chem. A 2013, 1, 585. (6) Miller, C. G. W., Graylon K. US2009199937 2009. (7) Huynh, M. H. V. WO2008US01904 2008. (8) Burchfield, H. P.; Gullstrom, D. K. US3054800 1962. (9) Wiley, R. H.; Smith, N. R. US3111524 1963. (10) Klapötke, T. M.; Penger, A.; Pflüger, C.; Stierstorfer, J.; Sućeska, M. Eur. J. Inorg. Chem. 2013, 2013, 4667. (11) Klapötke, T. M.; Piercey, D. G.; Stierstorfer, J. Dalton Trans. 2012, 41, 9451. (12) Kofman, T. P. Russ. J. Org. Chem. 2001, 37, 1158. 54 (13) Pevzner, M. S.; Kofman, T. P.; Kibasova, E. N.; Sushchenko, L. F.; Uspenskaya, T. L. Chem Heterocycl Compd 1980, 16, 194. (14) Stinecipher, M. M. Investigation of the Physical and Explosives Properties of the Eutectic Explosive Ammonium Nitrate/Ammonium 3,5-Dinitro-1,2,4-Triazolate; 1978 Annual Report, Los Alamos National Laboratory, 1978. (15) Haiges, R.; Jones, C. B.; Christe, K. O. Inorg Chem 2013, 52, 5551. (16) Bélanger-Chabot, G.; Rahm, M.; Haiges, R.; Christe, K. O. Angew. Chem. Int. Ed. 2013, 52, 11002. (17) Haiges, R.; Christe, K. O. Inorg. Chem. 2013, 52, 7249. (18) Rahm, M.; Bélanger-Chabot, G.; Haiges, R.; Christe, K. O. Angew. Chem. Int. Ed. 2014, 53, 6893. (19) Huynh, M. H. V.; Hiskey, M. A.; Gilardi, R. J. Energ. Mater. 2005, 23, 27. (20) Xue, H.; Gao, H.; Shreeve, J. M. J. Polym. Sci., Part A: Polym. Chem. 2008, 46, 2414. (21) Hiskey, M. A.; Stinecipher, M. M.; Brown, J. E. J. Energ. Mater. 1993, 11, 157. (22) Katritzky, A. R.; Singh, S.; Kirichenko, K.; Holbrey, J. D.; Smiglak, M.; Reichert, W. M.; Rogers, R. D. Chem. Commun. 2005, 868. (23) Katritzky, A. R.; Singh, S.; Kirichenko, K.; Smiglak, M.; Holbrey, J. D.; Reichert, W. M.; Spear, S. K.; Rogers, R. D. Chem. - Eur. J. 2006, 12, 4630. (24) Klapötke, T. M. P., A.;Schedlbauer, S.;Stierstorfer, J. Cent. Eur. J. Energ. Mater. 2013, 10, 151. (25) Lee, K. Y.; Ott, D. G.; Stinecipher, M. M. Ind. Eng. Chem. Process Des. Dev. 1981, 20, 358. (26) Sitzmann, M. E. J. Org. Chem. 1978, 43, 3389. (27) Smiglak, M.; Hines, C. C.; Wilson, T. B.; Singh, S.; Vincek, A. S.; Kirichenko, K.; Katritzky, A. R.; Rogers, R. D. Chem. Eur. J. 2010, 16, 1572. (28) Tao, G.-H.; Guo, Y.; Parrish, D. A.; Shreeve, J. M. J. Mater. Chem. 2010, 20, 2999. (29) Xue, H.; Gao, H.; Twamley, B.; Shreeve, J. M. Eur. J. Inorg. Chem. 2006, 2006, 2959. (30) Chernyshev, V. M. Z., N. D.; Taranushich, V. A. RU2174120 C1 2001. (31) Melnikov, V. V.; Stolpakova, V. V.; Khorkova, L. F.; Pevzner, M. S.; Melnikova, N. N. Chem. Heterocycl. Comp. 1972, 8, 111. (32) Melnikov, V. V.; Stolpakova, V. V.; Zacheslavskii, S. A.; Pevzner, M. S.; Gidaspov, B. V. Chem. Heterocycl. Comp. 1973, 9, 1289. 55 CHAPTER 4: SYNTHESIS OF BORANYL-BASED COMPLEX ANIONS OF NITRO-SUBSTITUTED AZOLATES 4.1 Introduction The search for environmentally benign, high-performance energetic materials has attracted significant interest in recent years. 1 More than 10 years ago, Christe et al. proposed the use of Group 13 complex anions with energetic ligands, such as dinitrotriazolate, as potential green replacements for ammonium perchlorate in solid rocket propellant formulations. The synthesis of a limited number of energetic boron compounds with nitrate, perchlorate and trinitromethanide ligands had been explored before. 2-5 In recent years, notable contributions include the first structural characterization of the (trinitromethyl)trihydroborate anion 6 (CHAPTER 5) and the synthesis of several energetic poly(azolyl)borate salts, 7,8 high-nitrogen azolyl-borate systems 9-14 and trinitromethylalkoxyborates. 15-18 No energetic azole complexes with BH 3 have been isolated so far. While tetrakis(nitroazolyl)borate and bis(nitroazolyl)dihydroborate salts have been studied by the Christe-Haiges group and others, the less substituted boron derivatives mono(nitroazolyl)trihydroborates have not been thoroughly studied. The considerable reducing character of the BH 3 moiety raises the question whether borane adducts with oxidizer groups-bearing ligands are prohibitively sensitive and unstable. The preparation of (trinitromethyl)trihydroborate salts showed that this type of compounds can be surprisingly stable. 6 Lastly, since the use of azole-borane and azolylborohydrides as hydrogen storage intermediates has been considered at least conceptually, 19 it is interesting to probe the influence of the electron withdrawing nitro and polynitromethyl moieties on the stability of the N-B bond formed in high-oxygen azolyl-BH 3 compounds. Herein the study of the reaction chemistry of the BH 3 moiety with the 3,5-dinitro-1,2,4- triazolate, 3-nitro-5-(trinitromethyl)-1,2,4-triazolate, 3-(fluorodinitromethyl)-5-nitro-1,2,4-triazolate, 5- (trinitromethyl)tetrazolate and 5-(fluorodinitromethyl)tetrazolate anions is described. 4.2 Detection of Boranyl-azolate Adducts by NMR The coordination chemistry of the nitroazolate anions with borane in acetonitrile solutions was probed using NMR spectroscopy. An excess of the phosphonium salt of the corresponding azolate (triazolate or tetrazolate), Az - , was reacted with a Me 2 S solution of BH 3 . SMe 2 and analyzed. An excess of the borane 56 solution was then added and the mixture was analyzed again. Figure 4.1 summarizes the species observed in the course of these studies and Table 4.1 lists their characteristic NMR chemical shifts. Figure 4.1: Coordination chemistry of the azolate ligands of this study with BH 3 . With the three triazolates (Trz - ) studied, only monosubstituted [Trz-BH 3 ] - species and no disubstituted species of the type [Trz(BH 3 ) 2 ] - could be observed, even in the presence of an excess of borane. For the (trinitromethyl)nitrotriazolate [TNMeNTrz-BH 3 ] and (fluorodinitromethyl)nitrotriazolate [FDNMeNTrz- BH 3 ] - complexes, however, two species were observed at the beginning of the reaction (Table 4.1), which are attributed to the two isomers with the borane moiety at the N 1 or N 2 positions, where the protonation usually occurs for similar triazole rings. 20 For the [TNMeNTrz-BH 3 ] - complex, the two isomers 2,3 were found in a 1:0.7 ratio at the beginning of the reaction, while for the [FDNMeNTrz-BH 3 ] - adduct , the isomers 4,5 were found in a 1:0.2 ratio. As expected from the symmetry-equivalent N 1 and N 2 positions in DNT - , only one isomer of [DNT-BH 3 ] - (1) was observed throughout the reaction. The three triazolate salts used in this study formed one major isomer of the borane complex [Trz-BH 3 ] - at the end of the reaction, 57 which were identified as the isomer substituted at the nitrogen atom in α-position to the bulkiest substituent by X-ray crystallography (vide infra). Even in the presence of an excess of tetrazolate (Tz - ) ligand, two major [Tz-BH 3 ] - species were observed, which were attributed to the 1-boranyl and 2-boranyl species. In the case of trinitromethyltetrazolate (TNTz - ), the 2-boranyl isomer 6 was found as the major species with smaller amounts of the 1-boranyl isomer 7 in solution. This assignment was confirmed by X-ray crystallography. In contrast, the isomers 8 and 9 were found in nearly identical amounts in the case of (fluorodinitromethyl)tetrazolate (FDNTz - ). X-ray crystallography showed that species 8 was the 2- boranyl isomer. It is noteworthy that it took several hours to several days for each reaction to reach completion and that the exact isomer ratio at a given time depended strongly on the solvent and conditions. Table 4.1: NMR chemical shifts (ppm) of the observed BH 3 -azole species in deuterated acetonitrile solution. δ( 11 B) δ( 14 N) δ( 1 H) δ( 19 F) [DNT-BH 3 ] - 1 -17.3 -26.0 2.23 - [TNMeNTrz-BH 3 ] - 2 -18.2 -34.5 -26.2 2.10 - [TNMeNTrz-BH 3 ] - 3 -17.3 -31.8 a Obscured - [FDNMeNTrz-BH 3 ] - 4 -18.5 -24.4 -26.0 2.01 -93.0 [FDNMeNTrz-BH 3 ] - 5 -17.4 Too weak Obscured -97.3 [TNTz-BH 3 ] - (2-boranyl isomer) 6 -17.2 -31.4 2.43 - [TNTz-BH 3 ] - (1-boranyl isomer) 7 -19.6 -34.4 Obscured - [FDNTz-BH 3 ] - (2-boranyl isomer) 8 -17.4 -21.8 2.40 -96.8 [FDNTz-BH 3 ] - (1-boranyl isomer) 9 -20.0 -24.1 Obscured -94.2 [TNTz(BH 3 ) 2 ] - 10 -16.2 -18.5 -37.8 Obscured - [FDNTz(BH 3 ) 2 ] - 11 -16.4 -19.1 -27.8 Obscured -95.3 a Other expected signal was obscured 58 When several equivalents of BH 3 . SMe 2 were added to solutions of the tetrazolate ligands, the formation of an additional species was observed by 14 N NMR spectroscopy. The 11 B NMR spectrum of such reaction mixtures displayed two new resonances (Table 4.1), which are attributed to one isomer of the 2:1 adducts bis(boranyl) tetrazolate, [Tz(BH 3 ) 2 ] - . There are four conceivable isomers of such adducts, two of which having a symmetry plane perpendicular to the plane of the tetrazole ring and two of which having no such symmetry. The observation of two non-equivalent boron chemical environments by 11 B NMR indicates that only one of the two isomers lacking a symmetry plane perpendicular to the plane of the tetrazole ring ( which are depicted in Figure 4.2) was formed. [TNTz(BH 3 ) 2 ] - (10) could not be isolated as a solid and attempts to determine the structure of [FDNTz(BH 3 ) 2 ] - (11) by X-ray crystallography were unsuccessful. Since the 14 N NMR resonances of the five-membered tetrazolate rings are too broad to unambiguously assign individual resonances, it is only possible to conjecture on which of the two possible isomers of Figure 4.2 was observed in solution. Based on steric arguments, the isomer substituted at the 1- and 3- positions would be the most likely. The consumption of both isomers 8 and 9 of [FDNTz-BH 3 ] - was observed during the formation of 11. In acetonitrile solution, [Tz-BH 3 ] - and [Tz(BH 3 ) 2 ] - were always in equilibrium and their respective concentration was strongly influenced by the presence of strong Lewis- basic ligands, such as dimethylsulfide. When similar reactions were performed in the presence of less Lewis basic solvent, such as dichloromethane or tetrahydrofuran (THF) and in the absence of SMe 2 , the formation of [Tz-(BH 3 ) 2 ] - species was much more favorable. Figure 4.2: Two possible isomers for the bis(boranyl)tetrazolate species observed based on the two signals observed in 11 B NMR spectroscopy. 4.3 Isolation While the reaction of BH 4 - with two equivalents of the parent azoles was successfully applied to the synthesis of several [BH 2 Az 2 ] - salts, 7 the reaction of a single equivalent of the azole with BH 4 - yielded only mixtures of [BH n Az 4-n ] - salts (n=2,3) (APPENDIX 3). PPN 59 (bis(triphenylphosphoranylidene)ammonium) and/or PPh 4 salts of the [Az-BH 3 ] - anions were therefore prepared by the addition of an excess (1.2 to 1.5 equivalents) of commercially available BH 3 . THF solutions in THF to either solid Az - salts or their solution in dichloromethane. A NMR sample of the reaction mixture was analyzed and the mixure was taken to dryness in vacuo. The obtained solids were analyzed by NMR and vibrational spectroscopy and their thermal stabilities determined by differential temperature analysis (DTA). With the exception of [TNTz-BH 3 ] - , all of the different [Az-BH 3 ] - salts were obtained in nearly quantitative yields but contained variable amounts of organic impurities, such as residual solvent, BH 3 . THF or BH 3 . THF decomposition products. The use of BH 3 . THF that was purified by vacuum transfer and kept cold did not prevent the presence of these impurities. Most of the impurities could be removed by washing the crude solids with water. However, this resulted in the isolation of materials that contained significant amounts of water, which could not be removed in vacuo. PPh 4 [DNT-BH 3 ] (1a), PPN[DNT-BH 3 ] (1b), and PPh 4 [TNMeNTrz-BH 3 ] (2a) were isolated in as pale yellow solids. PPh 4 [FDNMeNTrz-BH 3 ] (4a) was obtained as an off-white solid. PPh 4 [TNTz-BH 3 ] (6a) was isolated as a bright yellow solid by concentrating a dichloromethane solution in vacuo below 0°C and then dried in vacuo at ambient temperature for several hours. 14 N NMR spectra of this crude material revealed the presence of small amounts of the second isomer of PPh 4 [TNTz- BH 3 ] (7a) and the disubstituted species PPh 4 [TNTz-(BH 3 ) 2 ] (10a) as well as significant amounts (ca 10 mol%) of unidentified impurities with signals between -32.3 and -33.3 ppm, which can be correlated with decomposition. Subsequent studies revealed that 6, 7 and 10 are only marginally stable at room temperature. Attempts to remove these impurities by washing the solids with water were unsuccessful. Therefore, the synthesis of salts of 6, 7 and 10 was not pursued any further. PPh 4 [FDNTz-BH 3 ] (8a) was isolated as a white solid containing small amounts of PPh 4 [FDNTz(BH 3 ) 2 ] (11a) and ca 15 mol% of the second isomer 9a. A sample containing only 8a and 9a was obtained by washing the crude solid product with water, extracting the solids with dichloromethane, drying the organic phase over MgSO 4 and taking the solution to dryness in vacuo. The PPN and/or PPh 4 salts of [Tz(BH 3 ) 2 ] - were prepared by adding 2-3 equivalents of BH 3 . THF/THF either to the neat Tz - salt or a dichloromethane solution. The solvent was then removed by prolonged pumping at room temperature. PPh 4 [TNTz-(BH 3 ) 2 ] (10a) was observed both in dichloromethane and THF reaction mixtures, in equilibrium with [TNTz-BH 3 ] - and BH 3 . THF. This is indicative of the decreased strength of the second N- B bond in comparison to the first bond of this type. Upon drying of such reaction mixtures under vacuum 60 at room temperature for 24 h, a sticky orange paste was obtained. Washing this paste with pentane yielded a free-flowing glassy solid, which contained ca 45 mol% of 6a and ca 5 mol% of 10a and 50 mol% of various impurities as estimated by 11 B and 14 N NMR spectroscopy. Even though 10 was the major species in solution prior to the evaporation of the solvent, only minor amounts of it remained in the isolated material, either because of the removal of the second BH 3 molecule under vacuum and/or because of the decomposition of the complex anion. Attempts at isolating PPN[TNTz(BH 3 ) 2 ] (10b) yielded similar results, but did not require the use of pentane to obtain a free-flowing solid. PPh 4 [FDNTz(BH 3 ) 2 ] (11a) was observed both in dichloromethane and THF solutions. 11a was found to be in equilibrium with BH 3 . solvent + 8a/9a (11a being favored) in THF, while in dichloromethane no such equilibrium was observed. 11a was thus obtained from such a dichloromethane solution as a white solid which contained no detectable amounts of 8a or 9a. Attempts to obtain crystals of 11a of suitable quality for X-ray diffraction have been unsuccessful. 4.4 Stability As might be expected, all of the [Az-BH 3 ] - salts are significantly less stable than the corresponding boron- free azolate parent compounds. As indicated earlier, the PPh 4 + and PPN + salts of the [Az-BH 3 ] - complexes studied were marginally hydrolytically stable. Indeed, solids and solutions in hydrophobic solvents could be exposed to air without significant decomposition. The solids could also be suspended in water without significant hydrolysis. In water-miscible solvents, however, all of the compounds underwent hydrolysis at a significant rate. 11a, however, quickly yields 8a and 9a when exposed to water. It can be assumed that alkali metal salts are significantly less hydrolytically stable. 6 The compounds obtained in this work did not explode when provoked with a Teflon spatula. It should be assumed, however, that the alkali metal salts of compound 1-11 are more sensitive to friction and/or impact (see APPENDIX 3 on K[DNT-BH 3 ]). Some of the solids slowly decomposed when stored at room temperature and left very small amounts of white residue upon re-dissolution. This phenomenon has also been observed for salts of the [BH 3 C(NO 2 ) 3 ] - anion 6 (CHAPTER 5), and is usually associated with the formation of higher, insoluble boron hydrides. Upon storage at room temperature, salts of 1, 6/7 and 8/9 and their solutions decomposed significantly. After several weeks of storage at ambient temperature, solid samples of [FDNTz-BH 3 ] - , [DNT-BH 3 ] - and [TNTz-BH 3 ] - contained significant amounts of free Az - , unidentified boron products and unidentified nitro-containing species as identified by NMR spectroscopy. It is likely that 2a, 4a and 11a slowly decompose in a similar fashion upon storage at room temperature. Resonances very close to those 61 reported for [BH 2 (TNTz) 2 ] -7 were observed by NMR spectroscopy in [TNTz-BH 3 ] - solutions in acetonitrile. This suggests that [Az-BH 3 ] - can undergo scrambling reactions such as the one shown in Equation 4.1. 2 [TNTz-BH 3 ] → [BH 2 (TNTz) 2 ] - + BH 4 - (4.1) The thermal stability of the compounds was determined by DTA. The thermogram of compound 1a displays a decomposition onset of 173 °C, with a sharp exotherm at 183 °C, which is interestingly rather close to the melting point of PPh 4 DNT of 180 °C. 21 1a is therefore significantly less stable than the PPh 4 DNT which decomposes at 350 °C. The thermogram of 1b displays an endotherm at ca 125 °C and poorly defined decomposition onsets between 120 and 170 °C. A sample of crude 2a showed a decomposition onset of 70 °C with a very sharp exotherm at 98 °C, while the compound washed with water, which contained significant residual moisture, had an onset at ca 80 °C. Compound 4a has a decomposition onset of 113 °C, significantly lower than that of PPh 4 FDNMeNTrz (192 °C). As is the case for PPh 4 FDNMeNTrz and PPh 4 TNMeNTrz, the replacement of a nitro group by a fluorine atom in the trinitromethyl group significantly improves the stability of 4a compared to that of 2a. Samples containing 6, 7 and 10 had poorly defined decomposition onsets between 35 and 80 °C, which are consistent with observations that the compounds are marginally stable at room temperature. Interestingly, tetraphenylphosphonium dinitro(1-H-tetrazol-5-yl)methanide monohydrate (12) crystals (Figure 4.9) were obtained from the aqueous washings of crude 6a. This hints at one possible decomposition pathway for 6. Indeed, since dinitro(1-H-tetrazol-5-yl)methanide derivatives can be obtained by reduction of (trinitromethyl)tetrazole, it can reasonably assumed that similar reactions could occurs in the case of 6/7. It is however surprising that no evidence for such decomposition was observed for 2a. The DTA trace of a 1: 0.15 mixture of 8a/9a, which contained significant amounts of residual water (as observed by 1 H NMR) showed an endotherm onset of 85 °C with a decomposition onset of 112 °C. This decomposition onset is significantly lower than that observed for the parent compound PPh 4 FDNTz at 175 °C. 22 The thermogram of 11 displays an endotherm at 116 °C (onset 103 °C) and a decomposition onset between 116 and 126 °C with a major exotherm at 151 °C, suggesting similar thermal stabilities for 62 8, 9 and 11. The fluorodinitromethyl-substituted compounds are again significantly more stable than their trinitromethyl analogue. 4.5 Structural Characterization Crystals of compounds 1a, 2a, 4a, 6a, and 8a were obtained by slow evaporation of dichloromethane solutions in air. Crystals of 1b were obtained by recrystallization from dichloromethane/pentane solutions. Several different batches of crystals of 1a and 1b grown under various conditions were studied by X-ray diffraction but always yielded disordered structures. Table 4.2: Crystallographic data for the salts 1a, 1b, 2a, 4a, 6a, 8a and 12 obtained at 100(2) K. 1a 1b 2a 4a 6a 8a 12 Formula C 26 H 23 BN 5 O 4 P C 38 H 33 BN 6 O 4 P 2 C 27 H 23 BN 7 O 8 P C 27 H 23 BFN 6 O 6 P C 26 H 23 BN 7 O 6 P C 26 H 23 BFN 6 O 4 P C 26 H 23 N 6 O 5 P Mol wt [g/mol] 511.27 710.45 615.30 588.29 571.29 544.28 530.47 Crystal system Monoclinic Monoclinic Orthorhombic Orthorhombic Triclinic Triclinic Orthorhombic Space group Cc P2 1 /n Fdd2 P2 1 2 1 2 1 P P P2 1 2 1 2 1 a [Å] 7.22590(10) 12.1574(6) 38.9762(7) 7.3045(4) 7.70210(10) 7.55540(10) 7.3432(3) b [Å] 16.4562(2) 21.1776(10) 39.8619(7) 18.1495(10) 12.4067(3) 12.9653(2) 11.7912(5) c [Å] 20.8053(3) 14.0475(7) 7.38940(10) 20.4393(12) 14.9270(4) 14.2531(2) 29.3176(12) α [°] 90 90 90 90 99.739(2) 101.1770(10) 90 β [°] 96.6270(10) 103.0280(10) 90 90 98.257(2) 95.7900(10) 90 γ [°] 90 90 90 90 102.742(2) 103.1080(10) 90 V [Å 3 ] 2457.45(6) 3523.6(3) 11480.7(3) 2709.7(3) 1346.71(5) 1318.62(3) 2538.47(18) Z 4 4 16 4 2 2 4 λ[Å] 1.54178 0.71073 1.54178 1.54178 1.54178 1.54178 0.71073 ρ calc [g/cm 3 ] 1.382 1.339 1.424 1.442 1.409 1.371 1.388 μ [mm -1 ] 1.360 0.174 1.395 1.435 1.382 1.369 0.158 F(000) 1064 1480 5088 1216 592 564 1104 Reflections collected 14293 73768 41733 44702 34096 20769 64191 Independent reflections 4422 8084 5420 2031 5095 4908 7771 R int 0.0497 0.0692 0.0643 0.1098 0.0584 0.0688 0.0595 Number of parameters 368 487 401 501 374 374 362 R 1 [I > 2σ(I)] 0.0378 0.0409 0.0382 0.0555 0.0409 0.0687 0.0418 wR 2 [I > 2σ(I)] 0.0949 0.0945 0.0977 0.1511 0.1103 0.1828 0.0921 GOF 1.101 1.031 1.068 1.089 1.065 1.098 1.073 63 Table 4.3: Selected bond lengths (Å) and angles (°) for the [Trz-BH 3 ] - salts 1a, 1b, 2a and 4a. 1a R 1 = -NO 2 1b R 1 = -NO 2 2a R 1 =TNMeNTrz 4a R 1 =FDNMeNTrz N1-N2 1.342(12) 1.355(4) 1.354(4) 1.356(12) N2-C1 1.321(4) 1.330(7) 1.332(5) 1.423(17) C1-N3 1.331(4) 1.314(7) 1.324(4) 1.299(18) N3-C2 1.361(12) 1.322(5) 1.334(4) 1.336(12) C2-N1 1.341(8) 1.340(4) 1.348(4) 1.353(13) C1-N4 1.458(4) 1.441(5) 1.451(5) 1.425(9) N4-O1 1.219(5) 1.214(2) 1.226(4) 1.204(7) N4-O2 1.222(5) 1.225(2) 1.231(4) 1.228(7) N1-B 1.585(12) 1.608(4) 1.591(5) 1.596(14) N1-N2-C1 103.2(5) 102.1(3) 102.5(3) 102.5(9) N2-C1-N3 117.8(3) 118.0(3) 117.8(3) 113.8(9) C1-N3-C2 98.7(6) 99.3(3) 99.3(3) 103.2(10) N3-C2-N1 113.0(11) 114.3(4) 113.8(3) 113.3(11) C2-N1-N2 107.2(10) 106.3(3) 106.5(3) 107.1(10) O1-N4-O2 125.9(3) 125.06(18) 125.4(3) 124.5(6) C1-N4-O1 117.4(3) 117.78(17) 116.2(3) 116.8(6) C1-N4-O2 116.6(3) 117.17(16) 118.4(3) 118.6(5) The crystallographic data for compounds 1a, 1b, 2a, 4a, 6a, 8a and 12 are listed in Table 4.2. Several of the structures, namely that of the two [DNT-BH 3 ] - salts, that of the two salts containing the fluorodinitromethyl moiety and that of dinitromethylenetetrazole, show disorder, which prevents detailed, accurate structural discussions for these compounds. The quality of the data is nevertheless sufficient to confirm the connectivity in these compounds and for qualitative comparisons. Excluding a few outlying values obviously due to disordered moieties, most of the structural parameters for the triazole and tetrazole rings, including boron-nitrogen bond lengths, are identical or nearly identical. The structural parameters common to the four [Trz-BH 3 ] - salts (1a, 1b, 2a and 4a) are summarized in Table 4.3. Most of the parameters are very similar to those of the parent tetrazolate salts and only the N3-C2-N1 bond angle in the [Trz-BH 3 ] - salts appear to be appreciably narrower (113.0(11)- 114.3(4) °) than in the borane-free DNT - salts (ca 117°). 21 The B-N bond lengths in [Trz-BH 3 ] - range 64 between 1.585(12) and 1.608(4) Å, appreciably longer than the B-N distances in [BH 2 (Az) 2 ] - , which range between 1.572(2)-1.586(3) Å, 7 indicative of a more dative character of the B-N bond in [Trz-BH 3 ] - . The structural parameters common to the two [Tz-BH 3 ] - salts (6a, 8a) are summarized in Table 4.4. As in [Trz-BH 3 ] - , the bond angles for the atoms neighboring the BH 3 moiety in [Tz-BH 3 ] - are appreciably different from those of TNTz - . The N1-N2-N3 and N2-N3-N4 bond angles in [Tz-BH 3 ] - range between 111.8(3)-112.25(13) and 107.66(13)-107.6(3)°, respectively, while the ones in TNTz - range between 109.68(11)-110.8(2) and 108.9(4)-109.96(11)°, respectively. 23 The coordination of BH 3 thus results in a slight broadening of the N1-N2-N3 angle and a slight compression of the N2-N3-N3 angle. 23 In contrast to the [Trz-BH 3 ] - salts, the B-N distances in [Tz-BH 3 ] - (1.579(5)-1.583(2) Å) are for all practical purposes identical to those observed in [BH 2 (Az) 2 ] - , 7 suggesting a more covalent B-N bond in [Tz-BH 3 ] - than in [Trz-BH 3 ] - . Table 4.4: Selected bond lengths (Å) and angles (°) for the [Tz-BH 3 ] - salts 6a and 8a 1a crystallizes in the monoclinic space group Cc with four formula units per unit cell. In the solid-state structure (Figure 4.3), the [DNT-BH 3 ] - anion shows a positional disorder with a refined ratio of 1:4 in 6a R= -NO 2 8a R= -F 6a R= -NO 2 8a R= -F C1-N1 1.325(2) 1.329(4) C1-N1-N2 101.84(14) 102.5(3) N1-N2 1.3272(19) 1.318(4) N1-N2-N3 112.25(13) 111.8(3) N2-N3 1.324(2) 1.333(4) N2-N3-N4 107.66(13) 107.6(3) N3-N4 1.324(2) 1.324(5) N3-N4-C1 104.41(14) 104.7(3) N4-C1 1.335(2) 1.330(5) N4-C1-N1 113.83(15) 113.4(3) N2-B1 1.583(2) 1.579(5) O1-N5-O2 127.23(19) 128.5(8) C1-C2 1.480(2) 1.470(5) O3-N6-O4 129.92(19) 124.4(8) C2-N5 1.535(3) 1.537(9) O1-N5-C2 119.10(17) 114.9(8) C2-N6 1.521(3) 1.594(8) O2-N5-C2 113.68(18) 115.7(7) O1-N5 1.198(2) 1.212(9) O3-N6-C2 115.50(16) 115.7(5) O2-N5 1.205(2) 1.262(10) O4-N6-C2 117.53(19) 119.6(7) O3-N6 1.210(2) 1.231(8) N5-C2-N6 108.38(15) 100.1(5) O4-N6 1.212(2) 1.225(9) N5-C2-C1 113.89(15) 114.0(5) N6-C2-C1 109.30(15) 117.4(4) 65 which the five-membered ring is rotated by 180° along the C1-N4 bond, interchanging N3-C2-NO 2 and N2-N1-BH 3 . While the plane of the NO 2 group in -position to the BH 3 moiety is virtually coplanar with the five-membered ring, the nitro group in the -position is significantly rotated out of the triazole plane, with a O2-N4-C1-N2 torsion angle of 33(2)°. This is the apparent result of the repulsion from the BH 3 group. Figure 4.3: ORTEP plot of the asymmetric unit of PPh 4 [DNT-BH 3 ] (1a) showing only one part of the two-part positional disorder along the C1-N4 axis. Thermal ellipsoids are drawn at the 50% probability level. Hydrogen atom positions were idealized. Hydrogen atoms are depicted as spheres of arbitrary radius. Selected structural parameters: Bond distances (Å) : N1-N2 1.342(12), N2-C1 1.321(4), C1-N3 1.331(4), N3-C2 1.361(12), C2-N1 1.341(8), C1-N4 1.458(4), N4-O1 1.219(5), N4-O2 1.222(5), N1-B1 1.585(12), C2-N5 1.460(13), N5-O3 1.208(7), N5-O4 1.219(7). Bond angles (°): N1-N2-C1 103.2(5), N2-C1-N3 117.8(3), C1-N3-C2 98.7(6), N3-C2-N1 113.0(11), C2-N1-N2 107.2(10), O1-N4-O2 125.9(3), C1-N4-O1 117.4(3), C1-N4-O2 116.6(3), O3-N5-O4 126.1(5), C2-N5-O3 116.1(8), C2-N5-O4 117.8(7), O2-N4-C1-N2 0.2(5), O4-N5-C2-N1 33(2). 1b crystallizes in the monoclinic space group P2 1 /n with four formula units per unit cell. The crystal structure is depicted in Figure 4.4. The [DNT-BH 3 ] - anion shows the same pattern of disorder as for 1b. The structure of the anion is virtually identical to that of the anion in 1a. 66 Figure 4.4: ORTEP plot of the asymmetric unit of PPN[DNT-BH 3 ] (1b) showing only one part of the disordered DNT anion. Thermal ellipsoids are drawn at the 50% probability level. Hydrogen atom positions were idealized. Hydrogen atoms are depicted as spheres of arbitrary radius. Selected structural parameters: Bond distances (Å): N2-N1 1.355(4), N1-C2 1.330(7), C2-N3 1.314(7), N3-C1 1.322(5), C1- N2 1.340(4), C2-N5 1.441(5), N5-O3 1.214(2), N5-O4 1.225(2), N2-B1 1.608(4), C1-N4 1.441(5), N4- O1 1.228(3), N4-O2 1.210(3). Bond angles (°): N2-N1-C2 102.1(3), N1-C2-N3 118.0(3), C2-N3-C1 99.3(3), N3-C1-N2 114.3(4), C1-N2-N1 106.3(3), O3-N5-O4 125.06(18), C2-N5-O3 117.78(17), C2-N5- O4 117.17(16), C1-N4-O1 117.1(2), C1-N4-O2 117.9(3), O2-N4-C1-N2 29.6(6), O4-N5-C2-N1 15.7(4). 2a crystallizes with 16 formula units per unit cell in the orthorhombic space group Fdd2. Its crystal structure is depicted in Figure 4.5. The BH 3 group is coordinated in the α-position to the trinitromethyl group and in the β-position to the NO 2 group. 67 Figure 4.5: ORTEP plot of the asymetric unit of PPh 4 [TNMeNTrz-BH 3 ] (2a). Thermal ellipsoids are drawn at the 50% probability level. Hydrogen atom positions were idealized. Hydrogen atoms are depicted as spheres of arbitrary radius. Selected structural parameters: Bond distances (Å): N1-N2 1.354(4), N2-C1 1.332(5), C1-N3 1.324(4), N3-C2 1.334(4), C2-N1 1.348(4), C1-N4 1.451(5), N4-O1 1.226(4), N4-O2 1.231(4), N1-B1 1.591(5), C2-C30 1.483(5), C30-N5 1.521(4), C30-N6 1.526(4), C30- N7 1.540(5), N5-O3 1.213(4), N5-O4 1.221(4), N6-O5 1.217(5), N6-O6 1.208(4), N7-O7 1.218(4), N7- O8 1.220(4). N1-N2-C1 102.5(3), N2-C1-N3 117.8(3), C1-N3-C2 99.3(3), N3-C2-N1 113.8(3), C2-N1- N2 106.5(3), O1-N4-O2 125.4(3), C1-N4-O1 116.2(3), C1-N4-O2 118.4(3). Bond angles (°): O3-N5-O4 126.9(3), O5-N6-O6 128.1(3), O7-N7-O8 126.3(3), O3-N5-C30 114.3(3), O5-N6-C30 117.2(3), O8-N7- C30 116.7(3), N5-C30-N6 106.5(3), N6-C30-N7 104.5(3), N7-C30-N5 109.7(3), O2-N4-C1-N2 -6.5(5). 4a crystallizes in the orthorhombic space group P2 1 2 1 2 1 with four formula units per unit cell. In the solid- state structure, shown in Figure 4.6, the [FDNMeNTrz-BH 3 ] - anion is part of a positional disorder, where the anion is present in two slightly different orientations with a refined ratio of 0.6:0.4. As found for 2a, the substitution occurs at the nitrogen in α-position to the bulky group. 68 Figure 4.6: ORTEP plot the asymmetric unit of PPh 4 [FDNMeNTrz-BH 3 ] (4a) showing only one part of the disordered FDNMeNTrz anion. Thermal ellipsoids are drawn at the 50% probability level. Hydrogen atom positions were idealized. Hydrogen atoms are depicted as spheres of arbitrary radius. Selected structural parameters: Bond distances (Å):N1-N2 1.356(12), N2-C1 1.423(17), C1-N3 1.299(18), N3-C2 1.336(12), C2-N1 1.353(13), C1-N4 1.425(9), N4-O2 1.204(7), N4-O1 1.228(7), N1-B1 1.596(14), C2- C3 1.470(14), C3-F1 1.367(15), C3-N6 1.524(12), C3-N5 1.549(13), N5-O3 1.277(15), N5-O4 1.250(16), N6-O5 1.208(17), N6-O6 1.208(16). Bond angles (°): N1-N2-C1 102.5(9), N2-C1-N3 113.8(9), C1-N3-C2 103.2(10), N3-C2-N1 113.3(11), C2-N1-N2 107.1(10), O2-N4-O1 124.5(6), C1-N4- O2 116.8(6), C1-N4-O1 118.6(5), O3-N5-O4 128.4(11), O5-N6-O6 126.0(13), O3-N5-C3 110.9(11), O4- N5-C3 120.6(11), O5-N6-C3 119.2(12), O6-N6-C3 114.8(12), F1-C3-N5 107.4(12), F1-C3-N6 107.0(11), N5-C3-N6 104.3(8), O1-N4-C1-N2 7(1). 6a crystallizes in the triclinic space group P with two formula units per unit cell (Figure 4.7). The geometry of the N6 nitro group deviates significantly from the expected geometry, with the sum of the angles around the N6 atom being 362.95(18)°. The relative proximity of the nitro groups in neighboring [TNTz-BH 3 ] - anions (O6-O6 distance of 2.860(2) Å) and the instability of 6/7 suggests that such a distortion of the N6 nitro group might be an actual feature of the anion and not a disorder refinement 69 artefact. As found for trinitromethyltetrazole 23 and [BH 2 (TNTz) 2 ] - , 7 the N-substitution occurs at the β- position to the ring-carbon atom. Figure 4.7: ORTEP plot of the asymmetric unit of PPh 4 [TNTz-BH 3 ] (6a). Thermal ellipsoids are drawn at the 50% probability level. Hydrogen atom positions were idealized. Hydrogen atoms are depicted as spheres of arbitrary radius. Selected structural parameters: Bond distances (Å): C1-N4(1.325(2), N4- N3(1.3272(19), N3-N2(1.324(2), N2-N1(1.324(2), N1-C1(1.335(2), N3-B1(1.583(2), C1-C2(1.480(2), C2-N5(1.535(3), C2-N6(1.521(3), O1-N5(1.198(2), O2-N5(1.205(2), O3-N6(1.210(2), O4-N6(1.212(2), C2-N7(1.527(3), O5-N7(1.210(3), O6-N7(1.206(3). Shortest intermolecular distance : O6-O6 2.860(2) Å. Bond angles (°): C1-N4-N3(101.84(14), N4-N3-N2(112.25(13), N3-N2-N1(107.66(13), N2-N1- C1(104.41(14), N1-C1-N4(113.83(15), O1-N5-O2(127.23(19), O3-N6-O4(129.92(19), O2-N5- C2(113.68(18), O3-N6-C2(115.50(16), N5-C2-N6(108.38(15), O1-N5-C2(119.10(17), O4-N6- C2(117.53(19), N5-C2-C1(113.89(15), N6-C2-C1(109.30(15), O6-N7-C2(116.59(17), O5-N7- O6(127.5(2), N5-C2-N7(104.91(16), N6-C2-N7(107.32(16), O5-N7-C2(115.94(19), N7-C2- C1(112.73(15). 8a crystallizes in the triclinic space group P with two formula units per unit cell (Figure 4.8). The fluorodinitromethyl group is disordered and found in two slightly different orientations, related by a rotation along the C1-C2 axis, with a refined ratio of 0.6: 0.4. Substitution occurs again at the β-position to the ring-carbon atom. 70 Figure 4.8: ORTEP plot of the asymmetric unit of PPh 4 [FDNTz-BH 3 ] (8a) showing only one part of the disordered FDNTz anion. Thermal ellipsoids are drawn at the 50% probability level. The positions of the B-H hydrogen atoms were determined from the difference electron density map. All other hydrogen positions were idealized. The hydrogen atoms are depicted as spheres of arbitrary radius. Selected structural parameters: Bond distances (Å): C1-N4 1.329(4), N4-N3 1.318(4), N3-N2 1.333(4), N2-N1 1.324(5), N1-C1 1.330(5), B1-H1a 1.1586(fixed), B1-H1b 1.1547(fixed), B1-H1c 1.1709(fixed), N3-B1 1.579(5), C1-C2 1.470(5), C2-N6 1.537(9), C2-N5 1.594(8), O3-N6 1.212(9), O4-N6 1.262(10), O2-N5 1.231(8), O1-N5 1.225(9), C2-F1 1.374(6). Bond angles (°): C1-N4-N3 102.5(3), N4-N3-N2 111.8(3), N3-N2-N1 107.6(3), N2-N1-C1 104.7(3), N1-C1-N4 113.4(3), O3-N6-O4 128.5(8), O1-N5-O2 124.4(8), O3-N6-C2 114.9(8), O4-N6-C2 115.7(7), O2-N5-C2 115.7(5), O1-N5-C2 119.6(7), N6-C2-N5 100.1(5), N6-C2-C1 114.0(5), N5-C2-C1 117.4(4), F1-C2-C1 116.1(4), N6-C2-F1 106.3(5), N5-C2-F1 100.9(4). Tetraphenylphosphonium dinitro(1-H-tetrazol-5-yl)methanide monohydrate (12) crystallizes in the orthorhombic space group P2 1 2 1 2 1 with four formula units per unit cell (Figure 4.9). In the solid-state structure, the dinitro(1-H-tetrazol-5-yl)methanide anion is part of a positional disorder in which the two nitro groups are found in two different positions with a refined ratio of 0.54:0.46. The observed geometry 71 of the dinitromethanide moiety in the PPh 4 + salt is in good agreement with the one reported previously for other salts with this anion. 23-25 Figure 4.9: ORTEP plot of the asymmetric unit of tetraphenylphosphonium dinitro(1-H-tetrazol-5- yl)methanide monohydrate (12), showing only one part of the disordered dinitromethanide moiety. Thermal ellipsoids are drawn at the 50% probability level. Hydrogen atom positions were idealized. Hydrogen atoms are depicted as spheres of arbitrary radius. Selected structural parameters: Bond distances (Å): C1-N1 1.322(3), N1-N2 1.364(3), N2-N3 1.301(4), N3-N4 1.334(3), N4-C1 1.337(3), C1- C2 1.465(3), C2-N6 1.477(10), C2-N5 1.291(12), O1-N5 1.238(11), O2-N5 1.246(9), O3-N6 1.247(10), O4-N6 1.252(7), N4-H5N 0.9926(fixed). Bond angles (°): C1-N1-N2 104.9(2), N1-N2-N3 110.7(2), N2- N3-N4 106.9(2), N3-N4-C1 108.2(2), N4-C1-N1 109.3(2), O1-N5-O2 122.7(11), O1-N5-C2 125.2(7), O2-N5-C2 112.2(11), O3-N6-O4 119.3(9), O3-N6-C2 121.8(6), O4-N6-C2 118.9(9), N5-C2-N6 121.7(6), N5-C2-C1 125.7(5), N6-C2-C1 112.5(4), O2-N5-C2-C1 1(2), O4-N6-C2-C1 -5(1). 4.6 Vibrational Spectroscopy The vibrational spectra of the [Az-BH 3 ] - salts of this work are dominated by bands belonging to the organic cations as well as the nitro groups. In their Raman spectra, all of the [Az-BH 3 ] - compounds 72 displayed characteristic B-H stretching modes at ca 2380 and 2270 cm -1 . Instead of a single broad band at 2380 cm -1 , 1a and 2a display two sets of bands at 2382, 2363 and 2394, 2359 cm -1 , respectively. Samples of 11a display three bands at 2407, 2386 and 2371cm -1 in addition to a weaker band at 2266 cm -1 . Mixtures containing 6,7 showed an additional relatively strong band at 2451 cm -1 . The presence of such band correlates with the presence of major decomposition products at -20 ppm by 11 B NMR and around - 32 ppm by 14 N NMR. In their IR spectra, all of the compounds display bands centered at ca 2370 cm -1 , 2269 and 2314 (weak) cm -1 . Most of the bands observed at 2370 cm -1 are relatively broad, while for 2a four bands can be distinguished at 2392, 2371, 2360, 2350 cm -1 . Compound 11a displays three bands in that immediate region at 2406, 2389, 2363 cm -1 . The additional band at 2460 observed by Raman spectroscopy for 6,7 is also observed in the IR spectrum. 4.7 NMR Spectroscopy The 1 H NMR resonances for all of the BH 3 groups in [Az-BH 3 ] - were observed as broad quartets between 2.0 and 2.4 ppm in acetonitrile solution. These values are close to those observed for borane complexes with relatively strong Lewis bases, such as dimethylsulfide. For 11a, the resonance of the two BH 3 groups in [FDNMeTz(BH 3 ) 2 ] - was observed at 1.67 ppm in dichloromethane. This is at significantly higher field than that for the [Az-BH 3 ] - anions. Because of the broadness of the signals and their overlap with the solvent or impurities, it was not possible to distinguish between the different isomers or between the mono- and disubstituted species when more than one species was present in solution. The 11 B chemical shifts of the [Az-BH 3 ] - anions 1-9 were found between -17 and -20 ppm. The substitution of the azole rings with nitro-, fluorodinitro- or trinitromethyl groups have remarkably little effect on the 11 B chemical shift in [Az-BH 3 ]. Indeed, one isomer for each of the [Az-BH 3 ] complex displays one 11 B NMR shift at -17.3±0.1 ppm. The substitution position of the BH 3 group on the ring, however, has a greater influence. The 1-boranyl isomer can be found over 2 ppm upfield from the 2- boranyl isomer in the case of [Tz-BH 3 ] - , while the isomer with the boranyl substituted in β to the bulkiest substituent has a chemical shift ca 1 ppm downfield from that of the isomer substituted in α to the bulkiest group for [Trz-BH 3 ] - . This effect is also observed for the two non-equivalent boron atoms in [Tz(BH 3 ) 2 ] - . Indeed, the two 11 B resonances for each of the anions 10 and 11 were found at ca -16 and -20 ppm. All of these resonances are in the range expected for MeNH 2 . BH 3 26 but are significantly de-shielded in comparison to predicted values for tetrazolyl-BH 3 complexes, 19 in agreement with the "electron-poor" character of the azoles compounds of this study. 73 The 14 N NMR resonances of the nitro groups for compounds 1-11 were found between -20 and -37 ppp and are always at higher field than those observed for the corresponding uncoordinated azolate anion. The 14 N NMR resonances of the fluorodinitromethyl groups in 4, 8, 9 and 11 were observed as doublets with 2 J( 14 N- 19 F) ranging between 8-10 Hz due to coupling to the neighboring fluorine atom. The 14 N NMR signals for the nitro groups in the [Tz(BH 3 ) 2 ] - anions are at higher field than the ones observed for [Tz- BH 3 ] - . Due to their different chemical environments, two signals are expected for the two nitro groups in 1, 2, and 3. In the case of 1, the two non-equivalent nitro groups could not be readily distinguished from the relatively broad signal observed at -26 ppm. For 2, the resonance for the ring-nitro group was found at -26.2, very close to that of 1, and a resonance at -34.5 ppm was observed for the trinitromethyl group. As expected, the 14 N NMR resonances of the nitrogen atoms of the azole ring are very broad, making them much more difficult to observe and the expected number of signals could not be observed in all cases. The coordination of BH 3 has only a modest effect on the 13 C chemical shifts of the carbon atom of the five-membered ring as well as the fluorodinitro- and trinitromethyl groups. In 8a, 9a, 11a, the 1 J( 13 C- 19 F) (280-293 Hz) and 2 J( 13 C- 19 F) (24-28 Hz) couplings between the carbon atoms of FDNTz and the neighboring fluorine atom could be observed. The coordination of BH 3 to FDNMeNTrz and FDNTz has also only a modest effect on the 19 F chemical shift of the ligand. The 19 F NMR resonances for the fluorodinitromethyl group in 4, 5, 8, 9 and 11 were found in the range between -93.0 and -97.3 ppm, which is expected for fluorodinitromethyl compounds. 22 These resonances were typically observed as poorly resolved broad multiplets expected from the 2 J coupling with the two neighboring nitro groups. 4.8 Conclusion The formation of the complex anions dinitrotriazolato-trihydroborate, (trinitromethyl)nitrotriazolato- trihydroborate, (fluorodinitromethyl)nitrotriazolato-trihydroborate, (trinitromethyl)tetrazolato- trihydroborate, (fluorodinitromethyl)tetrazolato-trihydroborate and bis(boranyl)(fluorodinitromethyl)tetrazolate was studied by multinuclear NMR spectroscopy. While in the case of dinitrotriazolato-trihydroborate the formation of only one isomer was observed, two isomers were observed for all the other azolate anions. With the exception of the trinitromethyltetrazolate derivatives, it was possible to isolate a salt of one isomer of each complex anion in good yields and purities. The isolated compounds were fully characterized and the structures of the major isomers were confirmed by X-ray crystallography. Most of the compounds are reasonably stable at room temperature, while the 74 trinitromethyltetrazolate derivatives were found to be only marginally stable in the same conditions. The observed formation of dinitro(1-H-tetrazol-5-yl)methanide from (trinitromethyl)tetrazolato-trihydroborate suggests a possible decomposition mechanism in which the trinitromethyl moiety is reduced by BH 3 . All of the salts of the BH 3 -containing complex anions were less thermally stable than the corresponding azolate salts. Further experimental details, crystal packing diagrams and crystallographic information can be found in APPENDIX 3. 4.9 References (1) Talawar, M. B.; Sivabalan, R.; Mukundan, T.; Muthurajan, H.; Sikder, A. K.; Gandhe, B. R.; Rao, A. S. J. Hazard. Mater. 2009, 161, 589. (2) Titova, K. V.; Kolmakova, E. I.; Rosolovskii, V. Y. Bull. Acad. Sci. USSR, Div. Chem. Sci. (Engl. Transl.) 1977, 26, 1822. (3) Titova, K. V. Russ. J. Inorg. Chem. 2002, 47, 1121. (4) Ioffe, S. L. S., O. P.; Tartakowskij, V. A. "The Unusual Reation of Trinitromethane with BH 3 . THF- complex" IUPAC IMEBORON VIII, Knoxville, Tennessee, July 11-15, 1993. (5) Guibert, C. R.; Marshall, M. D. J. Am. Chem. Soc. 1966, 88, 189. (6) Bélanger-Chabot, G.; Rahm, M.; Haiges, R.; Christe, K. O. Angew. Chem. Int. Ed. 2013, 52, 11002. (7) Haiges, R.; Jones, C. B.; Christe, K. O. Inorg. Chem. 2013, 52, 5551. (8) Klapötke, T. M.; Rusan, M.; Sproll, V. Z. Anorg. Allg. Chem. 2014, 640, 1892. (9) Snyder, C. J.; Martin, P. D.; Heeg, M. J.; Winter, C. H. Chem. Eur. J. 2013, 19, 3306. (10) Janiak, C. J. Chem. Soc., Chem. Commun. 1994, 545. (11) Janiak, C.; Scharmann, T. G.; Green, J. C.; Parkin, R. P. G.; Kolm, M. J.; Riedel, E.; Mickler, W.; Elguero, J.; Charamunt, R. M.; Sanz, D. Chem. - Eur. J. 1996, 2, 992. (12) Lu, D.; Winter, C. H. Inorg. Chem. 2010, 49, 5795. (13) Groshens, T. J. J. Coord. Chem. 2010, 63, 1882. (14) Wang, Y.-L.; Cao, R.; Bi, W.-H. Polyhedron 2005, 24, 585. (15) Koch, E.-C.; Klapötke, T. M. Propellants, Explos., Pyrotech. 2012, 37, 335. (16) Klapötke, T. M.; Krumm, B.; Moll, R. Chem. - Eur. J. 2013, 19, 12113. (17) Klapötke, T. M.; Krumm, B.; Rusan, M.; Sabatini, J. J. Chem. Commun. 2014, 50, 9581. (18) Klapötke, T. M.; Rusan, M.; Sproll, V. Z. Anorg. Allg. Chem. 2013, 639, 2433. (19) Vasiliu, M.; Arduengo, A. J.; Dixon, D. A. J. Phys. Chem. C 2012, 116, 22196. (20) Kofman, T. P. Russ. J. Org. Chem. 2002, 38, 1231. 75 (21) Haiges, R.; Bélanger-Chabot, G.; Kaplan, S. M.; Christe, K. O. Dalton Trans. 2015, 44, 2978. (22) Haiges, R.; Christe, K. O. Dalton Trans. 2015, 44, 10166. (23) Haiges, R.; Christe, K. O. Inorg. Chem. 2013, 52, 7249. (24) Goiko, E. A.; Grigor’eva, N. V.; Margolis, N. V.; Mel’nikov, A. A.; Strochkina, T. K.; Tselinskii, I. V. Zh. Strukt. Khim. 1980, 21, 177−180. (25) Klapötke, T. M.; Steemann, F. X. Propellants, Explos., Pyrotech. 2010, 35, 114. (26) Nöth, H.; Vahrenkamp, H. Chem. Ber. 1966, 99, 1049. 76 CHAPTER 5: (TRINITROMETHYL)BORATE COMPLEX ANIONS 5.1 Introduction Ever since the first report on the trinitromethanide, or "nitroformate" anion, -C(NO 2 ) 3 - , 1 trinitromethyl derivatives have been a subject of significant interest. 2-10 The varied reactivity of the nitroformate anion and of the parent molecule, nitroform (HC(NO 2 ) 3 ) 11-14 and the high oxygen balance of the trinitromethyl moiety make these types of compounds attractive candidates for green high-energy density material (HEDM) applications. A major challenge in trinitromethyl chemistry is the limited thermal stability of many of the compounds. 6,15,16 In view of the relatively high average bond energy of B-C bonds (ca. 89 kcal/mol), 17 (trinitromethyl)borates can be expected to be stable compounds. In addition, the low atomic weight of boron, the low toxicity and the high formation enthalpy of B 2 O 3(g) (-200 kcal/mol) 18 would make (trinitromethyl)borates interesting prospects for green HEDMs. 19-25 The goal of this study was to explore methods for the stabilization of trinitromethyl substituted boron compounds. Surprisingly, only one previous report was found on the synthesis of Group 13-trinitromethyl compounds. Titova et al. studied the reaction of BCl 3 with [NR 4 ][C(NO 2 ) 3 ] salts and [NR 4 ][BCl 4 ] salts with Ag[C(NO 2 ) 3 ]. 15,26 The authors characterized salts of the postulated [BCl 3 C(NO 2 ) 3 ] - anion and observed its decomposition at -20 °C to BCl 3 and [C(NO 2 ) 3 ] - and at 20 °C, to BCl 4 - and N 2 O 4 . At lower temperatures, a marginally stable solid was obtained and studied by IR spectroscopy. 15 Since highly Lewis acidic fragments, such as BCl 3 , are known to induce the decomposition of oxygen rich ligands, 27,28 it was interesting to study whether the use of less Lewis acidic groups, 29 such as BH 3 , might provide kinetically more stable (trinitromethyl)borates. In this chapter, the synthesis and structural characterization of [BH 3 C(NO 2 ) 3 ] - , the first stable (trinitromethyl)borate anion, and the detection of [BH 2 (C(NO 2 ) 3 ) 2 ] - in solution are reported. 5.2 Detection of (trinitromethyl)borate Anions In solution, the [BH 3 C(NO 2 ) 3 ] - anion is formed by the elimination of H 2 from the reaction between the acidic proton of nitroform (pKa = 0.17 at 20 °C in water) 30 and a hydrido ligand of BH 4 - (Equation 5.1). [BH 4 ] - + HC(NO 2 ) 3 → [BH 3 C(NO 2 ) 3 ] - + H 2 (5.1) Figure 5.1 summarizes the reaction conditions for the formation of the various (trinitromethyl)borate salts and hydrolysis products. The best conditions for the synthesis of the (trinitromethyl)borate anion were the 77 use glyme (1,2-dimethoxyethane) as solvent and of a 10 mol% excess of nitroform. The multinuclear NMR spectra of the [BH 3 C(NO 2 ) 3 ] - anion are given in Figure 5.2. Figure 5.1: Synthesis of [BH 3 C(NO 2 ) 3 ] - salts (PPN = [Ph 3 P=N=PPh 3 ] + ). Although the monosubstituted anion, [BH 3 C(NO 2 ) 3 ] - , was the main product, even when a fourfold excess of nitroform was used in the reaction, small amounts of the disubstituted anion, [BH 2 (C(NO 2 ) 3 ) 2 ] - , were also observed by 11 B and 14 N NMR spectroscopy (see APPENDIX 4) after one to two days. The formation of [BH 2 (C(NO 2 ) 3 ) 2 ] - can be suppressed by extraction of the excess nitroform shortly after [BH 3 C(NO 2 ) 3 ] - has been formed. The slower formation of [BH 2 (C(NO 2 ) 3 ) 2 ] - indicates that the strong electron-withdrawing effect of the nitroformate ligand decreases the hydridic character and reactivity of the remaining H-substituents in [BH 3 C(NO 2 ) 3 ] - . However, these H-substituents still possess a hydridic character, as demonstrated by the formation of [C(NO 2 ) 3 ] - and boric acid upon exposure of [BH 3 C(NO 2 ) 3 ] - to water (Figure 5.1). 78 Figure 5.2: Multinuclear NMR spectra of [BH 3 C(NO 2 ) 3 ]- measured in CD 3 CN for 1 H and 11 B and in glyme for 13 C. 5.3 Isolation and Properties The sodium salt of [BH 3 C(NO 2 ) 3 ] - , obtained as a green-yellow powder, contained two glyme molecules per sodium atom, which could only be removed by pumping at room temperature. The removal of glyme from [Na(glyme) 2 ][BH 3 C(NO 2 ) 3 ] favored the dissociation of the BH 3 -nitroformate adduct and generated significant amounts of sodium nitroformate. Since no crystals of the sodium salt of [BH 3 C(NO 2 ) 3 ] - could be obtained, the PPh 4 + and PPN + salts of [BH 3 C(NO 2 ) 3 ] - were prepared by a metathetical reaction of glyme/dichloromethane solutions of Na[BH 3 C(NO 2 ) 3 ] with PPh 4 Cl and PPNCl, respectively (Figure 5.1). The compounds were obtained as neat, pale yellow and orange-yellow solids, respectively, by pumping off the solvent for 12 to 48 h between -40 and 0 °C. Crystalline materials were obtained from dichloromethane/pentane solutions, but only the PPN salt yielded crytals suitable for X-ray crystal structure determination. All solids were characterized by IR and Raman spectroscopy, as well as by solution NMR. The observed spectra agree well with those predicted by DFT calculations (see APPENDIX 4). 79 5.3.1 Structural Characterization PPN[BH 3 C(NO 2 ) 3 ] crystallizes in the monoclinic space group P2 1 /n with two symmetry-independent formula units in the unit cell. One of the two [BH 3 C(NO 2 ) 3 ] - anions is shown in Figure 5.3. The anions display a pseudo-tetrahedral arrangement of a -BH 3 and three nitro groups around the carbon atom. The nitro groups are arranged in a propeller-like fashion, as has been observed for trinitroethanol 31 and nitroform. 32 The observed geometries and conformations of the NO 2 groups in [BH 3 C(NO 2 ) 3 ] - are in excellent agreement with those predicted by M06-2X/aug-cc-pVTZ calculations (Figure 5.3, Table 5.1). Only B-C connectivity is observed in [BH 3 C(NO 2 ) 3 ] - , in contrast to some organic nitroformates where both -C(NO 2 ) 3 and -O-N(O)-C(NO 2 ) 3 linkages have been observed. 33 Figure 5.3: ORTEP plot for one of the two symmetry-independent anions in the asymmetric unit of PPN[BH 3 C(NO 2 ) 3 ]. Thermal ellipsoids are shown at the 50% probability level (hydrogen atom positions are idealized). Selected structural parameters with M06-2X/aug-cc-pVTZ calculated values in parentheses: Bond distances (Å): B1a-C1a 1.629(3) (B-C 1.645); C1a-N3a 1.513(2); C1a-N1a 1.517(2); C1a-N2a 1.527(2) (C-N 1.523); N1a-O1a 1.215(3); N1a-O2a 1.220(3); N2a-O3a 1.223(2); N2a-O4a 1.211(2); N3a-O5a 1.225(2); N3a-O6a 1.215(2) (N-O 1.201; 1.210). Bond angles (°). O1a-N1a-O2a 125.1(2); O6a-N3a-O5a 125.82(18); O4a-N2a-O3a 124.98(19) (O-N-O 125.75). A comparison of selected averaged structural parameters between the [BH 3 C(NO 2 ) 3 ] - anions in PPN[BH 3 C(NO 2 ) 3 ] and several ionic and covalent trinitromethanide derivatives is given in Table 5.1. It is interesting to note that the average ONO angle for every compound in Table 5.1 deviates from the ideal 120° angle (ONO angles in PPN[BH 3 C(NO 2 ) 3 ] : 125.1(2)°; 125.82(18)°; 124.98(19)° in one anion; 126.05(19)°; 124.8(2)°; 124.5(2)° in the other anion). Within each nitro group of each compound, the two CNO angles vary significantly, except in one nitro group (119.96(18)°, 114.92(18)°; 116.21(16)°, 80 117.85(16)°; 116.64(18)°, 118.37(17)°; 116.46(16)°, 117.48(18)°; 116.6(2)°, 118.6(2)°; 116.15(17)°, 119.31(18)° in PPN[BH 3 C(NO 2 ) 3 ]). Since the sum of ONO and CNO angles for each nitro group is always 360°, within the experimental errors (360.0 (6)°, 359.99 (54)°, 359.88 (50)°, 360.0 (6)°, 360.0 (6)°, 359.99 (53)° in PPN[BH 3 C(NO 2 ) 3 ]), this implies that the nitro groups are always tilted while remaining planar. The smallest CNO angle for each nitro group is the one corresponding to the oxygen atom that is the closest to the X group (-BH 3 , -CH 2 OH, -H). In each [BH 3 C(NO 2 ) 3 ] - anion, at least one NCB angle is significantly different from the other two (NCB angles : 111.52(15)°, 114.43(16)°, 114.93(16)°; 114.88(19)°, 112.09(16)°, 111.88(17)°). The NCN angles in PPN[BH 3 C(NO 2 ) 3 ] are smaller than the ideal tetrahedral angle (106.71(15)°, 105.23(14)°, 103.14(14)°, 106.52(16)°, 104.12(16)°, 106.69(15)°). These observations suggest varying degrees of steric interaction between the groups surrounding the carbon atom. The average ONO and CNO angles in PPN[BH 3 C(NO 2 ) 3 ] are between those for the covalent compounds and those for the ionic trinitromethanide compounds (potassium nitroformate 10 in the monoclinic structure and [PMePh 3 ]C(NO 2 ) 3 34 selected as examples). This implies an intermediate tilting of the NO 2 group towards X between that of neutral covalent trinitromethyl compounds and ionic trinitromethyl salts. The average NCN angle in PPN[BH 3 C(NO 2 ) 3 ] is smaller than the average NCN angle of nitroform and trinitroethanol. This is reflected, expectedly, by a larger average NCX angle in [BH 3 C(NO 2 ) 3 ] - . Table 5.1: Comparison of selected averaged structural parameters of PPN[BH 3 C(NO 2 ) 3 ] and several covalent and ionic trinitromethanide derivatives. PPN[BH 3C(NO 2) 3] [BH 3C(NO 2) 3] - Calculated a HC(NO 2) 3 32 (NO 2) 3CCH 2OH 31 KC(NO 2) 3 (P2 1/n) 10 [PPh 3Me]C(NO 2) 3 34 N-O (Å) 1.215(2) 1.206 1.206 1.2094 1.227 1.204 N-C (Å) 1.520(2) 1.523 1.505 1.5197 1.388 1.397 ONO (°) 125.21(19) 125.8 127.2 127.4 121.77 122.40 CNO (°) 117.38(18) 117.1 116.4 116.29 119.11 118.30 NCX (°) 113.29(17) 113.6 110.4 112.04 - - NCN (°) 105.40(15) 105.1 108.6 106.24 119.92 120.0 a M06-2X/aug-cc-pVTZ gas-phase geometry of [BH 3 C(NO 2 ) 3 ] - . 81 5.3.2 NMR Spectroscopy Figure 5.2 shows the multiplicity of the NMR signals observed for [BH 3 C(NO 2 ) 3 ] - salts. The 11 B- 1 H couplings (94 Hz) observed in 1 H and 11 B spectra and the signal multiplicity is consistent with a BH 3 moiety. The coupling observed in 13 C NMR is consistent with a boron-carbon direct bond in contrast to the possible [H 3 B-O-N(O)C(NO 2 ) 2 ] - . The 1 H NMR signal at δ 1.09 ppm in CD 3 CN is also consistent with an anionic BH 3 adduct (the signal is at higher field and sharper than the BH 3 -CD 3 CN adduct). Moreover, the 11 B shift signal (δ -23.8 ppm) in glyme is at much higher field and is significantly sharper than the one expected for a BH 3 . glyme adduct (around δ 2 ppm). 35 Additionally, its chemical shift is in a range expected for XBH 3 - species. 36 These observations indicate that no observable dissociation equilibrium exists at room temperature for the [BH 3 C(NO 2 ) 3 ] - anion. The observed 14 N NMR shift for the nitro groups of [BH 3 C(NO 2 ) 3 ] - (δ -7.2 ppm) is in excellent agreement with the predicted value (δ -6 ppm) and is at significantly lower field compared to C(NO 2 ) 3 - (exp. δ -28.6 ppm in CD 3 CN, predicted δ -32.8 ppm) and nitroform (δ -34 ppm in CD 3 CN). This implies that the nitro groups in the adduct reside in a significantly different chemical environment compared to the nitroformate anion or to nitroform, as is expected for a nitroformate adduct with a relatively strong Lewis acid. In addition, the 14 N NMR signal for the nitro groups in [BH 3 C(NO 2 ) 3 ] - is significantly broader (τ 1/2 = 14 Hz) than for nitroform (τ 1/2 = 7 Hz) CD 3 CN), which is likely due to the proximity of the 11 B and 10 B NMR-active quadrupolar nuclei. 37 The 13 C NMR chemical shift for [BH 3 C(NO 2 ) 3 ] - (δ 151.1 ppm in glyme) is at slightly lower field than for C(NO 2 ) 3 - (δ 147.2 ppm in acetonitrile), suggesting a lower electron density due to electron donation into the formally empty p z orbital of boron in BH 3 . The broadness of the 13 C signal is likely in part due to the proximity of four quadrupolar nuclei. 38,39 5.3.3 Vibrational Spectroscopy Quantum chemical calculations for the [BH 3 C(NO 2 ) 3 ] - anion agree well with the experimental spectra and helped in the assignment of the vibrational modes. Raman intensities are notoriously difficult to use as a quantifying tool, 40 which might explains why very small amounts of nitroformate, as determined by 14 N NMR, produce relatively intense bands in the Raman spectra of [BH 3 C(NO 2 ) 3 ] - salts. Comparisons between selected observed and predicted vibrational frequencies of the [BH 3 C(NO 2 ) 3 ] - salts and the ones for the nitroformate anion can be found in APPENDIX 4. Several vibrational bands display a shift towards higher frequencies from anionic C(NO 2 ) 3 - (with NaC(NO 2 ) 3 taken as a reference) 41 to covalently bonded nitroformate in [BH 3 C(NO 2 ) 3 ] - , as noted for [BCl 3 C(NO 2 ) 3 ] - in the report by Titova et al. 15 This is consistent with the decreased negative charge in the nitroformate moiety upon coplexation with the BH 3 82 group. It is interesting to note that the NO 2 symmetric in-phase stretching mode has roughly the same frequency in both the anion and [BH 3 C(NO 2 ) 3 ] - anions or even in the neutral covalent nitroform molecule. 42 The Raman spectra of [Na(glyme) 2 ][BCl 3 C(NO 2 ) 3 ], PPh 4 [BCl 3 C(NO 2 ) 3 ] and PPN[BCl 3 C(NO 2 ) 3 ] are practically identical with respect to the anion bands. They are also very similar to the glyme solution spectra of Na[BCl 3 C(NO 2 ) 3 ], suggesting relatively weak cation/anion interactions both in solution (in coordinating solvents) and in the solid state for the isolated compounds. 5.3.4 UV-Vis Spectroscopy UV-Vis spectra of a glyme solution of Na[BH 3 C(NO 2 ) 3 ] showed three absorptions at 257, 291 and 353 nm. Spiking that solution with sodium nitroformate resulted in a strong enhancement of the 353 nm absorption and a small variation in the other two. This suggests that small amounts of nitroformate impurities could contribute to the 353 nm absorbtion observed and to the color of the solid [BH 3 C(NO 2 ) 3 ] - salts isolated. This was also demonstrated by the fact that crystals of pure PPN[BH 3 C(NO 2 ) 3 ] are colorless. The 353 nm band has been interpreted as being caused by a n-π* transition in the C(NO 2 ) 3 - anion. 39 5.3.5 Sensitivity and Stability None of the salts, PPh 4 [BH 3 C(NO 2 ) 3 ], PPN[BH 3 C(NO 2 ) 3 ], or [Na(glyme) 2 ][BH 3 C(NO 2 ) 3 ] could be detonated by hitting them with a hammer or by vigorously scraping them with a metal spatula, while exposure to the flame of a Bunsen burner resulted in deflagrations. In an inert atmosphere, salts of [BH 3 C(NO 2 ) 3 ] - can be stored for months without any signs of decomposition when kept at -20 °C, both in the solid state and in solution. At room temperature, however, a slow decomposition of the anion under formation of nitroformate was observed for the solid PPN + and PPh 4 + salts. However, this decomposition was so slow that, even after several months, [BH 3 C(NO 2 ) 3 ] - was still the major species. The isolated [Na(glyme) 2 ][BH 3 C(NO 2 ) 3 ] salt was much less stable and completely decomposed within a week at room temperature. Its instability at ambient temperature might be due to an inherent loss of glyme resulting in the formation of the naked sodium [BH 3 C(NO 2 ) 3 ] - salt which, by analogy with naked sodium nitroformate, might possess only marginal stability at room temperature. 16 The lower stability of alkali metal salts compared to salts containing bulky cations, such as PPh 4 + and PPN + , is common and has been observed by Christe et al. for many polyazido salts. 43,44 83 The very slow decomposition of solid PPh 4 [BH 3 C(NO 2 ) 3 ] and PPN[BH 3 C(NO 2 ) 3 ] at room temperature contrasts with the DTA results, which show that the compounds seem to be thermally stable up to ca. 120 °C. This apparent difference might be due to the very fast heating rates typically used for DTA measurements and/or to potential photolytic instability of the complex anion, which exhibits absorptions between 200 and 400 nm in the UV/Vis spectrum. 5.4 Theoretical Insight The successful synthesis of [BH 3 C(NO 2 ) 3 ] - as room temperature stable PPh 4 [BH 3 C(NO 2 ) 3 ] and PPN[BH 3 C(NO 2 ) 3 ] salts contrasts with the reported instability of [NR 4 ][BCl 3 C(NO 2 ) 3 ]. 15 Consequently, quantum mechanical (QM) calculations using basis set extrapolated CCSD(T) and M06-2X DFT methods were carried out to explore the relative strengths of the C-BH 3 and C-BCl 3 bonds and analyze possible decomposition pathways. Theoretical predictions agree with the observed significantly higher stability of [BH 3 C(NO 2 ) 3 ] - relative to that of [BCl 3 C(NO 2 ) 3 ] - with respect to dissociation (Figure 5.4). The adiabatic enthalpies of dissociation in acetonitrile solution of [BCl 3 C(NO 2 ) 3 ] - (25.5 kcal/mol) and [BH 3 C(NO 2 ) 3 ] - (35.0 kcal/mol) differ by 10 kcal/mol. Assuming these values to represent the decomposition barriers and assuming first-order decomposition kinetics, a difference of 10 kcal/mol corresponds to a 9 million fold rate difference at ambient conditions. This is in good agreement with the experimental observations, which show significant dissociation of [BCl 3 C(NO 2 ) 3 ] - already at -20 °C 15 while, based on DTA results, [BH 3 C(NO 2 ) 3 ] - is stable up to about 120 °C. 84 Figure 5.4: (Top) Relative energies of C(NO 2 ) 3 - , [BH 3 C(NO 2 ) 3 ] - , [H 3 BO-N(O)-C(NO 2 )] - and [BCl 3 C(NO 2 ) 3 ] - in the gas phase (black) and in acetonitrile solution (1M, red, within parenthesis) shown in kcal/mol. (Bottom) Electrostatic surface potentials (ESPs) plotted on constant 0.001 e/bohr 3 electron density isosurfaces, derived from M06-2X/aug-cc-pVTZ wave functions in vacuum. The energy of the [H 3 BO-N(O)-C(NO 2 ) 2 ] - isomer (3) is predicted to be roughly 10 kcal/mol higher than that of [BH 3 C(NO 2 ) 3 ] - (Figure 5.4). This explains why, contrary to some organic derivatives of nitroformates, 33 no mixture of isomers is observed. The relatively low adiabatic dissociation enthalpy of 25.5 kcal/mol of [BCl 3 C(NO 2 ) 3 ] - explains its dissociation even at low temperature. It should be understood that this value does not imply that the B-C bond strength in this anion has such a low value, because B-C bonds generally have considerably larger values than 20 to 30 kcal/mol. 17 The reason for this low value is the fact that adiabatic dissociation energies can contain very large contributions from reorganization energies of the products. Reorganization energies have negative signs 45 and, therefore, decrease the value of the dissociation energy. In the case of the dissociation of the [BX 3 C(NO 2 ) 3 ] - anions, the structures of the two fragments, BX 3 and C(NO 2 ) 3 - , are reorganized from tetrahedral to planar. A second interesting aspect is the fact that the B-C bond dissociation energy in [BCl 3 C(NO 2 ) 3 ] - is about 10 kcal/mol lower than that in [BH 3 C(NO 2 ) 3 ] - . This is surprising because [BX 3 C(NO 2 ) 3 ] - can be considered as a Lewis acid/Lewis base adduct between the Lewis acid BX 3 and the Lewis base C(NO 2 ) 3 - . Therefore, based on Lewis acidity considerations alone, the stronger Lewis acid BCl 3 should form a stronger adduct with the C(NO 2 ) 3 - anion than BH 3 . 29 The reversal of this stability order can be partially attributed to the difference in the reorganization energies of BCl 3 and BH 3 and to steric factors. With a 85 bulky ligand, such as C(NO 2 ) 3 - , the bulkier BCl 3 group will experience considerably more ligand/ligand repulsion than the much smaller BH 3 group. [BCl 3 C(NO 2 ) 3 ] - was also reported to decompose to BCl 4 - and N 2 O 4 at 20 °C, 15 implying that dissociation alone might not fully account for the instability of that compound. An additional decomposition pathway involves a rapid ligand exchange reaction in [BCl 3 C(NO 2 ) 3 ] - to form BCl 4 - and the less stable [B(C(NO 2 ) 3 ) 4 ] - anion (Equation 5.2) 15,28 Preliminary results obtained by Titova et al. indicate that the subsequent decomposition of [B(C(NO 2 ) 3 ) 4 ] - is almost instantaneous at -30 °C. 15 4M[BXCl 3 ] → 3M[BCl 4 ] + M[BX 4 ] (5.2) Following that reasoning, the higher adiabatic B-C bond dissociation energy in [BH 3 C(NO 2 ) 3 ] - might make the cleavage of the B-C bonds more difficult and render the formation and subsequent decomposition of [B(C(NO 2 ) 3 ) 4 ] - less likely. This would also explain why no N 2 O 4 evolution nor any other sign of decomposition of the trinitromethyl group in the PPN + and PPh 4 + salts of [BH 3 C(NO 2 ) 3 ] - could be observed. The variation of the electrostatic potential over the molecular surface (ESP) relates to intermolecular interactions and is known to correlate with several macroscopic properties, such as lattice enthalpy, heat of sublimation, and boiling point. 46,47 Because the BH 3 and BCl 3 moieties in [BH 3 C(NO 2 ) 3 ] - and [BCl 3 C(NO 2 ) 3 ] - exhibit markedly different electrostatic features (Figure 5.4), they will affect overall macroscopic properties and influence the stability of the bulk material differently. For instance, visual inspection of the ESP of [BCl 3 C(NO 2 ) 3 ] - shows less variation and less negative potentials, compared to those of [BH 3 C(NO 2 ) 3 ] - and [C(NO 2 ) 3 ] - . In the solid state, this should decrease the heat of sublimation of the former, relative to the latter two, 48 and as a consequence lower the relative stability of the [BCl 3 C(NO 2 ) 3 ] - salt. The search for potential decomposition pathways indicated that the most likely route is the heterolytic B-C bond dissociation (see APPENDIX 4). Because the largest part of the entropic gain associated with such a process can be realized only after the barrier is passed, the relative enthalpy of the infinitely separated fragments represents an upper boundary to the free energy barrier (~35 kcal/mol in the case of [BH 3 C(NO 2 ) 3 ] - ). The only process found with a lower barrier than the direct dissociative pathway is TS1, which corresponds to the isomerization of [BH 3 C(NO 2 ) 3 ] - to [H 3 BO-N(O)-C(NO 2 ) 2 ] - with a free energy barrier of 29.2 kcal/mol (Figure 5.5). In TS1, the BH 3 group is almost completely separated from the nitroformate group with a C-B distance of >2.8Å. This barrier could possibly be further reduced if the leaving BH 3 molecule experienced specific interactions with a surrounding 86 environment. Such interactions are not treated in the implicit solvation model employed in this study. The experimental results support a dissociation-like decomposition pathway. Subjecting [Na(glyme) 2 ][BH 3 C(NO 2 ) 3 ] to a dynamic vacuum for a day at room temperature allowed the detection of B 2 H 6 and nitroformate as decomposition products. DTA measurements indicate that no significant decomposition occurs below 120 °C for PPN + and PPh 4 + [BH 3 C(NO 2 ) 3 ] - salts (see APPENDIX 4). This is in agreement with a barrier of about 29 kcal/mol, which, assuming first-order kinetics, would suggest a half-life of several years at room temperature but only of a few minutes at 120 °C. Figure 5.5: The intramolecular isomerization of [BH 3 C(NO 2 ) 3 ] - to [H 3 BO-N(O)-C(NO 2 ) 2 ] - offers an approximate free energy estimate of the dissociation barrier toward BH 3 elimination. Relative energies are shown in kcal/mol and bond lengths in Ångström (Å). Values for the process in acetonitrile solution (implicitly treated) are shown in parentheses. 5.5 Conclusion In conclusion, the BH 3 group in [BH 3 C(NO 2 ) 3 ] - provides a significant increase in stability compared to the previously studied BCl 3 -analog. 15 The [Na(glyme) 2 ] + , PPN + and PPh 4 + salts of [BH 3 C(NO 2 ) 3 ] - were successfully isolated and characterized by NMR and vibrational spectroscopy. The PPN + salt was structurally characterized, confirming the theoretically predicted B-C connectivity. In addition, evidence was obtained for the slower formation of the disubstituted [BH 2 (C(NO 2 ) 3 ) 2 ] - anion. While the dissociation barrier of [BH 3 C(NO 2 ) 3 ] - is too low for practical applications, it was demonstrated for the first time, both 87 experimentally and theoretically, that kinetically stable boron-trinitromethyl compounds can exist at room temperature. The most remarkable feature of the [BH 3 C(NO 2 ) 3 ] - salts is the coexistence of a strongly oxidizing trinitromethyl and a strongly reducing BH 3 moiety in a single complex ion, a marriage between fire and water. Complete experimental details, further data analysis, vibrational frequency comparisons, complete decomposition pathway analysis and crystallographic information can be found in APPENDIX 4. 5.6 References (1) Hantzsch, A.; Rinckenberger, A. Chem. Ber. 1899, 32, 628. (2) Novikov, S. S. G., T.I.; Tartakovskii, V.A., Izv. Akad. Nauk SSSR, Otd. Khim. Nauk 1960, 863. (3) Ioffe, S. L.; Mkarenkova, L. M.; Shitkin, V. M.; Kashutina, M. V.; Tartakovskii, V. A. Izv. Akad. Nauk SSSR, Ser. Khim. 1973, 1, 203. (4) Erashko, V. I.; Shevelev, S. A. Izv. Akad. Nauk SSSR, Ser. Khim. 1985, 2, 439. (5) Thottempudi, V.; Kim, T. K.; Chung, K.-H.; Kim, J. S. Bull. Korean Chem. Soc. 2009, 30, 2152. (6) Thottempudi, V.; Gao, H.; Shreeve, J. M. J. Am. Chem. Soc. 2011, 133, 6464. (7) Macbeth, A. K.; Orr, W. B. J. Chem. Soc. 1932, 0, 534. (8) Hammond, G. S.; Emmons, W. D.; Parker, C. O.; Graybill, B. M.; Waters, J. H.; Hawthorne, M. F. Tetrahedron 1963, 19, Supplement 1, 177. (9) Gakh, A. A.; Bryan, J. C.; Burnett, M. N.; Bonnesen, P. V. J. Mol. Struct. 2000, 520, 221. (10) Göbel, M.; Klapötke, T. M. Z. Anorg. Allg. Chem. 2007, 633, 1006. (11) Kaplan, L. A.; Kamlet, M. J. J. Org. Chem. 1962, 27, 780. (12) Erashko, V. I.; Shevelev, S. A.; A.A., F. b. Izv. Akad. Nauk SSSR, Ser. Khim. 1968, 9, 2117. (13) Shevelev, S. A.; Erashko, V. I.; Sankov, B. G.; Fainzil'berg, A. A. Izv. Akad. Nauk SSSR, Ser. Khim. 1968, 2, 382. (14) Gidaspov, A. A.; Bakharev, V. V.; Kukushkin, I. K. Russ. Chem. Bull. 2009, 58, 2154. (15) Titova, K. V.; Kolmakova, E. I.; Rosolovskii, V. Y. Bull. Acad. Sci. USSR, Div. Chem. Sci. (Engl. Transl.) 1977, 26, 1822. (16) Huang, Y.; Gao, H.; Twamley, B.; Shreeve, J. n. M. Eur. J. Inorg. Chem. 2007, 2007, 2025. (17) Holleman, A. F.; Wiberg, E. In Inorg. Chem.; First English ed.; Wiberg, N., Ed.; ACADEMIC PRESS: San Diego, 2001, p 131. (18) Chase, M.W., Jr, NIST-JANAF Thermochemical Tables Fourth Editions, J. Phys Chem Ref. Data, Monograph 9, 1998, 1-1951. 88 (19) Koch, E.-C.; Klapötke, T. M. Propellants, Explos., Pyrotech. 2012, 37, 335. (20) Haiges, R.; Jones, C. B.; Christe, K. O. Inorg. Chem. 2013, 52, 5551. (21) Snyder, C. J.; Martin, P. D.; Heeg, M. J.; Winter, C. H. Chem. Eur. J. 2013, 19, 3306. (22) Klapötke, T. M.; Krumm, B.; Moll, R. Chem. - Eur. J. 2013, 19, 12113. (23) Klapötke, T. M.; Rusan, M.; Sproll, V. Z. Anorg. Allg. Chem. 2013, 639, 2433. (24) Klapötke, T. M.; Krumm, B.; Rusan, M.; Sabatini, J. J. Chem. Commun. 2014, 50, 9581. (25) Klapötke, T. M.; Rusan, M.; Sproll, V. Z. Anorg. Allg. Chem. 2014, 640, 1892. (26) After the publication of this manuscript, we were made aware through private correspondence that graduate work had been done on (trinitromethyl)-hydroborane and -borate species in 1966-1967 and presented at a IUPAC conference in 1993 (Ioffe, S. L.; Shitov, O. P.; Tartakowskij, V. A. "The Unusual Reation of Trinitromethane with BH3.THF-complex", IUPAC IMEBORON VIII, Knoxville, Tennessee, July 11-15 1993). No reference to the species could be found in the English language literature. (27) Luk'yanov, O. A.; Anikin, O. V.; Gorelik, V. P.; Tartakovsky, V. A. Russ. Chem. Bull. 1994, 43, 1457. (28) Titova, K. V. Russ. J. Inorg. Chem. 2002, 47, 1121. (29) Grant, D. J.; Dixon, D. A.; Camaioni, D.; Potter, R. G.; Christe, K. O. Inorg. Chem. 2009, 48, 8811. (30) Novikov, S. S.; Slovetskii, V. I.; Shevelev, S. A.; Fainzil'berg, A. A. Bull. Acad. Sci. USSR, Div. Chem. Sci. (Engl. Transl.) 1962, 11, 552. (31) Göbel, M.; Klapötke, T. M. Acta Crystallogr. Sect. C: Cryst. Struct. Commun. 2007, 63, o562. (32) Schodel, H.; Dienelt, R.; Bock, H. Acta Crystallogr. Sect. C: Cryst. Struct. Commun. 1994, 50, 1790. (33) Tartakovskii, V. A.; Nikonova, L. A.; Novikov, S. S. Izv. Akad. Nauk SSSR, Ser. Khim. 1967, 706. (34) Scherfise, K. D.; Weller, F.; Dehnicke, K. Z. Naturforsch., B: Chem. Sci. 1985, 40b, 906. (35) Young, D. E.; McAchran, G. E.; Shore, S. G. J. Am. Chem. Soc. 1966, 88, 4390. (36) Singaram, B.; Cole, T. E.; Brown, H. C. Organometallics 1984, 3, 774. (37) Mason, J. In Multinuclear NMR; Mason, J., Ed.; Plenum Press: New York 1987, p 227. (38) Coburn, M. D.; Storm, C. B.; Moore, D. W.; Archibald, T. G. Magn. Reson. Chem. 1990, 28, 16. (39) Göbel, M.; Klapötke, T. M.; Mayer, P. Z. Anorg. Allg. Chem. 2006, 632, 1043. (40) Smith, E.; Dent, G. In Modern Raman Spectroscopy A Practical Approach; Wiley: West Sussex, 2005, p 23. (41) Shlyapochnikov, V. A.; Oleneva, G. I.; Novikov, S. S. Russ Chem Bull 1971, 20, 2477. (42) Brookes, M. J.; Jonathan, N. J. Chem. Soc. A 1968, 2266. (43) Haiges, R.; Boatz, J. A.; Vij, A.; Gerken, M.; Schneider, S.; Schroer, T.; Christe, K. O. Angew. Chem. Int. Ed. 2003, 42, 5847. 89 (44) Haiges, R.; Boatz, J. A.; Schneider, S.; Schroer, T.; Yousufuddin, M.; Christe, K. O. Angew. Chem. Int. Ed. 2004, 43, 3148. (45) Grant, D. J.; Matus, M. H.; Switzer, J. R.; Dixon, D. A.; Francisco, J. S.; Christe, K. O. J. Phys. Chem. A 2008, 112, 3145. (46) Murray, J. S.; Brinck, T.; Politzer, P. Chem. Phys. 1996, 204, 289. (47) Politzer, P.; Murray, J. S. J. Phys. Chem. A 1998, 102, 1018. (48) Rice, B. M.; Pai, S. V.; Hare, J. Combust. Flame 1999, 118, 445. 90 CHAPTER 6: REACTION OF DINITROAMINE WITH AMMONIA- BORANE 6.1 Introduction The dinitramide anion (DN - ), discovered in 1971 by Luk'yanov et al. 1 and independently in the 1980s by Bottaro et al., 2 has since generated much interest in the scientific community. While several covalent dinitramide compounds have been synthesized, including Group 14 compounds, 3,4 Group 16 compounds, 3 fluorodinitroamine 5 (CHAPTER 9) and trinitroamine, 6 no compound with Group 13 elements has so far been structurally characterized. The ongoing work on boron-derived green energetic materials in the Christe-Haiges group prompted the exploration of the synthesis of dinitramido substituted boron compounds. Herein the reaction chemistry of dinitroamine, HN(NO 2 ) 2 (HDN), with ammonia-borane is reported. 6.2 Detection of Dinitramide-substituted Ammonia-borane Species In glyme or acetonitrile solution, the reaction of one equivalent of HDN with NH 3 BH 3 yields a mixture of species detected by 14 N and 11 B NMR (Figure 6.1). The most dominant signals, a 11 B NMR signal (triplet) at -9 ppm and a sharp 14 N NMR signal at -34 ppm, were assigned to NH 3 . BH 2 DN (1). This assignment is further confirmed by a broad quartet at 2.77 ppm with 1 J( 1 H- 11 B)= 111 Hz and a broad triplet at 4.69 ppm with 1 J( 1 H- 14 N)=50 Hz observed by 1 H NMR, consistent with a deshielded BH 2 group and a deshielded boron-coordinated ammonia ligand. Other important signals were detected, namely two multiplets at ca -4 ppm in the 11 B NMR spectrum and a singlet at -29.6 ppm in the 14 N NMR spectrum. Variable amounts of ammonium dinitramide were also formed, depending on the reaction conditions. When three to four equivalents of HDN were used, the formation of other species could be observed by NMR spectroscopy. However, the equilibrium between HDN and the DN - by-product caused the 14 N NMR chemical shifts to be highly variable, which made the identification of new species difficult. Nevertheless, two new species could be observed by 11 B NMR spectroscopy, as doublets at -0.7 ppm and 1.6 ppm, respectively, indicating that a second hydrogen elimination had occurred at the boron atom. The only 14 N NMR signal which had a sufficiently high intensity comparable to that of the new 11 B NMR resonances at -0.7 ppm was found at -32.7 ppm, suggesting the presence of a BH(DN) 2 group. The absence of a 14 N NMR signal in the range typical for nitro groups of sufficient intensity to correlate with 91 that of the 11 B NMR signal at 1.6 ppm indicates that that boron species is not bound to a nitro-containing moiety. A new set of resonances was found at higher field in the 1 H NMR spectrum at 3.8 ppm (qt, 1 J( 1 H- 11 B)= 143 Hz) and 5.3 ppm (t, 1 J( 1 H- 14 N)=45 Hz). These resonances are consistent with NH 3 . BH(DN) 2 (2). The number of signals observed by 14 N NMR at each step of the reaction strongly suggests that the B-N rather than the B-O linkage isomer was present in these solutions. Indeed, the B-O isomer is expected to generate two relatively sharp signals in 14 N NMR for each boron species observed by 11 B NMR spectroscopy. After several days in the presence of excess dinitroamine, the disappearance of 1 and 2 was accompanied by the appearance of a signal at +20.6 ppm in the 11 B NMR spectrum. The presence of a 1 H NMR signal at ca 6 ppm combined with this signal characteristic for tri-coordinate boron strongly indicates the formation of boric acid, suggesting that the expected B(DN) 3 or NH 3 . B(DN) 3 species are unstable and decomposed quickly after being formed. Figure 6.1: Reaction of ammonia borane with dinitroamine in glyme or acetonitrile at room temperature. The last step might also involve some hydrolysis. 6.3 Synthesis and Isolation The isolation of NH 3 . BH 2 DN (1) was studied under various conditions. When reacted in stoichiometric amounts in glyme between -60 °C and ambient temperature, 1 was formed in ca 60 mol% yield, with ammonium dinitramide and a group of unidentified species accounting for the rest of the mixture. Evaporating the solvent under vacuum between 0 and 20 °C and extracting the products with dichloromethane afforded a solution containing only 1 (ca 80 mol %) and an unidentified species with two overlapping 11 B NMR signals at ca -4 ppm and a 14 N NMR signal at -29.6 ppm. When such a solution was evaporated, colorless crystals formed on the walls of the vessel, which were identified as 1 by X-Ray diffraction. The isolated solid was not completely soluble in CD 3 CN, indicating that some decomposition had occurred during the isolation process. The major unidentified side-products could not be removed by recrystallizations. Since a significant excess of HDN was required to form NH 3 . BH(DN) 2 (2) at a reasonable rate, the isolation of 2 was not pursued any further, because concentrating solutions containing excess dinitroamine can lead to explosions. 7 92 6.4 Theoretical Insight Six conformers of the B-O isomer were initially screened in the gas-phase. The lowest energy conformer was found to be near degenerate with the B-N isomer (Figure 6.2). Upon consideration of solvation effects, estimates for the Gibbs energy difference between the isomers suggest the B-N isomer to be favored by ca 1.6 kcal/mol in solution. This translates into an expected 90-95 % prevalence of the B-N isomer close to ambient conditions. This is in qualitative agreement with the fact that only the B-N isomer could be detected with sufficient confidence. Figure 6.2: Relative energies of the most stable B- N and B-O isomers of 1 in the gas phase and in acetonitrile solution (red), obtained using solvent corrected coupled-cluster calculations (CBS- QB3+SMD-M06-2X/cc-pVTZ). The predicted NMR chemical shifts for 1 are in excellent agreement with the major species observed for all three nuclei (Table 6.1). The vibrational spectra of the isolated compounds, known to contain significant amounts of ammonium dinitramide, agree well with the predicted data (APPENDIX 5). Based on the fit with the predicted vibrational and NMR data, it is reasonable to assume that the major compound observed by NMR is also the one constituting the crystal for which a crystal structure was obtained. ΔG 0 = -0.1 [+1.6] ΔH 0 = -0.3 [+1.0] N N N N O O O O O O O O N N N N B B H H H H H H H H H 93 Table 6.1: Predicted NMR chemical shifts for the two linkage isomers of 1. 1 H and 11 B chemical shifts were referenced to the experimentally observed values for NH 3 . BH 3 . 14 N chemical shifts were referenced to MeNO 2 . Predicted δ (ppm) Experimental δ (ppm) NH 3 -BH 2 -DN 4.6 4.69 NH 3 -BH 2 -DN 3.0 2.77 NH 3 -BH 2 -O-N(O)N-NO 2 4.5 - NH 3 -BH 2 -O-N(O)N-NO 2 2.9 - NH 3 -BH 2 -DN -8.2 -9.3 NH 3 -BH 2 -O-N(O)N-NO 2 -2.1 - NH 3 -BH 2 -N(NO 2 ) 2 -33.7 -34.3 NH 3 -BH 2 -N(NO 2 ) 2 -88.4 -91 (Et 2 O) NH 3 -BH 2 -N(NO 2 ) 2 -358.3 -361 NH 3 -BH 2 -O-N(O)N-NO 2 -23.1 - NH 3 -BH 2 -O-N(O)N-NO 2 -28.0 - NH 3 -BH 2 -O-N(O)N-NO 2 -38.9 - NH 3 -BH 2 -O-N(O)N-NO 2 -356.0 - 6.5 Structural Characterization Crystals obtained from dichloromethane solutions containing ca 80 mol% of 1 proved to be of the B-N isomer. The crystallographic data pertaining to 1 are listed in Table 6.2. 1 crystallizes in the P2 1 /n space group with 8 molecules per unit cell (Figure 6.3). This is the first example of a structurally characterized Group 13-dinitramide compound. The asymmetric unit contains two conformers of 1 differing mostly by the O-N-N-B dihedral angles. As seen for the dinitramide anion, the nitro groups are tilted (the two O-N- N angles differ significantly). This tilted geometry is also seen in trinitromethyl-containing compounds 8 (CHAPTER 5). The H-B-H angles deviate slightly from the ideal tetrahedral angle (ca 115 °). Both B-N distances are statistically equal for each molecule of the asymmetric unit (1.5798(19) and 1.5895(18) Å). 94 Figure 6.3: One of the two molecules of the asymmetric unit of 1. N-H hydrogen atoms positions were idealized. B-H hydrogen atoms were determined from the electron density map and then restrained. Selected structural parameters: Bond distances (Å): N1-N2 1.3926(15), N1-N3 1.4046(15), O1-N2 1.2157(15), O2-N2 1.2214(16), O3-N3 1.2160(15), O4-N3 1.2197(15), N1-B1 1.5798(19), B1-N4 1.5895(18), B1-H1 1.100(14), B1-H2 1.080(14), Bond angles (°): O1-N2-O2 125.20(12), O3-N3-O4 126.04(12), O4-N3-N1 113.26(11), O3-N3-N1 120.61(11), O1-N2-N1 115.57(11), O2-N2-N1 119.16(11), N2-N1-N3 115.64(11), N2-N1-B1 125.69(11), N3-N1-B1 118.47(11), N1-B1-N4 107.23(11), H1-B1-H2 115.7(13), H1-B1-N4 107.8(9), H1-B1-N1 107.7(9), H2-B1-N4 110.7(9), H2-B1-N1 107.4(9), O1-N2-N1-B1 -9.4(2), O4-N3-N1-B1 -33.4(2). 95 Table 6.2: Crystallographic details for 1. 1 Formula BH 5 N 4 O 4 Mol wt [g/mol] 135.89 Temp (K) 100(2)K Crystal system Monoclinic Space group P2 1 /n a [Å] 12.5970(2) b [Å] 6.54390(10) c [Å] 12.6505(2) α [°] 90 β [°] 91.7660(10) γ [°] 90 V [Å 3 ] 1042.33(3) Z 8 λ[Å] 1.54178 ρ calc [g/cm 3 ] 1.732 μ [mm -1 ] 1.499 F (000) 560 Reflections collected 12001 Independent reflections 2063 R int 0.0275 Number of parameters 177 R1 [I > 2σ(I)] 0.0324 wR2 [I > 2σ(I)] 0.0883 GOF 1.057 6.6 Vibrational Spectroscopy The vibrational spectra recorded for a sample of 1 containing significant amounts of ammonium dinitramide and ca 20 mol% of the unidentified impurities show a relatively good match with the predicted ones (APPENDIX 5). The main features of the vibrational spectra are the two strongest bands at ca 2520 and 2460 cm -1 , which correspond to the antisymmetric and symmetric B-H stretching modes, respectively, which are in the same range observed for NH 3 -BH 3 (2200-2500 cm -1 ) 9 and at similar frequencies as those found for the [BH 2 Azolate 2 ] - anions (2450-2550 cm -1 ). 10 The spectrum is otherwise dominated by the NH 3 bands at ca 3500 cm -1 and the dinitramide modes in the 700-1600 cm -1 range. 96 6.7 Stability 1 proved to be only marginally stable under ambient conditions. It underwent extensive decomposition upon isolation in the solid state, with an insoluble boron-containing colorless material and ammonium dinitramide as the major decomposition products. Volatile materials collected from the evaporation of solutions containing 1 contained only small amounts of N 2 O detected by IR spectroscopy, implying a very slow and negligible decomposition of the dinitramide moiety. The compound is particularly air-sensitive. From dichloromethane solutions of 1 in air, a tacky white solid started to precipitate and bubbles originated from its surface. In one instance, one of the resulting decomposition products crystallized during its formation and was shown to be ammonium dinitramide by X-ray diffraction. In one instance, a concentrated glyme solution containing HDN and 1 exposed to light decomposed significantly while a similar sample kept in darkness did not, suggesting that 1 and/or HDN are photosensitive. 6.8 B-O vs B-N Isomers The similarity of the predicted chemical shifts for the B-O linkage isomer of 1 with those of the major unidentified NMR signals suggests that this species might have been observed as a by-product. Two 11 B NMR signals were indeed observed in variable ratios in the vicinity of -3 ppm, very close to the predicted 11 B chemical shift for the B-O isomer of 1 (-2.1 ppm). One of the nitro groups for that isomer is also predicted to have a 14 N chemical shift of -28 ppm, very close to the most significant unidentified resonance at -29.6 ppm observed in reaction mixtures containing 1. The second nitro group is predicted to be at -23.1 ppm, somewhat close to a resonance frequently observed at ca -25 ppm. These resonances were however never observed in a ca 1:1 ratio expected for such a compound, and were found in variable ratios, suggesting they do not belong to the same compound. While it remains possible that some of the unidentified resonances could be attributed to the B-O isomer of 1, the isomer could not be conclusively associated with the major side-products observed besides 1. More details on the possible identity of the by-products are given in APPENDIX 5. 6.9 Conclusion A dinitramide-boron compound was structurally characterized for the first time. An analysis of the dominant observed NMR signals combined with quantum chemical calculations established that the given crystal structure is indeed that of the major compound formed in the reactions studied. Calculations confirm that in solution the observed B-N isomer is thermodynamically favored over the B-O isomer, 97 while the opposite is true for the gas phase. The formation of NH 3 . BH(DN) 2 was also observed by NMR spectroscopy. The observed compounds are very rare examples of a powerful redox pairs, in this case hydrido and dinitramido ligands, in single small molecules. 8 While NH 3 . BH 2 DN is only marginally stable under ambient conditions, its major decomposition pathway involves the elimination of ammonium dinitramide, suggesting that the instability does not arise from redox reactions of the dinitramide and hydride groups. More likely, the compound might be undergoing oligomerization reactions similar to those found for ammonia-borane. This indicates that, contrary to previous reports on the instability of dinitramide compounds with Lewis acids, 7 a careful choice of substrates can allow the isolation of dinitramide-substituted boranes. Further experimental details, remarks, NMR spectra, crystal packing diagrams and crystallographic information can be found in APPENDIX 5. 6.10 References (1) Luk'yanov, O. A.; Gorelik, V. P.; Tartakovskii, V. A. Russ. Chem. Bull. 1994, 43, 89. (2) Bottaro, J. C.; Penwell, P. E.; Schmitt, R. J. Synth. Commun. 1991, 21, 945. (3) Klapötke, T. M.; Krumm, B.; Scherr, M. Z. Anorg. Allg. Chem. 2009, 635, 885. (4) Luk'yanov, O. A.; Shlykova, N. I.; Tartakovsky, V. A. Russ. Chem. Bull. 1994, 43, 1680. (5) Christe, K. O.; Wilson, W. W.; Bélanger-Chabot, G.; Haiges, R.; Boatz, J. A.; Rahm, M.; Prakash, G. K. S.; Saal, T.; Hopfinger, M. Angew. Chem. Int. Ed. 2015, 54, 1316. (6) Rahm, M.; Dvinskikh, S. V.; Furó, I.; Brinck, T. Angew. Chem. Int. Ed. 2011, 50, 1145. (7) Luk'yanov, O. A.; Anikin, O. V.; Gorelik, V. P.; Tartakovsky, V. A. Russ. Chem. Bull. 1994, 43, 1457. (8) Bélanger-Chabot, G.; Rahm, M.; Haiges, R.; Christe, K. O. Angew. Chem. Int. Ed. 2013, 52, 11002. (9) Ziparo, C.; Colognesi, D.; Giannasi, A.; Zoppi, M. J. Phys. Chem. A 2012, 116, 8827. (10) Haiges, R.; Jones, C. B.; Christe, K. O. Inorg. Chem. 2013, 52, 5551. 98 CHAPTER 7: ON THE EXISTENCE OF DINITRAMIDO-BORATES 7.1 Introduction Since their discovery, 1,2 dinitramides have been recognized as potentially useful energetic materials. While several salts 1,3,4 and a few covalent 5-8 compounds of dinitramide have been synthesized, no Group 13-dinitramide complex anions have been reported so far. Since the Christe-Haiges group is interested in boron-based green solid propellants as replacements for ammonium perchlorate, studying the formation dinitramido-borate complexes was an attractive prospect. Herein NMR studies on the formation of several BX n [N(NO 2 ) 2 ] 4-n - anions (Where X= F, Cl, H) are described. 7.2 NMR Observations The possibility of isolating B[N(NO 2 ) 2 ] 4 - from the reaction of alkali metal dinitramide salts and boron trihalides was tested. While experiments conducted with BF 3 at -40 °C indicated that a transient boron dinitramide had formed, the stability of the species was too low even at these temperatures for a conclusive characterization, and NO 2 and N 2 O were obtained as the major decomposition products (APPENDIX 6). Since it had previously been demonstrated that [BH 3 C(NO 2 ) 3 ] - salts 9 (CHAPTER 5) were more stable than the corresponding [BCl 3 C(NO 2 ) 3 ] - , 10 it was relevant to verify whether (dinitramido)hydroborates, BH n [N(NO 2 ) 2 ] 4-n - , would be stable enough to be observable in solution. Dinitroamine (HDN) was thus reacted with various borohydride salts (Figure 7.4). When one equivalent of HDN was reacted with salts of BH 4 - in glyme at -30 °C, an equilibrium between free dinitramide (DN - ) + BH 3 . glyme and the new [BH 3 DN] - (1) species was observed. The 11 B NMR spectrum of 1 displayed a quartet at -14.5 ppm and its 14 N NMR spectrum displayed a singlet at -24.7 ppm. The 11 B NMR shift of - 14.5 ppm is significantly deshielded compared to other nitrogen base adducts of borane, consistent with the strong electron-withdrawing character of the dinitramide anion. The existence of the dissociation equilibrium for 1 clearly indicates that the N-B bond in 1 is quite weak, comparable in strength to that of the glyme . BH 3 adduct. 1 persisted at ambient temperature and only the ratios of 1, glyme . BH 3 , B 2 H 7 - and BH 4 - changed significantly with the temperature. 1 could also be generated from BH 3 . THF and KDN in a THF solution. It is interesting to note that the strength of the B-DN bond is so weak that an excess of BH 4 - competes with DN - for BH 3 . Indeed, significant amounts of B 2 H 7 - were observed when 1 was formed in 99 solution (Figure 7.1). Moreover, it was shown that dinitramide is unable to displace SMe 2 from BH 3 . SMe 2 /SMe 2 in an acetonitrile solution at room temperature, as no trace of 1 could be detected under these conditions. Figure 7.1: 11 B NMR spectrum of NaBH 4 + HDN in glyme at 0°C. When two equivalents of HDN were reacted with BH 4 - salts at room temperature, a new major species characterized by a 11 B NMR resonance (triplet) at ca -7 ppm and a 14 N NMR resonance at -34.5 ppm was observed, which was attributed to [BH 2 (DN) 2 ] - (2). An intense 11 B NMR resonance at -2 ppm, a triplet (or sometimes a triplet of doublets), was observed with no comparably intense 14 N NMR counterpart apart from a very intense one at -10.8 ppm, which was attributed to the dinitramide anion. The species with a -2 ppm 11 B NMR resonance was thus presumed not to contain dinitramide, but instead to be a solvated boronium cation. Interestingly, this 11 B NMR resonance displayed finer coupling (triplet of doublets) under certain conditions (Figure 7.2), which would be expected for a hydrogen-bridged, symmetrical BH 2 moiety, suggesting that this BH 2 + -type cation might be B 2 H 5 + . A sizeable amount of BH 3 . glyme was detected at 2.2 ppm by 11 B NMR. 100 Figure 7.2: 11 B NMR spectrum of TMABH 4 + excess HDN at room temperature, showing 2 and probably the glyme solvated B 2 H 5 + cation. Beyond two equivalents of HDN, the observed signals became increasingly difficult to reproduce. The best results were obtained by reacting TMABH 4 with 4 to 6 equivalents of HDN. After one week at room temperature, the solution contained only traces of 2 and displayed two new overlapping resonances (doublets) at 4.7 and 5.8 ppm in the 11 B NMR spectrum. The 14 N spectrum was less informative as all of the major species appeared to collapse into an odd shaped, broad singlet at ca -33 ppm. These signals were tentatively assigned to [BH(DN) 3 ] - (3) and glyme . BH(DN) 2 (4) (Figure 7.3). Figure 7.3: 11 B NMR spectrum showing species tentatively assigned as 3 and 4 from TMABH 4 + excess HDN in glyme which had reacted at room temperature for several days. 101 The failure to isolate TMA[BH 3 DN] or PPh 4 [BH 2 (DN) 2 ] in reasonable amounts (APPENDIX 6) demonstrated that neither 1 nor 2 are stable at room temperature in the solid state. Indeed, no salt of 1 could be isolated and only dinitramide and unidentified species are formed. Small amounts of PPh 4 [BH 2 (DN) 2 ] could be found in the solid state, but dinitramide and unknown boron species were found as the major products. Figure 7.4: Reactions of HDN with BH 4 - as observed by multinuclear NMR spectroscopy. Based on the number of 14 N resonances observed in relationship to the number of 11 B resonances observed in this study, it is very unlikely that a B-O-N(O)N-NO 2 linkage isomers are formed, which would have generated two relatively sharp 14 N resonances in an approximately 1:1 ratio for each boron species observed by 11 B NMR. 7.3 Conclusion It was demonstrated that the formation of marginally stable dinitramidoborates is possible. As previously observed, energetic ligand containing hydroborates are more stable than the corresponding chloroborates, which allowed the observation of dinitramidoborates in solution. The weakness of the B-N bond in the observed dinitramidoborates suggest that the dinitramide anion could be used as a weakly coordinating ligand, especially since it appears to tolerate the presence of strongly reducing hydrides. Further experimental details and remarks can be found in APPENDIX 6. 102 7.4 References (1) Luk'yanov, O. A.; Gorelik, V. P.; Tartakovskii, V. A. Russ. Chem. Bull. 1994, 43, 89. (2) Bottaro, J. C.; Penwell, P. E.; Schmitt, R. J. Synth. Commun. 1991, 21, 945. (3) Klapötke, T. M.; Krumm, B.; Scherr, M. Eur. J. Inorg. Chem. 2008, 2008, 4413. (4) Luk'yanov, O. A.; Anikin, O. V.; Gorelik, V. P.; Tartakovsky, V. A. Russ. Chem. Bull. 1994, 43, 1457. (5) Klapötke, T. M.; Krumm, B.; Scherr, M. Z. Anorg. Allg. Chem. 2009, 635, 885. (6) Luk'yanov, O. A.; Shlykova, N. I.; Tartakovsky, V. A. Russ. Chem. Bull. 1994, 43, 1680. (7) Christe, K. O.; Wilson, W. W.; Bélanger-Chabot, G.; Haiges, R.; Boatz, J. A.; Rahm, M.; Prakash, G. K. S.; Saal, T.; Hopfinger, M. Angew. Chem. Int. Ed. 2015, 54, 1316. (8) Rahm, M.; Dvinskikh, S. V.; Furó, I.; Brinck, T. Angew. Chem. Int. Ed. 2011, 50, 1145. (9) Bélanger-Chabot, G.; Rahm, M.; Haiges, R.; Christe, K. O. Angew. Chem. Int. Ed. 2013, 52, 11002. (10) Titova, K. V.; Kolmakova, E. I.; Rosolovskii, V. Y. Bull. Acad. Sci. USSR, Div. Chem. Sci. (Engl. Transl.) 1977, 26, 1822. 103 CHAPTER 8: SYNTHESIS OF NITRYL CYANIDE, NCNO 2 8.1 Introduction Being an intriguing example of a compound containing a strong oxidizer (NO 2 group) adjacent to an excellent fuel (CN group) 1 , nitryl cyanide, NCNO 2 (1), represents a very promising high energy density material by itself and as a building block for other energetic compounds. Furthermore, nitryl cyanide might play an important role in atmospheric and interstellar chemistry in view of its simplicity and the known multitude of nitriles in interstellar space. 2,3 Surprisingly, NCNO 2 has not been prepared to date. In this chapter, the successful synthesis, and spectroscopic and computational characterizations of nitryl cyanide are reported and its potential as a promising reagent in high energy density material (HEDM) chemistry is explored. Klapötke et al. were the first to report the attempted synthesis of 1. 4 In 1996, they explored the nitration of hydrogen cyanide with nitronium tetrafluoroborate, and observed signals of an unstable compound at -62.2 and -171.4 ppm in the low-temperature 14 N NMR spectrum. These were attributed to the nitrile isocyanide isomer, CNNO 2 (2), in spite of the fact that this isomer was calculated to lie 37 kcal/mol higher than NCNO 2 at the MP2/6-31G(d) level of theory. This report was followed by computational studies by Bartlett et al. who investigated a number of CN 2 O 2 isomers using coupled- cluster CCSD(T)/TZ2P calculations 5 and Zhang et al. who studied reaction pathways of nitrogen dioxide with the cyanogen radical. 6 They proposed that nitryl cyanide should be kinetically stable with respect to uni-molecular decomposition, 5 based on a single transition state, which lies 53.6-53.9 kcal/mol above the ground state. In the present study, the homolytic bond dissociation into neutral CN and NO 2 radical fragments was predicted to correspond to an enthalpy change of 62.0 kcal/mol at the CBS-QB3 (basis set extrapolated CCSD(T)) level of theory (Equation 8.2). Thus, assuming that intermolecular decomposition pathways can be neglected, it appears that nitryl cyanide might indeed possess significant kinetic stability under ambient conditions and represent an attractive target for synthesis. 8.2 Synthesis and Properties For the synthesis of NCNO 2 an extensive range of experimental approaches was investigated, which are discussed at length in APPENDIX 7. The high reactivity of NCNO 2 and the separation of reaction by- products required extensive method and solvent screening as well as careful control of the reaction 104 conditions. Ultimately, the best reaction system for the isolation of NCNO 2 in substantial amounts proved to be that of t-BuMe 2 SiCN with NO 2 BF 4 in nitromethane at -30 °C (Equation 8.1) NO 2 BF 4 + t-BuMe 2 SiCN → NCNO 2 + t-BuMe 2 SiF + BF 3 (8.1) The in situ yields are ca 50%, based on IR and NMR spectroscopic estimates, and are strongly dependent upon the reaction conditions. When an excess of the trialkylsilyl cyanide is used, the yields drop dramatically and side-products such as cyanogen are formed as the major species, apparently from the reaction of NCNO 2 with the silyl cyanide starting material. To decrease the amounts of cyanogen formed, the silyl cyanide needs to be condensed at -196 °C above a frozen mixture of NO 2 BF 4 in MeNO 2 . This mixture must then be carefully thawed and only small portions of the silyl cyanide must be dissolved at a time. Even under optimized conditions, reaction by-products (t-BuMe 2 SiF, BF 3 ) and side-products (NO 2 , N 2 O, CO 2 , cyanogen) need to be removed by fractional condensation. The initial separation of BF 3 is critical, since BF 3 reacts with NCNO 2 in the liquid phase to form a white solid (most likely NOBF 4 ), significantly lowering the yields. Pumping the volatile compounds through a succession of -80, -96, -112, -126 and - 196 °C traps allows the collection of relatively pure NCNO 2 at -112 °C. NCNO 2 could be obtained in good purity (>95% by IR spectroscopy) after three fractional condensation runs, in ca 40% isolated yield. Small amounts of NO 2 and traces of BF 3 and t-BuMe 2 SiF may remain after the purification steps. NCNO 2 is colorless in the solid-state. It melts to a colorless liquid at -85 ± 1 °C. Small amounts of nitrogen oxides can give it a pale yellow or bluish color. A density of 1.24 ± 0.08 g/mL was measured for the liquid at -79 °C, comparable to that of ClNO 2 (1.37 g/mL at 0 °C). 7 NCNO 2 is colorless in the gas- phase. Vapor pressures of NCNO 2 were measured between -96 and 0 °C, giving an extrapolated boiling point of 7 °C and an enthalpy of evaporation of 6.8±0.2 kcal/mol, in excellent agreement with the theoretically predicted value of 6.4 kcal/mol (APPENDIX 7). Gas density measurements gave a molecular weight of 70 ± 2 g/mol for NCNO 2 (calculated: 72.00 g/mol). Its relatively high melting point of -85 °C, compared to -145 °C for ClNO 2 7 , suggests a somewhat increased association in the liquid phase. This is also reflected by the Trouton constants (molar heat of evaporation divided by absolute boiling point), 24.3 cal K -1 mol -1 for NCNO 2 and 23.9 cal K -1 mol -1 for ClNO 2 . 8 Pure NCNO 2 is stable at room temperature. In the gas phase, it could be kept at low pressures in a glass vessel for several days with only trace amounts of CO 2 detectable by IR spectroscopy. In the liquid phase, color changes to greenish blue could be observed after several hours even at -80 °C due to trace amounts of impurities. Given the predicted high stability of monomeric NCNO 2 , it appears that in the condensed phase it might undergo more facile decomposition through intermolecular pathways. 105 Calculations suggest that trace amounts of NO 2 radicals might also significantly lower the activation barrier for NCNO 2 decomposition (ΔG ‡ : 54  28 kcal/mol). Reaction with NO 2 could result in the formation of an unstable radical intermediate, NCN(O)O-NO 2 , which presumably could decompose into NCNO, a well-known species of blue color. 9,10 NCNO was observed by IR spectroscopy during the synthesis of NCNO 2 under certain reaction conditions (APPENDIX 7). Bartlett et al. have predicted the heat of formation (ΔH 0 f(1,gas) ) of NCNO 2 as 60 kcal/mol, using MBPT(2) calculations of the isodesmic reaction shown in Equation 8.3. The most recent ΔH f 0 -NIST data was used for the compounds in Equations 8.2-8.4 combined with reliable CBS-QB3 reaction enthalpies, a more accurate value of ΔH 0 f(1,gas) = 50.7 ± 2 kcal/mol was thus obtained for NCNO 2 . This is an extraordinarily high value for such a small molecule, almost rivaling that of HN 3 (70 kcal/mol). 11 Below, the ΔH 0 f values of the individual compounds used for the calculations are shown in kcal/mol in parentheses. NCNO 2 (1) → ･CN (+104.0) + ･NO 2 (+7.9) (8.2) (ΔH 0 r,CBS-QB3 = +62.0 kcal/mol, ΔH 0 f,1 = +49.9 kcal/mol) NCNO 2 (1) + C 2 H 6 (-20.0) →CH 3 CN (+17.7) + CH 3 NO 2 (-19.4) (8.3) (ΔH 0 r,CBS-QB3 = -32.4 kcal/mol , ΔH 0 f,1 = +50.8 kcal/mol ) NCNO 2 (1) → CO 2 (-94.1) + N 2 (0.0) (8.4) (ΔH 0 r,CBS-QB3 = -145.6 kcal/mol, ΔH 0 f,1 = +51.5 kcal/mol) Our predicted heat of formation for NCNO 2 in the liquid phase is obtained by subtracting ΔH 1,vap(exp) from ΔH 0 f(1,gas) , i.e. ΔH 0 f(1,l) = 50.7 – 6.8 = 43.9 ± 2 kcal/mol. EOM-CCSD/Def2-TZVPP calculations suggest that NCNO 2 does not have any spin-allowed electronic transition states in the visible spectrum, and two significant transitions in the UV region at 185 nm ( 1 A 1 → 1 B 2 ) and 161 nm ( 1 A 1 → 1 A 1 ). Geometry optimizations of the excited states at the TD-ωB97X-D/aug-cc-pVTZ level show neither to be dissociative. Consequently, the compound is not expected to be light sensitive. The adiabatic ionization potential of 1 to 12.7 eV was estimated at the CBS-QB3 level. The adiabatic electron affinity of 1 was calculated to 2.0 eV, which in comparison to 2.7 eV for the much larger C 60 fullerene exemplifies a high electrophilicity. NCNO 2 is very reactive. In the presence of BF 4 - , it was found to react with nitriles to form the corresponding acyl fluorides. Since these were difficult to separate, this hampered the use of acetonitrile 106 for the synthesis of NCNO 2 . It also appears to react with the trialkylsilyl cyanide starting materials to form cyanogen. It hydrolyzes slowly below 0 °C to ammonium nitrate which was identified by 14 N NMR. The mechanism of the hydrolysis is best described by the hydrolysis of NCNO 2 to HCN and HNO 3 , followed by the well-known acidic hydrolysis of HCN to NH 3 and formic acid or CO, and NH 3 reacting with HNO 3 to give NH 4 NO 3 (Equations 8.5-8.8). NCNO 2 + H 2 O → HCN + HNO 3 (8.5) HCN + 2 H 2 O → NH 3 + HCOOH (8.6) HCOOH → CO + H 2 O (8.7) NH 3 + HNO 3 → NH 4 NO 3 (8.8) The thermal decomposition of NCNO 2 was studied by heating the compound in a stainless steel cylinder, separating the products by fractional condensation, and analyzing them by IR spectroscopy. Heating for one hour to 50 and 100 °C, respectively, resulted only in little decomposition with most of the NCNO 2 being recovered unchanged, and heating to 140 °C for several hours was required to achieve complete decomposition. The major decomposition products were CO 2 (80 mol% based on carbon) and N 2 (Equation 8.4), but smaller amounts of NO and some NO 2 and N 2 O were also observed, suggesting the possible involvement of a radical mechanism with the breakage of the C-N bond (Equation 8.2) as the first step. NCNO 2 was also found to be incompatible with fluorides, Lewis acids (BF 3 and B(C 6 F 5 ) 3 ), elemental mercury, amines and K 3 PO 4 . The observed vibrational spectra of NCNO 2 are in excellent agreement with the predicted ones (Table 8.1), and disagree with the alternative isocyanide isomer, CNNO 2 (APPENDIX 7), thus removing any doubt regarding the identity of the observed compound. The gas-phase IR spectrum (Figure 8.2) exhibits one C≡N stretching band at 2238 cm -1 , close to that of Cl-CN (2219 cm -1 ) 12 . The symmetric and antisymmetric ONO stretching modes at 1300 and 1580 cm -1 , respectively, are comparable to those of NO 2 (1325, 1610 cm -1 ). 12 In addition to the bands observed in the IR spectrum, the Raman spectra (Figure 8.3) show two bands at 216 and 267 cm -1 , which were out of the range the IR optics used. Also, the band at 1575 cm -1 is relatively broad in the liquid phase, but separates into two sharper bands in the solid state (Figure 8.3), indicative of some interaction in the condensed phase. 107 Figure 8.1: 14 N (230 K) and 13 C (238 K) NMR spectra of nitryl cyanide in SO 2 solution. The 13 C NMR spectrum shows one signal at 106.6 ppm, in the range of compounds such as Cl-CN (95.5) and HCN (111.5 ppm) 13 and features the expected 1:1:1 triplet pattern due to coupling with 14 N (I=1) with 1 J( 13 C- 14 N) = 25 Hz (Figure 8.1). In the 14 N NMR spectrum (Figure 8.1), however, this coupling could not be observed, as the quadrupole moment of 14 N broadens the signals. Both 14 N signals at -62.6 (NCNO 2 ) and -175 (NCNO 2 ) ppm are in good agreement with the calculated values (-55 and -168 ppm, respectively). The chemical shift for the nitro group (-62.6 ppm) is close to that in ClNO 2 (-70 ppm in CD 3 CN at -33 °C). The chemical shift for the cyanide group (-175 ppm) is relatively close to that of NCNH 2 (-188 ppm in CD 3 CN). Figure 8.2: Calculated and observed infrared spectra of NCNO 2 in the gas-phase. 108 Figure 8.3: Observed Raman spectra of NCNO 2 in the solid state and liquid phases, together with the calculated spectrum. Table 8.1: Fundamental vibrational modes of nitryl cyanide Vibrational assign. in C 2v symmetry calcd. freq. a (IR) [Ra] int. obsd. b IR (gas) obsd c Ra (liquid) ν C≡N (A 1) 2227(15) [10.0] 2238 w 2242 [10.0] ν asym NO 2 (B 2) ν sym NO 2 (A 1) ν C-N (A 1) δ wag NO 2 (B 1) δ rock NO 2 (B 2) δ scissor NO 2 (A 1) δ oop NCN (B 1) δ ip NCN (B 2) 1586(100) [1.0] 1300(33) [2.7] 891 (10) [0.7] 739 (2) [0.1] 600 (0) [0.0] 572 (1) [0.3] 278 (5) [0.1] 209 (2) [0.4] 1580 s 1300 m 890 m 720 w (≈590) [d] 576 vw x x 1577 [1.2]* 1304 [4.9] 896 [4.1] 721 [1.4] - 583 [1.4] 272 [1.0]* 216 [4.7] a B2PLYP/Def2-TZVPP frequencies (cm -1 ) scaled by 0.989. ip = in plane; oop = out of plane. B 1 modes are perpendicular to and B 2 modes are in the plane of the molecule. IR intensity (km/mol) and Raman activity (Å 4 /amu) are shown as fraction of the strongest band being 10. Raman activities were calculated at the B3LYP/aug-cc-pVTZ level. b +25 °C, ca 5 torr. “x” denotes out of range of instrument. Intensity: s = strong; m = medium; w = weak; vw = very weak c -90 °C, Raman intensities within brackets are expressed relative to the most intense band. *indicates broad or complex bands. d Estimated from the observed combination band of 216 + ca 590 at 802 cm -1 in the IR spectrum of the gas (APPENDIX 7). The close agreement between the NMR data for NCNO 2 and those previously attributed to CNNO 2 4 prompted the reinvestigation of the alleged synthesis of the latter. The nitration of HCN with NO 2 BF 4 was thus performed in Freon-11 at 0 °C. Aside from the 14 N NMR signal of unreacted HCN at -118.8 ppm, a sharp signal at -64.2 ppm and a broad signal at -171.9 ppm were observed, in agreement with the values reported by Klapötke et al. (-62.2, -120.0 and -171.4 ppm). 4 However, the chemical shifts of these signals are in poor agreement with the predicted values for CNNO 2 , -80 ppm (NO 2 ) and -164 ppm (CN), 109 suggesting that it had not been generated in the previously reported experiment, and that only the thermodynamically favored compound NCNO 2 had been produced. The Raman spectrum of the CFCl 3 solution at -60 °C and a gas phase IR spectrum of the purified reaction mixture revealed spectral features identical to those observed for isolated NCNO 2 . 8.3 Potential Applications Since NCNO 2 has an excellent oxygen balance (Equation 8.4) and a large positive heat of formation, it might be of interest as a high energy density material (HEDM). The specific impulse (I sp , rocket propellant performance) of NCNO 2 in vacuum was estimated, and a value of 343 s is predicted. This value can be compared to 240 s for hydrazine. It even exceeds that of a standard (NO 2 ) 2 /hydrazine bipropellant by 10 s. The potential applications of NCNO 2 in synthesis are many, and include homo- and co-polymerization resulting in energetic polymers, azide to nitrotetrazole transformation, and modification of various dienes through 1,3- and 1,4-dipolar cycloaddition reactions. Politzer has considered the cyclo-oligomerization of NCNO 2 as a pathway to cubane-like high energy density molecules. 14,15 The trimer of NCNO 2 , 2,4,6-trinitro-1,3,5-triazine (3) is particularly interesting (Figure 8.4). The first literature reference to the latter compound is over 106 years old, 16 however it has never been prepared. It is possible that it could be formed from NCNO 2 in a manner analogous to the known trimerization of other nitriles. 17 It has previously been predicted to be a powerful HEDM, with a density of 1.99 g/cm 3 . 18,19 Using the same approach as for NCNO 2 , the heats of formation were estimated for 2,4,6-trinitro-1,3,5-triazine: ΔH 0 f(3,gas) = 62.6 ± 2 kcal/mol, ΔH 3,vap ≈ 13.7 kcal/mol, ΔH 3,sub ≈ 18.0 kcal/mol and ΔH 0 f(3,s) = 44.6 ± 5 kcal/mol. Using the latter value, a detonation pressure of 42.1 GPa and a detonation velocity of 9770 m/s are predicted for 2,4,6-trinitro-1,3,5-triazine, which would outperform those of the standard high-explosive HMX by 13 and 6 %, respectively. Due to the high oxygen content of 2,4,6-trinitro-1,3,5-triazine, it might also have potential as an ingredient in rocket propellants. Several other higher oligomeric structures, such as the “boat” and “adamantane”-type tetramers are thermodynamically viable relative to 1, whereas more strained ones, such as the “cubane”-analogue, are not (APPENDIX 7). 110 Figure 8.4: Nitryl cyanide (1), Nitryl isocyanide (2) and 2,4,6-trinitro-1,3,5-triazine (3). Calculated relative enthalpies and Gibbs free energies are calculated at the M06-2X/cc-pVTZ level (kcal/mol, 1 atm, 298 K). Geometries (Å) are from M06-2X/cc-pVTZ and RI-B2PLYP/def2-TZVPP (*) optimizations. 8.4 Conclusion In conclusion, the elusive nitryl cyanide, NCNO 2 , has been synthesized and characterized. It was prepared in good yield, isolated by fractional condensation, and characterized by NMR and vibrational spectroscopy, and theoretical calculations. Nitryl cyanide holds promise as a high energy density material and might also prove useful as a HEDM building block. The simplicity and inherent stability of nitryl cyanide, together with the known multitude of nitriles in interstellar space 2,3 suggests that the compound might also be a potential candidate for observation in atmospheric and interstellar chemistry. Further experimental and computational details, a comprehensive account on the various synthesis and purification methods explored, vibrational analysis and an analysis of possible decomposition pathways can be found in APPENDIX 7. 8.5 References (1) Bélanger-Chabot, G.; Rahm, M.; Haiges, R.; Christe, K. O. Angew. Chem. Int. Ed. 2013, 52, 11002. (2) Chyba, C.; Sagan, C. Nature 1992, 355, 125. (3) Pilling, M. J. Astrochem Astrobiol. 2013, 73-113. (4) Klapötke, T. M.; McIntyre, G.; Schulz, A. J. Chem. Soc., Dalton Trans. 1996, 3237. (5) Korkin, A. A.; Leszczynski, J.; Bartlett, R. J. J. Phys. Chem. 1996, 100, 19840. (6) Zhang, J.-x.; Li, Z.-s.; Liu, J.-y.; Sun, C.-c. J. Phys. Chem. A 2005, 109, 10307. (7) Kaplan, R.; Shechter, H.; Castorina, T. C.; Tomlinson, W. R. In Inorg. Synth.; John Wiley & Sons, Inc.: 2007, p 52. (8) Ray, J. D.; Ogg, R. A. J. Chem. Phys. 1959, 31, 168. (9) Bak, B.; Nicolaisen, F. M.; Nielsen, O. J.; Skaarup, S. J. Mol. Struct. 1979, 51, 17. 111 (10) Nadler, I.; Pfab, J.; Reisler, H.; Wittig, C. J. Chem. Phys. 1984, 81, 653. (11) Rogers, D. W.; McLafferty, F. J. J. Chem. Phys. 1995, 103, 8302. (12) Nakamoto, K. Infrared and Raman Spectra of Inorganic and Coordination Compounds, Part A: Theory Applications in Inorganic Chemistry; Wiley Inter-Science, 1997. (13) Henderson, T. J.; Cullinan, D. B. Magn. Reson. Chem. 2007, 45, 954. (14) Politzer, P.; Pat, L.; Wiener, J. J. M. Carbocyclic and Heterocyclic Cage Compounds and Their Building Blocks: Synthesis, Structure, Mechanism, and Theory (Advances in Strained and Interesting Organic Molecules); JAI Press: Stamford, 1999. (15) Politzer, P.; Murray, J. S.; Seminario, J. M.; Lane, P.; Edward Grice, M.; Concha, M. C. Journal of Molecular Structure: THEOCHEM 2001, 573, 1. (16) Finger, H. J. Prakt. Chem. 1907, 75, 103. (17) Cairns, T. L.; Larchar, A. W.; McKusick, B. C. J. Am. Chem. Soc. 1952, 74, 5633. (18) Korkin, A. A.; Bartlett, R. J. J. Am. Chem. Soc. 1996, 118, 12244. (19) Li, J. Propellants, Explos., Pyrotech. 2008, 33, 443. 112 CHAPTER 9: SYNTHESIS OF FLUORODINITRAMINE 9.1 Introduction Trifluoramine, NF 3 , is a very stable compound and has been known for almost a century. It was first prepared in 1928 by Otto Ruff 1 by the electrolysis of NH 4 F/HF and is well characterized. 2 By contrast, trinitroamine, N(NO 2 ) 3 , is thermally unstable and decomposes above -40 °C. It has not been isolated as a neat material and has been identified by 14 N NMR spectroscopy and by several weak low-temperature infrared absorptions in CH 3 CN solution as a minor component in a complex reaction mixture. 3 Figure 9.1: The simple fluoro- and nitro-substituted amines Since the closely related -CF(NO 2 ) 2 group is more stable than the –C(NO 2 ) 3 group, it was interesting to synthesize and characterize mixed fluoronitroamines (Figure 9.1) and to explore whether trinitroamine can also be stabilized by partial fluorine substitution. Although alkylfluoronitroamines, RNF(NO 2 ), have been known for many years, 4 the unsubstituted fluoronitroamines, F n N(NO 2 ) (3-n) , where n = 1 or 2, have been unknown. Herein, the work on the synthesis of fluorodinitroamine, carried out during the past 19 years at the Edwards Air Force Base and at the University of Southern California, is summarized. In previous work, Christe et. al. have demonstrated the usefulness of NF 4 + salts for the oxidative fluorination of anions. These reactions involve the low-temperature metatheses of NF 4 SbF 6 with a cesium or potassium salt of an OX - anion, resulting in the formation of an intermediate thermally unstable [NF 4 + OX - ] salt, followed by the thermal decomposition of [NF 4 + OX - ] to NF 3 and the desired hypofluorite. In this manner, high yield syntheses of FOClO 3 , 5 FONO 2 , 6 FOSO 2 F, 7 FOTeF 5 , 8 and FOIOF 4 9,10 have been achieved. The preferred solvent for these reactions was anhydrous HF, and CsSbF 6 was the by- product of choice due to its low solubility in HF. If, however, the anion was incompatible with HF, other solvents, such as SO 2 , could also be used. For SO 2 , the use of potassium salts was preferred due to the lower solubility of KSbF 6 in this solvent. The intermediate formation of the NF 4 + OX - salts was established by the isolation of [NF 4 ][ClO 4 ] 5 and [NF 4 ][SO 3 F]. 7 Since the dinitramide anion, N(NO 2 ) 2 - , is well known, 11-16 it was a potential starting material for the synthesis of the yet unknown FN(NO 2 ) 2 molecule. 113 9.2 Synthesis and Properties Numerous experimental conditions for the synthesis of FN(NO 2 ) 2 were investigated, including the use of KN(NO 2 ) 2 or CsN(NO 2 ) 2 as starting materials, of NF 4 SbF 6 , NF 4 BF 4 , F 2 or FOSO 2 F as fluorinating agents, and of HF, SO 2 , CH 3 CN, C 2 H 5 CN, CH 3 NO 2 , SO 2 ClF, CH 2 Cl 2 , CHF 3 , or C 3 F 7 H as solvents. SO 2 ClF, CH 3 F, C 3 F 7 H, and CH 2 Cl 2 could not be used due to the low solubilities of the starting materials in these solvents at low temperature. With CH 3 NO 2 explosions were encountered at ambient temperature. The preferred combinations were KN(NO 2 ) 2 and NF 4 SbF 6 in either SO 2 solution at -64 °C or CH 3 CN solution at -30 °C. As with [NF 4 ][NO 3 ], 6 [NF 4 ][TeF 5 O] 8 and [NF 4 ][IF 4 O 2 ], 9,10 the [NF 4 ][N(NO 2 ) 2 ] intermediate could not be isolated due to its thermal instability. The insoluble alkali metal SbF 6 - salts could be filtered off at low temperature, weighed and identified by Raman spectroscopy. However, for the isolation of FN(NO 2 ) 2 the separation of the MSbF 6 precipitate from the other reaction products is not required. In SO 2 or CH 3 CN solutions, the dinitramide anion is directly fluorinated by the NF 4 + cation (Equation 9.1, M = K). NF 4 + SbF 6 - + M + N(NO 2 ) 2 - → NF 3 + FN(NO 2 ) 2 + MSbF 6 ↓ (9.1) The temperature dependence of the formation of FN(NO 2 ) 2 according to Equation 9.1 was followed by 19 F NMR spectroscopy (Figure 9.2). As can be seen, the formation of FN(NO 2 ) 2 starts already at -60 °C, proceeds well between -50 °C and -20 °C. The build-up of the FN(NO 2 ) 2 signal is accompanied by the disappearance of the NF 4 + signal and the build-up of the NF 3 signal. Above -20 °C, the concentration of FN(NO 2 ) 2 falls off due to the onset of decomposition. 114 Figure 9.2: 19 F NMR spectra of the KN(NO 2 ) 2 + NF 4 SbF 6 system in SO 2 solution recorded as a function of temperature. In addition, two minor side reactions are also observed. The fluorination of SO 2 (Equation 9.2), and the formation of a small amount of FONO 2 (Equation 9.3), due to the presence of a small amount of NO 3 - , a common impurity in dinitramide salts. NF 4 + SbF 6 - + SO 2 → NF 3 + SO 2 F 2 + SbF 5 (9.2) K + NO 3 - + NF 4 + SbF 6 - → NF 3 + FONO 2 + KSbF 6 (9.3) The formation of FN(NO 2 ) 2 in these solutions was also confirmed by low-temperature Raman spectroscopy. After removal of most of the SO 2 or CH 3 CN solvents by fractional condensation through - 64, -78 and -90 °C traps, the -78 and -90 °C traps contained the desired FN(NO 2 ) 2 and smaller amounts of solvent and N 2 O 4 . Complete removal of solvent and N 2 O 4 in this manner was not achieved. In HF solution, the reactions are much more complicated due to the fast reaction of the dinitramide anion with HF. The results from a detailed study of this system are beyond the scope of this paper and will be reported in a separate publication. For the synthesis of FN(NO 2 ) 2 , reactions of NF 4 SbF 6 with either KN(NO 2 ) 2 in SO 2 or CH 3 CN solutions or CsN(NO 2 ) 2 in HF solution were investigated (Equation 9.1, M= K, Cs). This reaction could be best controlled in SO 2 solution. The solvent was pumped off at -64 °C, and the volatile products, generated by the decomposition of the unstable intermediate [NF 4 ][N(NO 2 ) 2 ] between -60 °C and -56 °C, were pumped off at -78 °C trap. As can be seen from Figure 9.2, FN(NO 2 ) 2 is formed in SO 2 solution in 115 high yield. Several side reactions were also observed in this system. First of all, FN(NO 2 ) 2 starts to decompose already at relatively low temperature yielding N 2 O, NO 2 /N 2 O 4 , FNO 2 and some trans-N 2 F 2 , indicative of a decomposition mechanism involving NO 2 and NF radicals. These products were identified by their gas-phase infrared spectra. The formation of SO 2 F 2 was also observed by a slow attack of SO 2 by NF 4 + . Furthermore, FNO 2 can react with SO 2 resulting in the formation of solid [NO][SO 3 F], which was identified by its Raman spectrum. Its formation was verified in a separate experiment (Equation 9.4) FNO 2 + SO 2 → [NO][SO 3 F] (9.4) Although the in situ yield of FN(NO 2 ) 2 from the reaction in SO 2 solution, as determined by NMR spectroscopy, is nearly quantitative, the complete separation of FN(NO 2 ) 2 from the SO 2 solvent presents a major problem due to their similar volatilities and the thermal instability of FN(NO 2 ) 2 . To circumvent this separation problem, the possibility of using different solvents was investigated. As in the case of the recently discovered NCNO 2 17 (CHAPTER 8), compatibility, solubility, volatility and liquid range problems render this a very difficult task. For example, the use of CH 3 CN suffers from its higher melting point of -41°C and in some cases the formation of CH 3 COF. Thus, when a reaction of NF 4 BF 4 with KN(NO 2 ) 2 was carried out in this solvent at -22 °C, the CH 3 CN reacted with the nitramide anion and NF 4 + resulting in the formation of acetyl fluoride which was identified by its gas-phase IR spectrum. 18 When the reaction of KN(NO 2 ) 2 with NF 4 SbF 6 was carried out in CH 3 CN at -30 °C, no CH 3 COF was observed, however complete separation of FN(NO 2 ) 2 from the solvent and N 2 O 4 by fractional condensation was not achieved as shown by low-temperature Raman spectroscopy. Care must be taken to predissolve the separate reagents in CH 3 CN when scaling up the reaction, otherwise deflagration can occur. While the desired reaction also proceeds well in propionitrile, complete separation of FN(NO 2 ) 2 from the solvent by fractional condensation was not possible, because the C 2 H 5 CN was retained as a liquid in a trap warmer than that used for trapping the FN(NO 2 ) 2 and absorbed most of the FN(NO 2 ) 2 . The fluorination of N(NO 2 ) 2 - is not restricted to the use of NF 4 + salts as the fluorinating agent. For example, F 2 or FOSO 2 F were also used as the fluorinating agent, but these modifications did not alleviate the separation problems. Although the use of HF as a solvent suffers from its competing reactions with the dinitramide anion, an essentially pure sample of FN(NO 2 ) 2 could be isolated on one occasion from this system, which allowed its conclusive analysis by Raman spectroscopy. When KN(NO 2 ) 2 was combined with NF 4 SbF 6 in anhydrous HF at -78 °C, and all volatile products and the HF solvent were pumped off at -64 °C, the resulting residue was allowed to react further at -45 °C under a dynamic vacuum and FN(NO 2 ) 2 was trapped at -95 °C. Based on its low-temperature Raman spectrum (Figure 9.3) the resulting product was 116 essentially pure FN(NO 2 ) 2 containing a small amount of N 2 O 4 as the only detectable impurity. The exact nature of this reaction is only poorly understood and was difficult to duplicate. Figure 9.3: Raman spectrum of solid FN(NO 2 ) 2 recorded at -130 °C with the 4880-Å exciting line of an Ar-ion laser at two different attenuations; the bands marked by an asterisk are due to a small amount of N 2 O 4 . FN(NO 2 ) 2 is a colorless solid at low temperatures and melts at about -94 °C to a colorless liquid. Its composition was established by multinuclear NMR and Raman spectroscopy and quantum chemical calculations. 0 200 400 600 800 1000 1200 1400 1600 1800 Intensity Wavenumbers [cm -1 ] * * * * * * 117 Figure 9.4: 14 N NMR spectrum of FN(NO 2 ) 2 in SO 2 solution recorded at -43 °C; the weak signal at -19.7 ppm is due to a trace of N 2 O 4 . 19 The 19 F NMR spectrum (Figure 9.2) shows a single, somewhat broadened resonance at 53 ppm, in good agreement with those previously observed for the similar compounds F 2 NNF 2 (60.4 ppm), 20 FN=C(CN)(CH 3 ) (60 ppm), and FN=C(CF 3 ) 2 (48.3 ppm). 21 The 14 N NMR spectrum (Figure 9.4) showed a broad resonance at -6.1 ppm for the amidic nitrogen and a sharp doublet at -44.9 ppm with a 2 J( 14 N- 19 F) of 8 Hz for the nitro groups. The resonance for the amidic nitrogen (-6.1 ppm) is similar to that of -14.3 ppm found 22 for NF 3 in accord with the similar electronegativities of fluorine and the nitro group. The chemical shift for the nitro groups (-44.9 ppm) is similar to that observed by us at -46.3 ppm for HN(NO 2 ) 2 in diethylether. The 2 J( 14 N- 19 F) coupling constant of 8 Hz for the nitro groups is very similar to that of 9.8 Hz found for FC(NO 2 ) 3 . 23 The 1 J( 14 N- 19 F) coupling could not be observed in the 14 N spectrum due to the broadness of the amidic nitrogen resonance. Attempts to observe this 1 J coupling constant in the 15 N NMR spectrum of FN(NO 2 ) 2 with natural 15 N abundance were also unsuccessful. Overall, the 19 F and 14 N NMR spectra are in agreement with the proposed FN(NO 2 ) 2 structure. The Raman spectrum of solid FN(NO 2 ) 2 is shown in Figure 9.3, and the observed frequencies and their assignments are listed in Table 9.1. As can be seen, the agreement between observed and calculated vibrational spectra is very good, particularly when keeping in mind that the observed spectrum is for the solid where solid state effects can influence some of the frequencies and cause additional splittings of some of the bands. The only observed impurities in the spectrum were small amounts of the decomposition product N 2 O 4 . Thus, the Raman spectrum clearly establishes the identity of this compound as FN(NO 2 ) 2 . 118 Table 9.1 :Observed and calculated Raman spectra of FN(NO 2 ) 2 Vibrational assign. in C s symmetry, Approx. mode description Obsd. freq., rel. Ra int. a Calcd.freq. rel. Ra int. a,b A’ ν 1 ν as NO 2 ip c ν 2 ν sym NO 2 ip 1743 [1] 1351 [3], 1335 [2], 1332 [14] 1715 [16] 1329 [100] ν 3 ν NF ν 4 δ sciss NO 2 ip ν 5 ν sym N 3 ν 6 δ rock NF ν 7 δ sciss N 3 ν 8 δ rock NO 2 + δ rock NF ip ν 9 δ wag NO 2 ip ν 10  NO 2 ip A’’ ν 11 ν as NO 2 oop c ν 12 ν sym NO 2 oop ν 13 δ sciss NO 2 oop ν 14 ν asym N 3 ν 15 δ wag NF ν 16 δ rock NO 2 + δ rock NF oop ν 17 ν sym NO 2 oop ν 18  NO 2 oop lattice vibrations 1064 [1], 1054 [.5] 842 [2],825 [31] 794 [22] 615 [4], 606 [9] 430 sh, 425 [88] 329 [70] 207 [26], 200 [15] 77 [30] 1682 sh, 1680 [4], 1668 [.5] 1254 [1], 1249 [3] 758 [3] 680 sh, 675 [5] 593 [9], 589 [9] 337 [10] 324 [100] Not obsd. 165 [20], 99 [20], 87 [8] 1064 [8] 831 [63] 798 [14] 607 [15] 428 [86] 318 [24] 198 [16] 59 [9] 1693 [44] 1244 [7] 743 [0] 692 [3] 604 [17] 328 [3] 306 [50] (33) [5] a Frequencies in cm -1 ; uncorrected intensities based on peak heights in percent based on the most intense band being 100. b Calculated at the COSMO-mPW2PLYP/Def2-TZVPP level using individual anharmonicity corrections for each mode obtained by comparing harmonic and anharmonic B3LYP/aug-cc-pVTZ frequency calculations; Raman intensities were calculated at the B3LYP/aug-cc-pVTZ level of theory. c ip and oop stand for in phase and out of phase, respectively. In the absence of a crystal structure, the good agreement between the observed and calculated vibrational spectrum lends strong support to the minimum energy structure predicted by calculations for free gaseous FN(NO 2 ) 2 (Figure 9.5). The structure is derived from a pseudo-tetrahedron with the four ligand positions being occupied by two nitro groups, one fluorine atom and a sterically active free valence pair. The only symmetry element is a symmetry plane bisecting the fluorine ligand, the free valence electron pair and the central nitrogen atom. As can be seen from the N-N-N and F-N-N bond angles of 103.7° and 104.7°, respectively, the sterically active lone pair is arguably more voluminous than the remaining ligands compressing their bond angles from the ideal tetrahedral angle of 109.5° by about 5°. This view is supported by a quantum chemical topological study of the lone pair domain by the HELP method, 24 APPENDIX 8). Due to the planarity of the N(NO 2 ) group, the N-N-O bond angles are 113.9°. The predicted N-F bond length of 1.34 Å is similar but slightly shorter than that of 1.37 Å experimentally found for NF 3 , 25 and the N-O bond lengths are as expected for a normal NO 2 group. The N-N bonds of 1.55 Å are predicted to be considerably longer than those of 1.35-1.44 Å typically found in organic nitramines, 26 in accord with the decreased stability of FN(NO 2 ) 2 . Therefore, the predicted structure is in agreement with the observed decomposition mode, i.e., the strong N-F and N-O bonds and weak N-N 119 bonds result in an easy loss of NO 2 groups producing N 2 O 4 , and in the formation of NF radicals producing trans-N 2 F 2 . For a meaningful comparison of the trends within the NF 3 , F 2 N(NO 2 ), FN(NO 2 ) 2 , and N(NO 2 ) 3 series, it was necessary to calculate the structures of all the members at the same level of theory, since only the structure of NF 3 is experimentally known. 25 The results are shown in Figure 9.5. A comparison of the observed structure of NF 3 , rN-F = 1.37 Å, F-N-F = 102.1°, 25 with that predicted by us, rN-F = 1.36 Å, F-N-F = 101.9°, indicates that the predicted structures are good approximations to the actual structures. As can be seen from Figure 9.5, substitution of a fluorine ligand in NF 3 by a nitro group slightly shortens the N-F bonds. The N-N bonds in N(NO 2 ) 3 become increasingly longer and weaker with increasing fluorine substitution. Thus, the calculated N-N bond dissociation enthalpies for N(NO 2 ) 3 , FN(NO 2 ) 2 and F 2 N(NO 2 ) are 28.2, 22.1 and 14.2 kcal/mol, respectively, and F 2 N(NO 2 ) is predicted to be the least stable compound within this series. As expected, the tetrahedral angle is the smallest for NF 3 and increases with an increasing number of the more bulky nitro goups. Figure 9.5: Minimum energy structures of NF 3 , F 2 N(NO 2 ), FN(NO 2 ) 2 , and N(NO 2 ) 3 predicted at the M06- 2X/aug-cc-pVTZ level (bond lengths in Å) viewed along the sterically active free valence electron pair on the central nitrogen atom. 9.3 Conclusion FN(NO 2 ) 2 , one of the two missing mixed fluoronitroamines has successfully been prepared and characterized by multinuclear NMR and Raman spectroscopy. It is a thermally unstable compound that 120 readily decomposes to N 2 O 4 , trans-N 2 F 2 , N 2 O, and FNO 2 . It is shown that, in contrast to the closely related trinitromethyl compounds, fluorine substitution weakens the relatively labile N-N bonds in N(NO 2 ) 3 , and that the yet unknown F 2 N(NO 2 ) molecule will be even less stable than FN(NO 2 ) 2 , but might be accessible by low-temperature fluorination of the known FN(NO 2 ) - anion. 27 Further experimental and computational details, vibrational analysis and bond dissociation energies can be found in APPENDIX 8. 9.4 References (1) Ruff, O.; Fischer, J.; Luft, F. Z. Anorg. Allg. Chem. 1928, 172, 417. (2) For an extensive summary of the properties of NF 3 see the two volume USAF Propellant Handbook on Nitrogen Trifluoride AFRPL-TR-77-72, Aerojet Liquid Rocket Company, 28 October, 1977. (3) Rahm, M.; Dvinskikh, S. V.; Furó, I.; Brinck, T. Angew. Chem. 2011, 123, 1177. (4) Graff, M.; Gotzmer, C.; McQuistion, W. E. J. Org. Chem. 1967, 32, 3827. (5) Christe, K. O.; Wilson, W. W.; Wilson, R. D. Inorg. Chem. 1980, 19, 1494. (6) Hoge, B.; Christe, K. O. J. Fluorine Chem. 2001, 110, 87. (7) Christe, K. O.; Wilson, R. D.; Schack, C. J. Inorg. Chem. 1980, 19, 3046. (8) Schack, C. J.; Wilson, W. W.; Christe, K. O. Inorg. Chem. 1983, 22, 18. (9) Christe, K. O.; Wilson, R. D. Inorg. Nucl. Chem. Lett. 1979, 15, 375. (10) Christe, K. O.; Wilson, R. D.; Schack, C. J. Inorg. Chem. 1981, 20, 2104. (11) Bottaro, J. C.; Penwell, P. E.; Schmitt, R. J. Synth. Commun. 1991, 21, 945. (12) Gilardi, R.; Flippen-Anderson, J.; George, C.; Butcher, R. J. J. Am. Chem. Soc. 1997, 119, 9411. (13) Luk'yanov, O. A.; Anikin, O. V.; Gorelik, V. P.; Tartakovsky, V. A. Russ. Chem. Bull. 1994, 43, 1457. (14) Luk'yanov, O. A.; Gorelik, V. P.; Tartakovsky, V. A. Izv. Akad. Nauk Ser. Khim. 1994, 94. (15) Luk'yanov, O. A.; Konnova, Y. V.; Klimova, T. A.; Tartakovsky, V. A. Izv. Akad. Nauk Ser. Khim. 1994, 1264. (16) Titova, K. V. Russ. J. Inorg. Chem. 2002, 47, 1121. (17) Rahm, M.; Bélanger-Chabot, G.; Haiges, R.; Christe, K. O. Angew. Chem. Int. Ed. 2014, 53, 6893. (18) Shimanouchi, T. J. Phys. Chem. Ref. Data, 1977, 6, 1076. (19) Makhova, N. N.; Ovchinnikov, I. V.; Dubonos, V. G.; Strelenko, Y. A.; Khmel'nitsky, L. I. Russ. Chem. Bull. 1993, 42, 131. (20) Colburn, C. B.; Johnson, F. A.; Haney, C. J. Chem. Phys. 1965, 43, 4526. 121 (21) Berger, S.; Braun, S.; Kalinowski, H. O. NMR Spectroscopy of the Non-metallic Elements; John Wiley & Sons, 1997. (22) Mason, J.; Christe, K. O. Inorg. Chem. 1983, 22, 1849. (23) Grakauskas, V.; Baum, K. J. Org. Chem. 1968, 33, 3080. (24) Rahm, M.; Christe, K. O. ChemPhysChem 2013, 14, 3714. (25) Holleman-Wiberg, Inorganic Chemistry, Academic Press, English Edition 2001, p. 642. (26) Nazin, G. M.; Prokudin, V. G.; Dubikhin, V. V.; Aliev, Z. G.; Zbarskii, V. L.; Yudin, N. V.; Shastin, A. V. Russ. J. Gen. Chem. 2013, 83, 1071. (27) Bottaro, J. C.; Gilardi, R.; Penwell, P. E.; Petrie, M.; Malhotra, R. Synthesis 2007, 2007, 1151. 122 CHAPTER 10: SUMMARY AND OUTLOOK 10.1 Summary, Relevance of Results, and Research Outlook In CHAPTER 2, an improved method for the synthesis of 3,5-dinitro-1H-1,2,4-triazole (HDNT) was described. HDNT is one of several promising nitroazole building blocks, such as 5-amino-3-nitro-1H- 1,2,4-triazole (ANTA) 1 and 5-nitro-2H-tetrazole 2 (Figure 10.1). The purity of the HDNT synthesized allowed the structural characterization of the compound for the first time, as well as the study of its sensitivity and thermal stability, which showed that the material is moderately sensitive (IS 35 J; FS 155 N; decomposition onset 170 °C), demonstrating the desirable properties of the material. The compound is however highly hygroscopic, suggesting that salts and derivatives of HDNT might be more suitable for energetic material applications. The improved method yielded anhydrous HDNT, which allowed its use in reactions requiring anhydrous conditions (APPENDIX 3). The impurities identified throughout the synthesis helped developing improvements to the reported synthetic methods. In addition, the 5-azido-3- nitro-1,2,4-triazolate impurities, significantly more sensitive than HDNT, 3 identified in one particular version of a literature method, stressed the danger of using in situ-prepared diaminotriazole as a starting material. These experimental details will be of crucial importance in the event of a large-scale application of HDNT. Figure 10.1: HDNT along other useful nitroazole energetic building blocks In CHAPTER 3, the synthesis and full characterization of numerous salts of the 3,5-dinitro-1,2,4- triazolate anion (DNT), prepared from HDNT (CHAPTER 2). These salts include the complete alkali metal series, two salts of alkali earth metals, the ammonium and several organic ammonium salts, and bulky phosphonium salts. This also includes several salts of energetic cations, such as aminotetrazolium, guanidinium and aminoguanidinium. The diversity of salts studied showcased the coordination chemistry 123 of the DNT anion and the generally high thermal stability and very low sensitivity of most of its salts. These desirable properties should prompt further research for the development of potential applications. CHAPTER 4-CHAPTER 7 describe the synthesis of boron compounds bearing energetic ligands. Energetic boron compounds are generating interest because of the high heat of formation of boron oxide as a combustion product and its low environmental impact. Boron also produces a green flame when burning, which is an attractive property for pyrotechnic applications. In CHAPTER 4 the synthesis of several high-oxygen content nitroazolate-borane complex anion salts is described. While the salts display relatively low thermal stabilities, they demonstrate the feasibility of synthesizing BH 3 -azolate complexes with electron-withdrawing, oxidizing nitro groups. Moreover, the synthesized salts bring fundamental insight into the formation of other, more thermally and hydrolytically stable borohydride-containing species, [BH 2 (Azolyl) 2 ] - (where the azolyl moiety is DNT, 5-(fluorodinitomethyl)tetrazolate, 5-(trinitromethyl)tetrazolate, 5-nitrotetrazolate, 3-nitro-1,2,4-triazolate, 4-nitroimidazolate, 4,5-dinitroimidazolate, 3,4,5-trinitropyrazolate) 4-6 (Figure 10.2). Figure 10.2: Representative examples of [BH 2 (Azolyl) 2 ] - (Top: Haiges et al.; Bottom: Klapötke et al.). In CHAPTER 5, the synthesis and full characterization of a trinitromethylborate salt was described. This species is a proof-of-concept showing that stable compounds in which a strong oxidizer and a strong reducing moiety are directly bonded can be formed. This contrasts with the published report on 124 (trinitromethyl)chloroborates, which suggested that these compounds were unstable at room temperature. 7 The publication of this compound appears to have spurred the disclosure of closely related research, such as the synthesis of numerous neutral trinitromethylborane adducts and boronium salts of (trinitromethyl)trihydroborate. 8 The formation of trinitromethylborates is in sharp contrast with the absence of reactivity between the cyanodinitromethanide anion 9 and BH 3 . SMe 2 observed by the author of this dissertation. Although trinitromethyl compounds tend to have lower thermal stabilities, the large positive oxygen balance of the trinitromethyl moiety might warrant the search for more stable derivatives. CHAPTER 6 and CHAPTER 7 described the synthesis of boron-based dinitramide compounds, a previously undescribed type of compounds and further proofs that relatively stable compounds can be made where redox pairs are directly bonded in a single small molecule or anion. CHAPTER 6 described the synthesis of NH 3 . BH 2 [N(NO 2 ) 2 ] and the observation of the high oxygen content species NH 3 . BH[N(NO 2 ) 2 ] 2 . The compounds are only marginally stable in the solid state. These stability issues are however not due to the presence of a redox pair within the molecule, and the number of side-products hint at oligomerization reactions that would likely be worth exploring. The relative stability of the compounds is impressive, considering their rather large energy content. In CHAPTER 7 the detection of derivatives of [BH n (DN) 4-n ] - (n= 1,2,3) demonstrated the weakly coordinating character of dinitramide. Intriguing boron hydride species where observed as a result of this low coordinative power and further work in this field will likely yield more fascinating species. In CHAPTER 8 the synthesis of the spectacular nitryl cyanide, the fruit of nearly two years of work, was described. The very high reactivity of the molecule made its isolation a significant challenge and numerous reaction systems and conditions needed to be explored. The challenges were however met with proportional rewards. The isolation of nitryl cyanide in purities over 95 mol% revealed that its actual intrinsic stability is surpisingly high. In addition, it is the compound with the highest energy density ever predicted for a stable monopropellant. Its stability and its predicted very high performance makes further research into its large scale synthesis, stabilization and purification very attractive. Beyond its potential revolutionary impact on rocket propulsion, nitryl cyanide is an extremely reactive molecule and academically very interesting. Preliminary results obtained by the author of this dissertation show that it behaves as a source of cyanogen radicals (Figure 10.3). Controlled conditions could perhaps allow the cyano nitration of carbon-carbon multiple bonds (Figure 10.4). 125 Figure 10.3: NCNO 2 was shown to behave as a cyanogen radical source. Figure 10.4: Possible cyano nitration reaction of NCNO 2 at low temperature In CHAPTER 9, the synthesis of fluorodinitramide was described. This fundamental research effort yielded the first example of a mixed fluoronitroamine. Related compounds include trinitroamine, 10 fluoronitramide, 11 nitrogen trifluoride and the yet unknown difluoronitroamine. Although fluorodinitroamine and trinitroamine are too unstable for industrial applications, the knowledge acquired on these compounds is of relevance to the general field of nitroamines, which are an important class of explosophores (e.g. HMX and RDX). In addition, the synthesis of fluorodinitramide using NF 4 + as a fluorine source suggests the possibility of forming other halogen derivatives of dinitramide by direct halogenation using, for example, Cl 2 , Br 2 , I 2 or ICl, which has been shown to yield halogenotrinotromethane compounds in a similar fashion. 12,13 Further exciting work in this general field is conducted in the Christe-Haiges group by the author of this dissertation, Prof. Ralf Haiges, Dr. Martin Rahm and Thomas Saal, including an improved synthesis of trinitroamine (Figure 10.6) and the elucidation of the structure of dinitroamine, which are likely to yield more exciting results. Figure 10.5: (From left to right) fluorodinitroamine and the known trinitroamine, fluoronitramide, trifluoramine, and the still unknown difluoronitroamine. 126 Figure 10.6: Preliminary results indicate that significantly better yields of trinitroamine are obtained by this method than the published method, which involves the reaction of NO 2 BF 4 with KDN. 10.2 Conclusion It is my hope that the work presented in this dissertation will have laid the foundations for further syntheses of species which would have been deemed too unstable to deserve even exploratory work. The very recent publication of work on trinitromethylboranes and borates performed in the 1960s, in an apparent reaction to the publication of the work presented in CHAPTER 5, suggests that our work stimulated new interest in this type of compounds. Beyond the contribution of this dissertation to basic research in the field of energetic materials, the molecule nitryl cyanide, presented in CHAPTER 8, is predicted to have a spectacular performance and its stability warrants further research into potential applications. The creative research conducted by several groups worldwide, aided by the increasingly accurate predictions provided by skilled theoreticians, is likely to yield many more exciting compounds in this field. 10.3 References (1) Kofman, T. P. Russ. J. Org. Chem. 2002, 38, 1231. (2) Klapötke, T. M.; Miro Sabate, C.; Rasp, M. J. Mater. Chem. 2009, 19, 2240. (3) Izsák, D.; Klapötke, T. M. Crystals 2012, 2, 294. (4) Haiges, R.; Jones, C. B.; Christe, K. O. Inorg. Chem. 2013, 52, 5551. (5) Klapötke, T. M.; Rusan, M.; Sproll, V. Z. Anorg. Allg. Chem. 2013, 639, 2433. (6) Klapötke, T. M.; Rusan, M.; Sproll, V. Z. Anorg. Allg. Chem. 2014, 640, 1892. 127 (7) Titova, K. V.; Kolmakova, E. I.; Rosolovskii, V. Y. Bull. Acad. Sci. USSR, Div. Chem. Sci. (Engl. Transl.) 1977, 26, 1822. (8) Shitov, O. P.; Tartakovskii, V. A.; Ioffe, S. L. Chem Heterocycl Compd 2015, 50, 1647. (9) Parker, C. O.; Emmons, W. D.; Rolewicz, H. A.; McCallum, K. S. Tetrahedron 1962, 17, 79. (10) Rahm, M.; Dvinskikh, S. V.; Furó, I.; Brinck, T. Angew. Chem. Int. Ed. 2011, 50, 1145. (11) Bottaro, J. C.; Gilardi, R.; Penwell, P. E.; Petrie, M.; Malhotra, R. Synthesis 2007, 2007, 1151. (12) Macbeth, A. K.; Pratt, D. D. J. Chem. Soc., Trans. 1921, 119, 354. (13) Klapötke, T. M.; Krumm, B.; Moll, R.; Rest, S. F.; Vishnevskiy, Y. V.; Reuter, C.; Stammler, H.-G.; Mitzel, N. W. Chem. - Eur. J. 2014, 20, 12962. 128 BIBLIOGRAPHY Klapötke, T. M. Chemistry of High-energy Materials; de Gruyter: Berlin, 2012. Rahm, M. PhD dissertation, Royal Institute of Technology, 2010. Stierstorfer, J. PhD dissertation, Ludwig-Maximilian University, 2009. Brinck, T. Green Energetic Materials; Wiley: Chichester, 2014. Darwent, B. deB. Bond Dissociation Energies in Simple Molecules, National Standard Reference Data System, 1970. Talawar, M. B.; Agrawal, A. P.; Anniyappan, M.; Wani, D. S.; Bansode, M. K.; Gore, G. M. J. Hazard. Mater. 2006, 137, 1074. Stratton, H. "Theoretical Performance of Rocket Propellant Combinations", Advanced Programs, Rocketdyne Division, Rockwell International, 537A-2 rev. 2-88. Klapötke, T. M. High Energy Density Materials; Springer: Berlin, 2007; Vol. 125. Pearce, J. M. S. Eur. Neurol. 2007, 57, 118. Hartwig, A. In Cadmium: From Toxicity to Essentiality; Sigel, A., Sigel, H., Sigel, R. K. O., Eds.; Springer Netherlands: 2013; Vol. 11, p 491. Doig, A. T. Thorax 1976, 31, 30. Panz, K.; Miksch, K. J. Environ. Manage. 2012, 113, 85. Braverman, L. E.; He, X.; Pino, S.; Cross, M.; Magnani, B.; Lamm, S. H.; Kruse, M. B.; Engel, A.; Crump, K. S.; Gibbs, J. P. J. Clin. Endocrinol. Metab. 2005, 90, 700. Vernot, E. H.; MacEwen, J. D.; Bruner, R. H.; Haun, C. C.; Kinkead, E. R.; Prentice, D. E.; Hall III, A.; Schmidt, R. E.; Eason, R. L.; Hubbard, G. B.; Young, J. T. Fundam. Appl. Toxicol. 1985, 5, 1050. Shriver, D. F.; Drezdzon, M. A. The Manipulation of Air-Sensitive Compounds; J. Wiley and Sons: New York, 1986. Talawar, M. B.; Sivabalan, R.; Mukundan, T.; Muthurajan, H.; Sikder, A. K.; Gandhe, B. R.; Rao, A. S. J. Hazardous Mat. 2009, 161, 589. Langlet, A.; Oestmark, H. WO9402434A1 1994. Lee, K. Y.; Storm, C. B.; Hiskey, M. A.; Coburn, M. D. J. Energ. Mater. 1991, 9, 415. Simpson, R. L.; Pagoria, P. F.; Mitchell, A. R.; Coon, C. L. Propellants, Explos., Pyrotech. 1994, 19, 174. Zhang, Y.; Parrish, D. A.; Shreeve, J. n. M. J. Mater. Chem. A 2013, 1, 585. 129 Huynh, M. H. V. WO2008US01904 2008. Miller, C. G.; Williams, G. K. US2009199937 2009. Burchfield, H. P.; Gullstrom, D. K. US3054800 1962. Wiley, R. H.; Smith, N. R. US3111524 1963. Haiges, R.; Jones, C. B.; Christe, K. O. Inorg Chem 2013, 52, 5551. Klapötke, T. M.; Penger, A.; Pflüger, C.; Stierstorfer, J.; Sućeska, M. Eur. J. Inorg. Chem. 2013, 2013, 4667. Klapötke, T. M.; Piercey, D. G.; Stierstorfer, J. Dalton Trans. 2012, 41, 9451. Kofman, T. P. Russ. J. Org. Chem. 2001, 37, 1158. Pevzner, M. S.; Kofman, T. P.; Kibasova, E. N.; Sushchenko, L. F.; Uspenskaya, T. L. Chem Heterocycl Compd 1980, 16, 194. Stinecipher, M. M. Investigation of the Physical and Explosives Properties of the Eutectic Explosive Ammonium Nitrate/Ammonium 3,5-Dinitro-1,2,4-Triazolate; 1978 Annual Report, Los Alamos National Laboratory, 1978. Haiges, R.; Bélanger-Chabot, G.; Kaplan, S. M.; Christe, K. O. Dalton Trans 2015, 44, 2978. Garcia, E.; Lee, K.-Y. Acta Crystallogr. Sect. C: Cryst. Struct. Commun. 1992, 48, 1682. Izsák, D.; Klapötke, T. M. Crystals 2012, 2, 294. Bian, C.; Zhang, M.; Li, C.; Zhou, Z. J. Mater. Chem. A 2015, 3, 163. Larina, L.; Lopyrev, V. Nitroazoles: Synthesis, Structure and Applications; Springer: Dordrecht ; New York, 2009. Chernyshev, V. M.; Zemlyakov, N. D.; Il'in, V. B.; Taranushich, V. A. Russ. J. Appl. Chem. 2000, 73, 839. Roemer, J. J.; Kaiser, D. W. US2648671 1953. Thiele, J. Chem. Ber. 1908, 41, 2681. Miller, C. G. W., Graylon K. US2009199937 2009. Bélanger-Chabot, G.; Rahm, M.; Haiges, R.; Christe, K. O. Angew. Chem. Int. Ed. 2013, 52, 11002. Haiges, R.; Christe, K. O. Inorg. Chem. 2013, 52, 7249. Rahm, M.; Bélanger-Chabot, G.; Haiges, R.; Christe, K. O. Angew. Chem. Int. Ed. 2014, 53, 6893. 130 Huynh, M. H. V.; Hiskey, M. A.; Gilardi, R. J. Energ. Mater. 2005, 23, 27. Xue, H.; Gao, H.; Shreeve, J. M. J. Polym. Sci., Part A: Polym. Chem. 2008, 46, 2414. Hiskey, M. A.; Stinecipher, M. M.; Brown, J. E. J. Energ. Mater. 1993, 11, 157. Katritzky, A. R.; Singh, S.; Kirichenko, K.; Holbrey, J. D.; Smiglak, M.; Reichert, W. M.; Rogers, R. D. Chem. Commun. 2005, 868. Katritzky, A. R.; Singh, S.; Kirichenko, K.; Smiglak, M.; Holbrey, J. D.; Reichert, W. M.; Spear, S. K.; Rogers, R. D. Chem. - Eur. J. 2006, 12, 4630. Klapötke, T. M. P., A.;Schedlbauer, S.;Stierstorfer, J. Cent. Eur. J. Energ. Mater. 2013, 10, 151. Lee, K. Y.; Ott, D. G.; Stinecipher, M. M. Ind. Eng. Chem. Process Des. Dev. 1981, 20, 358. Sitzmann, M. E. J. Org. Chem. 1978, 43, 3389. Smiglak, M.; Hines, C. C.; Wilson, T. B.; Singh, S.; Vincek, A. S.; Kirichenko, K.; Katritzky, A. R.; Rogers, R. D. Chem. Eur. J. 2010, 16, 1572. Tao, G.-H.; Guo, Y.; Parrish, D. A.; Shreeve, J. M. J. Mater. Chem. 2010, 20, 2999. Xue, H.; Gao, H.; Twamley, B.; Shreeve, J. M. Eur. J. Inorg. Chem. 2006, 2006, 2959. Chernyshev, V. M. Z., N. D.; Taranushich, V. A. RU2174120 C1 2001. Melnikov, V. V.; Stolpakova, V. V.; Khorkova, L. F.; Pevzner, M. S.; Melnikova, N. N. Chem. Heterocycl. Comp. 1972, 8, 111. Melnikov, V. V.; Stolpakova, V. V.; Zacheslavskii, S. A.; Pevzner, M. S.; Gidaspov, B. V. Chem. Heterocycl. Comp. 1973, 9, 1289. Titova, K. V.; Kolmakova, E. I.; Rosolovskii, V. Y. Bull. Acad. Sci. USSR, Div. Chem. Sci. (Engl. Transl.) 1977, 26, 1822. Titova, K. V. Russ. J. Inorg. Chem. 2002, 47, 1121. Ioffe, S. L. S., O. P.; Tartakowskij, V. A. "The Unusual Reation of Trinitromethane with BH 3 . THF- complex" IUPAC IMEBORON VIII, Knoxville, Tennessee, July 11-15, 1993. Guibert, C. R.; Marshall, M. D. J. Am. Chem. Soc. 1966, 88, 189. Klapötke, T. M.; Rusan, M.; Sproll, V. Z. Anorg. Allg. Chem. 2014, 640, 1892. Snyder, C. J.; Martin, P. D.; Heeg, M. J.; Winter, C. H. Chem. Eur. J. 2013, 19, 3306. Janiak, C. J. Chem. Soc., Chem. Commun. 1994, 545. Janiak, C.; Scharmann, T. G.; Green, J. C.; Parkin, R. P. G.; Kolm, M. J.; Riedel, E.; Mickler, W.; Elguero, J.; Charamunt, R. M.; Sanz, D. Chem. - Eur. J. 1996, 2, 992. 131 Lu, D.; Winter, C. H. Inorg. Chem. 2010, 49, 5795. Groshens, T. J. J. Coord. Chem. 2010, 63, 1882. Wang, Y.-L.; Cao, R.; Bi, W.-H. Polyhedron 2005, 24, 585. Koch, E.-C.; Klapötke, T. M. Propellants, Explos., Pyrotech. 2012, 37, 335. Klapötke, T. M.; Krumm, B.; Moll, R. Chem. - Eur. J. 2013, 19, 12113. Klapötke, T. M.; Krumm, B.; Rusan, M.; Sabatini, J. J. Chem. Commun. 2014, 50, 9581. Klapötke, T. M.; Rusan, M.; Sproll, V. Z. Anorg. Allg. Chem. 2013, 639, 2433. Vasiliu, M.; Arduengo, A. J.; Dixon, D. A. J. Phys. Chem. C 2012, 116, 22196. Kofman, T. P. Russ. J. Org. Chem. 2002, 38, 1231. Haiges, R.; Bélanger-Chabot, G.; Kaplan, S. M.; Christe, K. Dalton Trans. 2015, 44, 2978. Goiko, E. A.; Grigor’eva, N. V.; Margolis, N. V.; Mel’nikov, A. A.; Strochkina, T. K.; Tselinskii, I. V. Zh. Strukt. Khim. 1980, 21, 177−180. Klapötke, T. M.; Steemann, F. X. Propellants, Explos., Pyrotech. 2010, 35, 114. Nöth, H.; Vahrenkamp, H. Chem. Ber. 1966, 99, 1049. Hantzsch, A.; Rinckenberger, A. Chem. Ber. 1899, 32, 628. Novikov, S. S. G., T.I.; Tartakovskii, V.A., Izv. Akad. Nauk SSSR, Otd. Khim. Nauk 1960, 863. Ioffe, S. L.; Mkarenkova, L. M.; Shitkin, V. M.; Kashutina, M. V.; Tartakovskii, V. A. Izv. Akad. Nauk SSSR, Ser. Khim. 1973, 1, 203. Erashko, V. I.; Shevelev, S. A. Izv. Akad. Nauk SSSR, Ser. Khim. 1985, 2, 439. Thottempudi, V.; Kim, T. K.; Chung, K.-H.; Kim, J. S. Bull. Korean Chem. Soc. 2009, 30, 2152. Thottempudi, V.; Gao, H.; Shreeve, J. M. J. Am. Chem. Soc. 2011, 133, 6464. Macbeth, A. K.; Orr, W. B. J. Chem. Soc. 1932, 0, 534. Hammond, G. S.; Emmons, W. D.; Parker, C. O.; Graybill, B. M.; Waters, J. H.; Hawthorne, M. F. Tetrahedron 1963, 19, Supplement 1, 177. Gakh, A. A.; Bryan, J. C.; Burnett, M. N.; Bonnesen, P. V. J. Mol. Struct. 2000, 520, 221. Göbel, M.; Klapötke, T. M. Z. Anorg. Allg. Chem. 2007, 633, 1006. Kaplan, L. A.; Kamlet, M. J. J. Org. Chem. 1962, 27, 780. 132 Erashko, V. I.; Shevelev, S. A.; A.A., F. b. Izv. Akad. Nauk SSSR, Ser. Khim. 1968, 9, 2117. Shevelev, S. A.; Erashko, V. I.; Sankov, B. G.; Fainzil'berg, A. A. Izv. Akad. Nauk SSSR, Ser. Khim. 1968, 2, 382. Gidaspov, A. A.; Bakharev, V. V.; Kukushkin, I. K. Russ. Chem. Bull. 2009, 58, 2154. Huang, Y.; Gao, H.; Twamley, B.; Shreeve, J. n. M. Eur. J. Inorg. Chem. 2007, 2007, 2025. Holleman, A. F.; Wiberg, E. In Inorg. Chem.; First English ed.; Wiberg, N., Ed.; ACADEMIC PRESS: San Diego, 2001, p 131. M.W., Jr, NIST-JANAF Thermochemical Tables Fourth Editions, J. Phys Chem Ref. Data, Monograph 9, 1998, 1-1951. Luk'yanov, O. A.; Anikin, O. V.; Gorelik, V. P.; Tartakovsky, V. A. Russ. Chem. Bull. 1994, 43, 1457. Grant, D. J.; Dixon, D. A.; Camaioni, D.; Potter, R. G.; Christe, K. O. Inorg. Chem. 2009, 48, 8811. Novikov, S. S.; Slovetskii, V. I.; Shevelev, S. A.; Fainzil'berg, A. A. Bull. Acad. Sci. USSR, Div. Chem. Sci. (Engl. Transl.) 1962, 11, 552. Göbel, M.; Klapötke, T. M. Acta Crystallogr. Sect. C: Cryst. Struct. Commun. 2007, 63, o562. Schodel, H.; Dienelt, R.; Bock, H. Acta Crystallogr. Sect. C: Cryst. Struct. Commun. 1994, 50, 1790. Tartakovskii, V. A.; Nikonova, L. A.; Novikov, S. S. Izv. Akad. Nauk SSSR, Ser. Khim. 1967, 706. Scherfise, K. D.; Weller, F.; Dehnicke, K. Z. Naturforsch., B: Chem. Sci. 1985, 40b, 906. Young, D. E.; McAchran, G. E.; Shore, S. G. J. Am. Chem. Soc. 1966, 88, 4390. Singaram, B.; Cole, T. E.; Brown, H. C. Organometallics 1984, 3, 774. Mason, J. In Multinuclear NMR; Mason, J., Ed.; Plenum Press: New York 1987, p 227. Coburn, M. D.; Storm, C. B.; Moore, D. W.; Archibald, T. G. Magn. Reson. Chem. 1990, 28, 16. Göbel, M.; Klapötke, T. M.; Mayer, P. Z. Anorg. Allg. Chem. 2006, 632, 1043. Smith, E.; Dent, G. In Modern Raman Spectroscopy A Practical Approach; Wiley: West Sussex, 2005, p 23. Shlyapochnikov, V. A.; Oleneva, G. I.; Novikov, S. S. Russ Chem Bull 1971, 20, 2477. Brookes, M. J.; Jonathan, N. J. Chem. Soc. A 1968, 2266. Haiges, R.; Boatz, J. A.; Vij, A.; Gerken, M.; Schneider, S.; Schroer, T.; Christe, K. O. Angew. Chem. Int. Ed. 2003, 42, 5847. 133 Haiges, R.; Boatz, J. A.; Schneider, S.; Schroer, T.; Yousufuddin, M.; Christe, K. O. Angew. Chem. Int. Ed. 2004, 43, 3148. Grant, D. J.; Matus, M. H.; Switzer, J. R.; Dixon, D. A.; Francisco, J. S.; Christe, K. O. J. Phys. Chem. A 2008, 112, 3145. Murray, J. S.; Brinck, T.; Politzer, P. Chem. Phys. 1996, 204, 289. Politzer, P.; Murray, J. S. J. Phys. Chem. A 1998, 102, 1018. Rice, B. M.; Pai, S. V.; Hare, J. Combust. Flame 1999, 118, 445. Luk'yanov, O. A.; Gorelik, V. P.; Tartakovskii, V. A. Russ. Chem. Bull. 1994, 43, 89. Bottaro, J. C.; Penwell, P. E.; Schmitt, R. J. Synth. Commun. 1991, 21, 945. Klapötke, T. M.; Krumm, B.; Scherr, M. Z. Anorg. Allg. Chem. 2009, 635, 885. Luk'yanov, O. A.; Shlykova, N. I.; Tartakovsky, V. A. Russ. Chem. Bull. 1994, 43, 1680. Christe, K. O.; Wilson, W. W.; Bélanger-Chabot, G.; Haiges, R.; Boatz, J. A.; Rahm, M.; Prakash, G. K. S.; Saal, T.; Hopfinger, M. Angew. Chem. Int. Ed. 2015, 54, 1316. Rahm, M.; Dvinskikh, S. V.; Furó, I.; Brinck, T. Angew. Chem. Int. Ed. 2011, 50, 1145. Ziparo, C.; Colognesi, D.; Giannasi, A.; Zoppi, M. J. Phys. Chem. A 2012, 116, 8827. Klapötke, T. M.; Krumm, B.; Scherr, M. Eur. J. Inorg. Chem. 2008, 2008, 4413. Chyba, C.; Sagan, C. Nature 1992, 355, 125. Pilling, M. J. Astrochem Astrobiol. 2013, 73-113. Klapötke, T. M.; McIntyre, G.; Schulz, A. J. Chem. Soc., Dalton Trans. 1996, 3237. Korkin, A. A.; Leszczynski, J.; Bartlett, R. J. J. Phys. Chem. 1996, 100, 19840. Zhang, J.-x.; Li, Z.-s.; Liu, J.-y.; Sun, C.-c. J. Phys. Chem. A 2005, 109, 10307. Kaplan, R.; Shechter, H.; Castorina, T. C.; Tomlinson, W. R. In Inorg. Synth.; John Wiley & Sons, Inc.: 2007, p 52. Ray, J. D.; Ogg, R. A. J. Chem. Phys. 1959, 31, 168. Bak, B.; Nicolaisen, F. M.; Nielsen, O. J.; Skaarup, S. J. Mol. Struct. 1979, 51, 17. Nadler, I.; Pfab, J.; Reisler, H.; Wittig, C. J. Chem. Phys. 1984, 81, 653. Rogers, D. W.; McLafferty, F. J. J. Chem. Phys. 1995, 103, 8302. 134 Nakamoto, K. Infrared and Raman Spectra of Inorganic and Coordination Compounds, Part A: Theory Applications in Inorganic Chemistry; Wiley Inter-Science, 1997. Henderson, T. J.; Cullinan, D. B. Magn. Reson. Chem. 2007, 45, 954. Politzer, P.; Pat, L.; Wiener, J. J. M. Carbocyclic and Heterocyclic Cage Compounds and Their Building Blocks: Synthesis, Structure, Mechanism, and Theory (Advances in Strained and Interesting Organic Molecules); JAI Press: Stamford, 1999. Politzer, P.; Murray, J. S.; Seminario, J. M.; Lane, P.; Edward Grice, M.; Concha, M. C. Journal of Molecular Structure: THEOCHEM 2001, 573, 1. Finger, H. J. Prakt. Chem. 1907, 75, 103. Cairns, T. L.; Larchar, A. W.; McKusick, B. C. J. Am. Chem. Soc. 1952, 74, 5633. Korkin, A. A.; Bartlett, R. J. J. Am. Chem. Soc. 1996, 118, 12244. Li, J. Propellants, Explos., Pyrotech. 2008, 33, 443. Ruff, O.; Fischer, J.; Luft, F. Z. Anorg. Allg. Chem. 1928, 172, 417. USAF Propellant Handbook on Nitrogen Trifluoride AFRPL-TR-77-72, Aerojet Liquid Rocket Company, 28 October, 1977. Rahm, M.; Dvinskikh, S. V.; Furó, I.; Brinck, T. Angew. Chem. 2011, 123, 1177. Graff, M.; Gotzmer, C.; McQuistion, W. E. J. Org. Chem. 1967, 32, 3827. Christe, K. O.; Wilson, W. W.; Wilson, R. D. Inorg. Chem. 1980, 19, 1494. Hoge, B.; Christe, K. O. J. Fluorine Chem. 2001, 110, 87. Christe, K. O.; Wilson, R. D.; Schack, C. J. Inorg. Chem. 1980, 19, 3046. Schack, C. J.; Wilson, W. W.; Christe, K. O. Inorg. Chem. 1983, 22, 18. Christe, K. O.; Wilson, R. D. Inorg. Nucl. Chem. Lett. 1979, 15, 375. Christe, K. O.; Wilson, R. D.; Schack, C. J. Inorg. Chem. 1981, 20, 2104. Gilardi, R.; Flippen-Anderson, J.; George, C.; Butcher, R. J. J. Am. Chem. Soc. 1997, 119, 9411. Luk'yanov, O. A.; Gorelik, V. P.; Tartakovsky, V. A. Izv. Akad. Nauk Ser. Khim. 1994, 94. Luk'yanov, O. A.; Konnova, Y. V.; Klimova, T. A.; Tartakovsky, V. A. Izv. Akad. Nauk Ser. Khim. 1994, 1264. Shimanouchi, T. J. Phys. Chem. Ref. Data, 1977, 6, 1076. 135 Makhova, N. N.; Ovchinnikov, I. V.; Dubonos, V. G.; Strelenko, Y. A.; Khmel'nitsky, L. I. Russ. Chem. Bull. 1993, 42, 131. Colburn, C. B.; Johnson, F. A.; Haney, C. J. Chem. Phys. 1965, 43, 4526. Berger, S.; Braun, S.; Kalinowski, H. O. NMR Spectroscopy of the Non-metallic Elements; John Wiley & Sons, 1997. Mason, J.; Christe, K. O. Inorg. Chem. 1983, 22, 1849. Grakauskas, V.; Baum, K. J. Org. Chem. 1968, 33, 3080. Rahm, M.; Christe, K. O. ChemPhysChem 2013, 14, 3714. Holleman-Wiberg, Inorganic Chemistry, Academic Press, English Edition 2001, p. 642. Nazin, G. M.; Prokudin, V. G.; Dubikhin, V. V.; Aliev, Z. G.; Zbarskii, V. L.; Yudin, N. V.; Shastin, A. V. Russ. J. Gen. Chem. 2013, 83, 1071. Bottaro, J. C.; Gilardi, R.; Penwell, P. E.; Petrie, M.; Malhotra, R. Synthesis 2007, 2007, 1151. Klapötke, T. M.; Miro Sabate, C.; Rasp, M. J. Mater. Chem. 2009, 19, 2240. Haiges, R.; Boatz, J. A.; Yousufuddin, M.; Christe, K. O. Angew. Chem. Int. Ed. 2007, 46, 2869. Sheldrick, G. Acta Crystallogr. Sect. A: Found. Crystallogr. 2015, 71, 3. Farrugia, L. J. Appl. Crystallogr. 1997, 30, 565. Haiges, R.; Boatz, J. A.; Schneider, S.; Schroer, T.; Yousufuddin, M.; Christe, K. O. Angew. Chem. 2004, 116, 3210. Chernyshev, V. M. Z., N. D.; Taranushich, V. A. RU Patent 2174120 C1 2001. Sheldrick, G. Acta Crystallogr. Sect. C: Cryst. Struct. Commun. 2015, 71, 3. Farrugia, L. J. Appl. Crystallogr. 2012, 45, 849. Burchfield, H.P.; Gullstrom, D.K. US Patent US3054800 A, 1962. V. M. Chernysev, N. D. Z., V. B. Il'in, V. A. Taranushich Russ. J. Appl. Chem. 2000, 73, 839. Thottempudi, V.; Gao, H.; Shreeve, J. n. M. J. Am. Chem. Soc. 2011, 133, 6464. Lingam, H. K.; Chen, X.; Zhao, J.-C.; Shore, S. G. Chem. - Eur. J. 2012, 18, 3490. Grimme, S. J. Chem. Phys. 2006, 124, 034108. Weigend, F.; Ahlrichs, R. Phys. Chem. Chem. Phys. 2005, 7, 3297-3305. Neese, F. Wiley Interdisciplinary Reviews: Computational Molecular Science 2012, 2, 73. 136 Christe, K. O.; Wilson, W. W.; Petrie, M. A.; Michels, H. H.; Bottaro, J. C.; Gilardi, R. Inorg. Chem. 1996, 35, 5068. Christe, K. O.; Wilson, W. W.; Sheehy, J. A.; Boatz, J. A. Angew. Chem. Int. Ed. 1999, 38, 2004. Bryce, D. L.; Wasylishen, R. E. Inorg. Chem. 2002, 41, 4131. Seyferth, D.; Kahlen, N. J. Org. Chem. 1960, 25, 809. Christe, K. O.; Wilson, W. W.; Wilson, R. D.; Bau, R.; Feng, J. A. J. Am. Chem. Soc. 1990, 112, 7619. Grimme, S. J. Chem. Phys. 2006, 124, 034108. Eichkorn, K.; Treutler, O.; Öhm, H.; Häser, M.; Ahlrichs, R. Chem. Phys. Lett. 1995, 240, 283. Weigend, F.; Ahlrichs, R. PCCP 2005, 7, 3297. Neese, F.; Schwabe, T.; Grimme, S. J. Chem. Phys. 2007, 126, 124115. Schwabe, T.; Grimme, S. PCCP 2006, 8, 4398. Frisch, M. J. et al., Gaussian 09, Revision A.02, Gaussian, Inc., Wallingford CT, 2009. Valero, R.; Gomes, J. R. B.; Truhlar, D. G.; Illas, F. J. Chem. Phys. 2008, 129, 124710. Zhao, Y.; Truhlar, D. G. Theor. Chem. Acc. 2008, 120, 215. Goerigk, L.; Grimme, S. PCCP 2011, 13, 6670. Zhao, Y.; Truhlar, D. G. J. Chem. Theory Comput. 2011, 7, 669. Montgomery, J. J. A.; Frisch, M. J.; Ochterski, J. W.; Petersson, G. A. J. Chem. Phys. 1999, 110, 2822. Montgomery, J. J. A.; Frisch, M. J.; Ochterski, J. W.; Petersson, G. A. J. Chem. Phys. 2000, 112, 6532. Guner, V.; Khuong, K. S.; Leach, A. G.; Lee, P. S.; Bartberger, M. D.; Houk, K. N. J. Phys. Chem. A 2003, 107, 11445. Ess, D. H.; Houk, K. N. J. Phys. Chem. A 2005, 109, 9542. Brinck, T. Hardsurf program. (HS95v09), 2009. Stanton, J. F.; Bartlett, R. J. J. Chem. Phys. 1993, 98, 7029. Kállay, M.; Gauss, J. J. Chem. Phys. 2004, 121, 9257. Chai, J.-D.; Head-Gordon, M. PCCP 2008, 10, 6615. Sorin, B.; Laurence, E. F.; Kurt, R. G.; Howard, W. M.; I-Feng, W. K.; Souers, P. C.; Vitello P. A., Cheetah 7.0, rev2280, Lawrence Livermore National Laboratory, 2012. 137 Klapötke, T. M.; Schulz, A. Inorg. Chem. 1996, 35, 7897. Christe, K. O.; Wilson, W. W.; Schack, C. J.; Wilson, R. D.; Bougon, R. In Inorg. Synth.; John Wiley & Sons, Inc.: 2007, p 39. Christe, K. O.; Wilson, W. W.; Schack, C. J. J. Fluorine Chem. 1978, 11, 71. Christe, K. O.; Schack, C. J.; Wilson, R. D. J. Fluorine Chem. 1976, 8, 541. Kohout, M., Dgrid, version 4.6, 2011. Bader, R. F. W. Atoms in Molecules: A Quantum Theory; Clarendon Press: Oxford, 1990. Kohout, M.; Savin, A. Int. J. Quantum Chem 1996, 60, 875. Glendenig, E. D.; Landis, C. R.; Weinhold, F. Wiley Interdiscip. Rev.: Comput. Mol. Sci. 2012, 2, 1. Reed, A. E.; Curtiss, L. A.; Weinhold, F. Chem. Rev. 1988, 88, 899. Reed, A. E.; Weinstock, R. B.; Weinhold, F. J. Chem. Phys. 1985, 83, 735. Rahm, M.; Brinck, T. Chem. Phys. 2008, 348, 53. Rahm, M.; Brinck, T. Chem. Eur. J. 2010, 16, 6590. Rahm, M.; Brinck, T. Chem. Commun. 2009, 2896. Rahm, M.; Brinck, T. J. Phys. Chem. A 2010, 114, 2845. Wiberg, K. B. Tetrahedron 1968, 24, 1083. Haiges, R.; Christe, K. O. Dalton Trans. 2015, 44, 10166. Parker, C. O.; Emmons, W. D.; Rolewicz, H. A.; McCallum, K. S. Tetrahedron 1962, 17, 79. Klapötke, T. M.; Krumm, B.; Moll, R.; Rest, S. F.; Vishnevskiy, Y. V.; Reuter, C.; Stammler, H.-G.; Mitzel, N. W. Chem. - Eur. J. 2014, 20, 12962. Macbeth, A. K.; Pratt, D. D. J. Chem. Soc., Trans. 1921, 119, 354. Ehemann, M.; Davies, N.; Nöth, H. Z. Anorg. Allg. Chem. 1972, 389, 235. Sheldrick, G. Acta Crystallogr. Sect. A: Found. Crystallogr. 2008, 64, 112. Guner, V. A.; Khuong, K. S.; Houk, K. N.; Chuma, A.; Pulay, P. J. Phys. Chem. A 2004, 108, 2959. Ess, D. H.; Houk, K. N. J. Phys. Chem. A 2005, 109, 9542. Marenich, A. V.; Cramer, C. J.; Truhlar, D. G. J. Phys. Chem. B 2009, 113, 6378. 138 APPENDIX 1: ADDITIONAL INFORMATION ON IMPROVED SYNTHESIS AND STRUCTURAL CHARACTERIZATION OF 3,5-DINITRO-1H-1,2,4- TRIAZOLE AND IDENTIFICATION OF IMPORTANT IMPURITIES (CHAPTER 2) A1.1 Experimental Details Caution! The compounds of this work are energetic materials that might explode under certain conditions (e.g., elevated temperature, impact, friction or electric discharge). Appropriate safety precautions, such as the use of shields or barricades in a fume hood and personal protection equipment (safety glasses, face shields, ear plugs, as well as gloves and suits made from leather and/or Kevlar) 1 should be taken at all times when handling these materials. Pure HDNT decomposes explosively when heated above 160 °C, and certain impurities might further lower the decomposition temperature. The sublimation of HDNT should be carried-out behind a blast shield and only on a small scale. Ignoring safety precautions may lead to serious injuries! Materials and apparatus. All chemicals and solvents were obtained from Sigma-Aldrich or Alfa-Aesar and were used as supplied. NMR spectra were recorded at 298 K on Bruker AMX500 or Varian VNMRS- 600s spectrometers using (CD 3 ) 2 CO or D 2 O solutions in standard 5 mm o.d. glass tubes. Chemical shifts are given relative to neat tetramethylsilane ( 1 H, 13 C) or neat CH 3 NO 2 ( 14 N, 15 N). Raman spectra were recorded at ambient temperatures in Pyrex glass tubes in the range of 4000–80 cm -1 on a Bruker Equinox 55 FT-RA spectrometer using a Nd-YAG laser at 1064 nm or a Cary 83 spectrometer using an Ar laser at 488 nm. Infrared spectra were recorded in the range 4000-400 cm -1 on a Midac, M Series spectrometer using KBr pellets or on a Bruker Optics Alpha FT-IR ATR spectrometer. KBr pellets were prepared carefully using an Econo mini-press (Barnes Engineering Co.). Differential thermal analysis (DTA) curves were recorded with a purge of dry nitrogen gas and a heating rate of 5 °C/min on an OZM Research DTA552-Ex instrument with the Meavy 2.2.0 software. The sample sizes were 3-15 mg. The impact and friction sensitivity data were determined with an OZM Research BAM Fall Hammer BFH-10 and an OZM Research BAM Friction apparatus FSKM-10, respectively, through five individual measurements that were averaged. Both instruments were calibrated using RDX. The samples were finely powdered materials that were not sifted. 139 X-ray Crystal Structure Determination. The single crystal X-ray diffraction data for HDNT-1, HDNT-2, (HDNT) 3 . 4 H 2 O, 3, 3·4 (co-crystals), and 5·H 2 O were collected on a Bruker SMART diffractometer, equipped with an APEX CCD detector, using Mo K  radiation (graphite monochromator) from a fine-focus tube. The single crystal X-ray diffraction data for the remaining structures were collected on a Bruker SMART APEX DUO diffractometer, equipped with an APEX II CCD detector, using Mo K  radiation (TRIUMPH curved-crystal monochromator) from a fine-focus tube or Cu K  from an I S micro-source. The frames were integrated using the SAINT algorithm to give the hkl files corrected for Lp/decay. 2 The absorption correction was performed using the SADABS program. 3 The structures were solved and refined on F 2 using the Bruker SHELXTL Software Package. 4,5 Non-hydrogen atoms were refined anisotropically. Structure drawings were prepared using the ORTEP-III for Windows V2.02 and Mercury 3.3.1 programs. 6,7 Further crystallographic details can be obtained from the Cambridge Crystallographic Data Centre (CCDC, 12 Union Road, Cambridge CB21EZ, UK (Fax: (+44) 1223-336-033; e-mail: deposit@ccdc.cam.ac.uk) on quoting the deposition no. CCDC 1009538, 1013935-1013938 and 1045452, and from the Fachinformationszentrum Karlsruhe, 76344 Eggenstein- Leopoldshafen, Germany (Fax: (+49) 7247-808-666, e-mail: crysdata@fiz-karlsruhe.de, http://www.fiz- karlsruhe.de/request_for_deposited_data.html) on quoting the deposition numbers CSD 427861-427864. Synthesis of potassium 3,5-dinitro-1,2,4-triazolate (method 1). In a 2000 mL three-necked round- bottom flask, equipped with a reflux condenser, an addition funnel and a mechanical stirrer, a mixture of sodium nitrite (220 g, 3.28 mol) in water (350 mL) was heated to 50 °C using a water bath until all sodium nitrite had dissolved. A solution of 3,5-diamino-1,2,4-triazole (40.0 g, 0.404 mol) in water (500 mL) and concentrated sulfuric acid (36 mL) was added slowly and carefully through the addition funnel while the reaction mixture was stirred vigorously. Immediately, the reaction mixture turned red, foamed and formed a dark red precipitate. In addition, brown-orange fumes of nitric oxide were produced. After about one to two hours, the addition was completed and 80% sulfuric acid (220 mL) was added carefully while the reaction mixture was stirred vigorously in order to avoid excessive foam formation. The reaction mixture was then refluxed for about 60 minutes and then allowed to cool to 40-50 °C. Activated decolorizing charcoal (10 g) was added and the mixture stirred at ambient temperature for eight hours. The reaction mixture was then filtered over Celite 545 and the filtrate extracted six times with ethyl acetate (150 mL each). The combined organic phases were dried over magnesium sulfate and the solvent removed immediately using a rotary evaporator without (!) heating the sample. The obtained yellow to orange oil or paste was dissolved immediately in acetone (200 mL) and the yellow solution poured onto potassium carbonate (60 g). Immediately, a gas was evolved and the mixture was stirred at ambient temperature. After about two hours, the mixture was filtered and the orange solid residue was washed 140 extensively with acetone. The combined yellow filtrates were taken to dryness on a rotary evaporator leaving behind a yellow solid. Recrystallization from water resulted in the isolation of yellow crystals of KDNT·2H 2 O that were dried in vacuo at 50 °C for eight hours, resulting in colorless to pale yellow KDNT (yield: 52.5 g, 65.9 %). DTA: 265 °C decomposition; NMR (CD 3 CN) δ(ppm): 13 C (125.76 MHz) 164.3 (C-NO 2 ); 14 N (36.14 MHz) -20.3 (s,  1/2 = 65Hz, 2N, C-NO 2 ), -52 (s,  1/2 = 500 Hz, DNT - ). Raman (200 mW) ῦ/cm -1 : 1545 (0.3), 1500 (0.2), 1425 (0.2), 1403 (10.0), 1388 (0.4), 1359 (1.6), 1310 (0.3), 1107 (0.3), 1100 (6.1), 1067 (0.1), 1028 (0.3), 834 (0.8), 769 (0.4), 515 (0.2). IR (ATR) ῦ/cm -1 : 3604 (m), 3414 (m), 3327 (w), 3233 (w), 2749 (w), 2694 (w), 2663 (w), 2459 (w), 2144 (w), 1672 (w), 1643 (m), 1552 (w), 1536 (s), 1493 (s), 1440 (w), 1415 (w), 1388 (s), 1354 (s), 1341 (w), 1299 (s), 1109 (m), 1100 (w), 1067 (w), 1050 (m), 997 (w), 877 (w), 846 (s), 835 (w), 770 (w), 749 (w), 684 (w), 648 (s), 605 (w), 515 (m), 502 (w), 483 (w), 472 (w), 457 (w), 404 (w). IR (AgCl pellet) ῦ/cm -1 : 2757 (w), 2754 (w), 2748 (w), 2691 (w), 2658 (w), 2457 (m), 2399 (w), 2186 (w), 2158 (w), 2139 (w), 1563 (w), 1556 (s), 1540 (w), 1504 (s), 1439 (w), 1414 (m), 1392 (s), 1359 (s), 1343 (w), 1320 (w), 1310 (m), 1301 (s), 1296 (m), 1107 (w), 1100 (s), 1066 (w), 1050 (m), 1044 (m), 1019 (w), 850 (s), 837 (s), 831 (m), 771 (m), 768 (w), 652 (s), 606 (m), 519 (w), 515 (w). Synthesis of potassium 3,5-dinitro-1,2,4-triazolate (method 2). In a 1000 mL three-necked round- bottom flask that was cooled by a water bath, concentrated nitric acid (80 mL) was slowly added to hydrazine hydrate (20.0 g, 0.40 mol). Water (140 mL) and 2-cyanoguanidine (dicyandiamide) (33.6 g, 0.40 mol) was added and the reaction mixture heated to 50 °C for one hour. A solution of concentrated sulfuric acid (35 mL) in water (300 mL) was added, and the resulting DAT solution transferred into an addition funnel from which it was added carefully to a vigorously stirred solution of sodium nitrite (220 g, 3.28 mol) in water (350 mL) at 50 °C. After about one to two hours, the addition was completed and 80% sulfuric acid (220 mL) was added carefully while the reaction mixture was stirred vigorously in order to avoid excessive foam formation. The reaction mixture was then refluxed for about 60 minutes and then allowed to cool to 40-50 °C. Activated decolorizing charcoal (10 g) was added and the mixture stirred at ambient temperature for eight hours. The reaction mixture was then filtered over Celite 545 and the filtrate extracted six times with ethyl acetate (150 mL each). The combined organic phases were dried over magnesium sulphate and the solvent removed immediately using a rotary evaporator without (!) heating the sample. The obtained yellow to orange oil or paste was dissolved immediately in acetone (200 mL) and the yellow solution poured onto potassium carbonate (60 g). Immediately, a gas was evolved and the mixture was stirred at ambient temperature. After about two hours, the mixture was filtered and the orange solid residue was washed extensively with acetone. The combined yellow filtrates were taken to 141 dryness on a rotary evaporator leaving behind in a yellow-orange solid. Recrystallization from water resulted in the isolation of yellow crystals that were dried in vacuo at 50 °C for eight hours, resulting in yellow KDNT containing various impurities (yield: 54.3 g, 69% based on KDNT). IR (ATR) ῦ/cm -1 : 2753 (vw), 2692 (vw), 2662 (vw), 2457 (vw), 2399 (vw), 2178 (vw), 2140 (m), 1553 (s), 1532 (s), 1494 (vs), 1413 (w), 1386 (vs), 1354 (vs), 1342 (s), 1296 (vs), 1099 (m), 1049 (m), 847 (vs), 835 (m), 787 (vw), 770 (w), 747 (vw), 730 (vw), 648 (s), 604 (m), 515 (w). Synthesis of 3,5-dinitro-1H-1,2,4-triazole (HDNT). A solution of KDNT (5.937 g, 30.11 mmol) in water (20 mL) was acidified with 20% sulfuric acid (50 mL) and the resulting yellow solution extracted four times with ethyl acetate (50 mL each). The combined organic phases were washed with water (50 mL), dried over magnesium sulfate and the solvent removed using a rotary evaporator. The resulting yellow oil was further dried in a high vacuum at 50-60 °C for eight hour, resulting in light-yellow, solid HDNT (yield: 4.502 g, 93.0%). HDNT of high purity was obtained as a white solid through careful sublimation of the crude compound at 100-105 °C in a vacuum of less than 0.1 Torr. The temperature was carefully monitored in order to avoid a potentially explosive decomposition of the HDNT. DTA: 170 °C (onset) explosive decomposition; Friction sensitivity: 144 N; Impact sensitivity: 35 J; NMR (CD 3 CN) δ(ppm): 1H (500.13 MHz) 13.6 (  1/2 = 100 Hz); 13 C (125.76 MHz) 157.1 (C-NO 2 ); 14 N (36.14 MHz) -32.4 (s, τ 1/2 = 20 Hz C-NO 2 ). Raman (50 mW) ῦ/cm -1 : 2863 (0.2), 1581 (0.8), 1574 (1.0), 1528 (0.4), 1499 (0.7), 1487 (0.9), 1451 (2.6), 1436 (10.0), 1382 (3.4), 1366 (1.1), 1317 (0.8), 1280 (0.6), 1181 (1.2), 1163 (0.6), 1048 (1.0), 1043 (1.1), 1027 (0.3), 1012 (0.7), 1007 (0.8), 829 (0.9), 826 (1.3), 774 (0.4), 762 (0.8), 510 (0.6), 357 (1.1), 290 (1.0), 182 (0.7), 171 (0.7), 111 (2.5), 95 (2.4). IR (ATR, 20°C) ῦ/cm -1 : 3039 (vw), 2992 (vw), 2934 (vw), 2883 (vw), 2839 (vw), 2802 (vw), 2748 (vw), 2639 (vw), 2557 (vw), 2162 (vw), 1698 (vw), 1563 (vs), 1530 (s sh), 1485 (s), 1430 (w), 1377 (s sh), 1363 (s sh), 1313 (vs), 1281 (m sh), 1174 (m), 1038 (m), 1023 (m sh), 1008 (m sh), 839 (vs), 825 (vs), 764 (vw), 730 (vw), 649 (m), 634 (m sh), 593 (m), 509 (m), 500 (m sh). Synthesis of PPN + 3,5-dinitro-1,2,4-triazolate PPN[DNT]. PPN[DNT] was prepared as described previously 8 by adding an aqueous solution of HDNT to an aqueous solution of PPNCl. The resulting PPN[DNT] precipitate was filtered, washed with water and recrystallized from acetone. On one occasion, reddish crystals were obtained from the aqueous filtrate. These crystals were identified by X-ray diffraction as PPN[H(AzNT) 2 ] (6). 142 The vibrational spectra of the isolated PPN[DNT] exhibited vibrational bands due to azido compounds. An X-ray structural analysis of some of the crystals showed an substitutional disorder in which the about 25% of the DNT - anions were relaced by AzNT - anions. Contaminated PPN[DNT]: IR (ATR, 20°C) ῦ/cm -1 : 3055 (w), 2122 (m), 1587 (w), 1531 (s), 1509 (w), 1480 (s), 1436 (s), 1423 (w), 1397 (w), 1373 (s), 1339 (m), 1325 (m), 1296 (w), 1281 (m), 1239 (vs), 1181 (w), 1162 (w), 1111 (s), 1073 (m), 1038 (w), 1024 (w), 997 (m), 932 (w), 858 (w), 838 (s), 828 (w), 803 (m), 760 (vw), 746 (s), 722 (vs), 690 (vs), 666 (vw), 616 (w), 597 (w), 553 (vs), 531 (w), 526 (vs), 498 (vs), 487 (w), 462 (w), 450 (w), 437 (w), 419 (vw). [PPN][H(AzNT) 2 ]: IR (ATR, 20°C) ῦ/cm -1 : 3063 (w), 2130 (s), 2017 (w), 1991 (vw), 1976 (w), 1896 (w), 1831 (vw), 1587 (w), 1573 (vw), 1556 (w), 1526 (m), 1479 (m), 1436 (m), 1383 (m), 1341 (w), 1294 (s), 1221 (w), 1183 (w), 1161 (w), 1113 (vs), 1024 (w), 1014 (w), 996 (m), 931 (vw), 839 (m), 793 (vw), 756 (vw), 749 (m), 721 (s), 690 (vs), 655 (w), 617 (vw), 549 (w), 528 (vs), 496 (s), 444 (m), 414 (vw). A1.2 Crystallographic Information A1.2.1 Crystal structure report for the monoclinic modification HDNT-1 Figure A1.1: Projection of the packing in the monoclinic modification HDNT-1 perpendicular to the 001 plane. 143 Table A1.1: Sample and crystal data for the monoclinic modification HDNT-1. Identification code HDNT-1 Chemical formula C 2 HN 5 O 4 Formula weight 159.08 g/mol Temperature 140(2) K Wavelength 0.71073 Å Crystal size 0.160 x 0.170 x 0.240 mm Crystal habit clear colourless prism Crystal system monoclinic Space group P 1 21/c 1 Unit cell dimensions a = 6.1585(14) Å α = 90° b = 9.083(2) Å β = 93.892(3)° c = 9.858(2) Å γ = 90° Volume 550.2(2) Å 3 Z 4 Density (calculated) 1.920 g/cm 3 Absorption coefficient 0.183 mm -1 F(000) 320 Table A1.2: Data collection and structure refinement for the monoclinic modification HDNT-1. Diffractometer Bruker SMART APEX Radiation source fine-focus tube, MoKα Theta range for data collection 3.05 to 29.48° Index ranges -8<=h<=8, -12<=k<=12, -13<=l<=13 Reflections collected 11484 Independent reflections 1470 [R(int) = 0.0308] Absorption correction multi-scan Max. and min. transmission 0.9710 and 0.9570 Structure solution technique direct methods Structure solution program SHELXTL XT 2014/4 (Bruker AXS, 2014) Refinement method Full-matrix least-squares on F 2 Refinement program SHELXTL XL 2014/7 (Bruker AXS, 2014) Function minimized Σ w(F o 2 - F c 2 ) 2 Data / restraints / parameters 1470 / 0 / 103 Goodness-of-fit on F 2 1.050 Final R indices 1305 data; I>2σ(I) R1 = 0.0327, wR2 = 0.0831 all data R1 = 0.0366, wR2 = 0.0865 Weighting scheme w=1/[σ 2 (F o 2 )+(0.0459P) 2 +0.1520P] where P=(F o 2 +2F c 2 )/3 Largest diff. peak and hole 0.317 and -0.203 eÅ -3 R.M.S. deviation from mean 0.046 eÅ -3 144 A1.2.2 Crystal structure report for the triclinic modification HDNT-2 Figure A1.2: Projection of the packing in the triclinic modification HDNT-2 perpendicular to the 100 plane. 145 Table A1.3: Sample and crystal data for the triclinic modification HDNT-2. Identification code HDNT (2) Chemical formula C 2 HN 5 O 4 Formula weight 159.08 g/mol Temperature 140(2) K Wavelength 0.71073 Å Crystal size 0.110 x 0.160 x 0.170 mm Crystal habit yellow prism Crystal system triclinic Space group P -1 Unit cell dimensions a = 8.7465(15) Å α = 111.927(2)° b = 8.9684(16) Å β = 96.726(3)° c = 11.942(2) Å γ = 93.853(2)° Volume 856.7(3) Å 3 Z 6 Density (calculated) 1.850 g/cm 3 Absorption coefficient 0.176 mm -1 F(000) 480 Table A1.4: Data collection and structure refinement for the triclinic modification HDNT-2. Diffractometer Bruker SMART APEX Radiation source fine-focus tube, MoKα Theta range for data collection 1.86 to 28.66° Index ranges -11<=h<=11, -11<=k<=11, -8<=l<=15 Reflections collected 5462 Independent reflections 3826 [R(int) = 0.0196] Absorption correction multi-scan Max. and min. transmission 0.9810 and 0.9710 Structure solution technique direct methods Structure solution program SHELXTL XT 2014/4 (Bruker AXS, 2014) Refinement method Full-matrix least-squares on F 2 Refinement program SHELXTL XL 2014/7 (Bruker AXS, 2014) Function minimized Σ w(F o 2 - F c 2 ) 2 Data / restraints / parameters 3826 / 0 / 302 Goodness-of-fit on F 2 1.036 Final R indices 2831 data; I>2σ(I) R1 = 0.0499, wR2 = 0.1170 all data R1 = 0.0716, wR2 = 0.1380 Weighting scheme w=1/[σ 2 (F o 2 )+(0.0601P) 2 +0.4374P] where P=(F o 2 +2F c 2 )/3 Largest diff. peak and hole 0.283 and -0.380 eÅ -3 R.M.S. deviation from mean 0.068 eÅ -3 146 A1.2.3 Crystal structure report for (HDNT) 3 ·4 H 2 O Figure A1.3: Projection of the packing in (HDNT) 3 ·4H 2 O perpendicular to the 100 plane. 147 Table A1.5: Sample and crystal data for (HDNT) 3 ·4H 2 O. Identification code HDNT_H2O Chemical formula C 6 H 11 N 15 O 16 Formula weight 549.30 g/mol Temperature 100(2) K Wavelength 0.71073 Å Crystal size 0.020 x 0.100 x 0.180 mm Crystal habit yellow plate Crystal system triclinic Space group P 1 Unit cell dimensions a = 6.1906(2) Å α = 111.377(2)° b = 9.5492(3) Å β = 93.467(2)° c = 9.5656(3) Å γ = 90.765(3)° Volume 525.22(3) Å 3 Z 1 Density (calculated) 1.737 g/cm 3 Absorption coefficient 0.169 mm -1 F(000) 280 Table A1.6: Data collection and structure refinement for (HDNT) 3 ·4H 2 O. Diffractometer Bruker APEX DUO Radiation source fine-focus tube, MoKα Theta range for data collection 2.29 to 23.19° Index ranges -6<=h<=6, -10<=k<=10, -10<=l<=10 Reflections collected 5789 Independent reflections 2617 [R(int) = 0.0212] Coverage of independent reflections 95.4% Absorption correction multi-scan Max. and min. transmission 0.9970 and 0.9700 Structure solution technique direct methods Structure solution program SHELXTL XT 2014/4 (Bruker AXS, 2014) Refinement method Full-matrix least-squares on F 2 Refinement program SHELXL-2014/6 (Sheldrick, 2014) Function minimized Σ w(F o 2 - F c 2 ) 2 Data / restraints / parameters 2617 / 3 / 343 Goodness-of-fit on F 2 1.015 Δ/σ max 0.003 Final R indices 2468 data; I>2σ(I) R1 = 0.0251, wR2 = 0.0554 all data R1 = 0.0288, wR2 = 0.0575 Weighting scheme w=1/[σ 2 (F o 2 )+(0.0347P) 2 ] where P=(F o 2 +2F c 2 )/3 Absolute structure parameter -0.2(8) Largest diff. peak and hole 0.146 and -0.222 eÅ -3 R.M.S. deviation from mean 0.036 eÅ -3 148 A1.2.4 Crystal Structure Report for 5-ethoxy-1-methyl-3-nitro-1H-1,2,4- triazole (1) Figure A1.4: Projection of the packing in 5-ethoxy-1-methyl-3-nitro-1H-1,2,4-triazole (1) perpendicular to the 001 plane. 149 Table A1.7: Data collection and structure refinement for 5-ethoxy-1-methyl-3-nitro-1H-1,2,4-triazole (1). Identification code cu_usc01_0m Empirical formula C 5 H 8 N 4 O 3 Formula weight 172.15 Temperature (K) 100.15 Crystal system triclinic Space group P-1 a/Å) 6.6336(8) b(Å) 7.6993(8) c(Å) 8.5129(9) α(°) 83.636(5) Β(°) 73.035(5) γ (°) 65.784(6) Volume (Å 3 ) 379.25(7) Z 2 ρ calc (g/cm 3 ) 1.508 Μ (mm -1 ) 1.086 F(000) 180.0 Crystal size (mm 3 ) 0.203 × 0.056 × 0.041 Radiation CuKα (λ = 1.54178) 2Θ range for data collection(°) 10.866 to 135.894 Index ranges -7 ≤ h ≤ 7, -9 ≤ k ≤ 9, -10 ≤ l ≤ 10 Reflections collected 8069 Independent reflections 1315 [R int = 0.0241, R sigma = 0.0145] Data/restraints/parameters 1315/0/141 Goodness-of-fit on F 2 1.055 Final R indexes [I>=2σ (I)] R 1 = 0.0300, wR 2 = 0.0823 Final R indexes [all data] R 1 = 0.0321, wR 2 = 0.0841 Largest diff. peak/hole (e Å -3 ) 0.20/-0.16 150 A1.2.5 Crystal Structure Report for 1-acetyl-3,5-diamino-1H-1,2,4-triazole (2) Figure A1.5: Projection of the packing in 1-acetyl-3,5-diamino-1H-1,2,4-triazole (2) perpendicular to the 100 plane. 151 Table A1.8: Sample and crystal data for 1-acetyl-3,5-diamino-1H-1,2,4-triazole (2). Identification code AcDNT Chemical formula C 4 H 7 N 5 O Formula weight 141.15 Temperature 100(2) K Wavelength 0.71073 Å Crystal system triclinic Space group P -1 Unit cell dimensions a = 5.25150(10) Å α = 67.2290(10)° b = 7.69060(10) Å β = 85.4460(10)° c = 8.49150(10) Å γ = 70.1300(10)° Volume 296.854(8) Å 3 Z 2 Density (calculated) 1.579 g/cm 3 Absorption coefficient 0.122 mm -1 F(000) 148 Table A1.9: Data collection and structure refinement for 2. Diffractometer Bruker APEX DUO Radiation source fine-focus tube, MoKα Theta range for data collection 2.61 to 36.40° Index ranges -8<=h<=8, -12<=k<=12, -14<=l<=14 Reflections collected 35219 Independent reflections 2865 [R(int) = 0.0256] Absorption correction multi-scan Structure solution technique direct methods Structure solution program SHELXTL XS 2013/1 (Sheldrick, 2013) Refinement method Full-matrix least-squares on F 2 Refinement program SHELXTL XLMP 2014/1 (Bruker AXS, 2013) Function minimized Σ w(F o 2 - F c 2 ) 2 Data / restraints / parameters 2865 / 0 / 109 Goodness-of-fit on F 2 1.123 Final R indices 2727 data; I>2σ(I) R1 = 0.0288, wR2 = 0.0807 all data R1 = 0.0299, wR2 = 0.0820 Weighting scheme w=1/[σ 2 (F o 2 )+(0.0463P) 2 +0.0425P] where P=(F o 2 +2F c 2 )/3 Extinction coefficient 0.0540(110) Largest diff. peak and hole 0.557 and -0.332 eÅ -3 R.M.S. deviation from mean 0.055 eÅ -3 152 A1.2.6 Crystal Structure Report for 1-(i-propyl)-3,5-dinitro-1H-1,2,4-triazole (3) Figure A1.6: Projection of the packing in 1-(i-propyl)-3,5-dinitro-1H-1,2,4-triazole (3) perpendicular to the 100 plane. 153 Table A1.10: Sample and crystal data for 1-(i-propyl)-3,5-dinitro-1H-1,2,4-triazole (3). Identification code HDNTimpurity Chemical formula C 5 H 7 N 5 O 4 Formula weight 201.16 Temperature 130(2) K Wavelength 0.71073 Å Crystal size 0.180 x 0.210 x 0.260 mm Crystal habit clear pale yellow prism Crystal system orthorhombic Space group P b c a Unit cell dimensions a = 9.4402(19) Å α = 90° b = 10.611(2) Å β = 90° c = 17.296(4) Å γ = 90° Volume 1732.5(6) Å 3 Z 8 Density (calculated) 1.542 g/cm 3 Absorption coefficient 0.134 mm -1 F(000) 832 Table A1.11: Data collection and structure refinement for 3. Diffractometer Bruker SMART APEX Radiation source fine-focus tube, MoKα Theta range for data collection 2.36 to 28.57° Index ranges -12<=h<=12, -14<=k<=14, -23<=l<=22 Reflections collected 13391 Independent reflections 2124 [R(int) = 0.0393] Coverage of independent reflections 95.9% Absorption correction multi-scan Max. and min. transmission 0.9760 and 0.9660 Structure solution technique direct methods Structure solution program SHELXTL XT 2013/6 (Sheldrick, 2013) Refinement method Full-matrix least-squares on F 2 Refinement program SHELXTL XLMP 2014/1 (Bruker AXS, 2013) Function minimized Σ w(F o 2 - F c 2 ) 2 Data / restraints / parameters 2124 / 0 / 129 Goodness-of-fit on F 2 1.024 Final R indices 1538 data; I>2σ(I) R1 = 0.0435, wR2 = 0.0943 all data R1 = 0.0691, wR2 = 0.1074 Weighting scheme w=1/[σ 2 (F o 2 )+(0.0450P) 2 +0.6327P] where P=(F o 2 +2F c 2 )/3 Largest diff. peak and hole 0.216 and -0.192 eÅ -3 R.M.S. deviation from mean 0.046 eÅ -3 154 A1.2.7 Crystal Structure Report for the Co-crystal of 1-(i-propyl)-3,5-dinitro- 1H-1,2,4-triazole and 3-nitro-1H-1,2,4-triazole (3·4). Figure A1.7: Projection of the packing in the co-crystal of 1-(i-propyl)-3,5-dinitro-1H-1,2,4-triazole and 3-nitro-1H-1,2,4-triazole (3·4).perpendicular to the 100 plane. 155 Table A1.12: Sample and crystal data for the co-crystal of 1-(i-propyl)-3,5-dinitro-1H-1,2,4-triazole and 3-nitro-1H-1,2,4-triazole (3·4). Identification code HDNTimpurity2 Chemical formula C 7 H 9 N 9 O 6 Formula weight 315.23 Temperature 130(2) K Wavelength 0.71073 Å Crystal size 0.130 x 0.220 x 0.270 mm Crystal habit clear pale yellow prism Crystal system orthorhombic Space group P b c a Unit cell dimensions a = 9.472(2) Å α = 90° b = 11.272(2) Å β = 90° c = 23.860(5) Å γ = 90° Volume 2547.5(9) Å 3 Z 8 Density (calculated) 1.644 g/cm 3 Absorption coefficient 0.144 mm -1 F(000) 1288 Table A1.13: Data collection and structure refinement for 3·4. Diffractometer Bruker SMART APEX Radiation source fine-focus tube, MoKα Theta range for data collection 1.71 to 27.53° Index ranges -12<=h<=12, -14<=k<=14, -30<=l<=14 Reflections collected 14568 Independent reflections 2921 [R(int) = 0.0323] Coverage of independent reflections 94.9% Absorption correction multi-scan Max. and min. transmission 0.9820 and 0.9620 Structure solution technique direct methods Structure solution program SHELXTL XT 2013/6 (Sheldrick, 2013) Refinement method Full-matrix least-squares on F 2 Refinement program SHELXTL XLMP 2014/1 (Bruker AXS, 2013) Function minimized Σ w(F o 2 - F c 2 ) 2 Data / restraints / parameters 2921 / 0 / 201 Goodness-of-fit on F 2 1.052 Final R indices 2252 data; I>2σ(I) R1 = 0.0564, wR2 = 0.1607 all data R1 = 0.0725, wR2 = 0.1752 Weighting scheme w=1/[σ 2 (F o 2 )+(0.0973P) 2 +1.3970P] where P=(F o 2 +2F c 2 )/3 Largest diff. peak and hole 0.784 and -0.566 eÅ -3 R.M.S. deviation from mean 0.070 eÅ -3 156 A1.2.8 Crystal Structure Report for Sodium 3-nitro-1,2,4-triazol-5-olate monohydrate (5·H 2 O) Figure A1.8: Projection of the packing in sodium 3-nitro-1,2,4-triazol-5-olate monohydrate (5·H 2 O) down the b-axis (blue= nitrogen, grey= carbon, red= oxygen, purple= sodium). 157 Table A1.14: Sample and crystal data for sodium 3-nitro-1,2,4-triazol-5-olate monohydrate (5·H 2 O). Identification code DNTimpurity Chemical formula C 2 H 3 N 4 NaO 4 Formula weight 170.07 Temperature 140(2) K Wavelength 0.71073 Å Crystal size 0.080 x 0.290 x 0.380 mm Crystal habit clear orange blade Crystal system monoclinic Space group P 1 21/c 1 Unit cell dimensions a = 10.7152(18) Å α = 90° b = 8.3768(14) Å β = 97.084(2)° c = 6.7473(11) Å γ = 90° Volume 601.01(17) Å 3 Z 4 Density (calculated) 1.880 g/cm 3 Absorption coefficient 0.232 mm -1 F(000) 344 Table A1.15: Data collection and structure refinement for 5·H 2 O. Diffractometer Bruker SMART APEX Radiation source fine-focus tube, MoKα Theta range for data collection 1.92 to 28.49° Index ranges -13<=h<=14, -11<=k<=11, -8<=l<=9 Reflections collected 6577 Independent reflections 1444 [R(int) = 0.0417] Coverage of independent reflections 94.4% Absorption correction multi-scan Max. and min. transmission 0.9820 and 0.9170 Structure solution technique direct methods Structure solution program SHELXTL XT 2013/6 (Sheldrick, 2013) Refinement method Full-matrix least-squares on F 2 Refinement program SHELXL-2014 (Sheldrick, 2014) Function minimized Σ w(F o 2 - F c 2 ) 2 Data / restraints / parameters 1444 / 0 / 112 Goodness-of-fit on F 2 1.055 Final R indices 1061 data; I>2σ(I) R1 = 0.0433, wR2 = 0.1008 all data R1 = 0.0692, wR2 = 0.1119 Weighting scheme w=1/[σ 2 (F o 2 )+(0.0529P) 2 +0.2158P] where P=(F o 2 +2F c 2 )/3 Largest diff. peak and hole 0.350 and -0.236 eÅ -3 R.M.S. deviation from mean 0.080 eÅ -3 158 A1.2.9 Crystal Structure Report for the Co-crystal of 5-azido-3-nitro-1,2,4- triazole and PPN + 5-azido-3-nitro-1,2,4-triazolate (6) Figure A1.9: Projection of the packing in co-crystal of 5-azido-3-nitro-1,2,4-triazole and PPN + 3-azido-5- nitro-1,2,4-triazolate (6) perpendicular to the 010 plane. Hydrogen atoms have been omitted for clarity. 159 Table A1.16: Sample and crystal data for co-crystal of 5-azido-3-nitro-1,2,4-triazole and PPN + 3-azido-5- nitro-1,2,4-triazolate (6). Identification code XGCMK4_02 Chemical formula C 40 H 31 N 15 O 4 P 2 Formula weight 847.74 Temperature 100(2) K Wavelength 0.71073 Å Crystal size 0.220 x 0.264 x 0.438 mm Crystal habit orange blade/prism Crystal system triclinic Space group P -1 Unit cell dimensions a = 11.1121(11) Å α = 70.424(2)° b = 12.5247(12) Å β = 72.600(2)° c = 16.0476(16) Å γ = 83.653(2)° Volume 2007.9(3) Å 3 Z 2 Density (calculated) 1.402 g/cm 3 Absorption coefficient 0.172 mm -1 F(000) 876 Table A1.17: Data collection and structure refinement for 6. Diffractometer Bruker APEX DUO Radiation source fine-focus tube, MoKα Theta range for data collection 1.40 to 30.67° Index ranges -15<=h<=15, -17<=k<=17, -22<=l<=22 Reflections collected 48323 Independent reflections 12028 [R(int) = 0.0605] Absorption correction multi-scan Max. and min. transmission 0.9680 and 0.9370 Structure solution technique direct methods Structure solution program SHELXTL XS 2013/1 (Sheldrick, 2013) Refinement method Full-matrix least-squares on F 2 Refinement program SHELXL-2013 (Sheldrick, 2013) Function minimized Σ w(F o 2 - F c 2 ) 2 Data / restraints / parameters 12028 / 0 / 550 Goodness-of-fit on F 2 1.026 Δ/σ max 0.001 Final R indices 7692 data; I>2σ(I) R1 = 0.0692, wR2 = 0.1441 all data R1 = 0.1154, wR2 = 0.1626 Weighting scheme w=1/[σ 2 (F o 2 )+(0.0651P) 2 +1.6487P] where P=(F o 2 +2F c 2 )/3 Largest diff. peak and hole 0.604 and -0.417 eÅ -3 R.M.S. deviation from mean 0.072 eÅ -3 160 A1.2.10 Crystal Structure Report for 5-azido-3-nitro-1,2,4-triazolate- containing PPN + 3,5-dinitro-1,2,4-triazolate Figure A1.10: Disordered anion in the crystal structure of 5-azido-3-nitro-1,2,4-triazolate-containing PPN + 3,5-dinitro-1H-1,2,4-triazolate with refined ratios of 1:3 (blue= nitrogen, grey= carbon, red= oxygen, orange= phosphorus). 161 Table A1.18: Sample and crystal data for 5-azido-3-nitro-1,2,4-triazolate-containing PPN + 3,5-dinitro- 1,2,4-triazolate. Identification code XGC4_MK4_01 Chemical formula C 38 H 30 N 6.51 O 3.49 P 2 Formula weight 695.60 g/mol Temperature 100(2) K Wavelength 0.71073 Å Crystal size 0.265 x 0.290 x 0.630 mm Crystal habit colorless prism Crystal system triclinic Space group P -1 Unit cell dimensions a = 10.8448(6) Å α = 67.1350(10)° b = 11.9539(7) Å β = 89.7880(10)° c = 14.4364(8) Å γ = 76.1090(10)° Volume 1665.59(16) Å 3 Z 2 Density (calculated) 1.387 g/cm 3 Absorption coefficient 0.182 mm -1 F(000) 723 162 Table A1.19: Data collection and structure refinement for 5-azido-3-nitro-1,2,4-triazolate-containing PPN + 3,5-dinitro-1,2,4-triazolate. Diffractometer Bruker APEX DUO Radiation source fine-focus tube, MoKα Theta range for data collection 1.54 to 31.12° Index ranges -15<=h<=15, -17<=k<=17, -20<=l<=20 Reflections collected 42188 Independent reflections 10304 [R(int) = 0.0367] Coverage of independent reflections 96.0% Absorption correction multi-scan Max. and min. transmission 0.9530 and 0.8940 Structure solution technique direct methods Structure solution program SHELXTL XT 2014/4 (Bruker AXS, 2014) Refinement method Full-matrix least-squares on F 2 Refinement program SHELXL-2014/7 (Sheldrick, 2014) Function minimized Σ w(F o 2 - F c 2 ) 2 Data / restraints / parameters 10304 / 0 / 659 Goodness-of-fit on F 2 0.900 Δ/σ max 0.027 Final R indices 8327 data; I>2σ(I) R1 = 0.0377, wR2 = 0.0931 all data R1 = 0.0516, wR2 = 0.1019 Weighting scheme w=1/[σ 2 (F o 2 )+(0.0530P) 2 +1.0479P] where P=(F o 2 +2F c 2 )/3 Largest diff. peak and hole 0.430 and -0.346 eÅ -3 R.M.S. deviation from mean 0.055 eÅ -3 A1.3 References (1) Haiges, R.; Boatz, J. A.; Yousufuddin, M.; Christe, K. O. Angew. Chem. Int. Ed. 2007, 46, 2869. (2) Bruker SHELXTL V2011.4-0; Bruker AXS: Madison, WI, 2011. (3) SADABS V2012/1 (4) SHELXTL V2012.4-3 (5) Sheldrick, G. Acta Crystallogr. Sect. A: Found. Crystallogr. 2015, 71, 3. (6) Farrugia, L. J. Appl. Crystallogr. 1997, 30, 565. (7) Haiges, R.; Boatz, J. A.; Schneider, S.; Schroer, T.; Yousufuddin, M.; Christe, K. O. Angew. Chem. 2004, 116, 3210. (8) Haiges, R.; Bélanger-Chabot, G.; Kaplan, S. M.; Christe, K. O. Dalton Trans. 2015, 44, 2978. 163 APPENDIX 2: ADDITIONAL INFORMATION ON SYNTHESIS AND CHARACTERIZATION OF SALTS OF THE 3,5-DINITRO-1,2,4- TRIAZOLATE ANION (CHAPTER 3) A2.1 Experimental details Caution! The compounds of this work are energetic materials that might explode under certain conditions (e.g. elevated temperatures, impact, friction or electric discharge). Appropriate safety precautions, 1 such as the use of shields or barricades in a fume hood and personal protection equipment (safety glasses, face shields, ear plugs, as well as gloves and suits made from leather and/or Kevlar) should be taken all the time when handling these materials. Ignoring safety precautions may lead to serious injuries! Unless otherwise specified, all chemicals were used as purchased from Alfa Aesar or Sigma- Aldrich without further purification. NMR spectra were recorded on Varian VNMRS-600S spectrometer or Bruker AMX-500 spectrometers. 13 C NMR spectra were referenced to the deuterated solvent signal and 1 H spectra were referenced to the residual protic signal. 14 N spectra were referenced to neat nitromethane and 31 P spectra were referenced to neat H 3 PO 4 . Raman spectra were recorded at ambient temperature in Pyrex glass melting point capillaries or J. Young NMR tubes on a Bruker Equinox 55 FT-RA spectrometer using a Nd:YAG laser at 1064 nm. Infrared spectra were recorded between 4000 and 400 cm−1 on Midac, M Series or Bruker Optics Alpha FT-IR spectrometers. The samples were either as KBr pellets or AgCl pellets or the neat solid on a ATR module (Bruker). KBr pellets were prepared with an Econo minipress (Barnes Engineering Co.). Differential thermal analysis (DTA) curves were recorded on an OZM Research DTA 552-Ex instrument under a dry nitrogen gas flow with a heating rate of 5 °C using the Meavy 2.2.0 software. Sensitivity measurements were performed on an OZM Research BAM fall hammer BFH-10 and an OZM Research BAM friction apparatus FSKM-10. The single-crystal X-ray diffraction data were collected on Bruker SMART APEX or APEX DUO diffractometers. The SMART APEX instrument was equipped with an APEX CCD detector and a Cryo Industries low-temperature device. This diffractometer was using graphite monochromatized Mo Kα radiation from a finefocus tube. The APEX DUO diffractometer was equipped with an APEX II CCD detector and an Oxford Cryostream 700 apparatus for low-temperature data collection. The diffractometer was using Mo Kα radiation (Triumph curved-crystal monochromator) from a finefocus tube or Cu Kα 164 radiation (multi-layer optics) from an I S microsource). The collected frames were integrated with the SAINT algorithm 2 to give the hkl files corrected for Lp/decay. Absorption correction was performed with the SADABS program. 3 The structures were solved by direct methods or intrinsic phasing and refined on F 2 by use of the Bruker SHELXTL software package. 4-7 All non-hydrogen atoms were refined anisotropically. Unless noted otherwise, the positions of hydrogen atoms have been located from the difference electron density map. ORTEP drawings were prepared with the ORTEP-III for Windows V2.02 program. 8 Short-contact, hydrogen-bonding and crystal packing drawings of most inorganic salts were prepared with the Mercury 3.1 Development (Build RC5) software. 9 Further crystallographic details can be obtained from the Cambridge Crystallographic Data Centre (CCDC, 12 Union Road, Cambridge CB21EZ, UK (Fax: (+44) 1223-336-033; e-mail: deposit@ccdc.cam.ac.uk) on quoting the deposition no. CCDC 1006182 ([HNEt 3 ][DNT]), 1006181 ([H 2 NEt 2 ][DNT] monoclinic) 1013941 ([H 2 NEt 2 ][DNT] triclinic), 1006639 ([CH 6 N 3 ][DNT]), 1014382 ([CH 7 N 4 ][DNT]), 1006408 ([C 5 H 6 N][DNT]), 1009537 (PPh 4 [DNT]), 1006184 (PPN[DNT]), 1031353 (NMe 4 [DNT]), and 1006180 (NMe 4 [DNT] ·HDNT) from the Fachinformationszentrum Karlsruhe, 76344 Eggenstein-Leopoldshafen, Germany (Fax: (+49) 7247- 808-666, e-mail: crysdata@fiz-karlsruhe.de, http://www.fiz- karlsruhe.de/request_for_deposited_data.html) on quoting the deposition numbers CSD 427905 (LiDNT·2H 2 O), 427851 (NaDNT·2H 2 O), 427971 (KDNT), 427853 (KDNT·2H 2 O), 427852 (RbDNT), 427854 (CsDNT), 428080 (Sr(DNT) 2 ·6H 2 O), 428122 (Ba(DNT) 2 ·11H 2 O), 427855 ([Ag(NH 3 )][DNT]), 428079 (NH 4 DNT·2H 2 O), 427856 ([CH 4 N 5 ][DNT]·H 2 O). 3,5-dinitro-1H-1,2,4-triazole (HDNT) was prepared according to a modified literature method. 10,11 All yields are reported with respect to the HDNT starting material. In all cases, the purity of the materials was assessed by 1 H and 13 C NMR spectroscopy. Preparation of LiDNT . 2H 2 O (1 . 2H 2 O). A solution of HDNT (0.318 g, 1.5 mmol) in acetone (5 mL) was added to Li 2 CO 3 (0.370 g, 5.00 mmol). The remaining suspension was stirred for 2 hours and then filtered. The yellow filtrate was taken to dryness by evaporation in air, resulting in the formation of yellow crystals of LiDNT . 2H 2 O formed (yield: 0.230 g, 95 %). DTA: 280 °C decomposition. NMR (CD 3 CN) δ(ppm): 13 C (100.54 MHz) 163.8 (s, C-NO 2 ); 14 N (36.14 MHz) -21.6 (s, τ 1/2 = 110 Hz, C-NO 2 ), - 57 (s, τ 1/2 > 500 Hz, DNT - ). IR (ATR, 20°C) ῦ/cm -1 : 3533 (s), 3364 (s), 3250 (m), 2920 (w), 2854 (w), 2774 (w), 2710 (w), 2677 (w), 2480 (w), 2425 (w), 2395 (w), 2217 (w), 2163 (w), 1639 (m), 1542 (s), 1497 (s), 1418 (w), 1395 (s), 1357 (s), 1305 (s), 1115 (m), 1049 (m), 830 (s), 642 (w), 617 (w), 542 (s), 503 (w), 416 (w). 165 Preparation of NaDNT . 2H 2 O (2 . 2H 2 O). A solution of HDNT (0.233 g, 1.47 mmol) in acetone (5 mL) was added to a suspension of Na 2 CO 3 (0.844 g, 7.96 mmol) (Macron) in acetone (30 mL), resulting in effervescence and a color change of the Na 2 CO 3 from white to yellow. Water (~0.5 mL) was added and the suspension was stirred vigorously for three days. The suspension was then vacuum filtered. The pale yellow filtrate was was taken to dryness by evaporation in air, resulting in the formation of pale yellow crystals. The crystals were dried in an evacuated desiccator for two hours resulting in pale yellow to dark orange microcrystalline NaDNT . 2H 2 O (yield: 0.321 g, 100 %). DTA: 260 °C decomposition. NMR (DMSO-d 6 ) δ(ppm): 13 C (125.76 MHz) 163.0 (C-NO 2 ); 14 N( 36.14 MHz) -20.7 (s, τ 1/2 = 200Hz, C-NO 2 ), - 52(s, τ 1/2 = 1000 Hz, DNT - ), -145 (s, τ 1/2 = 800 Hz, DNT - ). IR (ATR, 20°C) ῦ/cm -1 : 3554 (s), 3353 (s), 3260 (m), 3087 (w), 3060 (w), 2923 (w), 2851 (w), 2762 (w), 2699 (w), 2672 (w), 2472 (w), 2147 (w), 1641 (s), 1587 (w), 1537 (s), 1480 (s), 1439 (s), 1394 (m), 1377 (w), 1341 (s), 1307 (s), 1286 (w), 1191 (w), 1164 (w), 1107 (s), 1052 (m), 1037 (w), 1026 (w), 996 (m), 828 (s), 764 (w), 750 (w), 719 (m), 698 (w), 685 (s), 646 (w), 614 (w), 523 (s), 428 (w). Preparation of KDNT (3). HDNT (1.095 g, 6.90 mmol) in acetone (10 mL) was added to a K 2 CO 3 (6.119 g, 44.30 mmol) suspension in acetone (60 mL), resulting in the immediate orange coloration of the carbonate. The suspension was sonicated and then stirred overnight. The suspension was filtered and the orange solid washed with acetone, and the resulting pale yellow filtrate was taken to dryness by evaporation in air. Colorless needles of KDNT (3) crystallized out of the acetone solution. The solid residue was dried in a high vacuum at 50 °C overnight, resulting in colorless to pale yellow KDNT (yield: 1.279 g, 94 %). DTA: 265 °C decomposition. NMR (CD 3 CN) δ(ppm): 13 C (125.76 MHz) 164.3 (C- NO 2 ); 14 N (36.14 MHz) -20.3 (s, τ 1/2 = 65Hz, 2N, C-NO 2 ), -52 (s, τ 1/2 = 500 Hz, DNT - ). Raman (200 mW) ῦ/cm -1 : 1545 (0.3), 1500 (0.2), 1425 (0.2), 1403 (10.0), 1388 (0.4), 1359 (1.6), 1310 (0.3), 1107 (0.3), 1100 (6.1), 1067 (0.1), 1028 (0.3), 834 (0.8), 769 (0.4), 515 (0.2). IR (ATR) ῦ/cm -1 : 3604 (m), 3414 (m), 3327 (w), 3233 (w), 2749 (w), 2694 (w), 2663 (w), 2459 (w), 2144 (w), 1672 (w), 1643 (m), 1552 (w), 1536 (s), 1493 (s), 1440 (w), 1415 (w), 1388 (s), 1354 (s), 1341 (w), 1299 (s), 1109 (m), 1100 (w), 1067 (w), 1050 (m), 997 (w), 877 (w), 846 (s), 835 (w), 770 (w), 749 (w), 684 (w), 648 (s), 605 (w), 515 (m), 502 (w), 483 (w), 472 (w), 457 (w), 404 (w). IR (AgCl pellet) ῦ/cm -1 : 2757 (w), 2754 (w), 2748 (w), 2691 (w), 2658 (w), 2457 (m), 2399 (w), 2186 (w), 2158 (w), 2139 (w), 1563 (w), 1556 (s), 1540 (w), 1504 (s), 1439 (w), 1414 (m), 1392 (s), 1359 (s), 1343 (w), 1320 (w), 1310 (m), 1301 (s), 1296 (m), 1107 (w), 1100 (s), 1066 (w), 1050 (m), 1044 (m), 1019 (w), 850 (s), 837 (s), 831 (m), 771 (m), 768 (w), 652 (s), 606 (m), 519 (w), 515 (w). 166 Upon standing in air, the crystals of KDNT picked up moisture and turned yellow-orange. Preparation of (3 . 2H 2 O). KDNT (0.100 g, 0.51 mmol) was dissolved in water (5 mL). The yellow- orange solution was allowed to evaporate to dryness in air, resulting in orange crystals of KDNT . 2H 2 O (yield: 0.118 g, 100 %). Preparation of RbDNT (4). A solution of HDNT (0.244 g, 1.53 mmol) in acetone (5 mL) was added to a suspension of Rb 2 CO 3 (1.885 g, 8.16 mmol) in acetone (30 mL) resulting in effervescence and a color change of the Rb 2 CO 3 to yellow-orange. Water (~0.5 mL) was added and the suspension was stirred vigorously for 24 hours. The suspension was vacuum filtered and the pale yellow filtrate was allowed to evaporate to dryness. The resulting crystals were dried in an evacuated desiccator, yielding a colorless solid of RbDNT (0.369 g, 99 %). DTA : 170 °C melting, 250°C decomposition. NMR (DMSO-d 6 ) δ(ppm): 13 C (100.54 MHz) 162.9 (s, C-NO 2 ); 14 N (36.14 MHz) -21.0 (s, τ 1/2 = 220 Hz, NO 2 ), -50 (s, τ 1/2 > 1000 Hz, DNT - ), -145 (s, τ 1/2 > 1000 Hz, DNT - ). IR (ATR, 20°C) ῦ/cm -1 : 3396 (w), 2924 (m), 2852 (w), 2748 (w), 2684 (w), 2655 (w), 2449 (w), 2390 (w), 2147 (m), 1712 (w), 1526 (s), 1489 (s), 1444 (w), 1407 (w), 1381 (s), 1352 (s), 1340 (w), 1307 (w), 1293 (s), 1171 (w), 1098 (s), 1079 (w), 1041 (s), 984 (w), 873 (w), 845 (w), 835 (s), 770 (m), 674 (w), 658 (s), 593 (w), 515 (w), 436 (w). Synthesis of CsDNT (5). A solution of HDNT (0.228 g, 1.43 mmol) in acetone (5 mL) was added to a suspension of Cs 2 CO 3 (2.515 g, 7.72 mmol) in acetone (30 mL) resulting in effervescence and a color change of the Cs 2 CO 3 to yellow-orange. Water (~0.5 mL) was added and the suspension was stirred vigorously for 24 hours. The suspension was then vacuum filtered and the solvent of the filtrate allowed to evaporate, yielding a pale yellow solid of CsDNT (0.408 g, 98%). DTA : 250 °C decomposition. NMR (DMSO-d 6 ) δ(ppm): 13 C (100.54 MHz) 162.9 (s, C-NO 2 ); 14 N (36.14 MHz) -20.5 (s, τ 1/2 = 170 Hz, NO 2 ), - 49 (s, τ 1/2 > 1000 Hz, DNT - ), -146 (s, τ 1/2 > 800 Hz, DNT - ). IR (ATR, 20°C) ῦ/cm -1 : 2922 (w), 2851 (w), 2744 (w), 2685 (w), 2654 (w), 2449 (w), 2390 (w), 1548 (s), 1537 (w), 1488 (s), 1407 (m), 1384 (s), 1351 (s), 1339 (w), 1304 (w), 1294 (s), 1097 (s), 1042 (m), 1014 (w), 872 (w), 843 (s), 829 (s), 772 (w), 650 (s), 603 (m), 513 (w). Preparation of Sr(DNT) 2 . 6H 2 O (6 . 6H 2 O). A solution of HDNT (0.212 g, 1.34 mmol) in acetone (5 mL) was added to a suspension of SrCO 3 (1.018 g, 6.90 mmol) in acetone (30 mL). Water (~0.5 mL) was added, and the suspension was stirred vigorously for 24 hours. The suspension was vacuum filtered. The filtrate was was taken to dryness by evaporation in air resulting in an off-white solid of Sr(DNT) 2 . 6H 2 O (0.320 g, 46.8%). DTA: 290 °C explosion. NMR (DMSO-d 6 ) δ(ppm): 13 C (100.54 MHz) 162.9 (s, C- 167 NO 2 ); 14 N (36.14 MHz) -21.1 (s, τ 1/2 = 200 Hz, C-NO 2 ), -50 (s, τ 1/2 >1000 Hz, DNT - ), -147 (s, τ 1/2 >700 Hz, DNT - ). IR (ATR, 20°C) ῦ/cm -1 : 3589 (m), 3393 (s), 2922 (s), 2853 (m), 2483 (w), 2167 (w), 2159 (w), 2020 (w), 1737 (w), 1648 (s), 1556 (s), 1499 (s), 1419 (w), 1394 (s), 1349 (s), 1301 (s), 1125 (m), 1054 (m), 848 (s), 836 (w), 769 (w), 644 (s), 517 (w), 435 (w). Preparation of Ba(DNT) 2 . 11H 2 O (7 . 11H 2 O). A solution of HDNT (0.318 g, 2.00 mmol) in acetone (5 mL) was added to a suspension of BaCO 3 (1.200 g, 3.18 mmol) in acetone (10 mL). The suspension was stirred vigorously for one hour and then vacuum filtered. The pale yellow filtrate was taken to dryness by evaporation in air resulting in a pale yellow solid of Ba(DNT) 2 . 11H 2 O (yield: 0.567 g, 87 %). DTA: 245 °C decomposition. NMR (DMSO-d 6 ) δ(ppm): 13 C (100.54 MHz) 162.9 (s, C-NO 2 ); 14 N(36.14 MHz) -21.2 (s, τ 1/2 = 220 Hz, NO 2 ), -53 (s, τ 1/2 >1000 Hz, DNT - ), -147 (s, τ 1/2 = 800 Hz, DNT - ). IR (ATR, 20°C) ῦ/cm -1 : 3577 (w), 3415 (w), 2141 (w), 1727 (w), 1626 (w), 1555 (s), 1499 (s), 1384 (m), 1343 (s), 1296 (s), 1135 (w), 1055 (m), 847 (s), 833 (w), 767 (w), 645 (s), 596 (w), 584 (w), 516 (w). Preparation of AgDNT (8). Caution: AgDNT is an impact and friction sensitive explosive compound! A solution of AgNO 3 (0.357 g, 2.10 mmol) in water (10 mL) was added to a solution of HDNT (0.318 g, 2.00 mmol) in water (5 mL), resulting in the formation of a white precipitate. The suspension was carefully filtered and the filtration residue washed with water. The solid was dried in a vacuum in darkness, resulting in colorless AgDNT (yield: 0.253 g, 95 %). DTA: 240 °C decomposition. IR (ATR, 20°C) ῦ/cm -1 : 3377 (s), 2148 (w), 1644 (w), 1551 (s), 1494 (s), 1401 (m), 1364 (m), 1312 (s), 1130 (w), 1066 (w), 843 (s), 828 (w), 643 (m), 606 (w), 520 (w). Preparation of [Ag(NH 3 )][DNT] (9). A solution of AgNO 3 (0.357 g, 2.10 mmol) in water (10 mL) was added tp a solution of HDNT (0.318 g, 2.00 mmol) in water (5 mL), resulting in the formation of a white precipitate. The suspension was carefully filtered and the filtration residue washed with water. The filtration residue was then dissolved in concentrated aqueous ammonia (5 mL), and the solution was allowed to slowly evaporate in the air under exclusion of light, resulting in crystalline [Ag(NH 3 )][DNT] (yield: 0.526 g; 93 %). DTA: 155 °C (endotherm, loss of NH 3 ), 245 °C decomposition. NMR (DMSO-d 6 ) δ(ppm): 1 H (399.80 MHz) 3.34 (s, NH 3 ); 13 C (100.54 MHz) 162.12 (s, C-NO 2 ); 14 N (36.14 MHz) -22.9 (s, τ 1/2 = 300Hz, NO 2 ), -382 (s, τ 1/2 = 800 Hz, NH 3 ). IR (ATR, 20°C) ῦ/cm -1 : 3387 (m), 3297 (m), 2856 (w), 2488 (w), 2144 (w), 1607 (w), 1549 (s), 1492 (s), 1428 (w), 1400 (w), 1382 (m), 1358 (s), 1345 (w), 1304 (s), 1204 (w), 1158 (s), 1129 (m), 1063 (m), 168 1018 (w), 883 (w), 844 (s), 827 (m), 767 (w), 644 (s), 609 (w), 590 (w), 559 (m), 458 (w), 441 (w), 421 (w). Preparation of NH 4 [DNT] . 2H 2 O (10 . 2H 2 O). Concentrated aqueous ammonia (2 mL) was added to solution of HDNT (0.204 g, 1.31 mmol) water (5 mL). The solution was allowed to evaporate to dryness in air, resulting in colorless crystals of NH 4 DNT . 2H 2 O (0.259 g, 93 %). DTA: 140 °C decomposition. NMR (DMSO-d 6 ) δ(ppm): 1 H (399.80 MHz) 7.13 (s, NH 4 + ); 13 C (100.54 MHz) 162.9 (C-NO 2 ); 14 N (36.14 MHz) -20.9 (s, τ 1/2 = 200Hz, NO 2 ), -50 (s, τ 1/2 = 1000 Hz, DNT - ), - 147 (s, τ 1/2 = 800 Hz, DNT - ), - 358.5 (s, τ 1/2 = 5 Hz, NH 4 + ). IR (ATR, 20°C) ῦ/cm -1 : 3591 (m), 3290 (s), 3172 (s), 3051 (s), 2872 (s), 2465 (w), 2140 (w), 1634 (m), 1535 (s), 1494 (s), 1436 (s), 1389 (s), 1352 (s), 1300 (s), 1109 (s), 1049 (m), 1029 (w), 877 (w), 843 (s), 830 (m), 769 (w), 691 (w), 643 (s), 519 (w), 495 (s), 426 (w), 404 (w). Preparation of [HNEt 3 ][DNT] (11). Triethylamine (1.50 g, 14.84 mmol) was added to a solution of HDNT (0.398 g, 2.50 mmol) in water (5 ml). The solvent of the resulting yellow to orange solution was slowly evaporated in air, resulting in an orange crystalline solid of [HNEt 3 ][DNT] (yield 0.593 g, 92 %). DTA: 100 °C melting, 190 °C decomposition. NMR (DMSO-d 6 ) δ(ppm): 1 H (399.80 MHz) 8.88 (bs, 1H, NH + ), 3.10 (q, 3 J( 1 H 1 H) = 7 Hz, 6H, NCH 2 CH 3 ), 1.17 (t, 3 J( 1 H 1 H) = 7 Hz, 9H, NCH 2 CH 3 ); 13 C (100.54 MHz) 162.9 (s, C-NO 2 ), 45.8 (s, NCH 2 CH 3 ), 40.2 (s,NCH 2 CH 3 ); 14 N (36.14 MHz) -20.9 (s, τ 1/2 = 180Hz, NO 2 ), -51 (s, τ 1/2 = 1000Hz, DNT - ), -147 (s, τ 1/2 = 700 Hz, DNT - ), -325 (s, τ 1/2 = 300Hz, HNEt 3 + ). IR (ATR, 20°C) ῦ/cm -1 : 3011 (m), 2780 (w), 2707 (m), 2509 (w), 2456 (w), 1529 (s), 1487 (s), 1473 (w), 1405 (w), 1378 (s), 1346 (s), 1334 (w), 1304 (w), 1292 (s), 1163 (m), 1108 (m), 1073 (w), 1062 (w), 1047 (w), 1030 (s), 1012 (w), 903 (w), 866 (w), 840 (s), 828 (w), 796 (m), 770 (w), 750 (w), 648 (s), 598 (w), 554 (s), 510 (s), 466 (s), 454 (w). Preparation of [H 2 NEt 2 ][DNT] (12). Diethylamine (1.50 g, 20.54 mmol) was added to a solution of HDNT (0.398 g, 2.50 mmol) in water (5 ml). The solvent of the resulting yellow to orange solution was slowly evaporated in the air, resulting in an orange crystalline solid of monoclinic [HNEt 3 ][DNT] (12a) (yield 0.551 g, 95 %). Crystals of the triclinic polymorph (12b) were obtained when a saturated solution of [H 2 NEt 2 ][DNT] in diethylamine was evaporated in air. DTA: 150 °C melting, 190 °C decomposition. NMR (CD 3 CN) δ(ppm): 1 H (399.80 MHz) δ(ppm) 6.78 and 6.22 (bs, 2H, NH 2 + ), 3.09 (m, 4H, NCH 2 CH 3 ), 1.26 (m, 6H, NCH 2 CH 3 ); 13 C (100.54 MHz) 163.9 (s, C- 169 NO 2 ), 43.7 (s, NCH 2 CH 3 ), 11.6 (s, NCH 2 CH 3 ); 14 N (36.14 MHz) -21.5 (s, τ 1/2 = 80 Hz, NO 2 ), -330.2(s, τ 1/2 = 100 Hz, H 2 NEt 2 + ). IR (ATR, 20°C) ῦ/cm -1 : 3008 (m), 2933 (w), 2863 (w), 2822 (m), 2518 (m), 2420 (w), 2132 (w), 1682 (w), 1623 (m), 1543 (s), 1494 (s), 1476 (m), 1452 (w), 1416 (w), 1381 (s), 1352 (s), 1338 (w), 1297 (s), 1159 (m), 1124 (m), 1048 (w), 970 (w), 915 (w), 862 (w), 841 (s), 828 (m), 800 (m), 770 (w), 750 (w), 646 (s), 604 (w), 538 (w), 511 (m), 491 (w), 432 (w), 412 (w). Preparation of guanidinium DNT (13). A solution of HDNT (0.318 g, 2.00 mmol) in ethanol (5 mL) was added to a suspension of guanidinium carbonate (0.901 g, 5.00 mmol) in ethanol (5 mL). The suspension was stirred for 4 hours and then filtered. The yellow filtrate was taken to dryness by evaporation in air, resulting in colorless to yellow crystals of guanidinium DNT (yield: 0.371 g, 85 %). DTA: 200 °C melting and decomposition. NMR (DMSO-d 6 ) δ(ppm): 1 H (500.14 MHz) 6.94 (s, C(NH 2 ) 3 + ), 13 C (125.76 MHz) 162.9 (2C, C-NO 2 ), 158.0 (C(NH 2 ) 3 + ); 14 N (36.14 MHz) -20.9 (s, τ 1/2 = 200 Hz, NO 2 ), -50 (s, τ 1/2 >1000 Hz, DNT - ), -147 (s, τ 1/2 = 800Hz, DNT - ), -301(s, τ 1/2 = 2000Hz, C(NH 2 ) 3 + ). IR (ATR, 20°C) ῦ/cm -1 : 3470 (s), 3400 (m), 3357 (m), 3293 (w), 3268 (w), 3190 (m), 2790 (w), 2693 (w), 2464 (w), 2418 (w), 2230 (w), 2148 (w), 1647 (s), 1579 (w), 1526 (s), 1477 (s), 1413 (m), 1387 (s), 1339 (s), 1306 (w), 1294 (s), 1144 (w), 1120 (m), 1048 (m), 1015 (w), 922 (w), 871 (w), 845 (s), 832 (w), 819 (w), 768 (w), 737 (m), 646 (m), 554 (m), 533 (s), 432 (m). Preparation of aminoguanidinium DNT (14). A solution of HDNT (0.513 g, 3.22 mmol) in ethanol (5 mL) was added to a suspension of aminoguanidinium bicarbonate (0.438 g, 3.22 mmol) in ethanol (20 mL), resulting in effervescence. The suspension was stirred vigorously for two days and then vacuum filtered. The pale yellow filtrate was taken to dryness by evaporation of the solvent in air, resulting in a pale yellow solid of aminoguanidinium DNT (yield: 0.603 g, 80 %) as an orange-yellow crystalline material. DTA: 200 °C (sharp exotherm) decomposition. NMR (DMSO-d 6 ) δ(ppm): 1 H (399.80 MHz) 8.54 (s, 1H, NH-NH 2 ), 7.24 (bs, 2H, C-NH 2 ) 6.72 (bs, 2H, C-NH 2 ), 4.68 (s, 2H, NH-NH 2 ); 13 C (100.54 MHz) 162.9 (C-NO 2 ), 158.8 (CH 7 N 3 + ); 14 N (36.14 MHz) -21.1(s, τ 1/2 = 200 Hz, NO 2 ), -53 (s, τ 1/2 = 1000 Hz, DNT - ), -149 (s, τ 1/2 = 1000 Hz, DNT - ). IR (ATR, 20°C) ῦ/cm -1 : 3469 (w), 3418 (s), 3363 (w), 3326 (m), 3293 (w), 3217 (m), 2922 (m), 2852 (w), 2740 (w), 2693 (w), 2647 (w), 2468 (w), 2418 (w), 2147 (w), 1670 (s), 1543 (w), 1529 (s), 1484 (s), 1412 (w), 1386 (s), 1340 (s), 1308 (m), 1294 (m), 1198 (m), 1121 (w), 1078 (m), 1044 (w), 947 (s), 876 (w), 843 (s), 829 (w), 766 (w), 737 (w), 687 (w), 645 (m), 620 (w), 604 (w), 581 (m), 513 (w), 471 (m). 170 Preparation of pyridinium DNT·H 2 O (15·H 2 O). Pyridine (1.50 g, 18.97 mmol) was added to a solution of HDNT (0.398 g, 2.50 mmol) in water (5 ml). The resulting yellow to orange solution was taken to dryness by evaporation of the solvent in air, resulting in an orange crystalline solid of pyridinium DNT·H 2 O (yield: 0.621 g, 97 %). DTA: 120 °C melting, 160 °C decomposition. NMR (DMSO-d 6 ) δ(ppm): 1 H (399.80 MHz) 8.94 (m, 2H, pyridinium), 8.61 (m, 1H, pyridinium), 8.08 (m, 2H, pyridinium), 6.2 (vbs, 1H, py-H). 13 C (100.54 MHz) 162.9 (s, C-NO 2 ), 146.2 (s, pyridinium), 142.4 (s, pyridinium), 127.2 (s, pyridinium); 14 N (36.14 MHz) -20.9 (s, τ 1/2 = 160 Hz, NO 2 ), -51 (s, τ 1/2 = 800 Hz, DNT - ), -146 (s, τ 1/2 = 500 Hz, DNT - ), -168.8 (s, τ 1/2 = 90 Hz, pyridinium). IR (ATR, 20°C) ῦ/cm -1 : 3349 (m), 3141 (w), 3103 (m), 3075 (w), 2160 (m), 2040 (w), 1998 (w), 1724 (m), 1637 (m), 1537 (s), 1486 (s), 1410 (m), 1382 (s), 1348 (s), 1306 (m), 1294 (m), 1257 (m), 1198 (m), 1159 (m), 1120 (m), 1050 (m), 1023 (m), 999 (s), 947 (m), 864 (m), 839 (m), 827 (s), 757 (s), 682 (s), 648 (m), 580 (s). Preparation of aminotetrazolium DNT·H 2 O (16·H 2 O). A solution of HDNT (0.232 g, 1.46 mmol) in water (5 mL) was added to a suspension of aminotetrazole (0.124 g, 1.46 mmol) in water (10 mL), which led to the complete consumption of the aminotetrazole. The clear colorless solution was taken to dryness by evaporation of the solvent in air, resulting in colorless crystals of aminotetrazolium DNT·H 2 O (yield 0.367 g, 96 %). DTA: 160 °C (sharp exotherm) decomposition. NMR (DMSO-d 6 ) δ(ppm): 1 H (500.14 MHz) 9.00 (bs, 4H, CH 4 N 5 + ); 13 C (125.76 MHz) 162.5 (C-NO 2 ), 154.2 (s, CH 4 N 5 + ); 14 N (36.14 MHz) - 22.1 (s, τ 1/2 = 250Hz, NO 2 ), -58 (s, τ 1/2 = 2000Hz, DNT - ), -149 (s, τ 1/2 = 2000 Hz, DNT - ). IR (ATR, 20°C) ῦ/cm -1 : 3529 (m), 3316 (w), 3252 (w), 3135 (m), 2972 (w), 2825 (w), 2674 (w), 2160 (w), 1693 (s), 1657 (w), 1558 (w), 1541 (s), 1501 (s), 1398 (m), 1362 (m), 1306 (s), 1136 (m), 1105 (w), 1075 (w), 1037 (m), 986 (m), 896 (w), 844 (s), 826 (m), 770 (w), 736 (w), 673 (w), 643 (s), 609 (w), 516 (m). Preparation of PPh 4 [DNT] (17). A solution of HDNT (0.159 g, 1.00 mmol) in water (5 mL) was added to a solution of PPh 4 Cl (0.450 g, 1.20 mmol) in water (10 mL), resulting in the precipitation of a white solid. The mixture was filtered and the filtration residue washed with plenty of water. The solid was dried in a vacuum, resulting in an amorphous solid of PPh 4 [DNT] (yield: 0.487 g, 98 %). Single crystals suitable for crystal structure determination were grown from an acetone solution. DTA: 180 °C melting, 350 °C decomposition. NMR (CD 3 CN) δ(ppm): 1 H (599.81 MHz) 7.7 (m, 16H), 7.9 (m, 4H); 13 C (150.84 MHz) 164.3 (C-NO 2 ), 136.3 (PPh 4 ), 135.6 (PPh 4 ), 131.3 (PPh 4 ), 118.9 (d, 1 J( 13 C 31 P) = 90 Hz, PPh 4 ); 14 N (36.14 MHz) -19.8(s, τ 1/2 = 60 Hz NO 2 ), -51 (s, τ 1/2 = 500 Hz, DNT - ); 31 P{ 1 H} (242.82 MHz) 22.9 (s). 171 IR (ATR, 20°C) ῦ/cm -1 : 3086 (w), 3059 (w), 3026 (w), 1587 (m), 1535 (s), 1479 (s), 1434 (s), 1396 (w), 1377 (s), 1339 (s), 1326 (w), 1299 (w), 1285 (m), 1190 (m), 1164 (m), 1107 (s), 1037 (w), 1026 (w), 996 (m), 938 (w), 862 (w), 836 (s), 827 (w), 764 (m), 750 (s), 720 (s), 699 (m), 686 (s), 651 (w), 615 (m), 525 (s), 470 (w), 441 (w), 428 (w), 409 (w). Preparation of PPN[DNT] (18). A solution of HDNT (0.875 g; 5.50 mmol) in water (20 mL) was added to solution of PPNCl (3.158 g ; 5.501 mmol) in water (300 mL), resulting in the precipitation of a white solid. The mixture was vacuum filtered, and the filtration residue suspended in water (300 mL) and vacuum filtered again. The obtained solid was dried in a vacuum for 12 hours, resulting in a white amorphous solid of PPN[DNT] (3.075 g, 80 %). Crystals suitable for X-ray diffraction were grown from an acetone solution. DTA: 150 °C melting, 340 °C decomposition. NMR (CD 3 CN) δ(ppm): 1 H (599.81 MHz) 7.5 (m, 12H), 7.6 (m, 12H), 7.7 (m, 6H); 13 C (150.84 MHz) 164.5 (s, C-NO 2 ), 134.6 (PPh 3 ), 133.2 (PPh 3 ), 130.3 (PPh 3 ), 128.8 (PPh 3 ); 14 N (36.14 MHz) -19.9 (s, τ 1/2 = 50 Hz, NO 2 ), -51 (s, τ 1/2 = 300 Hz, DNT - ) 31 P{ 1 H} (242.82 MHz) 20.8. Raman (20 °C, 200 mW) ῦ/cm -1 3172 (0.12), 3147 (0.16), 3068 (0.17), 3060 (5.18), 3011 (0.11), 2957 (0.11), 1588 (2.09), 1576 (0.20), 1524 (0.56), 1483 (0.42), 1387 (10.00), 1369 (0.28), 1341 (1.33), 1329 (0.14), 1298 (0.36), 1284 (0.28), 1189 (0.29), 1161 (0.64), 1110 (0.31), 1096 (7.04), 1028 (1.22), 1004 (3.47), 829 (0.84), 767 (0.22), 727 (0.17), 667 (1.33), 616 (0.70), 509 (0.14). IR (ATR, 20°C) ῦ/cm -1 : 3052 (w), 2917 (w), 2038 (w), 1969 (w), 1910 (w), 1828 (w), 1586 (m), 1530 (s), 1479 (s), 1436 (s), 1397 (w), 1374 (m), 1339 (m), 1327 (w), 1296 (w), 1281 (m), 1237 (s), 1180 (m), 1161 (w), 1111 (s), 1075 (w), 1038 (w), 1024 (m), 996 (m), 932 (w), 856 (w), 839 (m), 827 (m), 802 (m), 759 (w), 746 (s), 721 (s), 689 (s), 615 (w), 552 (m), 526 (s), 497 (s), 486 (m), 461 (m), 450 (w), 437 (m). Preparation of TMA[DNT] (19). A solution of TMA[OH]·5H 2 O (0.725 g, 4.00 mmol) in water (20 mL) was added to a solution of HDNT (0.636 g, 4.00 mmol) in water (20 mL). The bright yellow solution was taken to dryness on a rotary evaporator. The yellow residue was recrystallized from ethanol, yielding colorless crystals of TMA[DNT] (0.780 g, 84%). DTA: 225 °C melting, 235 °C decomposition. NMR (acetone-d 6 ) δ(ppm): 1 H (599.81 MHz) 3.51 (12H); 13 C (150.84 MHz) 164.5 (s, C-NO 2 ), 56.0 (t, 1 J( 13 C 14 N) = 4 Hz, NMe 4 ); 14 N (36.14 MHz) -19.8 (s, τ 1/2 = 50 Hz, NO 2 ), -51 (s, τ 1/2 = 400 Hz, DNT - ), - 148 (s, τ 1/2 = 300 Hz, DNT - ). Raman (20 °C, 200 mW) ῦ/cm -1 : 3037 (0.8), 2987 (0.4), 2963 (0.2), 2929 (0.4), 1536 (0.5), 1451 (0.2), 1391 (10.0), 1341 (0.6), 1331 (0.6), 1301 (0.3), 1286 (0.5), 1097 (7.1), 1010 (0.4), 952 (0.3), 832 (1.0), 769 (0.1), 759 (1.2), 510 (0.2), 299 (0.3). 172 IR (ATR, 20°C) ῦ/cm -1 : 3039 (w), 2961 (vw), 2666 (vw), 2430 (w), 2374 (vw), 1539 (vs), 1518 (m), 1483 (vs), 1451 (w sh), 1418 (vw), 1397 (w), 1376 (s), 1342 (s), 1331 (s), 1298 (m), 1284 (m), 1095 (m), 1038 (w), 951 (s), 841 (vs), 828 (s), 771 (w), 648 (s), 599 (w), 457 (w). Vibrational Data for 3,5-dinitro-1H-1,2,4-triazole (HDNT). IR (ATR, 20°C) ῦ/cm -1 : 3039 (vw), 2992 (vw), 2934 (vw), 2883 (vw), 2839 (vw), 2802 (vw), 2748 (vw), 2639 (vw), 2557 (vw), 2162 (vw), 1698 (vw), 1563 (vs), 1530 (s sh), 1485 (s), 1430 (w), 1377 (s sh), 1363 (s sh), 1313 (vs), 1281 (m sh), 1174 (m), 1038 (m), 1023 (m sh), 1008 (m sh), 839 (vs), 825 (vs), 764 (vw), 730 (vw), 649 (m), 634 (m sh), 593 (m), 509 (m), 500 (m sh). A2.2 Crystallographic details A2.2.1 Crystal Structure Report for LiDNT·2H 2 O (2 . 2H 2 O) Figure A2.1: Projection of the packing in LiDNT·2H 2 O (2·2H 2 O) along the b-axis (blue=nitrogen, red=oxygen, black=carbon, gold=lithium). 173 Table A2.1: Sample and crystal data for LiDNT·2H 2 O (2·2H 2 O). Identification code LiDNT Chemical formula C2H4LiN5O6 Formula weight 201.04 Temperature 100(2) K Wavelength 0.71073 Å Crystal size 0.198 x 0.579 x 0.590 mm Crystal habit clear yellow prism Crystal system monoclinic Space group C 1 2/c 1 Unit cell dimensions a = 15.342(2) Å α = 90° b = 8.3048(11) Å β = 113.079(2)° c = 6.3573(8) Å γ = 90° Volume 745.17(17) Å3 Z 4 Density (calculated) 1.792 g/cm3 Absorption coefficient 0.172 mm-1 F(000) 408 Table A2.2: Data collection and structure refinement for 2·2H 2 O. Diffractometer Bruker APEX DUO Radiation source fine-focus tube, MoKα Theta range for data collection 2.85 to 30.53° Index ranges -21<=h<=21, -11<=k<=11, -9<=l<=9 Reflections collected 8446 Independent reflections 1124 [R(int) = 0.0269] Absorption correction multi-scan Max. and min. transmission 0.9670 and 0.9050 Structure solution technique direct methods Structure solution program SHELXTL XT 2013/6 (Sheldrick, 2013) Refinement method Full-matrix least-squares on F2 Refinement program SHELXTL XLMP 2014/1 (Bruker AXS, 2013) Function minimized Σ w(Fo2 - Fc2)2 Data / restraints / parameters 1124 / 0 / 71 Goodness-of-fit on F2 1.077 Final R indices 991 data; I>2σ(I) R1 = 0.0311, wR2 = 0.0850 all data R1 = 0.0359, wR2 = 0.0894 Weighting scheme w=1/[σ2(Fo2)+(0.0488P)2+0.4470P] where P=(Fo2+2Fc2)/3 Largest diff. peak and hole 0.365 and -0.291 eÅ-3 R.M.S. deviation from mean 0.072 eÅ-3 174 A2.2.2 Crystal Structure Report for NaDNT·2H 2 O (3·2H 2 O) Figure A2.2: Projection of the packing in NaDNT·2H 2 O (3·2H 2 O) down the b-axis. Table A2.3: Sample and crystal data for NaDNT·2H 2 O (3·2H 2 O). Identification code NaDNT_2H2O Chemical formula C 4 H 8 N 10 Na 2 O 12 Formula weight 434.18 Temperature 100(2) K Wavelength 0.71073 Å Crystal size 0.174 x 0.373 x 0.511 mm Crystal habit clear colourless prism Crystal system monoclinic Space group C 1 2/c 1 Unit cell dimensions a = 15.3444(7) Å α = 90° b = 8.7270(4) Å β = 114.9340(6)° c = 6.6379(3) Å γ = 90° Volume 806.04(6) Å 3 Z 2 Density (calculated) 1.789 g/cm 3 Absorption coefficient 0.216 mm -1 F(000) 440 175 Table A2.4: Data collection and structure refinement for (3·2H 2 O). Diffractometer Bruker APEX DUO Radiation source fine-focus tube, MoKα Theta range for data collection 2.75 to 30.45° Index ranges -21<=h<=21, -12<=k<=12, -9<=l<=9 Reflections collected 9609 Independent reflections 1231 [R(int) = 0.0207] Absorption correction multi-scan Max. and min. transmission 0.9630 and 0.8980 Structure solution technique direct methods Structure solution program SHELXTL XS 2013/1 (Sheldrick, 2013) Refinement method Full-matrix least-squares on F 2 Refinement program SHELXTL XLMP 2014/1 (Bruker AXS, 2013) Function minimized Σ w(F o 2 - F c 2 ) 2 Data / restraints / parameters 1231 / 0 / 72 Goodness-of-fit on F 2 1.095 Δ/σ max 0.001 Final R indices 1159 data; I>2σ(I) R1 = 0.0237, wR2 = 0.0662 all data R1 = 0.0254, wR2 = 0.0673 Weighting scheme w=1/[σ 2 (F o 2 )+(0.0342P) 2 +0.4360P] where P=(F o 2 +2F c 2 )/3 Extinction coefficient 0.0089(13) Largest diff. peak and hole 0.411 and -0.255 eÅ -3 R.M.S. deviation from mean 0.046 eÅ -3 176 A2.2.3 Crystal Structure Report for KDNT (3) Figure A2.3: Projection of the packing in KDNT (3) down the a-axis (blue= nitrogen, grey= carbon, red= oxygen, purple= potassium). Table A2.5: Sample and crystal data for KDNT (3). Identification code XGC4_144_imported Chemical formula C 2 KN 5 O 4 Formula weight 197.17 Temperature 100(2) K Wavelength 0.71073 Å Crystal size 0.076 x 0.234 x 0.353 mm Crystal habit colorless blade Crystal system monoclinic Space group P 1 21/n 1 Unit cell dimensions a = 6.741(2) Å α = 90° b = 12.997(4) Å β = 97.697(5)° c = 14.749(5) Å γ = 90° Volume 1280.6(7) Å 3 Z 8 Density (calculated) 2.045 g/cm 3 Absorption coefficient 0.813 mm -1 F(000) 784 Table A2.6: Data collection and structure refinement for 3. 177 Diffractometer Bruker APEX DUO Radiation source fine-focus tube, MoKα Theta range for data collection 2.10 to 30.35° Index ranges -9<=h<=9, -18<=k<=18, -20<=l<=17 Reflections collected 14946 Independent reflections 3713 [R(int) = 0.0395] Coverage of independent reflections 96.3% Absorption correction multi-scan Max. and min. transmission 0.9510 and 0.7990 Structure solution technique direct methods Structure solution program SHELXTL XS 2013/1 (Bruker AXS) Refinement method Full-matrix least-squares on F 2 Refinement program SHELXL-2013 (Sheldrick, 2013) Function minimized Σ w(F o 2 - F c 2 ) 2 Data / restraints / parameters 3713 / 0 / 217 Goodness-of-fit on F 2 1.277 Final R indices 3121 data; I>2σ(I) R1 = 0.0764, wR2 = 0.1578 all data R1 = 0.0900, wR2 = 0.1620 Weighting scheme w=1/[σ 2 (F o 2 )+7.1292P] where P=(F o 2 +2F c 2 )/3 Absolute structure parameter 0.0(0) Largest diff. peak and hole 1.247 and -0.759 eÅ -3 R.M.S. deviation from mean 0.120 eÅ -3 178 A2.2.4 Crystal Structure Report for KDNT·2H 2 O (3 . 2H 2 O) Table A2.7: Sample and crystal data for KDNT·2H 2 O (3·2H 2 O). Identification code KDNT2H2O_2 Chemical formula C 2 H 4 KN 5 O 6 Formula weight 233.20 Temperature 130(2) K Wavelength 0.71073 Å Crystal size 0.110 x 0.150 x 0.360 mm Crystal habit clear orange prism Crystal system triclinic Space group P 1 Unit cell dimensions a = 4.5220(15) Å α = 96.093(4)° b = 6.355(2) Å β = 92.533(4)° c = 7.770(3) Å γ = 108.569(4)° Volume 209.75(12) Å 3 Z 1 Density (calculated) 1.846 g/cm 3 Absorption coefficient 0.652 mm -1 F(000) 118 179 Table A2.8: Data collection and structure refinement for 3·2H 2 O. Diffractometer Bruker SMART APEX Radiation source fine-focus tube, MoKα Theta range for data collection 2.65 to 27.48° Index ranges -5<=h<=5, -8<=k<=7, -9<=l<=9 Reflections collected 2303 Independent reflections 1445 [R(int) = 0.0228] Coverage of independent reflections 95.9% Absorption correction multi-scan Max. and min. transmission 0.9320 and 0.7990 Structure solution technique direct methods Structure solution program SHELXTL XT 2013/6 (Sheldrick, 2013) Refinement method Full-matrix least-squares on F 2 Refinement program SHELXTL XLMP 2014/1 (Bruker AXS, 2013) Function minimized Σ w(F o 2 - F c 2 ) 2 Data / restraints / parameters 1445 / 3 / 127 Goodness-of-fit on F 2 1.138 Δ/σ max 0.011 Final R indices 1416 data; I>2σ(I) R1 = 0.0255, wR2 = 0.0714 all data R1 = 0.0263, wR2 = 0.0725 Weighting scheme w=1/[σ 2 (F o 2 )+(0.0385P) 2 +0.0115P] where P=(F o 2 +2F c 2 )/3 Absolute structure parameter 0.1(0) Largest diff. peak and hole 0.219 and -0.233 eÅ -3 R.M.S. deviation from mean 0.057 eÅ -3 180 A2.2.5 Crystal Structure Report for RbDNT (4) Figure A2.4: Projection of the packing of RbDNT (4) down the a-axis. Table A2.9: Sample and crystal data for RbDNT (4). Identification code RbDNT Chemical formula C 2 N 5 O 4 Rb Formula weight 243.54 Temperature 140(2) K Wavelength 0.71073 Å Crystal size 0.040 x 0.200 x 0.220 mm Crystal habit clear yellow plate Crystal system orthorhombic Space group P 21 21 21 Unit cell dimensions a = 7.6561(12) Å α = 90° b = 7.8172(12) Å β = 90° c = 11.0946(17) Å γ = 90° Volume 664.00(18) Å 3 Z 4 Density (calculated) 2.436 g/cm 3 Absorption coefficient 7.440 mm -1 F(000) 464 181 Table A2.10: Data collection and structure refinement for 4. Diffractometer Bruker SMART APEX Radiation source fine-focus tube, MoKα Theta range for data collection 3.19 to 28.72° Index ranges -9<=h<=10, -10<=k<=10, -14<=l<=9 Reflections collected 4210 Independent reflections 1583 [R(int) = 0.0400] Coverage of independent reflections 94.2% Absorption correction multi-scan Max. and min. transmission 0.7550 and 0.2910 Structure solution technique direct methods Structure solution program SHELXTL XT 2013/6 (Sheldrick, 2013) Refinement method Full-matrix least-squares on F 2 Refinement program SHELXTL XLMP 2014/1 (Bruker AXS, 2013) Function minimized Σ w(F o 2 - F c 2 ) 2 Data / restraints / parameters 1583 / 0 / 109 Goodness-of-fit on F 2 0.998 Δ/σ max 0.001 Final R indices 1421 data; I>2σ(I) R1 = 0.0335, wR2 = 0.0703 all data R1 = 0.0399, wR2 = 0.0729 Weighting scheme w=1/[σ 2 (F o 2 )+(0.0277P) 2 ] where P=(F o 2 +2F c 2 )/3 Absolute structure parameter 0.0(0) Largest diff. peak and hole 0.858 and -0.796 eÅ -3 R.M.S. deviation from mean 0.123 eÅ -3 182 A2.2.6 Crystal Structure Report for CsDNT (5) Figure A2.5: Projection of the packing in CsDNT (5) down the c-axis (blue= nitrogen, grey= carbon, red= oxygen, purple= cesium). Table A2.11: Sample and crystal data for CsDNT (5). Identification code CsDNT_2 Chemical formula C 2 CsN 5 O 4 Formula weight 290.98 Temperature 133(2) K Wavelength 0.71073 Å Crystal size 0.040 x 0.190 x 0.290 mm Crystal habit clear pale orange plate Crystal system monoclinic Space group P 1 21/n 1 Unit cell dimensions a = 7.974(3) Å α = 90° b = 10.088(3) Å β = 109.941(4)° c = 9.267(3) Å γ = 90° Volume 700.8(4) Å 3 Z 4 Density (calculated) 2.758 g/cm 3 Absorption coefficient 5.269 mm -1 F(000) 536 183 Table A2.12: Data collection and structure refinement for 5. Diffractometer Bruker SMART APEX Radiation source fine-focus tube, MoKα Theta range for data collection 2.92 to 28.63° Index ranges -9<=h<=10, -13<=k<=10, -12<=l<=12 Reflections collected 3944 Independent reflections 1589 [R(int) = 0.1014] Coverage of independent reflections 88.2% Absorption correction multi-scan Max. and min. transmission 0.8170 and 0.3100 Structure solution technique direct methods Structure solution program SHELXTL XT 2013/6 (Sheldrick, 2013) Refinement method Full-matrix least-squares on F 2 Refinement program SHELXTL XLMP 2014/1 (Bruker AXS, 2013) Function minimized Σ w(F o 2 - F c 2 ) 2 Data / restraints / parameters 1589 / 0 / 109 Goodness-of-fit on F 2 1.040 Final R indices 1433 data; I>2σ(I) R1 = 0.0332, wR2 = 0.0848 all data R1 = 0.0357, wR2 = 0.0869 Weighting scheme w=1/[σ 2 (F o 2 )+(0.0434P) 2 ] where P=(F o 2 +2F c 2 )/3 Largest diff. peak and hole 1.170 and -1.327 eÅ -3 R.M.S. deviation from mean 0.222 eÅ -3 184 A2.2.7 Crystal Structure Report for Sr(DNT) 2 ·6H 2 O (6 . 6H 2 O) Figure A2.6: Solid state structure of Sr(DNT) 2 . 6H 2 O (6·6H 2 O). Thermal ellipsoids are shown at the 50% probability level. Hydrogen atom positions were determined from the electron density map and are depicted as spheres of arbitrary radius. Figure A2.7: Projection of the packing of (6·6H 2 O) down the a-axis. 185 Table A2.13: Sample and crystal data for Sr(DNT) 2 . 6H 2 O (6·6H 2 O). Identification code SrDNT Chemical formula C 4 H 12 N 10 O 14 Sr Formula weight 511.86 Temperature 100(2) K Wavelength 0.71073 Å Crystal size 0.200 x 0.310 x 0.410 mm Crystal habit clear yellow-orange prism Crystal system triclinic Space group P -1 Unit cell dimensions a = 9.2895(3) Å α = 108.2150(10)° b = 12.8551(4) Å β = 90.8870(10)° c = 16.5552(7) Å γ = 111° Volume 1736.03(11) Å 3 Z 4 Density (calculated) 1.958 g/cm 3 Absorption coefficient 3.204 mm -1 F(000) 1024 Table A2.14: Data collection and structure refinement for 6·6H 2 O. Diffractometer Bruker APEX DUO Radiation source fine-focus tube, MoKα Theta range for data collection 1.83 to 30.62° Index ranges -13<=h<=13, -18<=k<=18, -23<=l<=23 Reflections collected 42154 Independent reflections 10328 [R(int) = 0.0331] Absorption correction multi-scan Max. and min. transmission 0.5670 and 0.3530 Structure solution technique direct methods Structure solution program SHELXTL XS 2013/1 (Sheldrick, 2013) Refinement method Full-matrix least-squares on F 2 Refinement program SHELXTL XLMP 2014/1 (Bruker AXS, 2013) Function minimized Σ w(F o 2 - F c 2 ) 2 Data / restraints / parameters 10328 / 0 / 619 Goodness-of-fit on F 2 1.032 Δ/σ max 0.001 Final R indices 8672 data; I>2σ(I) R1 = 0.0240, wR2 = 0.0491 all data R1 = 0.0351, wR2 = 0.0520 Weighting scheme w=1/[σ 2 (F o 2 )+(0.0239P) 2 +0.2726P] where P=(F o 2 +2F c 2 )/3 Largest diff. peak and hole 0.331 and -0.430 eÅ -3 R.M.S. deviation from mean 0.079 eÅ -3 A2.2.8 Crystal Structure Report for Ba(DNT) 2 ·11H 2 O (7·11H 2 O) 186 Figure A2.8: Solid state structure of Ba(DNT) 2 ·11H 2 O (7·11H 2 O). Thermal ellipsoids are shown at the 50% probability level. Hydrogen atom positions were determined from the electron density map and are depicted as spheres of arbitrary radius. No satisfactory hydrogen atoms for water molecules O5 through O9 could be found. These hydrogen atoms are therefore not depicted. Figure A2.9: Projection of the packing of (7·11H 2 O) down the a-axis blue= nitrogen, grey= carbon, red= oxygen, green= barium. Table A2.15: Sample and crystal data for Ba(DNT) 2 ·11H 2 O (7·11H 2 O). 187 Identification code MK16 Chemical formula C 4 H 22 BaN 10 O 19 Formula weight 651.65 g/mol Temperature 100(2) K Wavelength 0.71073 Å Crystal size 0.092 x 0.123 x 0.149 mm Crystal system monoclinic Space group P 1 21/m 1 Unit cell dimensions a = 6.9239(6) Å α = 90° b = 18.7743(15) Å β = 111.6350(10)° c = 9.2088(7) Å γ = 90° Volume 1112.73(16) Å 3 Z 2 Density (calculated) 1.945 g/cm 3 Absorption coefficient 1.891 mm -1 F(000) 648 Table A2.16: Data collection and structure refinement for 7·11H 2 O. Theta range for data collection 2.17 to 30.61° Index ranges -9<=h<=9, -26<=k<=26, -13<=l<=13 Reflections collected 26945 Independent reflections 3476 [R(int) = 0.0424] Max. and min. transmission 0.8450 and 0.7660 Structure solution technique direct methods Structure solution program SHELXTL XS 2013/1 (Bruker AXS) Refinement method Full-matrix least-squares on F 2 Refinement program SHELXL-2014 (Sheldrick, 2014) Function minimized Σ w(F o 2 - F c 2 ) 2 Data / restraints / parameters 3476 / 7 / 178 Goodness-of-fit on F 2 1.204 Final R indices 3214 data; I>2σ(I) R1 = 0.0370, wR2 = 0.0831 all data R1 = 0.0423, wR2 = 0.0846 Weighting scheme w=1/[σ 2 (F o 2 )+6.3309P] where P=(F o 2 +2F c 2 )/3 Absolute structure parameter 0.0(0) Largest diff. peak and hole 1.396 and -3.222 eÅ -3 R.M.S. deviation from mean 0.146 eÅ -3 188 A2.2.9 Crystal structure report for [Ag(NH 3 )][DNT] (9) Figure A2.10: Projection of the packing of [Ag(NH 3 )][DNT] (9) perpendicular to the 010 plane. Table A2.17: Sample and crystal data for [Ag(NH 3 )][DNT] (9). Identification code AgDNT_NH3 Chemical formula C 2 H 3 AgN 6 O 4 Formula weight 282.97 Temperature 140(2) K Wavelength 0.71073 Å Crystal size 0.130 x 0.150 x 0.490 mm Crystal habit clear orange prism Crystal system monoclinic Space group P 1 21/n 1 Unit cell dimensions a = 7.8061(19) Å α = 90° b = 5.5756(14) Å β = 101.686(3)° c = 17.034(4) Å γ = 90° Volume 726.0(3) Å 3 Z 4 Density (calculated) 2.589 g/cm 3 Absorption coefficient 2.773 mm -1 F(000) 544 189 Table A2.18: Data collection and structure refinement for 9. Diffractometer Bruker SMART APEX Radiation source fine-focus tube, MoKα Theta range for data collection 2.44 to 28.45° Index ranges -9<=h<=10, -7<=k<=6, -22<=l<=15 Reflections collected 4279 Independent reflections 1688 [R(int) = 0.0176] Coverage of independent reflections 91.8% Absorption correction multi-scan Max. and min. transmission 0.7140 and 0.3440 Structure solution technique direct methods Structure solution program SHELXTL XT 2013/6 (Sheldrick, 2013) Refinement method Full-matrix least-squares on F 2 Refinement program SHELXTL XLMP 2014/1 (Bruker AXS, 2013) Function minimized Σ w(F o 2 - F c 2 ) 2 Data / restraints / parameters 1688 / 6 / 130 Goodness-of-fit on F 2 1.084 Δ/σ max 0.001 Final R indices 1604 data; I>2σ(I) R1 = 0.0185, wR2 = 0.0476 all data R1 = 0.0195, wR2 = 0.0482 Weighting scheme w=1/[σ 2 (F o 2 )+(0.0245P) 2 +0.4039P] where P=(F o 2 +2F c 2 )/3 Largest diff. peak and hole 0.440 and -0.623 eÅ -3 R.M.S. deviation from mean 0.080 eÅ -3 A2.2.10 Crystal Structure Report for NH 4 [DNT] . 2H 2 O (10 . 2H 2 O) Figure A2.11: Projection of the packing of NH 4 [DNT] ·2H 2 O (10·2H 2 O) perpendicular to the 100 plane. 190 Table A2.19: Sample and crystal data for NH 4 [DNT] ·2H 2 O (10·2H 2 O) Identification code NH4DNT Chemical formula C 2 H 8 N 6 O 6 Formula weight 212.14 Temperature 100(2) K Wavelength 0.71073 Å Crystal size 0.080 x 0.230 x 0.290 mm Crystal habit clear orange plate Crystal system triclinic Space group P 1 Unit cell dimensions a = 4.653(3) Å α = 93.105(7)° b = 6.461(4) Å β = 93.948(7)° c = 7.619(4) Å γ = 108.360(7)° Volume 216.2(2) Å 3 Z 1 Density (calculated) 1.630 g/cm 3 Absorption coefficient 0.158 mm -1 F(000) 110 Table A2.20: Data collection and structure refinement for 10·2H 2 O Diffractometer Bruker SMART APEX Radiation source fine-focus tube, MoKα Theta range for data collection 2.69 to 28.49° Index ranges -6<=h<=5, -8<=k<=8, -9<=l<=9 Reflections collected 2547 Independent reflections 1599 [R(int) = 0.0122] Coverage of independent reflections 91.7% Absorption correction multi-scan Max. and min. transmission 0.9880 and 0.9560 Structure solution technique direct methods Structure solution program SHELXTL XT 2013/6 (Sheldrick, 2013) Refinement method Full-matrix least-squares on F 2 Refinement program SHELXTL XLMP 2014/1 (Bruker AXS, 2013) Function minimized Σ w(F o 2 - F c 2 ) 2 Data / restraints / parameters 1599 / 3 / 159 Goodness-of-fit on F 2 1.075 Final R indices 1510 data; I>2σ(I) R1 = 0.0264, wR2 = 0.0610 all data R1 = 0.0291, wR2 = 0.0627 Weighting scheme w=1/[σ 2 (F o 2 )+(0.0323P) 2 +0.0243P] where P=(F o 2 +2F c 2 )/3 Absolute structure parameter -1.0(7) Largest diff. peak and hole 0.182 and -0.155 eÅ -3 R.M.S. deviation from mean 0.037 eÅ -3 A2.2.11 Crystal Structure Report for [HNEt 3 ][DNT] (11) 191 Figure A2.12: Projection of the packing in [HNEt 3 ][DNT] (11) perpendicular to the 010 plane. Table A2.21: Sample and crystal data for [HNEt 3 ][DNT] (11). Identification code Et3NHDNT Chemical formula C 8 H 16 N 6 O 4 Formula weight 260.27 Temperature 100(2) K Wavelength 0.71073 Å Crystal size 0.232 x 0.293 x 0.402 mm Crystal habit clear pale yellow prism Crystal system monoclinic Space group P 1 21/n 1 Unit cell dimensions a = 7.0842(5) Å α = 90° b = 9.3195(6) Å β = 100.3110(11)° c = 19.3391(13) Å γ = 90° Volume 1256.17(15) Å 3 Z 4 Density (calculated) 1.376 g/cm 3 Absorption coefficient 0.111 mm -1 F(000) 552 Table A2.22: Data collection and structure refinement for 11. Diffractometer Bruker APEX DUO Radiation source fine-focus tube, MoKα 192 Theta range for data collection 2.14 to 30.65° Index ranges -10<=h<=9, -13<=k<=13, -27<=l<=27 Reflections collected 29946 Independent reflections 3841 [R(int) = 0.0325] Absorption correction multi-scan Max. and min. transmission 0.9750 and 0.9570 Structure solution technique direct methods Structure solution program SHELXTL XT 2013/6 (Sheldrick, 2013) Refinement method Full-matrix least-squares on F 2 Refinement program SHELXTL XLMP 2014/1 (Bruker AXS, 2013) Function minimized Σ w(F o 2 - F c 2 ) 2 Data / restraints / parameters 3841 / 0 / 169 Goodness-of-fit on F 2 1.042 Δ/σ max 0.001 Final R indices 3189 data; I>2σ(I) R1 = 0.0397, wR2 = 0.0959 all data R1 = 0.0508, wR2 = 0.1022 Weighting scheme w=1/[σ 2 (F o 2 )+(0.0434P) 2 +0.5128P] where P=(F o 2 +2F c 2 )/3 Largest diff. peak and hole 0.623 and -0.484 eÅ -3 R.M.S. deviation from mean 0.047 eÅ -3 193 A2.2.12 Crystal Structure Report for Monoclinic [H 2 NEt 2 ][DNT] (12a) Figure A2.13: Projection of the packing of monoclinic [H 2 NEt 2 ][DNT] (12a) perpendicular to the 100 plane. 194 Table A2.23: Sample and crystal data for [H 2 NEt 2 ][DNT] (12a). Identification code Et2NH2DNT Chemical formula C 6 H 12 N 6 O 4 Formula weight 232.22 Temperature 100(2) K Wavelength 0.71073 Å Crystal size 0.162 x 0.258 x 0.499 mm Crystal habit clear yellow prism Crystal system monoclinic Space group P 1 21/n 1 Unit cell dimensions a = 6.4659(4) Å α = 90° b = 8.9964(6) Å β = 99.5420(10)° c = 18.8357(12) Å γ = 90° Volume 1080.51(12) Å 3 Z 4 Density (calculated) 1.427 g/cm 3 Absorption coefficient 0.120 mm -1 F(000) 488 Table A2.24: Data collection and structure refinement for 12a. Diffractometer Bruker APEX DUO Radiation source fine-focus tube, MoKα Theta range for data collection 2.19 to 29.13° Index ranges -8<=h<=8, -12<=k<=12, -25<=l<=25 Reflections collected 24237 Independent reflections 2903 [R(int) = 0.0293] Absorption correction multi-scan Structure solution technique direct methods Structure solution program SHELXTL XT 2013/6 (Sheldrick, 2013) Refinement method Full-matrix least-squares on F 2 Refinement program SHELXTL XLMP 2014/1 (Bruker AXS, 2013) Function minimized Σ w(F o 2 - F c 2 ) 2 Data / restraints / parameters 2903 / 1 / 153 Goodness-of-fit on F 2 1.034 Δ/σ max 0.001 Final R indices 2518 data; I>2σ(I) R1 = 0.0318, wR2 = 0.0795 all data R1 = 0.0385, wR2 = 0.0842 Weighting scheme w=1/[σ 2 (F o 2 )+(0.0412P) 2 +0.2954P] where P=(F o 2 +2F c 2 )/3 Largest diff. peak and hole 0.243 and -0.255 eÅ -3 R.M.S. deviation from mean 0.042 eÅ -3 195 A2.2.13 Crystal Structure Report for Triclinic [H 2 NEt 2 ][DNT] (12b) Figure A2.14: Projection of the packing of triclinic [H 2 NEt 2 ][DNT] (12b).perpendicular to the 001 plane. Figure A2.15: Projection of the packing of 12b perpendicular to the 010 plane. 196 Figure A2.16: Projection of the packing of 12b perpendicular to the 100 plane. Table A2.25: Sample and crystal data for triclinic [H 2 NEt 2 ][DNT] (12b). Identification code XMK18 Chemical formula C 6 H 12 N 6 O 4 Formula weight 232.20 Temperature 100(2) K Wavelength 1.54178 Å Crystal size 0.132 x 0.209 x 0.257 mm Crystal habit yellow blade Crystal system triclinic Space group P -1 Unit cell dimensions a = 7.3101(9) Å α = 73.347(3)° b = 8.1876(10) Å β = 73.673(2)° c = 9.9470(12) Å γ = 77.130(3)° Volume 540.79(11) Å 3 Z 2 Density (calculated) 1.426 g/cm 3 Absorption coefficient 1.036 mm -1 F(000) 244 197 Table A2.26: Data collection and structure refinement for 12b. Diffractometer Bruker APEX DUO Radiation source IuS microsource, CuKα Theta range for data collection 4.77 to 63.18° Index ranges -8<=h<=8, -9<=k<=9, -10<=l<=11 Reflections collected 8580 Independent reflections 1606 [R(int) = 0.0255] Structure solution technique direct methods Structure solution program SHELXTL XT 2013/5 (Sheldrick, 2013) Refinement method Full-matrix least-squares on F 2 Refinement program SHELXTL XLMP 2014/1 (Bruker AXS, 2013) Function minimized Σ w(F o 2 - F c 2 ) 2 Data / restraints / parameters 1606 / 0 / 153 Goodness-of-fit on F 2 1.029 Final R indices 1486 data; I>2σ(I) R1 = 0.0267, wR2 = 0.0671 all data R1 = 0.0292, wR2 = 0.0689 Weighting scheme w=1/[σ 2 (F o 2 )+(0.0318P) 2 +0.1977P] where P=(F o 2 +2F c 2 )/3 Largest diff. peak and hole 0.135 and -0.153 eÅ -3 R.M.S. deviation from mean 0.032 eÅ -3 A2.2.14 Crystal Structure Report for Guanidinium DNT (13) Figure A2.17: Projection of the packing of guanidinium DNT (13) perpendicular to the 001 plane. 198 Table A2.27: Sample and crystal data for guanidinium DNT (13) Identification code GuaniDNT Chemical formula C 3 H 6 N 8 O 4 Formula weight 218.16 Temperature 100(2) K Wavelength 0.71073 Å Crystal size 0.274 x 0.295 x 0.801 mm Crystal system monoclinic Space group C 1 2/c 1 Unit cell dimensions a = 8.3235(7) Å α = 90° b = 16.7996(14) Å β = 102.3250(13)° c = 12.3068(10) Å γ = 90° Volume 1681.2(2) Å 3 Z 8 Density (calculated) 1.724 g/cm 3 Absorption coefficient 0.155 mm -1 F(000) 896 Table A2.28: Data collection and structure refinement for 13. Diffractometer Bruker APEX DUO Radiation source fine-focus tube, MoKα Theta range for data collection 2.42 to 30.52° Index ranges -11<=h<=11, -23<=k<=24, -17<=l<=17 Reflections collected 20271 Independent reflections 2562 [R(int) = 0.0333] Coverage of independent reflections 99.3% Absorption correction multi-scan Max. and min. transmission 0.9590 and 0.8860 Structure solution technique direct methods Structure solution program SHELXTL XT 2013/6 (Sheldrick, 2013) Refinement method Full-matrix least-squares on F 2 Refinement program SHELXTL XLMP 2014/1 (Bruker AXS, 2013) Function minimized Σ w(F o 2 - F c 2 ) 2 Data / restraints / parameters 2562 / 0 / 160 Goodness-of-fit on F 2 1.075 Final R indices 2221 data; I>2σ(I) R1 = 0.0420, wR2 = 0.1147 all data R1 = 0.0486, wR2 = 0.1189 Weighting scheme w=1/[σ 2 (F o 2 )+(0.0585P) 2 +1.5987P] where P=(F o 2 +2F c 2 )/3 Largest diff. peak and hole 0.486 and -0.262 eÅ -3 R.M.S. deviation from mean 0.055 eÅ -3 199 A2.2.15 Crystal Structure Report for Aminoguanidinium DNT (14) Figure A2.18: Projection of the packing of aminoguanidinium DNT (14) perpendicular to the 100 plane. Table A2.29: Sample and crystal data for aminoguanidinium DNT (14). Identification code XMK20b_01 Chemical formula C 3 H 7 N 9 O 4 Formula weight 233.18 g/mol Temperature 100(2) K Wavelength 1.54178 Å Crystal size 0.096 x 0.116 x 0.182 mm Crystal habit colorless blade Crystal system monoclinic Space group P 1 21/c 1 Unit cell dimensions a = 9.0616(3) Å α = 90° b = 16.2452(5) Å β = 110.189(3)° c = 6.5935(2) Å γ = 90° Volume 910.98(5) Å 3 Z 4 Density (calculated) 1.700 g/cm 3 Absorption coefficient 1.338 mm -1 F(000) 480 200 Table A2.30: Data collection and structure refinement for 14. Diffractometer Bruker APEX DUO Radiation source IuS microsource, CuKα Theta range for data collection 2.72 to 72.18° Index ranges -11<=h<=11, -19<=k<=18, -7<=l<=7 Reflections collected 12104 Independent reflections 1692 [R(int) = 0.0474] Coverage of independent reflections 93.7% Absorption correction multi-scan Max. and min. transmission 0.8820 and 0.7930 Structure solution technique direct methods Structure solution program SHELXTL XT 2014/3 (Sheldrick, 2014) Refinement method Full-matrix least-squares on F 2 Refinement program SHELXL-2014 (Sheldrick, 2014) Function minimized Σ w(F o 2 - F c 2 ) 2 Data / restraints / parameters 1692 / 35 / 201 Goodness-of-fit on F 2 1.220 Final R indices 1440 data; I>2σ(I) R1 = 0.0790, wR2 = 0.2765 all data R1 = 0.0884, wR2 = 0.2846 Weighting scheme w=1/[σ 2 (F o 2 )+(0.1444P) 2 +3.2934P] where P=(F o 2 +2F c 2 )/3 Largest diff. peak and hole 0.446 and -0.423 eÅ -3 R.M.S. deviation from mean 0.111 eÅ -3 A2.2.16 Crystal Structure Report for Pyridinium DNT·H 2 O (15·H 2 O) Figure A2.19:Projection of the packing of pyridinium DNT . H 2 O (15·H 2 O) perpendicular to the 100 plane. 201 Table A2.31: Sample and crystal data for pyridinium DNT . H 2 O (15·H 2 O). Identification code PyrDNT Chemical formula C 7 H 8 N 6 O 5 Formula weight 256.19 Temperature 100(2) K Wavelength 0.71073 Å Crystal size 0.200 x 0.361 x 0.485 mm Crystal habit clear orange prism Crystal system monoclinic Space group C 1 2/c 1 Unit cell dimensions a = 13.9051(6) Å α = 90° b = 26.2532(12) Å β = 111.9480(7)° c = 6.3689(3) Å γ = 90° Volume 2156.48(17) Å 3 Z 8 Density (calculated) 1.578 g/cm 3 Absorption coefficient 0.136 mm -1 F(000) 1056 Table A2.32: Data collection and structure refinement for 15·H 2 O. Diffractometer Bruker APEX DUO Radiation source fine-focus tube, MoKα Theta range for data collection 1.55 to 30.55° Index ranges -19<=h<=19, -37<=k<=36, -9<=l<=9 Reflections collected 26448 Independent reflections 3279 [R(int) = 0.0274] Coverage of independent reflections 99.0% Absorption correction multi-scan Max. and min. transmission 0.9730 and 0.9370 Structure solution technique direct methods Structure solution program SHELXTL XT 2013/6 (Sheldrick, 2013) Refinement method Full-matrix least-squares on F 2 Refinement program SHELXTL XLMP 2014/1 (Bruker AXS, 2013) Function minimized Σ w(F o 2 - F c 2 ) 2 Data / restraints / parameters 3279 / 0 / 174 Goodness-of-fit on F 2 1.064 Final R indices 2729 data; I>2σ(I) R1 = 0.0345, wR2 = 0.0974 all data R1 = 0.0425, wR2 = 0.1042 Weighting scheme w=1/[σ 2 (F o 2 )+(0.0581P) 2 +0.7631P] where P=(F o 2 +2F c 2 )/3 Largest diff. peak and hole 0.400 and -0.271 eÅ -3 R.M.S. deviation from mean 0.048 eÅ -3 A2.2.17 Crystal Structure Report for Aminotetrazolium DNT·H 2 O (16·H 2 O) 202 Figure A2.20: Projection of the packing of aminotetrazolium DNT·H 2 O (16·H 2 O) perpendicular to the 001 plane. 203 Table A2.33: Sample and crystal data for aminotetrazolium DNT·H 2 O (16·H 2 O). Identification code ATDNT Chemical formula C 3 H 6 N 10 O 5 Formula weight 262.18 Temperature 100(2) K Wavelength 0.71073 Å Crystal size 0.090 x 0.120 x 0.230 mm Crystal habit clear colourless prism Crystal system orthorhombic Space group P 21 21 21 Unit cell dimensions a = 6.0692(3) Å α = 90° b = 17.6181(8) Å β = 90° c = 19.4400(9) Å γ = 90° Volume 2078.68(17) Å 3 Z 8 Density (calculated) 1.676 g/cm 3 Absorption coefficient 0.153 mm -1 F(000) 1072 Table A2.34: Data collection and structure refinement for 16·H 2 O. Diffractometer Bruker APEX DUO Radiation source fine-focus tube, MoKα Theta range for data collection 1.56 to 30.51° Index ranges -8<=h<=8, -25<=k<=25, -27<=l<=27 Reflections collected 35316 Independent reflections 6290 [R(int) = 0.0344] Coverage of independent reflections 99.6% Absorption correction multi-scan Max. and min. transmission 0.9860 and 0.9660 Structure solution technique direct methods Structure solution program SHELXTL XT 2013/6 (Sheldrick, 2013) Refinement method Full-matrix least-squares on F 2 Refinement program SHELXL-2014 (Sheldrick, 2014) Function minimized Σ w(F o 2 - F c 2 ) 2 Data / restraints / parameters 6290 / 0 / 373 Goodness-of-fit on F 2 1.060 Final R indices 5417 data; I>2σ(I) R1 = 0.0372, wR2 = 0.0797 all data R1 = 0.0472, wR2 = 0.0836 Weighting scheme w=1/[σ 2 (F o 2 )+(0.0444P) 2 +0.2528P] where P=(F o 2 +2F c 2 )/3 Absolute structure parameter -0.2(4) Largest diff. peak and hole 0.296 and -0.318 eÅ -3 R.M.S. deviation from mean 0.060 eÅ -3 204 A2.2.18 Crystal Structure Report for PPh 4 [DNT] (17) Figure A2.21: Molecular structure of PPh 4 [DNT] (17). Thermal ellipsoids are shown at the 50% probability level. Hydrogen atoms were omitted for clarity. Bond distances (Å): N1-O3 1.2261(17), N1- O4 1.2258(17), N4-O1 1.2294(17), N4-O2 1.2306(17), C1-N1 1.4516(19), C2-N4 1.4494(19), C1-N2 1.3310(19), C2-N3 1.3330(19), C2-N5 1.3347(19), C1-N5 1.3348(19), N2-N3 1.3637(18). Bond angles (°): O1-N4-O2 124.61(14), O3-N1-O4 124.43(14), C1-N1-O3 117.63(13), C1-N1-O4 117.94(13), C2-N4- O1 117.88(13), C2-N4-O2 117.51(13), O1-N4-C2-N3 -8.1(2), O3-N1-C1-N2 0.5(2). 205 Figure A2.22: Projection of the packing of 17 perpendicular to the 100 plane. Hydrogen atoms were omitted for clarity. Table A2.35: Sample and crystal data for PPh 4 [DNT] (17). Identification code XGC2_83imp Chemical formula C 26 H 20 N 5 O 4 P Formula weight 497.44 Temperature 100(2) K Wavelength 0.71073 Å Crystal size 0.141 x 0.257 x 0.286 mm Crystal system monoclinic Space group P 1 21/c 1 Unit cell dimensions a = 11.2261(5) Å α = 90° b = 15.0086(7) Å β = 95.0410(10)° c = 13.8078(6) Å γ = 90° Volume 2317.45(18) Å 3 Z 4 Density (calculated) 1.426 g/cm 3 Absorption coefficient 0.164 mm -1 F(000) 1032 206 Table A2.36: Data collection and structure refinement for 17. Theta range for data collection 1.82 to 28.50° Index ranges -15<=h<=15, -20<=k<=20, -18<=l<=18 Reflections collected 50180 Independent reflections 5872 [R(int) = 0.0558] Absorption correction multi-scan Structure solution technique direct methods Structure solution program SHELXTL XS 2013/1 (Bruker AXS) Refinement method Full-matrix least-squares on F 2 Refinement program SHELXL-2013 (Sheldrick, 2013) Function minimized Σ w(F o 2 - F c 2 ) 2 Data / restraints / parameters 5872 / 0 / 325 Goodness-of-fit on F 2 1.010 Final R indices 4600 data; I>2σ(I) R1 = 0.0373, wR2 = 0.0837 all data R1 = 0.0557, wR2 = 0.0923 Weighting scheme w=1/[σ 2 (F o 2 )+(0.0397P) 2 +1.1073P] where P=(F o 2 +2F c 2 )/3 Largest diff. peak and hole 0.345 and -0.264 eÅ -3 R.M.S. deviation from mean 0.050 eÅ -3 207 A2.2.19 Crystal Structure Report for PPN[DNT] (18) Figure A2.23: Molecular structure of PPN[DNT] (18). Thermal ellipsoids are shown at the 50% probability level. Hydrogen atoms were omitted for clarity. Selected structural parameters: Bond lengths (Å): N4-O1 1.221(2), N4-O2 1.230(2), N5-O4 1.220(2), N5-O3 1.223(2), C1-N5 1.450(3), C2-N4 1.444(3), C1-N2 1.329(3), C2-N1 1.332(3), C1-N3 1.330(3), C2-N3 1.329(3), N1-N2 1.364(2); Bond angles (°): O1-N4-O2 124.03(19), O3-N5-O4 123.9(2), C1-N5-O3 118.20(18), C1-N5-O4 117.84(18), C2-N4-O1 117.92(19), C2-N4-O2 118.05(18), O2-N4-C2-N1 13.5(3), O4-N5-C1-N2 1.9(3). 208 Figure A2.24: Projection of the packing of 18 perpendicular to the 001 plane. Hydrogen atoms were omitted for clarity. Table A2.37: Sample and crystal data for PPN[DNT] (18). Identification code PNPDNT Chemical formula C 38 H 30 N 6 O 4 P 2 Formula weight 696.62 Temperature 100(2) K Wavelength 0.71073 Å Crystal size 0.140 x 0.180 x 0.220 mm Crystal habit clear colourless prism Crystal system triclinic Space group P -1 Unit cell dimensions a = 10.9099(7) Å α = 112.6130(8)° b = 11.9031(8) Å β = 90.0610(9)° c = 14.4603(10) Å γ = 103.5780(9)° Volume 1676.28(19) Å 3 Z 2 Density (calculated) 1.380 g/cm 3 Absorption coefficient 0.182 mm -1 F(000) 724 209 Table A2.38: Data collection and structure refinement for 18. Diffractometer Bruker SMART APEX Radiation source fine-focus tube, MoKα Theta range for data collection 1.53 to 28.67° Index ranges -8<=h<=14, -15<=k<=15, -19<=l<=18 Reflections collected 10859 Independent reflections 7529 [R(int) = 0.0198] Coverage of independent reflections 87.1% Absorption correction multi-scan Max. and min. transmission 0.9750 and 0.9610 Structure solution technique direct methods Structure solution program SHELXTL XT 2013/6 (Sheldrick, 2013) Refinement method Full-matrix least-squares on F 2 Refinement program SHELXTL XLMP 2014/1 (Bruker AXS, 2013) Function minimized Σ w(F o 2 - F c 2 ) 2 Data / restraints / parameters 7529 / 0 / 451 Goodness-of-fit on F 2 1.025 Final R indices 5805 data; I>2σ(I) R1 = 0.0463, wR2 = 0.1150 all data R1 = 0.0635, wR2 = 0.1273 Weighting scheme w=1/[σ 2 (F o 2 )+(0.0662P) 2 +0.3671P] where P=(F o 2 +2F c 2 )/3 Largest diff. peak and hole 0.454 and -0.278 eÅ -3 R.M.S. deviation from mean 0.058 eÅ -3 210 A2.2.20 Crystal Structure Report for TMA[DNT] (19) Table A2.39: Sample and crystal data for TMA[DNT] (19). Identification code TMADNT Chemical formula C 6 H 12 N 6 O 4 Formula weight 232.22 g/mol Temperature 100(2) K Wavelength 0.71073 Å Crystal size 0.122 x 0.217 x 0.398 mm Crystal habit clear colourless prism Crystal system monoclinic Space group P 1 21/c 1 Unit cell dimensions a = 6.1515(3) Å α = 90° b = 17.1906(10) Å β = 92.6300(10)° c = 9.8599(5) Å γ = 90° Volume 1041.57(10) Å 3 Z 4 Density (calculated) 1.481 g/cm 3 Absorption coefficient 0.124 mm -1 F(000) 488 211 Table A2.40: Data collection and structure refinement for 19. Diffractometer Bruker APEX DUO Radiation source fine-focus tube, MoKα Theta range for data collection 2.37 to 30.47° Reflections collected 3159 Independent reflections 3159 [R(int) = 0.0146] Coverage of independent reflections 99.6% Absorption correction multi-scan Max. and min. transmission 0.9850 and 0.9520 Structure solution technique direct methods Structure solution program SHELXTL XT 2014/4 (Bruker AXS, 2014) Refinement method Full-matrix least-squares on F 2 Refinement program SHELXTL XL 2014/6 (Bruker AXS, 2014) Function minimized Σ w(F o 2 - F c 2 ) 2 Data / restraints / parameters 3159 / 28 / 184 Goodness-of-fit on F 2 1.096 Final R indices 2867 data; I>2σ(I) R1 = 0.0406, wR2 = 0.1086 all data R1 = 0.0455, wR2 = 0.1120 Weighting scheme w=1/[σ 2 (F o 2 )+(0.0572P) 2 +0.2829P] where P=(F o 2 +2F c 2 )/3 Largest diff. peak and hole 0.425 and -0.249 eÅ -3 R.M.S. deviation from mean 0.055 eÅ -3 212 A2.2.21 Crystal Structure Report for TMA[DNT]·HDNT Figure A2.25: Molecular structure of TMA[DNT]·HDNT. Thermal ellipsoids are shown at the 50% probability level. Hydrogen atom H1 has a site occupancy factor of 50%. Selected structural parameters: Bond distances (Å): N4-O1 1.2155(17), N4-O2 1.2168(17), N5-O3 1.2214(16), N5-O4 1.2220(16), C1- N4 1.4524(17), C2-N5 1.4524(16), C1-N2 1.3241(17), C2-N1 1.3199(17), C1-N3 1.3145(16), C2-N3 1.3334(16), N1-N2 1.3503(15). Bond angles (°):O1-N4-O2 126.11(13), O3-N5-O4 125.11(12), C1-N4- O1 116.88(12), C1-N4-O2 117.01(12), C2-N5-O3 117.71(11), C2-N5-O4 117.18(12), O2-N4-C1-N2 - 3.2(2), O4-N5-C2-N1 5.3(2). 213 Figure A2.26: Projection of the packing of TMA[DNT]·HDNT perpendicular to the 100 plane. Table A2.41: Sample and crystal data for TMA[DNT]·HDNT Identification code TMADNT Chemical formula C 8 H 7 N 11 O 8 Formula weight 391.30 Temperature 130(2) K Wavelength 0.71073 Å Crystal size 0.230 x 0.280 x 0.440 mm Crystal habit clear colourless prism Crystal system monoclinic Space group C 1 2/c 1 Unit cell dimensions a = 14.336(3) Å α = 90° b = 10.931(2) Å β = 93.978(3)° c = 10.644(2) Å γ = 90° Volume 1664.0(6) Å 3 Z 4 Density (calculated) 1.562 g/cm 3 Absorption coefficient 0.138 mm -1 F(000) 808 Table A2.42: Data collection and structure refinement for TMA[DNT]·HDNT. 214 Diffractometer Bruker SMART APEX Radiation source fine-focus tube, MoKα Theta range for data collection 2.35 to 28.58° Index ranges -19<=h<=13, -14<=k<=13, -13<=l<=14 Reflections collected 5163 Independent reflections 1961 [R(int) = 0.0224] Coverage of independent reflections 92.0% Absorption correction multi-scan Max. and min. transmission 0.9690 and 0.9420 Structure solution technique direct methods Structure solution program SHELXTL XT 2013/6 (Sheldrick, 2013) Refinement method Full-matrix least-squares on F 2 Refinement program SHELXTL XLMP 2014/1 (Bruker AXS, 2013) Function minimized Σ w(F o 2 - F c 2 ) 2 Data / restraints / parameters 1961 / 0 / 125 Goodness-of-fit on F 2 1.040 Final R indices 1646 data; I>2σ(I) R1 = 0.0418, wR2 = 0.1111 all data R1 = 0.0499, wR2 = 0.1192 Weighting scheme w=1/[σ 2 (F o 2 )+(0.0670P) 2 +0.7354P] where P=(F o 2 +2F c 2 )/3 Largest diff. peak and hole 0.254 and -0.307 eÅ -3 R.M.S. deviation from mean 0.053 eÅ -3 A2.3 References (1) Haiges, R.; Boatz, J. A.; Yousufuddin, M.; Christe, K. O. Angew. Chem. Int. Ed. 2007, 46, 2869. (2) SAINT+ V8.27B (3) SADABS V2012/1 (4) Sheldrick, G. Acta Crystallogr. Sect. A: Found. Crystallogr. 2008, 64, 112. (5) SHELXL 2012-4 (6) SHELXS-97 (7) SHELXTL V2012.4-3 (8) Farrugia, L. J. Appl. Crystallogr. 1997, 30, 565. (9) CCSD 2012. (10) Chernyshev, V. M.; Zemlyakov, N. D.; Il'in, V. B.; Taranushich, V. A. Russ. J. Appl. Chem. 2000, 73, 839. (11) Chernyshev, V. M. Z., N. D.; Taranushich, V. A. RU Patent 2174120 C1 2001. 215 APPENDIX 3: ADDITIONAL INFORMATION ON SYNTHESIS OF BORANYL-BASED COMPLEX ANIONS OF NITRO-SUBSTITUTED AZOLATES (CHAPTER 4) A3.1 Experimental Details Caution ! The materials used and synthesized in this study are energetic. They should be handled in quantities not exceeding the milimolar scales. Manipulations should be carried out behind blast shields and with adequate personal safety gear (face shield, heavy leather jacket and gloves, ear protection). Alkali metal salts of the complex anions described in this work are expected to be highly sensitive to friction and shock and their isolation should be avoided. Indeed, a sample of K[DNT-BH 3 ] deflagrated when provoked with a Teflon rod. Reactions were carried out in Teflon-FEP reactors equipped with stainless steel valves. using standard Schlenk and vacuum line techniques. Non-volatile compounds were handled in the nitrogen atmosphere of a drybox. NMR spectra were recorded on Varian VNMRS-600S spectrometer or Bruker AMX-500 spectrometers. 13 C NMR spectra were referenced to the deuterated solvent signal and 1 H spectra were referenced internally to the residual protic signal whenever possible or externally to neat tetramethylsilane. 11 B, 14 N and 31 P spectra were referenced externally to BF 3 . OEt 2 in CDCl 3 , nitromethane and H 3 PO 4 , respectively. Raman spectra were recorded in in Pyrex J.Young NMR tubes in the range of 4000-80 cm -1 on a Bruker Equinox 55 FT-RA spectrometer, using a Nd:YAG laser at 1064nm. The IR spectra were recorded on a Bruker Optics Alpha FT-IR spectrometers. Differential thermal analysis (DTA) curves were recorded under a dry nitrogen gas flow with a 5 °C heating rate on an OZM Research DTA 552-Ex instrument using the Meavy 2.2.0 software. Single-crystal X-ray diffraction data were collected on a Bruker Smart Apex Duo 3-circle platform diffractometer, equipped with an Apex II charge-coupled device (CCD) detector with the χ-axis fixed at 54.74° and using Mo Kα radiation (Triumph curved-crystal monochromator) from a finefocus tube or Cu Kα radiation (multi-layer optics monochromator IuS microsource). The diffractometer was equipped with an Oxford Cryostream 700 apparatus for low-temperature data collection. The data acquisition was performed with the BIS software package. For Mo radiation datasets, a complete hemisphere of data was scanned on ω (0.5°) with run times of 1-20 s/frame at a detector resolution of 512 × 512 pixels. For Cu radiation, individual acquisition strategies were determined for each data sets. Frames were acquired with run times of 10-40 s and were scanned on ω (1°). The frames were then 216 integrated with the SAINT algorithm 1 to give the hkl files corrected for Lp/decay. Absorption correction was performed with the SADABS program. 2 The structures were solved by direct methods and refined on F 2 by use of the Bruker SHELXTL software package. 3-6 All non-hydrogen atoms were refined anisotropically. ORTEP drawings were prepared with the ORTEP-III for Windows V2014.1 program. 7 Short-contact distances and torsion angles were measured with the Mercury 3.1 Development (Build RC5) software. 8 Further crystallographic details for compounds 1a-12 can be obtained from the Cambridge Crystallographic Data Centre [CCDC, 12 Union Road, Cambridge CB2 1EZ, U.K.; fax (+44) 1223-336- 033; e-mail deposit@ccdc.cam.ac.uk] on quoting the deposition numbers CCDC 1047968 (1a), 1047961 (1b), 1047966 (2a), 1048118 (4a), 1048144 (6a), 1047960 (8a), 1047959 (12). BH 3 . THF (Sigma-Aldrich and Strem) was either used as received or vacuum-transferred and kept cold. The concentration was determined by measuring the amount of dihydrogen formed by hydrolysis through pressure measurements in a calibrated glass vacuum line using a Heise gauge. BH 3 . SMe 2 (Sigma- Aldrich) was vacuum-transferred, then collected in a -64 °C trap under a dynamic vacuum. Even then the compound was found to be a concentrated solution of the borane adduct in dimethylsulfide, as estimated from the integration ratio of the SMe 2 protons in the free ligand and in the complex. Sodium borohydride, PPh 4 Cl and PPNCl (bis(triphenylphosphoranylidene)ammonium chloride) were purchased from Sigma- Aldrich. Drysolv © acetonitrile was purchased from EMD. Tetrahydrofuran (THF), 1,2-dimethoxyethane (glyme) (Alfa Aesar) and pentanes (EMD) were distilled over sodium/benzophenone ketyl after the appearance of a deep purple color. Alternately, they were bought as Drysolv and stored over 3Å molecular sieves. CD 3 CN (Cambridge Isotope Laboratories) was dried over molecular sieves (3Å) and degassed using freeze-pump-thaw cycles. Dichloromethane (Macron) and acetonitrile (EMD) were refluxed over and distilled from P 2 O 5 (J.T. Baker). All solvents were stored over 3Å molecular sieves that were dried by heating in vacuo. 3,5-dinitro-1H-1,2,4-triazole (HDNT) and its salts were prepared following literature methods. 9-13 5-(trinitromethyl)tetrazole, 14 5-(Fluorodinitromethyl)tetrazole, (trinitromethyl)nitro-1,2,4-triazole 15 and (fluorodinitromethyl)nitro-1,2,4-triazole and their salts were prepared according to modified literature methods. Salts of PPN + and PPh 4 + cations were chosen for safety and stability concerns. Indeed, these bulky cations are known to stabilize energetic anions which would otherwise be either unstable 16 or very sensitive. 17,18 Indeed, a sample of K[DNT-BH 3 ] deflagrated when provoked with a Teflon spatula. The PPN[Tz-BH 3 ] salts were always isolated as glassy solids and attempts to recrystallize them lead to tacky 217 solids which precluded the growth of crystals of X-ray diffraction quality . The synthesis of PPh 4 + salts was therefore prefered for most of this study. Synthesis of potassium 3,5-dinitro-1,2,4-triazolato-trihydroborate, K[DNT-BH 3 ]. A large excess of BH 3 . THF 1M in THF (8.5 mL; 8.5 mmol) was syringed onto KDNT (282 mg; 1.43 mmol). The yellow solution was stirred for 30 minutes, after which time the solvent was removed in vacuo. The yellow solid was found to contain significant amounts of BH 3 . THF decomposition products, mostly B(n-BuO) 3 . The solid appeared to be somewhat unstable at room temperature as the material left a white residue when extracted with acetonitrile. The dried extract deflagrated when provoked with a teflon spatula and work on this compound was discontinued. NMR (CD 3 CN) δ(ppm): 1 H 2.2 (bqt, 1 J( 1 H- 11 B)=100 Hz); 11 B -17.3 (qt, 1 J( 11 B- 1 H)=100 Hz); 14 N -25.9 (s, τ 1/2 =38 Hz). Raman (200 mW) ῦ/cm -1 : 2937 (0.2), 2911 (1.7), 2877 (0.2), 2731 (0.2), 2393 (1.1), 1559 (0.4), 1424 (10.0), 1375 (1.7), 1323 (0.4), 1195 (1.5), 1113 (0.9), 1034 (0.9), 920 (0.4), 833 (1.3), 763 (0.4), 516 (0.2), 458 (<0.1), 351 (<0.1), 108 (<0.1). Synthesis of tetraphenylphosphonium 3,5-dinitro-1,2,4-triazolato-trihydroborate, PPh 4 [DNT-BH 3 ] (1a). Dichloromethane was condensed onto PPh 4 DNT (250 mg; 0.50 mmol) in a Teflon-FEP reactor. BH 3 . THF (0.9 mL; 0.6 mmol) was syringed into the reactor. The mixture was briefly shaken, then left standing for 2h. The bulk of the solvent was then removed under vacuum at -30 °C. The materials were then further dried under vacuum at room temperature overnight. The weight of the tared reactor was ca 5% higher than expected, suggesting some residual organic impurities. The solids were suspended in water, filtered, dissolved in dichloromethane, the solution dried over MgSO 4 and dried in vacuo. The materials obtained contained traces of organic impurities, significant amounts of residual water as well as ca 2 mol% of B 3 H 8 - as detected by 11 B NMR. Crystals of 1a were obtained from evaporation of acetone (not recommended as it eventually leads to hydrolysis !) and dichloromethane solutions. DTA 173 °C, decomposition. NMR (CD 3 CN) δ(ppm): 1 H 7.68 (m), 7.74 (m), 7.91 (m), 2.21 (bqt, 1 J( 1 H- 11 B)=90 Hz); 11 B -17.3 (qt, 1 J( 11 B- 1 H)= 98 Hz); 13 C{ 1 H} 136.3 (s, C-P), 135.6 ( 3 J( 13 C- 31 P) =11 Hz), 131.3 (d, 2 J( 13 C- 31 P) = 13 Hz), 118.8 (d, 1 J( 13 C- 31 P) = 90 Hz); 14 N -25.6 (s, τ 1/2 =40 Hz), 31 P{ 1 H}24.3. Raman (200 mW) ῦ/cm -1 : 3171 (0.2), 3146 (0.5), 3067 (8.8), 3010 (0.1), 2994 (0.2), 2957 (0.2), 2383 (1.2), 2363 (1.3), 2266 (0.1), 1587 (5.3), 1554 (0.4), 1416 (6.5), 1371 (2.0), 1311 (0.2), 1190 (2.6), 1165 (2.0), 1101 (2.9), 1026 (3.3), 1002 (10.0), 941 (0.2), 827 (1.2), 760 (0.3), 726 (0.4), 682 (1.9), 617 (1.2), 453 (0.2), 353 (0.5), 295 (0.5), 254 (3.0), 201 (2.4). IR (ATR, 20°C) ῦ/cm -1 : 3066(w), 2364(m), 2311(w), 2259(w), 1585(w), 1573(w), 1548(s), 1495(s), 1482(w), 1434(s), 1412(w), 1366(m), 1324(w), 1307(s), 218 1186(w), 1142(m), 1107(s), 1073(vw), 1025(w), 995(m), 929(m), 844(s), 825(m), 753(m), 722(s), 688(s), 652(w), 615(m), 524(vs), 450(m), 432(w), 420(vw). Synthesis of Bis(triphenylphosphoranylidene)ammonium 3,5-dinitro-1,2,4-triazolato- trihydroborate, PPN[DNT-BH 3 ] (1b). BH 3 . THF/THF (9 mL ; 1 mmol) was syringed onto solid PPNDNT (621 mg; 0.891 mmol). The reactor was sonicated for ca 5 minutes to dissolve the PPNDNT. The resulting pale yellow solution was left standing for several hours. The solvent was then removed under vacuum overnight, leaving behind a pale yellow solid (500 mg; 79 % recovered yield based on PPNDNT). The isolated materials contained ca 5 mol% of residual BH 3 . THF and organic impurities. Crystals of 1b were obtained from dichloromethane/pentane solutions. DTA Endotherm (onset 125 °C), poorly defined decomposition onset 125 to 140 °C . NMR (CD 3 CN) δ(ppm): 1 H 7.66(m), 7.59(m), 7.48 (m), 2.25 (bqt, 1 J( 1 H- 11 B)= 97 Hz); 11 B{ 1 H} -17.2; 13 C{ 1 H}134.5 (s, PPN), 105.6 (m, PPN), 130.3 (m, PPN), 128.2 (d, 1 J( 13 C- 31 P) = 109 Hz); 14 N - 25.5 (s, τ 1/2 = 40 Hz). NMR (dichloromethane) δ(ppm): 13 C{ 1 H}: 158.9 (C-NO 2 ), 156.2 (C-NO 2 ), 133.6 (PPN), 132.0 (m, PPN), 129.3 (m, PPN), 127.0 (d, 1 J( 13 C- 31 P) = 108 Hz); 14 N -24.3 (s, τ 1/2 = 40 Hz, -NO 2 ), -70 (vbs), -132 (vbs). Raman (200 mW) ῦ/cm -1 : 3176 (1.6), 3147 (2.3), 3062 (87.3), 3013 (1.0), 2994 (1.8), 2959 (2.1), 2914 (0.7), 2385 (7.0), 2273 (1.0), 1590 (36.4), 1577 (2.6), 1545 (3.3), 1415 (55.5), 1367 (8.6), 1314 (2.2), 1192 (15.9), 1165 (5.2), 1112 (21.5), 1028 (26.1), 1003 (71.8), 927 (1.6), 827 (6.6), 763 (1.7), 727 (2.8), 665 (20.4), 618 (12.4), 357 (5.8), 264 (1.9), 239 (16.0), 192 (5.2), 170 (5.1). IR (ATR, 20°C) ῦ/cm -1 : 3057 (w), 2372 (m), 2315 (sh), 2268 (w), 1588 (w), 1548 (s), 1494 (s), 1436 (s), 1410 (w), 1363 (m), 1312 (m), 1282 (w), 1260 (s), 1182 (w), 1151 (m), 1112 (s), 1072 (w), 1025 (w), 997 (m), 923 (w), 843 (s), 825 (w), 797 (w), 756 (w), 743 (s), 722 (s), 689 (vs), 663 (w), 616 (w), 547 (m), 532 (s), 523 (w), 497 (vs), 449 (m), 408 (w). Synthesis of tetraphenylphosphonium 5-nitro-3-(trinitromethyl)-1,2,4-triazolato-trihydroborate, PPh 4 [TNMeNTrz-BH 3 ] (2a). BH 3 . THF/THF (5.4 mL; 0.59 mmol) was syringed onto PPh 4 TNMeNTrz (313 mg ; 0.520 mmol) in Teflon-FEP reactor. After homogenizing the reaction mixture, the yellow solution was left standing for several hours after which time it was concentrated under vacuum. A precipitate formed and the supernatant was removed by cannulation. The solids were then dried under vacuum overnight, yielding 189mg (59% recovered yield based on PPh 4 TNMeNTrz) of greyish yellow solid. The isolated materials contained very small amounts of TNMeNTrz - found by 14 N NMR and the usual THF-derived organic impurities. Crystals of 2a were obtained by slow evaporation of a dichloromethane solution. 219 DTA 70 °C, decomposition (very sharp exotherm at 95 °C). NMR (CD 3 CN) δ(ppm): 1 H 7.91 (m), 7.74 (m), 7.68 (m), 1.87 (bqt, 1 J( 1 H- 11 B)= 85 Hz); 11 B -18.1 (qt, 1 J( 11 B- 1 H)= 95 Hz); 13 C{ 1 H} 163.3 (bs, likely C-NO 2 ), 143.9 (likely C-C(NO 2 ) 3 ), 136.4 (P-C), 135.7 (d, 3 J( 13 C- 31 P) = 10 Hz), 131.3(d, 2 J( 13 C- 31 P) = 13 Hz), 123.1 (bs, likely C(NO 2 ) 3 ), 118.9 (d, 1 J( 13 C- 31 P) = 90 Hz); 14 N -26.2 (s, τ 1/2 = 60 Hz, C-NO 2 ), -34.6 (s, τ 1/2 = 10 Hz, C(NO 2 ) 3 ), 31 P{ 1 H} 24.3 Raman (200 mW) ῦ/cm -1 : 3172 (0.2), 3146 (0.4), 3070 (7.4), 3011 (0.1), 2994 (0.2), 2960 (0.3), 2871 (0.2), 2565 (0.1), 2394 (0.8), 2358 (0.8), 2264 (0.3), 1624 (0.1), 1588 (4.8), 1547 (0.6), 1484 (0.6), 1424 (7.8), 1374 (4.3), 1343 (0.1), 1314 (0.3), 1287 (0.1), 1207 (2.3), 1190 (0.3), 1165 (1.8), 1111 (0.1), 1101 (2.0), 1028 (3.4), 1002 (10.0), 955 (0.4), 847 (1.1), 836 (0.3), 770 (0.3), 727 (0.4), 682 (1.9), 617 (1.3), 530 (0.1), 453 (0.2), 421 (0.2), 382 (0.5), 357 (0.6), 325 (1.7), 294 (0.5), 251 (3.1), 199 (2.0). IR (ATR, 20°C) ῦ/cm -1 : 3086 (vw), 3069 (w), 2390 (shoulder), 2352 (m), 2295 (shoulder), 2261 (m), 1620 (s), 1602 (s), 1588 (s), 1543 (s), 1482 (s), 1435 (s), 1409 (m), 1373 (m), 1340 (w), 1311 (s), 1286 (s), 1187 (w), 1158 (s), 1108 (vs), 1071 (w), 1026 (w), 995 (s), 916 (m), 845 (m), 833 (s), 800 (vs), 756 (s), 723 (vs), 688 (vs), 655 (w), 615 (m), 591 (vw), 524 (vs), 450 (m), 435 (vw). Synthesis of tetraphenylphosphonium 3-(fluorodinitromethyl)-5-nitro-1,2,4-triazolato- trihydroborate, PPh 4 [FDNMeNTrz-BH 3 ] (4a). THF was condensed onto a small sample of PPh 4 FDNMeNTrz (104 mg; approx. 0.18mmol), known to contain a fluorine-containing impurity. BH 3 . THF/THF (0.2 mL; 0.2 mmol) was syringed into the solution and the mixture was stirred for 1h. The solvent was removed in vacuo and the solid was washed with pentane. The solids were pumped on overnight, yielding 109 mg of beige solid. The compound contained small amounts of a fluorinated impurity ( 19 F NMR δ -151.93 ppm), THF and BH 3 . THF (5 mol%). Crystals of 4a were obtained both from the evaporation of a dichloromethane solution and of an acetone solution of an inadvertent side-product during the synthesis of PPh 4 [BH 2 (FDNMeNTrz) 2 ]. DTA: 113 °C, decomposition. NMR (CD 3 CN) δ(ppm): 1 H 1.96 (bqt, 1 J( 1 H- 11 B)=85 Hz); 11 B{ 1 H} -18.5; 14 N -24.3 (s, τ 1/2 =25 Hz, CF(NO 2 ) 2 ), -25.8? (bs, τ 1/2 =70 Hz, weak, C-NO 2 ); 13 C{ 1 H} 136.4 (d, 4 J( 13 C- 31 P) = 3 Hz), 135.7 ( 3 J( 13 C- 31 P) =10 Hz), 131.3 (d, 2 J( 13 C- 31 P) = 13 Hz), 118.8 (d, 1 J( 13 C- 31 P) = 97 Hz); 19 F - 93.0 (bm); 31 P{ 1 H} 24.3. Raman (200 mW) ῦ/cm -1 : 3172 (0.2), 3148 (0.4), 3071 (8.2), 3012 (0.2), 2995 (0.2), 2861 (0.6), 2373 (1.1), 2265 (0.3), 1588 (5.3), 1551 (0.5), 1495 (0.6), 1439 (3.7), 1414 (1.1), 1377 (5.0), 1313 (0.5), 1209 (1.0), 1190 (0.3), 1165 (1.4), 1111 (0.2), 1101 (2.1), 1028 (4.0), 1002 (10.0), 986 (0.1), 834 (1.2), 771 (0.3), 726 (0.4), 682 (1.9), 617 (1.3), 397 (0.6), 346 (0.9), 294 (0.5), 253 (2.7), 199 (1.8). 220 IR (ATR, 20°C) ῦ/cm -1 : 3064 (w), 2367 (m), 2265 (w), 2219 (vw), 1601 (s), 1545 (s), 1493 (w), 1484 (m), 1435 (s), 1412 (w), 1375 (m), 1340 (w), 1308 (s), 1242 (m), 1186 (vw), 1150 (m), 1107 (s), 1071 (w), 1026 (w), 996 (w), 981 (s), 929 (w), 839 (s), 832 (w), 795 (s), 753 (s), 721 (s), 687 (s), 657 (w), 615 (w), 524 (vs), 450 (m). Synthesis of tetraphenylphosphonium 5-trinitromethyltetrazole-trihydroborate, PPh 4 [TNTz-BH 3 ] (6a, 7a). Dichloromethane (ca 5 mL) was condensed onto PPh 4 TNTz (104 mg; 0.187 mmol) in a Teflon- FEP reactor. BH 3 . THF/THF (0.3 mL; 0.2 mmol) was syringed into the reactor. The reaction mixture was homogenized, then left standing for 2 h. The solvent was removed in vacuo below 0 °C overnight, yielding 122 mg (expected 107 mg, indicative of residual solvent) of yellow glassy and microcrystalline solids. The isolated materials contained ca 5 mol% BH 3 . THF, small amounts of 7a, traces of 10a small amounts of decomposition products by 14 N and 11 B NMR. Crystals of 6a were obtained by slow evaporation of a dichloromethane solution. NMR (CD 3 CN) δ(ppm): 1 H 7.91(m), 7.74(m), 7.68(m), 2.44 (bqt, 1 J( 1 H- 11 B)= 96 Hz); 11 B -17.3 (qt, 1 J( 11 B- 1 H)= 97 Hz); 13 C{ 1 H} 136.3 (C-P), 135.6 (d, 3 J( 13 C- 31 P)= 10 Hz), 131.2 (d, 2 J( 13 C- 31 P) = 12 Hz); 124.4 (bs, very weak, likely C(NO 2 ) 3 ; 118.9 (d, 1 J( 13 C- 31 P) = 90 Hz); 14 N 13.6 (bs, τ 1/2 = 300 Hz), -32.3 (s, τ 1/2 = 5 Hz), -48.7(bs, τ 1/2 = 250 Hz), 31 P{ 1 H}24.4 In an attempt to obtain 10a (vide infra), further spectroscopic data was obtained from a sample of 6a containing ca 10% of 10a and highly contaminated with unidentified decomposition products. DTA 35°C, decomposition. NMR (CD 3 CN) δ(ppm): 1 H 7.91 (m), 7.73(m), 7.68 (m), 2.43 (bqt, 1 J( 1 H- 11 B)= 97 Hz); 11 B{ 1 H} -17.3 (qt, 1 J( 11 B- 1 H)= Hz); 13 C{ 1 H}151.2 (C-C(NO 2 ) 3 ), 136.4 (P-C), 135.7 (d, 3 J( 13 C- 31 P) = 11 Hz), 131.3 (d, 2 J( 13 C- 31 P) = 13 Hz), 118.9 (d, 1 J( 13 C- 31 P) = 90 Hz); 14 N -31.5 (s, τ 1/2 = 4 Hz). NMR (dichloromethane) δ(ppm): 1 H 7.33 (m), 7.18 (m), 7.08 (m), 1.74 (bm); 11 B{ 1 H} -18.2; 14 N - 32.5 (s, τ 1/2 = 6 Hz). Raman (200 mW) ῦ/cm -1 : 3174 (0.2), 3148 (0.3), 3066 (8.9), 3013 (0.2), 2996 (0.1), 2962 (0.3), 2909 (0.1), 2861 (0.1), 2460 (br, decomposition product), 2380 (1.1), 2268 (0.1), 1589 (5.0), 1553 (0.1), 1443 (0.5), 1387 (0.1), 1355 (0.8), 1305 (0.4), 1249 (0.1), 1191 (0.5), 1167 (1.1), 1100 (2.2), 1029 (4.2), 1002 (10.0), 845 (0.6), 819 (1.1), 753 (0.1), 727 (0.4), 681 (1.9), 617 (1.6), 532 (0.1), 405 (0.1), 349 (0.9), 287 (0.2), 253 (2.6), 202 (1.3). IR (ATR, 20°C) ῦ/cm -1 : 3089 (sh), 3061 (w), 2477 (sh), 2369 (s), 2274 (vw), 1617 (vw), 1592 (s), 1548 (m), 1483 (m), 1472 (vw), 1436 (s), 1395 (vw), 1339 (vw), 1284 (vw), 1242 (s), 1186 (vw), 1139 (w), 1105 (s), 1049 (w), 1028 (vw), 995 (s), 930 (vw), 842 (s), 818 (vw), 799 (s), 752 (s), 719 (s), 686 (s), 615 (vw), 522 (vs), 448 (w). 221 Synthesis of Bis(triphenylphosphoranylidene)ammonium 5-(trinitromethyl)tetrazolato- trihydroborate, PPN[TNTz-BH 3 ] (6b, 7b). Dichloromethane (~2 mL) was condensed onto PPNTNTz (82 mg ; 0.108 mmol), yielding a pale yellow solution. BH 3 . SMe 2 (ca 0.2mmol) was condensed onto the solution. The reaction was allowed to proceed at room temperature for 3 hours. The reactor was then placed in a warm water bath at 35° C, and the solvent was slowly pumped off. Dichloromethane was condensed on the solid obtained and pentane was added. The mixture was thoroughly shaken, yielding a two-phase system, from which the supernatant pentane-rich phase, was removed by cannulation. The dichloromethane-rich phase was dried under vacuum, yielding an off-white foam-like solid. The solid contained ca 10 mol% of BH 3, either from the residual starting material or from 10b (10 indeed dissociates to BH 3 . NCCH 3 and 6/7 in the presence of acetonitrile). The solid was also severely contaminated with unidentified decomposition products. DTA 35 °C, decomposition (exotherm ca 100 °C). NMR (CD 3 CN) δ(ppm): 1 H 7.66 (m), 7.57(m), 7.47 (m), 2.43 (bqt, 1 J( 1 H- 11 B)= 97 Hz); 11 B{ 1 H} -17.3; 14 N -31.5 (s, τ 1/2 = 4 Hz). Raman (200 mW) ῦ/cm -1 : 3177 (0.2), 3147 (0.3), 3063 (9.3), 3013 (0.2), 2995 (0.2), 2959 (0.3), 2451 (br), 2381 (0.9), 2267 (0.1), 1591 (5.0), 1577 (0.3), 1441 (0.5), 1385 (0.1), 1355 (0.3), 1303 (0.5), 1186 (0.5), 1164 (1.2), 1112 (2.6), 1055 (0.1), 1030 (3.6), 1003 (10.0), 846 (0.5), 819 (0.6), 727 (0.3), 665 (2.5), 618 (2.1), 403 (0.1), 349 (1.0), 266 (0.3), 237 (1.9), 198 (0.5), 175 (0.2). Synthesis of tetraphenylphosphonium 5-(fluorodinitromethyl)tetrazolato-trihydroborate, PPh 4 [FDNTz-BH 3 ] (8a, 9a). BH 3 . THF/THF (9.6 mL ; ca 1.6 mmol) was syringed onto PPh 4 FDNTz (499 mg ; 0.941 mmol). The reactor was briefly sonicated in order to dissolve PPh 4 FDNTz. A sample of the reaction mixture was taken for NMR analysis, which revealed that an equilibrium largely in favor of 11 between 11 and 8+9 + BH 3 . THF took place in THF solution. The reaction was allowed to proceed for two hours, and then the solvent was pumped off and the product was dried under vacuum. The product suspended in water and then extracted with dichloromethane (~3 mL). The organic phase was washed thrice with water, and then dried with magnesium sulfate. The filtered product was dried under vacuum overnight, yielding a glassy product, which contained 8a and 9a in a 1 : 0.15 ratio as well as negligible amounts of hydrolysis products and 11a. Crystals of 8a were obtained by slow evaporation of a dichloromethane solution. DTA endotherm 85°C, decomposition onset 112 °C. NMR (CD 3 CN) δ(ppm): 1 H 7.91(m), 7.74(m), 7.68(m), 2.40 (bqt, 1 J( 1 H- 11 B)= 96 Hz); 11 B -17.4 (qt, 1 J( 11 B- 1 H)= 97 Hz, 8), -20.0 (qt, 1 J( 11 B- 1 H)= 97 Hz, 9); 13 C{ 1 H} 152.5 (d, 2 J( 13 C- 19 F)= 24 Hz, C-C-F, likely 9), 147.3 (d, 2 J( 13 C- 19 F)= 24 Hz, C-C-F likely 8), 222 136.4 (d, 4 J( 13 C- 31 P) = 3 Hz), 135.7(d, 3 J( 13 C- 31 P) = 11 Hz), 131.3 (d, 2 J( 13 C- 31 P) = 13 Hz), 118.9 (d, 1 J( 13 C- 31 P) = 90 Hz); 118.8 (d, 1 J( 13 C- 19 F)= 280 Hz, CF(NO 2 ) 2 ); 14 N - 21.8(d, 2 J( 14 N- 19 F)= 10 Hz, 8), -24.2 (m, weak, 9) 19 F -94.2 (bm, 9), -96.82 (bm, 8); 31 P{ 1 H} 24.3. Raman (200 mW) ῦ/cm -1 : 3173(0.2), 3148(0.3), 3063(9.4), 3012(0.2), 2996(0.1), 2959(0.2), 2566(br), 2386(1.4), 2273(0.1), 1589(4.7), 1483(1.2), 1440(0.3), 1404(0.1), 1363(0.1), 1341(0.2), 1317(0.1), 1290(0.7), 1191(0.6), 1168(1.1), 1111(0.2), 1101(2.1), 1057(0.5), 1027(2.8), 1001(10.0), 986(0.1), 956(0.1), 837(0.8), 756(0.1), 726(0.4), 682(1.7), 618(1.1), 532(0.1), 407(0.1), 358(0.7), 294(0.7), 252(2.7), 200(1.5). IR (ATR, 20°C) ῦ/cm -1 : 3062(w), 2376(m), 2314(sh), 2271(w), 1594(), 1483(m), 1436(s), 1402(w), 1360(w), 1338(vw), 1313(m), 1241(m), 1185(vw), 1157(m), 1108(s), 1082(w), 1055(m), 1026(w), 996(m), 976(sh), 900(w), 834(s), 797(s), 755(s), 722(vs), 686(s), 616(m), 586(w), 523(vs), 450(m), 434(w). Attempted synthesis of tetraphenylphosphonium bis(boranyl)-5-trinitromethyltetrazolate, PPh 4 [TNTz(BH 3 ) 2 ] (10a), and crystal growth of 12. PPh 4 TNTz (499 mg; 0.895 mmol) was placed in a Teflon-FEP reactor. BH 3 . THF/THF (1.98 g; 2.05 mmol) was weighed in a tared glass reactor. Dichloromethane was condensed onto the BH 3 . THF sample and the resulting solution was added to the TNTz - salt by cannulation. The mixture was homogenized and a sample was taken for NMR analysis, which revealed that even in dichloromethane, an equilibrium between 10 and 6/7 + BH 3 . THF exists, suggesting that the isolation of clean 10a should be impossible under ambient conditions: NMR (10a only)(dichloromethane) δ(ppm): 11 B{ 1 H} -17.1, -19.4; 14 N - 38.6 (s, τ 1/2 = 5 Hz). The pale yellow solution was concentrated under vacuum, resulting in a sticky orange solid, on top of which ca 15 mL of THF was condensed. The solvent was removed under vacuum overnight. The sticky material obtained was then washed with pentane, resulting in a glassy, free-flowing yellow solid (274 mg, ca 50% recovered yield based on PPh 4 TNTz). The solid contained mostly 6a and decomposition products and only ca 5 mol% of 10a and its characterization can be found under the synthesis of 6a/7a. In a similar synthesis, crystals of 12 were grown from the aqueous washings of 6a/7a in dichloromethane. Synthesis of tetraphenylphosphonium bis(boranyl)-5-(fluorodinitromethyl)tetrazolate, PPh 4 FDNTz(BH 3 ) 2 ] (11a). BH 3 . THF/THF (1.4 mL; 1.3 mmol) in dichloromethane was cannulated onto PPh 4 FDNTz (282 mg: 0.532 mmol) in a Teflon-FEP reactor. After homogenizing the mixture, a sample was taken for NMR analysis, which revealed that no measurable equilibrium between 11 and 8/9 + BH 3 . L occurred in dichloromethane even in the presence of small amounts of THF. The solvent was removed under vacuum. THF was condensed into the reactor and the solution was evaporated again. The solid was 223 then washed with pentane and dried under vacuum overnight, which yielded 206 mg of 11a as a white powder (69 % recovered yield based on PPh 4 FDNTz). The material obtained contained only trace amounts of impurities. Attempts at obtaining crystals of 11a have so far been unsuccessful. DTA 102 °C (endotherm), 116-125°C decomposition onset, largest exotherm 151 °C. NMR (dichloromethane) δ(ppm): 1 H 7.39 (m), 7.23(m), 7.12(m), 1.67 (broad pseudo-triplet); 11 B -17.3 (bqt, 1 J( 11 B- 1 H)= ca 90 Hz), -19.9 (bqt, 1 J( 11 B- 1 H)= ca 100 Hz); 13 C{ 1 H} 147.2 (d, 2 J( 13 C- 19 F)= 28 Hz, C-C-F), 135.6 (P-C), 134.4 ( 3 J( 13 C- 31 P) = 10 Hz), 130.5 (d, 2 J( 13 C- 31 P) = 13 Hz), 117.5(d, 1 J( 13 C- 31 P) = 90 Hz), 115.1 (d, 2 J( 13 C- 19 F)=293 Hz, C-C-F); 14 N -28.5 (bd, CF(NO 2 ) 2 ), -57 (vbs), -111.2 (2x very broad singlets). 19 F -94.3(bm). Raman (200 mW) ῦ/cm -1 : 3172(0.2), 3147(0.4), 3075(1.5), 3061(6.6), 3011(0.2), 2995(0.1), 2958(0.2), 2564(br), 2405(sh), 2386(2.4), 2374(sh), 2304(sh), 2267(0.4), 1611(1.3), 1588(5.3), 1503(1.3), 1439(0.5), 1342(0.2), 1305(2.0), 1234(0.4), 1188(0.8), 1165(1.9), 1110(0.2), 1100(2.3), 1058(0.7), 1029(3.4), 1003(10.0), 959(0.2), 836(1.0), 798(0.2), 752(0.1), 726(0.4), 681(2.0), 617(1.3), 539(0.2), 419(0.3).IR (ATR, 20°C) ῦ/cm -1 : 3099(vw), 3079(w), 2905(vw), 2410(sh), 2388(m), 2363(m), 2307(vw), 2264(w), 1606(s), 1503(w), 1483(m), 1436(s), 1340(w), 1304(m), 1265(m), 1232(w), 1207(w), 1186(w), 1141(s), 1107(s), 1056(w), 1026(w), 994(s), 932(m), 901(w), 834(m), 794(s), 760(w), 751(m), 722(s), 687(s), 666(w), 616(m), 608(vw), 523(vs), 445(m), 435(w). Attempted synthesis of BH 3 DNT - salts from HDNT and borohydride salts. In a typical synthesis, HDNT and MBH 4 (where M is Na + , PPh 4 + , PPN + and tetramethylammonium) were placed in equimolar amounts in a glass vessel equipped with a Teflon valve. Glyme was condensed at -196 °C. Gas evolution occurred as the solvent thawed. The non-condensable gases were measured in a calibrated glass vacuum line and found in the expected amounts. The solvent was removed in vacuo and the solids analyzed by Raman and NMR spectroscopy. They were found to contain variable amounts of [DNT-BH 3 ] - , [BH 2 (DNT) 2 ] - and other minor products. Alternatively, a solution of HDNT was cannulated onto a potassium borohydride suspension. The products were shown to contain very significant amounts of [BH 2 (DNT) 2 ] - . 224 A3.2 Crystallographic Information A3.2.1 Crystal Structure Report for PPh 4 [DNT-BH 3 ] (1a) Figure A3.1:ORTEP plot of the asymmetric unit of PPh 4 [DNT-BH 3 ] 1a showing the two part disorder. Hydrogen atoms were omitted for clarity. Thermal ellipsoids are drawn at the 50% probability level. 225 Figure A3.2: Projection of the packing showing one part of the disordered anion 1a perpendicular to the 100 plane. Table A3.1: Sample and crystal data for PPh 4 [DNT-BH 3 ] (1a) Identification code XGC4_212_01 Chemical formula C 26 H 23 BN 5 O 4 P Formula weight 511.27 g/mol Temperature 100(2) K Wavelength 1.54178 Å Crystal size 0.128 x 0.204 x 0.570 mm Crystal habit colorless needle Crystal system monoclinic Space group C 1 c 1 Unit cell dimensions a = 7.22590(10) Å α = 90° b = 16.4562(2) Å β = 96.6270(10)° c = 20.8053(3) Å γ = 90° Volume 2457.45(6) Å 3 Z 4 Density (calculated) 1.382 g/cm 3 Absorption coefficient 1.360 mm -1 F(000) 1064 226 Table A 3.2: Data collection and structure refinement for (1a) Diffractometer Bruker APEX DUO Radiation source IuS microsource, CuKα Theta range for data collection 4.28 to 71.64° Index ranges -8<=h<=8, -19<=k<=20, -25<=l<=25 Reflections collected 14293 Independent reflections 4422 [R(int) = 0.0497] Absorption correction multi-scan Max. and min. transmission 0.8450 and 0.5110 Structure solution technique direct methods Structure solution program SHELXTL XT 2014/4 (Bruker AXS, 2014) Refinement method Full-matrix least-squares on F 2 Refinement program SHELXL-2014/7 (Sheldrick, 2014) Function minimized Σ w(F o 2 - F c 2 ) 2 Data / restraints / parameters 4422 / 12 / 368 Goodness-of-fit on F 2 1.101 Δ/σ max 0.002 Final R indices 4349 data; I>2σ(I) R1 = 0.0378, wR2 = 0.0949 all data R1 = 0.0382, wR2 = 0.0952 Weighting scheme w=1/[σ 2 (F o 2 )+(0.0595P) 2 +0.2730P] where P=(F o 2 +2F c 2 )/3 Absolute structure parameter 0.0(0) Largest diff. peak and hole 0.394 and -0.382 eÅ -3 R.M.S. deviation from mean 0.067 eÅ -3 227 A3.2.2 Crystal Structure Report for PPN[DNT-BH 3 ] (1b) Figure A3.3:ORTEP plot of the asymmetric unit of PPN[DNT-BH 3 ] (1b) showing the disordered [DNT- BH 3 ] - ring. Hydrogen atoms were omitted for clarity. 228 Figure A3.4: Projection of the packing of one part of the disordered asymmetric unit of 1b perpendicular to the 001 plane. Hydrogen atoms were omitted for clarity. Table A 3.3: Sample and crystal data for PPN[DNT-BH 3 ] (1b) Identification code XMK22_01 Chemical formula C 38 H 33 BN 6 O 4 P 2 Formula weight 710.45 g/mol Temperature 100(2) K Wavelength 0.71073 Å Crystal size 0.396 x 0.409 x 0.474 mm Crystal habit colorless prism Crystal system monoclinic Space group P 1 21/n 1 Unit cell dimensions a = 12.1574(6) Å α = 90° b = 21.1776(10) Å β = 103.0280(10)° c = 14.0475(7) Å γ = 90° Volume 3523.6(3) Å 3 Z 4 Density (calculated) 1.339 g/cm 3 Absorption coefficient 0.174 mm -1 F(000) 1480 229 Table A3.4: Data collection and structure refinement for 1b Diffractometer Bruker APEX DUO Radiation source fine-focus tube, MoKα Theta range for data collection 1.77 to 27.48° Index ranges -15<=h<=15, -27<=k<=27, -18<=l<=18 Reflections collected 73768 Independent reflections 8084 [R(int) = 0.0692] Absorption correction multi-scan Max. and min. transmission 0.9340 and 0.9220 Structure solution technique direct methods Structure solution program SHELXTL XT 2013/1 (Bruker AXS, 2014) Refinement method Full-matrix least-squares on F 2 Refinement program SHELXTL XL 2014/7 (Bruker AXS, 2014) Function minimized Σ w(F o 2 - F c 2 ) 2 Data / restraints / parameters 8084 / 500 / 487 Goodness-of-fit on F 2 1.031 Δ/σ max 0.001 Final R indices 6008 data; I>2σ(I) R1 = 0.0409, wR2 = 0.0945 all data R1 = 0.0648, wR2 = 0.1077 Weighting scheme w=1/[σ 2 (F o 2 )+(0.0467P) 2 +1.6510P] where P=(F o 2 +2F c 2 )/3 Largest diff. peak and hole 0.344 and -0.324 eÅ -3 R.M.S. deviation from mean 0.053 eÅ -3 230 A3.2.3 Crystal Structure Report for PPh 4 [TNMeNTrz-BH 3 ] (2a) Figure A3.5: Projection of the packing of PPh 4 [TNMeNTrz-BH 3 ] (2a) perpendicular to the 001 plane. Table A3.5: Sample and crystal data for PPh 4 [TNMeNTrz-BH 3 ] (2a) Identification code XGC4_216_01 Chemical formula C 27 H 23 BN 7 O 8 P Formula weight 615.30 g/mol Temperature 100(2) K Wavelength 1.54178 Å Crystal size 0.043 x 0.097 x 0.198 mm Crystal habit colorless prism/needle Crystal system orthorhombic Space group F d d 2 Unit cell dimensions a = 38.9762(7) Å α = 90° b = 39.8619(7) Å β = 90° c = 7.38940(10) Å γ = 90° Volume 11480.7(3) Å 3 Z 16 Density (calculated) 1.424 g/cm 3 Absorption coefficient 1.395 mm -1 F(000) 5088 231 Table A 3.6: Data collection and structure refinement for 2a. Diffractometer Bruker APEX DUO Radiation source IuS microsource, CuKα Theta range for data collection 3.17 to 72.42° Index ranges -47<=h<=47, -48<=k<=46, -8<=l<=8 Reflections collected 41733 Independent reflections 5420 [R(int) = 0.0643] Coverage of independent reflections 98.3% Absorption correction multi-scan Max. and min. transmission 0.9420 and 0.7700 Structure solution technique direct methods Structure solution program SHELXTL XT 2014/4 (Bruker AXS, 2014) Refinement method Full-matrix least-squares on F 2 Refinement program SHELXL-2014/7 (Sheldrick, 2014) Function minimized Σ w(F o 2 - F c 2 ) 2 Data / restraints / parameters 5420 / 1 / 401 Goodness-of-fit on F 2 1.068 Δ/σ max 0.002 Final R indices 4814 data; I>2σ(I) R1 = 0.0382, wR2 = 0.0977 all data R1 = 0.0457, wR2 = 0.1024 Weighting scheme w=1/[σ 2 (F o 2 )+(0.0651P) 2 ] where P=(F o 2 +2F c 2 )/3 Absolute structure parameter 0.0(0) Largest diff. peak and hole 0.276 and -0.355 eÅ -3 R.M.S. deviation from mean 0.058 eÅ -3 232 A3.2.4 Crystal Structure Report for PPh 4 [FDNMeNTrz-BH 3 ] (4a) Figure A 3.6: Asymmetric unit of PPh 4 [FDNMeNTrz-BH 3 ] (4a) showing the two-part disorder in the [FDNMeNTrz-BH 3 ] - anion. 233 Figure A3.7: Projection of the packing of one part of the disordered asymmetric unit of 4a perpendicular to the 100 plane. Table A3.7: Sample and crystal data for PPh 4 [FDNMeNTrz-BH 3 ] (4a) Identification code SN21 Chemical formula C 27 H 23 BFN 6 O 6 P Formula weight 588.29 g/mol Temperature 100(2) K Wavelength 1.54178 Å Crystal system orthorhombic Space group P 21 21 21 Unit cell dimensions a = 7.3045(4) Å α = 90° b = 18.1495(10) Å β = 90° c = 20.4393(12) Å γ = 90° Volume 2709.7(3) Å 3 Z 4 Density (calculated) 1.442 g/cm 3 Absorption coefficient 1.435 mm -1 F(000) 1216 234 Table A3.8: Data collection and structure refinement for 4a Diffractometer Bruker APEX DUO Radiation source IuS microsource, CuKα Theta range for data collection 3.26 to 55.97° Reflections collected 44702 Independent Reflections 2031 [R(int)= 0.1098] Absorption correction Twin Structure solution technique direct methods Structure solution program SHELXTL XT 2014/4 (Bruker AXS, 2014) Refinement method Full-matrix least-squares on F 2 Refinement program SHELXTL XL 2014/6 (Bruker AXS, 2014) Function minimized Σ w(F o 2 - F c 2 ) 2 Data / restraints / parameters 2031 / 475 / 501 Goodness-of-fit on F 2 1.089 Δ/σ max 0.009 Final R indices 1706 data; I>2σ(I) R1 = 0.0555, wR2 = 0.1511 all data R1 = 0.0727, wR2 = 0.1655 Weighting scheme w=1/[σ 2 (F o 2 )+(0.1384P) 2 ] where P=(F o 2 +2F c 2 )/3 Absolute structure parameter 0.4(1) Extinction coefficient 0.0030(7) Largest diff. peak and hole 0.296 and -0.381 eÅ -3 R.M.S. deviation from mean 0.080 eÅ -3 A3.2.5 Crystal Structure Report for PPh 4 [TNTz-BH 3 ] (6a) Figure A3.8: Projection of the packing of PPh 4 [TNTz-BH 3 ] (6a) perpendicular to the 100 plane. 235 Table A3.9: Sample and crystal data for PPh 4 [TNTz-BH 3 ] (6a) Identification code XGC4_207_01 Chemical formula C 26 H 23 BN 7 O 6 P Formula weight 571.29 g/mol Temperature 100(2) K Wavelength 1.54178 Å Crystal size 0.124 x 0.162 x 0.222 mm Crystal habit colorless prism Crystal system triclinic Space group P -1 Unit cell dimensions a = 7.70210(10) Å α = 99.739(2)° b = 12.4067(3) Å β = 98.257(2)° c = 14.9270(4) Å γ = 102.742(2)° Volume 1346.71(5) Å 3 Z 2 Density (calculated) 1.409 g/cm 3 Absorption coefficient 1.382 mm -1 F(000) 592 Table A3.10: Data collection and structure refinement for 6a. Diffractometer Bruker APEX DUO Radiation source IuS microsource, CuKα Theta range for data collection 3.06 to 72.47° Index ranges -9<=h<=9, -15<=k<=14, -18<=l<=18 Reflections collected 34096 Independent reflections 5095 [R(int) = 0.0584] Coverage of independent reflections 95.3% Absorption correction multi-scan Max. and min. transmission 0.8470 and 0.7490 Structure solution technique direct methods Structure solution program SHELXTL XT 2014/4 (Bruker AXS, 2014) Refinement method Full-matrix least-squares on F 2 Refinement program SHELXL-2014/7 (Sheldrick, 2014) Function minimized Σ w(F o 2 - F c 2 ) 2 Data / restraints / parameters 5095 / 0 / 374 Goodness-of-fit on F 2 1.065 Final R indices 4451 data; I>2σ(I) R1 = 0.0409, wR2 = 0.1103 all data R1 = 0.0466, wR2 = 0.1147 Weighting scheme w=1/[σ 2 (F o 2 )+(0.0642P) 2 +0.4522P] where P=(F o 2 +2F c 2 )/3 Largest diff. peak and hole 0.396 and -0.342 eÅ -3 R.M.S. deviation from mean 0.053 eÅ -3 236 A3.2.6 Crystal Structure Report for PPh 4 [FDNTz-BH 3 ] (8a) Figure A3.9: ORTEP plot of the disordered asymmetric unit of PPh 4 [FDNTz-BH 3 ] (8a). Hydrogen atoms were omitted for clarity. Figure A3.10: Projection of the packing of one part of the disordered asymmetric unit of 8a perpendicular to the 100 plane. Hydrogen atoms were omitted for clarity. 237 Table A3.11: Sample and crystal data for PPh 4 [FDNTz-BH 3 ] (8a) Identification code XMK27b_01 Chemical formula C 26 H 23 BFN 6 O 4 P Formula weight 544.28 g/mol Temperature 100(2) K Wavelength 1.54178 Å Crystal size 0.158 x 0.206 x 0.384 mm Crystal habit colorless blade Crystal system triclinic Space group P -1 Unit cell dimensions a = 7.55540(10) Å α = 101.1770(10)° b = 12.9653(2) Å β = 95.7900(10)° c = 14.2531(2) Å γ = 103.1080(10)° Volume 1318.62(3) Å 3 Z 2 Density (calculated) 1.371 g/cm 3 Absorption coefficient 1.369 mm -1 F(000) 564 Table A3.12: Data collection and structure refinement for 8a Diffractometer Bruker APEX DUO Radiation source IuS microsource, CuKα Theta range for data collection 3.20 to 71.58° Index ranges -9<=h<=9, -15<=k<=15, -16<=l<=17 Reflections collected 20769 Independent reflections 4908 [R(int) = 0.0688] Absorption correction multi-scan Max. and min. transmission 0.8130 and 0.6210 Structure solution technique direct methods Structure solution program SHELXTL XT 2014/4 (Bruker AXS, 2014) Refinement method Full-matrix least-squares on F 2 Refinement program SHELXTL XL 2014/7 (Bruker AXS, 2014) Function minimized Σ w(F o 2 - F c 2 ) 2 Data / restraints / parameters 4908 / 15 / 374 Goodness-of-fit on F 2 1.098 Final R indices 4091 data; I>2σ(I) R1 = 0.0687, wR2 = 0.1828 all data R1 = 0.0770, wR2 = 0.1876 Weighting scheme w=1/[σ 2 (F o 2 )+(0.0822P) 2 +2.1673P] where P=(F o 2 +2F c 2 )/3 Largest diff. peak and hole 0.886 and -0.665 eÅ -3 R.M.S. deviation from mean 0.084 eÅ -3 238 A3.2.7 Crystal Structure Report for Tetraphenylphosphonium dinitro(1-H- tetrazol-5-yl)methanide monohydrate (12) Figure A3.11: ORTEP plot of the asymmetric unit of tetraphenylphosphonium dinitro(1-H-tetrazol-5- yl)methanide monohydrate (12) showing the two-part disordered anion. 239 Figure A3.12: Projection of the packing of one part of the disordered asymmetric unit of 12 perpendicular to the 100 plane. Hydrogen atoms were omitted for clarity. Table A3.13: Sample and crystal data for tetraphenylphosphonium dinitro(1-H-tetrazol-5-yl)methanide monohydrate (12) Identification code XGC4_207a_01 Chemical formula C 26 H 23 N 6 O 5 P Formula weight 530.47 g/mol Temperature 100(2) K Wavelength 0.71073 Å Crystal size 0.166 x 0.189 x 0.406 mm Crystal habit colorless blade/prism Crystal system orthorhombic Space group P 21 21 21 Unit cell dimensions a = 7.3432(3) Å α = 90° b = 11.7912(5) Å β = 90° c = 29.3176(12) Å γ = 90° Volume 2538.47(18) Å 3 Z 4 Density (calculated) 1.388 g/cm 3 Absorption coefficient 0.158 mm -1 F(000) 1104 Table A3.14: Data collection and structure refinement for 12 240 Diffractometer Bruker APEX DUO Radiation source fine-focus tube, MoKα Theta range for data collection 1.39 to 30.59° Index ranges -10<=h<=10, -16<=k<=16, -41<=l<=40 Reflections collected 64191 Independent reflections 7771 [R(int) = 0.0595] Coverage of independent reflections 99.8% Absorption correction multi-scan Max. and min. transmission 0.9740 and 0.9390 Structure solution technique direct methods Structure solution program SHELXTL XT 2013/1 (Bruker AXS, 2014) Refinement method Full-matrix least-squares on F 2 Refinement program SHELXTL XL 2014/7 (Bruker AXS, 2014) Function minimized Σ w(F o 2 - F c 2 ) 2 Data / restraints / parameters 7771 / 7 / 362 Goodness-of-fit on F 2 1.073 Final R indices 6589 data; I>2σ(I) R1 = 0.0418, wR2 = 0.0921 all data R1 = 0.0558, wR2 = 0.0987 Weighting scheme w=1/[σ 2 (F o 2 )+(0.0457P) 2 +0.5995P] where P=(F o 2 +2F c 2 )/3 Absolute structure parameter 0.1(0) Largest diff. peak and hole 0.445 and -0.255 eÅ -3 R.M.S. deviation from mean 0.051 eÅ -3 A3.3 References (1) SAINT+ V8.34A, Bruker AXS Madison, WI. (2) SADABS V2014/4, Bruker AXS Madison, WI. (3) SHELXTL V2014/6, Bruker AXS Madison, WI. (4) Sheldrick, G. Acta Crystallogr. Sect. A: Found. Crystallogr. 2008, 64, 112. (5) Sheldrick, G. Acta Crystallogr. Sect. A: Found. Crystallogr. 2015, 71, 3. (6) Sheldrick, G. Acta Crystallogr. Sect. C: Cryst. Struct. Commun. 2015, 71, 3. (7) Farrugia, L. J. Appl. Crystallogr. 2012, 45, 849. (8) CCSD 2014. (9) Burchfield, H.P.; Gullstrom, D.K. US Patent US3054800 A, 1962. (10) N. R. Smith, R. H. Wiley. US Patent US3111524, 1963. (11) Haiges, R.; Bélanger-Chabot, G.; Kaplan, S. M.; Christe, K. O. Dalton Trans. 2015, 44, 2978. (12) V. M. Chernysev, N. D. Z., V. B. Il'in, V. A. Taranushich Russ. J. Appl. Chem. 2000, 73, 839. (13) Haiges, R.; Bélanger-Chabot, G.; Kaplan, S. M.; Christe, K. O. Dalton Trans. 2015, 44, 7586. (14) Haiges, R.; Christe, K. O. Inorg. Chem. 2013, 52, 7249. (15) Thottempudi, V.; Gao, H.; Shreeve, J. n. M. J. Am. Chem. Soc. 2011, 133, 6464. 241 (16) Bélanger-Chabot, G.; Rahm, M.; Haiges, R.; Christe, K. O. Angew. Chem. Int. Ed. 2013, 52, 11002. (17) Haiges, R.; Boatz, J. A.; Vij, A.; Gerken, M.; Schneider, S.; Schroer, T.; Christe, K. O. Angew. Chem. Int. Ed. 2003, 42, 5847. (18) Haiges, R.; Boatz, J. A.; Schneider, S.; Schroer, T.; Yousufuddin, M.; Christe, K. O. Angew. Chem. Int. Ed. 2004, 43, 3148. 242 APPENDIX 4: ADDITIONAL INFORMATION ON (TRINITROMETHYL)BORATE COMPLEX ANIONS (CHAPTER 5) A4.1 Experimental Details Caution ! The materials used and synthesized in this study are energetic. They should be handled in quantities not exceeding the milimolar scale. Manipulations should be carried out behind blast shields and with adequate personal safety gear (face shield, heavy leather jacket and gloves, ear protection). Although no incident occured while handling nitroform, it is a strong oxidizer and care must be taken when handling organic mixtures, especially with aromatic compounds 1 . Nitroformate salts, especially alkali metal salts are unstable and should not be stored. Thoroughly drying [Na . (glyme) 2 ][BH 3 C(NO 2 ) 3 ] under vacuum at room temperature resulted in a mixture of sodium nitrate, sodium nitroformate and Na[BH 3 C(NO 2 ) 3 ], which can explode when provoked with a metal spatula. All reactions were carried out in glass ampules that were closed by Teflon-FEP valves and in Schlenk tubes using standard Schlenk and vacuum line techniques. Non-volatile compounds were handled in the nitrogen atmosphere of a drybox. Crystallization and drying operations were performed in Teflon-FEP reactors equipped with stainless steel valves. Raman spectra were recorded in Teflon-FEP reactors or in Pyrex J-Young NMR tubes in the range of 4000-80 cm -1 on a Bruker Equinox 55 FT-RA spectrometer, using a Nd:YAG laser at 1064nm. The IR spectra were recorded on a Midac M series spectrometer in the range of 4000-370 cm -1 ; spectra of solid compounds were recorded as KBr pellets and spectra of volatile compounds were recorded in a glass cell with silver chloride windows and a Teflon-FEP valve. UV-Vis spectra were recorded with an Olis Globalworks Carry-14 spectrophotometer in the range of 185 to 825 nm. The NMR spectra were recorded with Varian 400MR (400 MHz) and VNMRS-500 (500 MHz) instruments and with a Bruker AMX 500 MHz instrument. NMR signals were referenced externally to H 3 PO 4 for 31 P, to CH 3 NO 2 for 14 N, to Et 2 O . BF 3 for 11 B, to TMS for 1 H and 13 C and, when possible, internally to the CD 2 HCN residual signal for for 1 H, and to the CD 3 CN signal for 13 C. Reference NMR signals for BH 3 were obtained by dissolving BH 3 /THF (Sigma-Aldrich) in CD 3 CN ( 11 B NMR 400 MHz δ -25.9 ppm, qt; 1 H NMR 400 MHz δ 1.53 ppm, qt). DTA measurements were performed at a heating rate of 5°C/min using an OZM DTA 552-Ex differential thermal analyzer. Nitroform and nitroformate salts were prepared according to literature methods. 1,2 PPh 4 BH 4 was prepared according to literature methods. 3 Sodium borohydride, PPh 4 Cl and PPNCl were purchased from Sigma-Aldrich. PPh 4 Cl was dried under vacuum at 70 °C for several hours prior to use. Drysolv © acetonitrile was purchased from EMD. 1,2- 243 dimethoxyethane (glyme) (Alfa Aesar) and pentanes (EMD) were refluxed over sodium/benzophenone ketyl until a deep purple color was observed and then distilled. CD 3 CN (Cambridge Isotope Laboratories) was dried over molecular sieves (3Å) and degased using freeze-pump-thaw cycles. Dichloromethane (Macron) and acetonitrile (EMD) were refluxed over and distilled from P 2 O 5 (J.T. Baker). All solvents were stored over 3Å molecular sieves that were dried by heating in vacuo. Synthesis of [Na(glyme) 2 ][BH 3 C(NO 2 ) 3 ]. In a typical synthesis, a solution of nitroform (414 mg ; 2.74 mmol) in glyme (5 mL) was rapidly added to a suspension of NaBH 4 (94.3 mg ; 2.49 mmol) in glyme (5 mL), resulting in effervescence and a yellow solution. Na[BH 3 C(NO 2 ) 3 ] was formed quantitatively with respect to NaBH 4 according to 11 B NMR spectroscopy. The addition of pentane (< 5 mL) to the solution resulted in a two phase mixture in which a concentrated glyme solution of Na[BH 3 C(NO 2 ) 3 ] constituted the denser, viscous phase. Removing the lighter phase follwed by two more extractions allowed the removal of the excess nitroform. Dichloromethane (10 mL) was added to the resulting bright yellow liquid and the solution was filtered into a FEP reactor. The solvent was removed under vacuum between - 40 and -8 °C over the course of several days, yielding a green-yellow solid, which was characterized by NMR, Raman and IR spectroscopy. The 14 N NMR spectrum showed that the sample contained less than 1% sodium nitroformate. A 1 H NMR spectrum in CD 3 CN indicated that an approximately 2:1 glyme adduct had been isolated ([Na(glyme) 2 ][BH 3 C(NO 2 ) 3 ]). Attempts to remove the coordinated solvent under vacuum at room temperature yielded a mixture containing significant amounts of sodium nitroformate. The solid was stored without decomposition below 0 °C. 14 N NMR (CD 3 CN, 500 MHz) δ/ppm -7.2 (s, τ 1/2 = 14 Hz, [BH 3 C(NO 2 ) 3 ] - ); 11 B NMR (CD 3 CN, 500 MHz) δ/ppm -23.8 (qt, 1 J( 11 B- 1 H) = 95 Hz, τ 1/2 = 8 Hz [BH 3 C(NO 2 ) 3 ] - ); 1 H NMR (CD 3 CN, 500 MHz, referenced internally to CD 2 HCN) δ/ppm 3.46 (s, MeO- CH 2 -), 3.29 (s, CH 3 O-), 1.08 (qt, 1 J( 1 H- 11 B) = 95 Hz, τ 1/2 = 11 Hz, [BH 3 C(NO 2 ) 3 ] - ). Raman (-40 °C, 200 mW) ῦ/cm -1 : 3004 (5.2), 2953 (10), 2931 (9.9), 2907 (10.0), 2879 (sh), 2853 (9.0), 2832 (9.1), 2817 (sh), 2795 (sh), 2731 (1.2), 2386 (sh), 2361 (4.2), 2307 (0.5, broad), 2231 (0.4, broad), 1576 (1.0, broad), 1479 (2.6), 1455 (sh), 1448 (2.4), 1415 (0.3), 1379 (2.1), 1326 (0.6, broad), 1285 (1.4), 1245 (0.5), 1161 (1.0), 1130 (1.0), 1036 (1.0, broad), 1026 (1.2), 1003 (0.5, broad), 869 (4.0, nitroformate), 857 (3.9), 844 (1.1), 573 (0.4, broad), 479 (0.6, broad), 408 (1.3, broad), 376 (2.3, broad), 283 (0.6, broad), 212 (1.0), 122 (3.1); IR (KBr, 20°C) ῦ/cm -1 : 2901 m, 2711 vw, 2654 vw, 2423 s, 2401 s, 2375 s, 2345 m, 2249 w, 2001 vw, 1601 vs, 1577 vs, 1566 vs, 1488 sh, 1383 m, 1330 s, 1316 sh, 1268 sh (nitroformate), 1203 w, 1183 m, 1163 sh (nitroformate), 1134 s, 1089 w, 1003 m, 856 s, 848 sh, 773 s, 763 s, 734 vw (nitroformate), 673 w, 652 m, 635 m; UV-Vis (glyme) 257 nm (1.38), 291 nm (1.15, sh), 353 nm (1.26). DTA : Large exotherm at 121 °C (onset at 100 °C). 244 Detection of [BH 2 (C(NO 2 ) 3 ) 2 ]- and the 13 C NMR signal for [BH 3 C(NO 2 ) 3 ]-. A solution of nitroform (179 mg ; 1.19 mmol) in glyme (2-3 mL) was cannulated over NaBH 4 (40.8 mg ; 1.08 mmol), resulting in a bright yellow solution, which was then cannulated into an NMR tube. The solution was analyzed by NMR spectroscopy. The 14 N, 11 B and 1 H NMR spectra corresponded to the ones previously recorded for Na[BH 3 C(NO 2 ) 3 ]. The 13 C NMR spectrum was recorded overa period of approximately 2 days (pulse delay of 3 s, approximately 20000 transients), which allowed the observation of a boron-coupled signal. Under these conditions, only the the 13 C NMR signal of Na[BH 3 C(NO 2 ) 3 ] was observed. 13 C NMR (glyme, unlocked, 500 MHz, referenced externally to TMS) δ/ppm 151.1 (qt, 1 J( 11 B- 13 C) = 32 Hz). Figure A4.1: 14 N NMR spectrum in glyme showing trace amounts of [BH 2 (C(NO 2 ) 3 ) 2 ] - at ca -15 ppm. 245 Figure A4.2: 11 B NMR spectrum in glyme showing trace amounts of [BH 2 (C(NO 2 ) 3 ) 2 ] - at ca -17.5 ppm. During the course of the 13 C NMR experiment, the small excess of nitroform had slowly reacted with the [BH 3 C(NO 2 ) 3 ] - anion and a new species, identified as [BH 2 (C(NO 2 ) 3 ) 2 ] - could be observed by NMR (Figure A4.1 and Figure A4.2). 14 N NMR (glyme, unlocked, 500MHz, referenced externally to TMS) δ/ppm -15.0 (s, [BH 2 (C(NO 2 ) 3 ) 2 ] - ); 11 B NMR (glyme, unlocked, 500MHz, referenced externally to TMS) δ/ppm -17.6 (t, 1 J ( 11 B- 1 H) = 104 Hz, [BH 3 C(NO 2 ) 3 ] - ). Synthesis of PPh 4 [BH 3 C(NO 2 ) 3 ]. A solution of Na[BH 3 C(NO 2 ) 3 ] (1.53 mmol) in dichloromethane (10 mL) was obtained as described previously. It was added to a PPh 4 Cl (577 mg; 1.54 mmol) solution in dichloromethane (15-20 mL), yielding a fine white suspension. A portion of the solution was filtered into a FEP reactor and the solvent was removed under vacuum between -30 and 0 °C overnight, yielding [PPh 4 ][BH 3 C(NO 2 ) 3 ] as pale yellow flakes. Traces of residual solvents in the isolated solid could only be observed by by 1 H NMR spectroscopy. 2 mol% of nitroformate anions were detected by 14 N NMR. The remainder of the solution was used in crystallization experiments, which yielded crystalline solids of insufficient quality for X-Ray diffraction. Alternatively, the compound could also be prepared by reacting nitroform with PPh 4 BH 4 in acetonitrile. 31 P NMR (CD 3 CN, 400 MHz) δ/ppm 22.9 (s, PPh 4 ); 14 N NMR 246 (CD 3 CN, 500 MHz) δ/ppm -7.2 (s, τ 1/2 = 15 Hz, [BH 3 C(NO 2 ) 3 ] - ); 13 C (CD 3 CN, 400MHz, referenced internally to CD 3 CN signal) δ/ppm 119.3 (d, 1 J( 13 C- 31 P) = 89.6 Hz, PPh 4 ), 131.7 (d, 3 J( 13 C- 31 P) = 13.2 Hz, PPh 4 ), 136.1 (d, 2 J( 13 C- 31 P) = 10.3Hz, PPh 4 ), 136.8 (d, 4 J( 13 C- 31 P) = 3Hz, PPh 4 ), no signal observed for [BH 3 C(NO 2 ) 3 ] - within a practical timeframe; 11 B NMR (CD 3 CN, 400 MHz) δ/ppm -23.8 (qt, 1 J( 11 B- 1 H) = 94 Hz, BH 3 C(NO 2 ) 3 - ); 1 H (CD 3 CN, 400 MHz, referenced internally to CD 2 HCN signal) δ/ppm 1.09 (qt, 1 J( 1 H- 11 B) = 94 Hz, BH 3 C(NO 2 ) 3 - ), 7.68 (m, PPh 4 ), 7.74 (m, PPh 4 ), 7.92 (m, PPh 4 ); Raman (20 °C, 75 mW) ῦ/cm -1 : 3175 (0.4), 3150 (0.5), 3083 (sh), 3067 (10), 3058 (sh), 3027 (0.5), 3013, (0.7), 2997 (0.4), 2961 (0.4), 2889 (0.1, broad), 2566 (0.1, broad), 2384 (sh), 2360 (1.4), 2306 (0.1, broad), 2229 (0.1, broad), 1590 (5.0), 1579 (sh), 1486 (0.1), 1443 (sh), 1438 (0.4), 1378 (0.8, broad), 1343 (0.3), 1320 (0.3, broad), 1191 (0.9), 1165 (1.1), 1113 (sh), 1101 (2.0), 1075 (0.2), 1029 (3.6), 1002 (9.5), 988 (0.4, sh), 934 (0.1), 930 (sh, broad), 857 (1.0), 759 (0.1), 728 (0.3), 682 (1.6), 617 (1.0), 531 (0.1), 484 (0.2), 411 (sh), 382 (0.5), 296 (1.0), 283 (sh), 254 (2.0), 213 (sh), 200 (1.8); IR (KBr, 20 °C): 3080 w, 3064 w, 3020 vw, 2865 vw, 2687 vw, 2382 m, 2350 sh, 2305w, 2232 vw, 1992 vw, 1968 vw, 1903 vw, 1822 vw, 1777 vw, 1680 vw, 1578 sh, 1566 s, 1550 s, 1481 m, 1442 sh, 1436 s, 1374 w, 1340 sh, 1315 m, 1259 sh (nitroformate), 1182 w, 1163 sh, 1155 w (nitroformate), 1125 sh, 1108 s, 1082 sh, 1070 sh, 1029 w, 994 m, 983 sh, 930 vw, 878 vw, 854 m, 844 m, 758 m, 746 m, 723 s, 688 s, 615 w, 526 s, 445 w, 411 w; DTA : Strong exotherm at 135 °C (onset at 118 °C). Synthesis of PPN[BH 3 C(NO 2 ) 3 ]. A solution of Na[BH 3 C(NO 2 ) 3 ] (1.35 mmol) in dichloromethane (10 mL) was obtained as described previously. It was added to a PPNCl (PPN = bis- (triphenylphosphoranylidene) ammonium) (716 mg; 1.25 mmol) solution in dichloromethane (15-20 mL), yielding a fine white suspension. A portion of the solution was filtered into a FEP reactor and the solvent was removed under vacuum between -30 and 0 °C overnight, yielding a pale yellow-orange solid which was characterized by IR, Raman and NMR spectroscopy. Small amounts of residual solvents in the isolated solid could only be observed by 1 H NMR spectroscopy. 2 mol% of nitroformate anions were detected by 14 N NMR. The remainder of the solution was filtered and used for crystallization experiments. X-ray quality crystals of [PPN][BH 3 C(NO 2 ) 3 ] were grown from the filtrate added with pentane at -20 °C. 31 P NMR (CD 3 CN, 400 MHz) δ/ppm 20.8 (s, PPN); 14 N NMR (CD 3 CN, 500 MHz) δ/ppm -7.2 (s, τ 1/2 = 15 Hz); 13 C (CD 3 CN, 400 MHz, internally referenced to CD 3 CN signal) δ/ppm 128.6 (m, PPN), 130.8 (m, PPN), 133.7 (m, PPN), 135.0 (m, PPN), no signal could be observed for [BH 3 C(NO 2 ) 3 ] within a practical timeframe; 11 B NMR (CD 3 CN, 400 MHz) δ/ppm -23.8 (qt, 1 J( 11 B- 1 H) = 94 Hz); 1 H (CD 3 CN, 400 MHz internally referenced to CD 2 HCN signal) δ/ppm 1.09 (qt, 1 J( 1 H- 11 B) = 96 Hz), 7.48 (m, PPN), 7.58 (m, PPN), 7.66 (m, PPN); Raman (-40°C, 200 mW) ῦ/cm -1 : 3180 (0.3), 3151 (0.4), 3095 (sh), 3066 (10), 3042 (sh), 3028 (sh), 3015 (0.4), 2996 (0.3), 2960 (0.4), 2895 (0.1, broad), 2616 (0.1, broad), 2560 (0.1, broad), 247 2384 (sh), 2361 (0.8), 2305 (sh), 2229 (0.1), 1593 (3.1), 1578 (1.2), 1485 (0.1), 1442 (0.2), 1378 (0.4), 1340 (sh, broad), 1321 (0.2, broad), 1189 (0.6), 1165 (0.7), 1114 (1.6), 1077 (0.1), 1030 (2.0), 1003 (5.3), 986 (sh), 978 (sh), 931 (0.1), 857 (0.5), 728 (0.2, broad), 697 (0.1, broad), 665 (1.2, broad), 618 (0.8), 487 (0.1, broad), 404 (0.2), 378 (0.3), 351 (0.2), 283 (0.4), 269 (0.4), 255 (0.5), 243 (sh), 237 (0.7), 228 (0.4), 201 (0.5), 192 (sh), 181 (0.3), 176 (sh); IR (KBr pellet) ῦ/cm -1 : 3080 w, 3060 w, 2881 w, 2686 w, 2370 m, 2305 w, 2220 w, 1964 w, 1896 w, 1818 w, 1778 w, 1590 sh, 1569 s, 1553 s, 1483 m, 1438 s, 1372 w, 1320 s, 1251 s, 1180 w, 1153 w, 1117 s, 1072 w, 1027 w, 997 m, 976 w, 930 w, 855 w, 841 w, 803 w, 792 sh, 761 sh, 746 m, 723 s, 690 s, 646 sh, 614 w, 548 s, 534 s, 527 sh, 499 s, 446 w, 425 w, 411 w; DTA : Broad endotherm at 138 °C (onset at 125 °C); Strong exotherm at 168 °C (onset at 165 °C). Crystallography. The single crystal diffraction data were collected on a Bruker SMART APEX DUO three-circle platform diffractometer, equipped with an APEX II CCD detector with the χ-axis fixed at 54.74°, and using Mo Kα radiation (TRIUMPH curved-crystal monochromator) from a fine-focus tube. The diffractometer was equipped with an Oxford Cryosystems Cryostream 700 apparatus for low- temperature data collection at 100 K. A complete hemisphere of data was scanned on ω (0.5°) and φ (0.5°) with a run time of 1 s per frame at a detector resolution of 512x512 pixels using the BIS software package. 4-8 A total of 2520 frames were collected in 5 sets. The frames were then integrated using the SAINT algorithm 7[4] to give the hkl files corrected for LP/decay. The absorption correction was performed using the SADABS program. 6 The structures were solved by direct methods and refined on F 2 using the Bruker SHELXTL software package. 5 All non-hydrogen atoms were refined anisotriopically. Hydrogen atoms were calculated using the riding model. ORTEP drawings were prepared using the ORTEP-3 for Windows V2.02 program. 9 A4.2 Additional Remarks General remarks. The use of excess BH 4 - led to the formation of several unidentified products observable by 11 B NMR spectroscopy and significantly increased the nitroformate contamination of [BH 3 C(NO 2 ) 3 ] - compounds. Therefore, a 10 mol% excess of nitroform was used in order to ensure the complete consumption of the BH 4 - starting material. Choice of solvent. The two-step nature of the formation of [BH 3 C(NO 2 ) 3 ] - from nitroform and NaBH 4 was demonstrated by running an in situ variable temperature NMR experiment in CD 3 CN. At 265 K, a rapid dihydrogen elimination was observed, yielding the nitroformate anion and the CD 3 CN . BH 3 adduct. When brought to 295 K, the products slowly reacted to form the [BH 3 C(NO 2 ) 3 ] - anion. In glyme, the 248 reaction is significantly faster at room temperature. Attempts to prepare Na[BH 3 C(NO 2 ) 3 ] in dichloromethane or even in the more coordinating diethyl ether resulted in the formation of insoluble products from which only nitroformate salts could be extracted with glyme. However, one formed, salts of [BH 3 C(NO 2 ) 3 ] - are soluble in dichloromethane. Qualitative observations on stability. As expected for B-H compounds, [BH 3 C(NO 2 ) 3 ] - derivatives are moisture sensitive. However, the PPh 4 + and PPN + salts of [BH 3 C(NO 2 ) 3 ] - can, because of their low solubility in water, be handled (not stored) in moist air without significant hydrolysis. In an inert atmosphere, solutions and salts of [BH 3 C(NO 2 ) 3 ] - are indefinitely stable when kept below 0°C. At room temperature, however, an increase in nitroformate anion accompanied by the formation of a very fine colorless precipitate were noted in solutions of PPh 4 and PPN [BH 3 C(NO 2 ) 3 ]. Since no additional boron species could be detected in 11 B NMR for several months, it was concluded that this decomposition was caused by the formation of insoluble boron derivatives, which probably does not occur through unimolecular reactions. The solid [BH 3 C(NO 2 ) 3 ] - salts contained variable amounts of C(NO 2 ) 3 - when left at room temperature. However, PPN and PPh 4 [BH 3 C(NO 2 ) 3 ] remain the major species even after several months in these conditions. Solid [Na(glyme) 2 ][BH 3 C(NO 2 ) 3 ], however, becomes a pale yellow solid after a week or two, which contains significant amounts of NaNO 3 , indicative of the decomposition of the nitroformate moiety. A4.3 Further Analysis of Experimental Data Vibrational Spectroscopy. Quantum chemical calculations on the [BH 3 C(NO 2 ) 3 ] - anion agree well with the experimental spectra and helped in the assignment of vibrational modes. Table A4.1 compares selected vibrational modes, which could be unambiguously observed by Raman and IR spectroscopy, with the values observed for sodium nitroformate and the values predicted by gas-phase calculations. 249 Table A4.1: Selected vibrational data for [BH 3 C(NO 2 ) 3 ] - derivatives and for NaC(NO 2 ) 3 . [Na(glyme) 2][BH 3C(NO 2) 3] PPh 4[BH 3C(NO 2) 3] PPN[BH 3C(NO 2) 3] [BH 3C(NO 2) 3] - NaC(NO 2) 3 10 Raman cm -1 (Intensity) IR cm -1 Raman cm -1 (Intensity) IR cm -1 Raman cm -1 (Intensity) IR cm -1 Theoretical a (IR)[Ra] b Raman cm -1 (Intensity) IR cm -1 Assignment 2386 (sh) 2423 s 2401 s 2384 (sh) 2382 m 2384 (sh) 2370 m 2404 (183)[132] 2404 (183)[131] - - υ as BH 3 2361 (10) 2375s 2345 m 2360 (10) 2350 (sh) 2361 (10) Not Obs. 2367(84)[283] - - υ s BH 3 Not Obs. 1601 vs Obsc. 1566 s 1550 s Obsc. 1569 s 1553 s 1578(279)[8] 1578 (279)[8] 1512 (0.7) 1500 (0.7) 1517 s υ as NO 2 (o. o. ph.) 1576 (br, 2.4) 1577-1566 vs Obsc. Obsc. 1557 (334)[2] 1462 (1.3) 1502 s υ as NO 2 (i. ph.) 1379 (5.0) 1383 m 1378 (5.8) 1374 w 1378 (4.3) 1372 w 1384 (79)[27] 1402 (2.7) 1423 m 1385 w υ s NO 2 (i. ph.) 1326 (br, 1.3) 1330 s 1316 sh 1321 (2.4) 1340 sh 1315 m 1321 (2.1) 1320 s 1323 (128)[7] 1323 (128)[7] 1305 (5.3) 1270 s c υ s NO 2 (o. o. ph.) 857 (9.4) 856 s 857 (7.4) 854 m 857 (6.4) 855 w 850 (16)[21] 885 (10) 872 w δ NO 2 (i. ph.) sh = shoulder; Obsc. = obscured; br = broad; m = medium; s = strong; vs = very strong; w = weak; i. ph. = in phase; o. o. ph. = out of phase. a B3LYP/aug-cc-pVTZ frequencies using a 0.977 scaling factor. Not obs. = Not observed or could not unambiguously be attributed to the [BH 3C(NO 2) 3] - anion. b IR intensities are expressed in km/mol, Raman activities are expressed in Å 4 /amu. c Observed by IR in KBr pellets and by Raman by us but only by IR "solid phase on Ge" in the work cited. DTA. To confirm the validity of the DTA data, which indicates that [BH 3 C(NO 2 ) 3 ] - is intrinsically quite stable, PPh 4 [BH 3 C(NO 2 ) 3 ] was heated to 118 °C (two degrees below its decomposition onset) at a rate of 5 °C/min. The sample was then dissolved completely in CD 3 CN and was shown to contain pure PPh 4 [BH 3 C(NO 2 ) 3 ] by NMR spectroscopy. A4.4 Theoretical Data Computational Details. The multi-component complete basis set (CBS-QB3) 11,12 method was used for obtaining the energetics in the gas phase using the Gaussian 09 program. 13 CBS-QB3 uses coupled cluster (CCSD(T)) energies, which are extrapolated to the basis set limit using MP2 and MP4 energies and empirical corrections. Geometries and thermodynamic corrections are obtained at the B3LYP/6- 311G(d,p) level. CBS-QB3 has been extensively tested in several benchmarks and is expected to produce reliable results. 11,12,14-16 The mean absolute deviation in the G2 test set is reported to be 0.87 kcal/mol. 11 For an accurate fit with the X-ray structure the BH 3 -C(NO 2 ) 3 C 3 symmetric anion was optimized at the M06-2X/aug-cc-pVTZ level of theory in the gas phase. To account for possibly significant solvation-induced structural rearrangements, accurate solvation energies (ΔH solv and ΔG solv ) were obtained from separate vacuum and solution phase 250 optimizations at the M06-2X/cc-pVTZ level of theory, 17-19 using the SMD-PCM 20 method and ε = 35.69 (acetonitrile). The Hessian matrix was evaluated both for the gas-phase and SMD-PCM wave functions. This approach has been shown particularly important for intramolecular NO 2 -transfer transition states, which can switch between NO 2 -radical transfer in the gas-phase, towards NO 2 -cation character in polar solvents. 21,22 Final free energies in solution were obtained as, ΔG solv = ΔG(CBS-QB3) gas-phase, 1atm, 298K + ΔE(SMD-PCM-M06-2X/cc-pVTZ, solution geometry) + G corr (SMD-PCM-M06-2X, solution geometry) - ΔE(M06-2X/cc-pVTZ, vacuum geometry) - G corr (M06- 2X, vacuum geometry) + G corr (1atm → 1M) (A4.1) The larger size of the [BCl 3 C(NO 2 ) 3 ] - adduct prohibits calculations at the CBS-QB3 level. Its gas-phase energy relative to C(NO 2 ) 3 - + BCl 3 was instead estimated by correcting the corresponding [BH 3 C(NO 2 ) 3 ] - → C(NO 2 ) 3 - + BH 3 CBS-QB3 energy using M06-2X/cc-pVTZ calculations, ΔE ([BCl 3 C(NO 2 ) 3 ] - → C(NO 2 ) 3 - + BCl 3 ) = ΔE ([BH 3 C(NO 2 ) 3 ] - → C(NO 2 ) 3 - + BH 3 ) (CBS-QB3) + ΔE ([BCl 3 C(NO 2 ) 3 ] - → C(NO 2 ) 3 - + BCl 3 ) (M06-2X/cc-pVTZ) - ΔE ([BH 3 C(NO 2 ) 3 ] - → C(NO 2 ) 3 - + BH 3 ) (M06-2X/cc-pVTZ) (A4.2) Nuclear magnetic shielding tensors were calculated in acetonitrile at the GIAO-PCM-B3LYP/6- 311+(2d,p) level. Raman spectra were calculated at the B3LYP/aug-cc-pVTZ level, and scaled by 0.977 for best fit with the experimental results. Transition States. Species 5 (Figure A4.3) is the resulting radical anion following homolytic bond dissociation of NO 2 from nitroformate. The bond strenght of the C-N bond in 1 can be estimated from the relative enthalpy of 5 and NO 2 at infinite separation, giving a value of 60.9 kcal/mol in acetonitrile solution (Table A4.2). This is markedly higher than the corresponding bond strength of 2 (calculated relative to NO 2 and 6), which equals 47.8 kcal/mol (Figure A4.3). Since salts of nitroformate and of 2 have similar thermal stabilities and decompose at ~100-120 °C, 23 both values are clearly too high to account for decompositon at this temperature. Furthermore, the estimated difference of 13.1 kcal/mol is not in accord with the fact that both compounds are near equally stable. The decomposition of the nitroformate anion is not treated further in the current work. Instead, a series of pathways for the decomposition of 2 was investigated. Figure A4.3 illustrates the most relevant decomposition pathways investigated, together with their respective energies compiled in Table A4.2. Aside from the lowest transition state TS1, discussed in the main article, five additional transition states were found (TS2-6). The one lowest in energy, TS3, describes concerted proton transfer and NO 2 dissociation, and 251 corresponds to a 35.8 kcal/mol free energy barrier in solution at ambient conditions. This barrier is slightly higher than the highest estimate for the direct homolytic C-N bond dissociation (34.9 kcal/mol), and homolytic C-B bond dissociation (35.0 kcal/mol) calculated here. Therefore, the most likely pathway for the decomposition of 2 appears to be through solvent-assisted BH 3 dissociation, as discussed in the main article.The closest upper free energy barrier estimate for such a process is TS1, which describes the isomerization process to 3 and correspond to 29.2 kcal/mol in solution. The latter barrier is in good accord with the observed kinetic stability by DTA. Figure A4.3: Considered pathways for intramolecular decomposition of [C(NO 2 ) 3 ] - , [BH 3 C(NO 2 ) 3 ] - (2) and [BH 3 ON(O)C(NO 2 ) 2 ] - (3). Selected bond lengths are shown in Ångström (Å). Values for acetonitrile are shown within parentheses. 252 Table A4.2: Energetics of considered decomposition pathways. All energies are relative 2 unless otherwise stated. Free energies in acetonitrile (1 M and 298 K) and given within parentheses. Structure: Description: ΔH a ΔG a 1 + BH 3 C(NO 2 ) 3 - + BH 3 38.8 (35.0) 28.0 (24.2) 2 BH 3 -C(NO 2 ) 3 - 0.0 (0.0) 0.0 (0.0) 3 BH 3 -O-N(O)-C(NO 2 ) 2 - 13.3 (11.7) 11.6 (9.8) 4 BCl 3 -C(NO 2 ) 3 - -33.9 (-25.5) b -20.2 (-9.5) b TS1 2 →3 32.9 (32.0) 30.5 (29.2) TS2 C-NO 2 →O-NO 2 transfer in 2 73.5 (70.5) 73.6 (69.6) TS3 2 → NO 2 + HC(NO 2 ) 2 - 38.2 (37.5) 36.9 (35.8) TS4 2 → HONO + BH 2 C(NO2) 2 - 39.0 (40.9) 38.5 (39.6) TS5 3 → BH 2 OH + ONC(NO 2 ) 2 - 45.3 (44.2) 43.0 (42.1) TS6 3 → [BH 3 -NO 3 + ONC-NO 2 - 83.1 (81.0) 81.9 (79.3) 5 NO 2 dissociated from 1 65.7 (60.9) 50.9 (49.2) 6 NO 2 dissociated from 2 49.8 (47.8) 34.9 (33.7) 7 NO 2 dissociated from 3 75.4 (71.3) 60.7 (57.7) a Corrected CBS-QB3 energies (see computational details) b Energy relative to 1 + BCl 3 . 253 A4.5 Predicted Vibrational Data Table A4.3: Calculated vibrational spectra of the gas phase C 3 symmetric [BH 3 C(NO 2 ) 3 ] - anion (B3LYP/aug-cc-pVTZ level, unscaled frequencies). Mode Symmetry Freq (cm-1) IR act (km/mol) Raman act (Å 4 /amu) 1 E 37 0 1 2 E 37 0 1 3 A 117 0 3 4 E 197 0 1 5 E 197 0 1 6 A 239 0 0 7 A 257 0 0 8 E 268 6 2 9 E 268 6 2 10 A 362 0 7 11 E 396 0 4 12 E 396 0 4 13 A 453 0 6 14 E 640 9 1 15 E 640 9 1 16 A 670 6 1 17 E 774 18 1 18 E 774 18 1 19 E 850 35 1 20 E 850 35 1 21 A 870 16 21 22 A 954 5 1 23 E 991 0 7 24 E 991 0 7 25 A 1144 104 3 26 E 1175 8 5 27 E 1175 8 5 28 E 1354 128 7 29 E 1354 128 7 30 A 1417 79 27 31 A 1594 334 2 32 E 1616 279 8 33 E 1616 279 8 34 A 2423 84 283 35 E 2461 183 131 36 E 2461 183 132 254 A4.6 Crystallographic Data for PPN[BH 3 C(NO 2 ) 3 ] Figure A4.4: ORTEP plot of the two symmetrically independent [BH 3 C(NO 2 ) 3 ] - anions in the asymmetric unit of PPN[BH 3 C(NO 2 ) 3 ]. Thermal ellipsoids are drawn at the 50% probability level. Hydrogen atom positions were idealized. Figure A4.5: Projection of the packing of PPN[BH 3 C(NO 2 ) 3 ] perpendicular to the 010 plane. 255 Table A4.4: Sample and crystal data for PPN[BH 3 C(NO 2 ) 3 ]. Identification code XGC2_106b_ Chemical formula C 37 H 33 BN 4 O 6 P 2 Formula weight 702.42 Temperature 100(2) K Wavelength 0.71073 Å Crystal size 0.430 x 0.730 x 1.170 mm Crystal habit colorless prism Crystal system monoclinic Space group P 1 21/n 1 Unit cell dimensions a = 26.6005(18) Å α = 90° b = 10.1048(7) Å β = 104.3170(10)° c = 26.9873(18) Å γ = 90° Volume 7028.7(8) Å 3 Z 8 Density (calculated) 1.328 g/cm 3 Absorption coefficient 0.176 mm -1 F(000) 2928 Table A4.5: Data collection and structure refinement for PPN[BH 3 C(NO 2 ) 3 ]. Theta range for data collection 1.24 to 30.50° Index ranges -37<=h<=37, -14<=k<=14, -38<=l<=38 Reflections collected 169029 Independent reflections 21269 [R(int) = 0.0602] Max. and min. transmission 0.9279 and 0.8211 Structure solution technique direct methods Structure solution program SHELXS-97 (Sheldrick, 2008) Refinement method Full-matrix least-squares on F 2 Refinement program SHELXL-97 (Sheldrick, 2008) Function minimized Σ w(F o 2 - F c 2 ) 2 Data / restraints / parameters 21269 / 1 / 903 Goodness-of-fit on F 2 0.965 Δ/σ max 0.001 Final R indices 16153 data; I>2σ(I) R1 = 0.0486, wR2 = 0.1024 all data R1 = 0.0724, wR2 = 0.1127 Weighting scheme w=1/[σ 2 (F o 2 )+(0.0270P) 2 +11.2991P] where P=(F o 2 +2F c 2 )/3 Largest diff. peak and hole 0.571 and -0.540 eÅ -3 R.M.S. deviation from mean 0.065 eÅ -3 A4.7 References (1) Huang, Y.; Gao, H.; Twamley, B.; Shreeve, J. n. M. Eur. J. Inorg. Chem. 2007, 2007, 2025. (2) Göbel, M.; Klapötke, T. M.; Mayer, P. Z. Anorg. Allg. Chem. 2006, 632, 1043. (3) Ehemann, M.; Davies, N.; Nöth, H. Z. Anorg. Allg. Chem. 1972, 389, 235. 256 (4) Bruker Instrument Service v2011.4.0.0; Bruker AXS: Madison, WI, 2011. (5) Bruker SHELXTL V2011.4-0; Bruker AXS: Madison, WI, 2011. (6) SADABS V2008/1; Bruker AXS: Madison, WI, 2008. (7) SAINT V7.68A; Bruker AXS: Madison, WI, 2009. (8) Sheldrick, G. Acta Crystallogr. Sect. A: Found. Crystallogr. 2008, 64, 112. (9) Farrugia, L. J. Appl. Crystallogr. 1997, 30, 565. (10) Shlyapochnikov, V. A.; Oleneva, G. I.; Novikov, S. S. Russ Chem Bull 1971, 20, 2477. (11) Montgomery, J. J. A.; Frisch, M. J.; Ochterski, J. W.; Petersson, G. A. J. Chem. Phys. 1999, 110, 2822. (12) Montgomery, J. J. A.; Frisch, M. J.; Ochterski, J. W.; Petersson, G. A. J. Chem. Phys. 2000, 112, 6532. (13) Frisch, M. J. et al. Gaussian 09, revision A.02; Gaussian, Inc., Wallingford CT: 2009 (14) Guner, V.; Khuong, K. S.; Leach, A. G.; Lee, P. S.; Bartberger, M. D.; Houk, K. N. J. Phys. Chem. A 2003, 107, 11445. (15) Guner, V. A.; Khuong, K. S.; Houk, K. N.; Chuma, A.; Pulay, P. J. Phys. Chem. A 2004, 108, 2959. (16) Ess, D. H.; Houk, K. N. J. Phys. Chem. A 2005, 109, 9542. (17) Valero, R.; Gomes, J. R. B.; Truhlar, D. G.; Illas, F. J. Chem. Phys. 2008, 129, 124710. (18) Zhao, Y.; Truhlar, D. Theor. Chem. Acc. 2008, 120, 215. (19) Zhao, Y.; Truhlar, D. G. J. Chem. Theory Comput. 2011, 7, 669. (20) Marenich, A. V.; Cramer, C. J.; Truhlar, D. G. J. Phys. Chem. B 2009, 113, 6378. (21) Rahm, M.; Brinck, T. Chem. Eur. J. 2010, 16, 6590. (22) Rahm, M.; Dvinskikh, S. V.; Furó, I.; Brinck, T. Angew. Chem. Int. Ed. 2011, 50, 1145. (23) Titova, K. V.; Kolmakova, E. I.; Rosolovskii, V. Y. Bull. Acad. Sci. USSR, Div. Chem. Sci. (Engl. Transl.) 1977, 26, 1822. 257 APPENDIX 5: ADDITIONAL INFORMATION ON REACTION OF DINITROAMINE WITH AMMONIA-BORANE (CHAPTER 6) A5.1 Experimental Details Caution ! The materials used and synthesized in this study are energetic. They should be handled in quantities not exceeding the milimolar scales. Manipulations should be carried out behind blast shields and with adequate personal safety gear (face shield, heavy leather jacket and gloves, ear protection). Reactions were carried out either in glass vessels equipped with Teflon valves or in Teflon-FEP reactors equipped with stainless steel valves using standard Schlenk and vacuum line techniques. Non- volatile compounds were handled in the dry nitrogen atmosphere of a drybox. NMR spectra were recorded on a Bruker AMX-500 spectrometer. 1 H spectra were referenced externally to tetramethylsilane. 11 B spectra were referenced externally to BF 3 . OEt 2 /CDCl 3 . 14 N spectra were referenced externally to neat MeNO 2 . Raman spectra were recorded in Pyrex J.Young NMR tubes in the range of 4000-80 cm -1 on a Bruker Equinox 55 FT-RA spectrometer, using a Nd:YAG laser at 1064nm. The IR spectra were recorded on a MIDAC M Series in the range 4000-370 cm -1 on AgCl pellets for solid samples and in a glass cell equipped with AgCl windows and a Teflon valve for gaseous samples. Single-crystal X-ray diffraction data were collected on a Bruker Smart Apex Duo 3-circle platform diffractometer, equipped with an Apex II charge-coupled device (CCD) detector with the χ-axis fixed at 54.74° and using Cu Kα radiation (multi-layer optics monochromator IuS microsource). The diffractometer was equipped with an Oxford Cryostream 700 apparatus for low-temperature data collection. The data acquisition was performed with the BIS software package. For Cu radiation, an individual acquisition strategie was determined. Frames were acquired with run times of 20 s and were scanned on ω (1°). The frames were then integrated with the SAINT algorithm 1 to give the hkl files corrected for Lp/decay. Absorption correction was performed with the SADABS program. 2 The structures were solved by direct methods and refined on F 2 by use of the Bruker SHELXTL software package. 3-6 All non-hydrogen atoms were refined anisotropically. ORTEP drawings were prepared with the ORTEP-III for Windows V2.02 program. 7 Short-contact distances and torsion angles were measured with the Mercury 3.1 Development (Build RC5) software. 8 Further crystallographic details for compounds 1 can be obtained from the Cambridge Crystallographic Data Centre [CCDC, 12 Union Road, Cambridge CB2 1EZ, U.K.; fax (+44) 1223-336- 033; e-mail deposit@ccdc.cam.ac.uk] on quoting the deposition numbers CCDC 104993. 258 Drysolv © acetonitrile and glyme were purchased from EMD and were stored over 3Å molecular sieves. CD 3 CN (Cambridge Isotope Laboratories) was dried over molecular sieves (3Å) and degassed using freeze-pump-thaw cycles. Dichloromethane (Macron) and acetonitrile (EMD) were refluxed over and distilled from P 2 O 5 (J.T. Baker). All solvents were stored over 3Å molecular sieves that were dried by heating in vacuo. NH 3 . BH 3 was purchased from Sigma-Aldrich. In the most successful syntheses, the ammonia-borane was extracted in glyme, then dried under vacuum overnight prior to use. Potassium dinitramide (KDN) was generously donated by EURENCO Bofors. It was dried by dissolving in dry acetonitrile or glyme and evaporating the solvent under vacuum overnight. Dinitroamine (HDN) was prepared using a modified literature method 9 by condensing a stoichiometric amount of hydrogen chloride onto a glyme suspension of KDN. The exact nature of HDN made in this way extends beyond the scope of this work. Synthesis of ammonia-dinitramidoborane, NH 3 . BH 2 N(NO 2 ) 2 (1). In a typical synthesis, glyme (5-10 mL) was condensed onto NH 3 . BH 3 (17 mg; 0.55 mmol). The solution was frozen at -196 °C and a solution of known HDN concentration in glyme (1.25g of solution; 0.53 mmol HDN) was cannulated onto the frozen mixture. After evacuation of the vessel, the mixture was thawed and gas evolution was observed. The non-condensable gases at -196 °C were measured on a calibrated glass vacuum line and indicated that the reaction reached completion (ca 0.5 mmol of H 2 measured). An aliquot of the reaction mixture was cannulated into a J. Young NMR tube, and its NMR analysis indicated that 1 was the major species in solution, with only ca 20 mol% NO 2 -containing side-products detected by 14 N NMR. The reaction mixture was then concentrated under vacuum until only a drop of viscous liquid remained. Dichloromethane was condensed onto the resulting liquid, yielding an oily deposit at the bottom of the reactor. The supernatant dichloromethane phase was filtered into a Teflon-FEP reactor and the extraction was repeated with an additional 5 mL of dichloromethane. The solid residue in the reaction vessel contained mostly ammonium dinitramide and some insoluble solid in CD 3 CN (the insoluble product, upon hydrolysis, displayed a 11 B NMR shift at ca 15 ppm, which is not the expected value for boric acid). The dichloromethane extract was concentrated in vacuo to about one half of the initial volume. Some white material started precipitating, and the suspension was cooled to -80 °C and left standing overnight. The supernatant was transferred into a J. Young NMR tube and contained relatively clean 1 along with the unidentified impurity. In one instance, nice colorless prismatic crystals suitable for X-ray diffraction were obtained from the evaporation of such dichloromethane extract. The white solid was dried in vacuo and recuperated in a glovebox. The sample still contained significant amounts of 1, but was severely contaminated with ammonium dinitramide. 259 NMR (1 only) (CD 3 CN) δ(ppm): 1 H 2.77 (bqt, 1 J( 1 H- 11 B)=111 Hz), 4.69 (bt 1 J( 1 H- 14 N)=50 Hz; 11 B -9.3 (t, 1 J( 11 B- 1 H)= 113 Hz); 14 N -34.3 (s, τ 1/2 =6 Hz, NO 2 ), -360.9(bm, NH 3 -B); (dichloromethane solution) (1 only) δ(ppm): 11 B -9.6 (t); 14 N -36.8 (s, NO 2 ), -358.7 (bm, NH 3 -B); (Et 2 O solution) (1 only) δ(ppm): 11 B - 8.9 (t); 14 N -34.1 (s, NO 2 ), -91 (broad, B-N-NO 2 ), -361.1 (bs, NH 3 -B); (Other signals) (CD 3 CN) δ(ppm): 1 H, 4.83 (overlapping t, 1 J( 1 H- 14 N) ~50 Hz), 6.14 (1:1:1 t, 1 J( 1 H- 14 N)=55 Hz, NH 4 + ); 11 B -14.8 (t, 1 J( 11 B- 1 H)= 119 Hz), -4.1 (overlapping multiplets), -3.5 (t, 1 J( 11 B- 1 H)= 113 Hz). 14 N -32.0 (bs), -29.6 (s, τ 1/2 = 7 Hz), -8.8 (s, τ 1/2 = 8Hz, N 3 O 4 - ). Raman (20 °C, 400 mW) ῦ/cm -1 : 3325 (0.9), 3254 (3.6), 3213 (0.2), 2522 (1.0), 2462 (7.3), 2379 (0.6), 1603 (0.5), 1589 (3.4), 1577 (0.5), 1564 (0.7), 1548 (0.7), 1526 (0.3), 1503 (0.4), 1455 (0.6), 1389 (0.2), 1346 (8.4), 1226 (3.0), 1194 (1.9), 1163 (1.6), 1152 (0.2), 1097 (0.9), 1053 (2.2), 918 (0.7), 876 (0.2), 856 (10.0), 826 (0.3), 792 (1.5), 767 (0.2), 544 (0.7), 475 (0.1), 463 (5.9), 411 (0.3), 342 (2.2), 319 (0.1), 311 (0.3), 288 (0.6), 107 (1.1). IR (AgCl pellet, 20°C) ῦ/cm -1 : 3323 (s), 3261 (s), 2827 (w), 2759 (w), 2511 (s), 2460 (s), 2379 (m), 2310 (m), 2087 (w), 1699 (w), 1602 (s), 1533 (s), 1390 (s), 1348 (m), 1312 (m), 1233 (s), 1188 (s), 1161 (s), 1046 (s), 1023 (sh), 956 (sh), 916 (m), 885 (s), 853 (m), 825 (m), 794 (m), 766 (s), 749 (m), 725 (s), 694 (m), 653 (w), 550 (sh), 536 (m), 475 (m). Detection of ammonia-bis(dinitramido)borane, NH 3 . BH[N(NO 2 ) 2 ] 2 (2). An in situ-generated HDN CD 3 CN solution (0.27 mmol) was transferred over ammonia-borane (2.1 mg; 0.068 mmol). After the vigorous effervescence subsided, the solution was analyzed by NMR spectroscopy and monitored over ca one week. The solution initially contained 1 along with side-products but contained significant amounts of new species, including 2, after ca two days. NMR (2 only) (CD 3 CN) δ(ppm): 1 H 3.78 (bqt, 1 J( 1 H- 11 B)=143 Hz), BH 2 ), 5.33 (bt 1 J( 1 H- 14 N)= 45 Hz); 11 B -0.7 (d, 1 J( 11 B- 1 H)=149 Hz); 14 N -32.7 (s, τ 1/2 = 9 Hz, NO 2 ); (Other signals) (CD 3 CN) δ(ppm): 1 H 2.80 (bqt, 1); 4.74 (bt, 1); 5.20 (bt, overlapping); 5.49 (bt, overlapping); 6.27 (1:1:1 t, NH 4 ); 11 B -14.9 (impurity often observed with 1), -9.4 (t, 1); -4.9 (impurity often observed with 1), -3.4 (impurity often observed with 1), -2.8 (impurity often observed with 1), -0.04, 0.8, 1.46 (d, 1 J( 11 B- 1 H)=148 Hz, major species), 20.3 (bs, likely B(OH) 3 ); 14 N -34.3 (1), -33.7 (bs), -29.4 (s, HDN/DN - ). After a week at room temperature, the solution contained mostly boric acid, N 2 O and ammonium dinitramide as determined by 14 N and 11 B NMR spectroscopy. 260 A5.2 Additional Remarks Synthesis. The use of excess ammonia borane lead to the formation of a larger number and larger quantities of unidentified species. The use of ammonia-borane purified by extraction in glyme allowed the isolation of higher purity materials. On the identity of the by-products. Ammonia borane itself decomposes in solution (and in the solid state) over time, yielding several unidentified species by 11 B NMR. 1 is unlikely to be more stable than ammonia-borane and the decomposition pathway at play in the latter could alone account for most the minor unidentified species encountered in this study. Several reasonable structures can be proposed to account for the most significant side-product found. A signal was often observed at -14.8 ppm by 11 B NMR spectroscopy, which is very close to that observed for NH 3 . BH 2 Cl. 10 Since HDN was prepared from HCl and KDN, this would be a reasonable assumption. However, this resonance reappeared upon isolation of 1 from a dichloromethane solution which did not display the resonance. A more satisfying assignment involves the auto-ionisation of 1 (Figure A5.1) would also allow for the formation of a boronium diammoniate cation, which also has a chemical shift very close to -15 ppm. 10 The most important side-products, seen as two resonances in variable ratios at ca -3 ppm in 11 B NMR and one resonance at -29.6 ppm in 14 N NMR, cannot be straightforwardly assigned to such reaction product. The 11 B signals are two overlapping triplets, and one of them does not seem to correlate with the unidentified resonance in 14 N NMR spectroscopy. Since the species are soluble in dichloromethane, this could likely exclude polymeric products. The multiplicity and chemical shift also exclude borazine analogues. The nitro-free species could perhaps be a symmetric oligomer of NH 3 . BH 3 . The remaining unknown species could then be a solvent-coordinated BH 2 species, such as glyme . BH 2 DN as a possible candidate. Such species could be formed through the dissociation of the complex anions of Figure A5.1 and Figure A5.2. Figure A5.1: Possible auto-ionisation of 1. As pointed out in the main article, the formation of ADN from 1 is not as readily explainable as its formation by hydrolysis. It is possible that 1 undergoes the elimination of ADN under the formation of more complex boron-hydride ion pairs (Figure A5.2). Such ion pair would presumably quickly 261 decompose, potentially yielding insoluble products, possibly accounting for the experimental observations. Figure A5.2: Possible formation of ammonium dinitramide from the decomposition of 1 A5.3 NMR Spectra Figure A5.3: 11 B{ 1 H} NMR spectrum of a DCM extract containing 1 as the major product. Signals at ca - 4 ppm are associated to the major unassigned by-products. 262 Figure A5.4: 14 N NMR spectrum of a dichloromethane extract containing 1 as the major species. The signal at ca -33 ppm is associated to one of the major unassigned by-products. Figure A5.5: 11 B NMR spectrum of isolated 1 in CD 3 CN, showing the triplet B-H coupling pattern. Signals at ca -4 ppm are associated to the major unassigned by-products, while the triplet at ca -15 is likely (NH 3 ) 2 BH 2 + . 263 Figure A5.6: 11 B NMR spectrum of a reaction mixture of NH 3 BH 3 + excess HDN after two days at room temperature, which shows significant amounts of a species assigned to 2. The doublets at ca +2 and ca -3 ppm belong to unassigned species. The very intense signal at ca +2 ppm only has the HDN/DN - signal as a possible match (with respect to intensities) in the 14 N NMR spectrum, suggesting it does not belong to a dinitramide-substituted species. Figure A5.7: 14 N NMR spectrum of a reaction mixture of NH 3 BH 3 + excess HDN after two days of reaction at room temperature, showing significant amounts of a species assigned to 2. The intense signal at ca -29 ppm moves towards lower field as the concentration of DN - increases. 264 A5.4 Computational Details Table A5.1: Calculated and experimental vibrational frequencies (in cm -1 ) for 1. Harmonic frequencies were calculated at the B2PLYP 11 /Def2-TZVPP 12 level of theory using the ORCA 3.1 code, 13 with implicit consideration of CH 3 CN solution, as treated by the COSMO method. Frequencies were scaled by 0.98 for best fit with experimental values observed for 1 its B-O isomer. Calculated Scaled calculated (IR) [RA] Experimental IR (int) 1 Experimental Raman [int.] 1 49 48 (11) [4] - - 77 75 (5) [1] - - 106 104 (8) [3] - 107 [1.1] 149 146 (16) [3] - Not observed 178 175 (10) [~0] - Not observed 260 255 (18) [~0] - Not observed 288 282 (8) [1] - 288 [0.6] 316 310 (12) [2] - 311 [0.3]; 319 [0.1]; 342 [2.2] 451 442 (4) [8] - 463 [5.9];411 [0.3] 508 498 (13) [4] 475 (m) 475 [0.1] 544 533 (26) [2] 550 (sh); 536 (m) 544 [0.7] 713 699 (1) [0] 694 (m); 653 (w) Not observed 741 726 (47) [2] 725 (s) Not observed 757 742 (18) [5] 749 (m) Not observed 771 756 (68) [4] 766 (s) 767 [0.2] 803 787 (62) [1] 794 (m) 792 [1.5] 857 840 (13) [4] 853 (m); 825 (m) 856 [10.0], 826 [0.3] 894 876 (39) [20] 885 (s) 876 [0.2] 930 911 (66) [3] 916 (m) 918 [0.7] 1063 1042 (18) [6] 1053 [2.2] 1104 1082 (47) [9] 1046 (s); 1023 (sh); 956 (sh) 1127 1104 (4) [7] 1097 [0.9] 1178 1155 (47) [4] 1161 (s) 1163 [1.6]; 1152 [0.2] 1231 1207 (19) [8] 1188 (s) 1194 [1.9] 1245 1220 (42) [38] 1233 (s) 1226 [3.0] 1366 1339 (72) [0] 1348 (m); 1312 (m) Not observed 1408 1380 (21) [56] 1390 (s) 1346 [8.4];1389 [0.2]; 1455 [0.6] 1560 1529 (9) [1] 1533 (s) 1503 [0.4]; 1526 [0.3]; 1548 [0.7]; 1565 [0.7]; 1577 [0.5] 1629 1596 (61) [4] 1602 (s); 1699 (w) 1589 [3.4]; 1603 [0.5] 1636 1603 (62) [3] 1662 1629 (56) [4] 2570 2519 (40) [326] 2460 (s) 2462 [7.3]; 2379 [0.6] 2645 2593 (20) [96] 2511 (s) 2522 [1.0] 2759 (w) Not observed 2827 (w) Not observed 3473 3403 (41) [172] 3261 (s) 3254 [3.6]; 3213 [0.2] 3562 3491 (99) [51] 3323 (s) 3325 [0.9] 3573 3501 (21) [46] 265 Table A5.2: Calculated vibrational frequencies (in cm -1 ) for the B-O isomer of 1. Calculated (IR) [RA] Scaled Calculated (IR) [RA] Scaled 72 (11) [1] 70 1042 (267) [26] 1021 132 (10) [0] 130 1043 (13) [5] 1022 181 (0) [2] 177 1100 (38) [16] 1078 294 (22) [2] 288 1184 (517) [9] 1161 318 (3) [1] 312 1235 (32) [9] 1210 375 (6) [4] 367 1275 (428) [5] 1249 431 (2) [5] 422 1326 (87) [25] 1300 579 (1) [2] 567 1428 (388) [27] 1399 648 (0) [1] 635 1546 (768) [20] 1515 674 (19) [2] 660 1621 (1001) [3] 1588 736 (30) [1] 721 1640 (57) [5] 1607 745 (55) [7] 730 1771 (75) [3] 1736 798 (21) [2] 782 2574 (147) [358] 2523 834 (402) [1] 818 2606 (306) [104] 2553 868 (16) [14] 851 3480 (47) [178] 3410 937 (18) [5] 918 3569 (109) [58] 3498 955 (119) [4] 936 3601 (108) [40] 3529 266 A5.5 Crystal Structure Report for NH 3 BH 2 N 3 O 4 (1) Figure A5.8: ORTEP plot of the complete asymmetric unit of 1. Hydrogen atoms were found in the Fourrier difference map. N-H hydrogen atoms positions were idealized, while the B-H distances were restrained. Selected structural parameters: Bond lengths (Å): N1-N2 1.3926(15), N1-N3 1.4046(15), O1- N2 1.2157(15), O2-N2 1.2214(16), O3-N3 1.2160(15), O4-N3 1.2197(15), N1-B1 1.5798(19), B1-N4 1.5895(18), B1-H1 1.100(14), B1-H2 1.080(14), N6-N5 1.4137(16), N7-N5 1.3992(15), O5-N6 1.2199(15), O6-N6 1.2091(15), O7-N7 1.2135(15), O8-N7 1.2261(15), N5-B2 1.5825(18), N8-B2 1.5809(18), B2-H6 1.097(13), B2-H7 1.106(13); Bond angles (°): O1-N2-O2 125.20(12), O3-N3-O4 126.04(12), O4-N3-N1 113.26(11), O3-N3-N1 120.61(11), O1-N2-N1 115.57(11), O2-N2-N1 119.16(11), N2-N1-N3 115.64(11), N2-N1-B1 125.69(11), N3-N1-B1 118.47(11), N1-B1-N4 107.23(11), H1-B1-H2 115.7(13), H1-B1-N4 107.8(9), H1-B1-N1 107.7(9), H2-B1-N4 110.7(9), H2-B1-N1 107.4(9), O1-N2-N1-B1 -9.4(2), O4-N3-N1-B1 -33.4(2), O5-N6-O6 125.79(12), O7-N7-O8 126.03(11), O5-N6-N5 114.52(11), O6-N6-N5 119.65(11), O7-N7-N5 120.63(11), O8-N7-N5 113.22(10), N6-N5-N7 115.89(10), N6-N5-B2 123.38(10), N7-N5-B2 120.61(11), N5-B2-N8 109.01(11), H6-B2-H7 115.5(13), H6-B2-N8 108.6(9), H6-B2-N5 105.1(9), H7-B2-N8 109.6(9), H7-B2-N5 108.8(9), O5-N6-N5-B2 - 30.4(2), O8-N7-N5-B2 -12.2(2). 267 Figure A5.9: Projection of the packing of 1 perpendicular to the 001 plane. Figure A5.10: Projection of the packing of 1 perpendicular to the 010 plane. 268 Table A5.3: Sample and crystal data for 1. Identification code XGC4_202_02 Chemical formula BH 5 N 4 O 4 Formula weight 135.89 g/mol Temperature 100(2) K Wavelength 1.54178 Å Crystal size 0.116 x 0.150 x 0.269 mm Crystal habit colorless blade Crystal system monoclinic Space group P 1 21/n 1 Unit cell dimensions a = 12.5970(2) Å α = 90° b = 6.54390(10) Å β = 91.7660(10)° c = 12.6505(2) Å γ = 90° Volume 1042.33(3) Å 3 Z 8 Density (calculated) 1.732 g/cm 3 Absorption coefficient 1.499 mm -1 F(000) 560 269 Table A5.4: Data collection and structure refinement for 1. Diffractometer Bruker APEX DUO Radiation source IuS microsource, CuKα Theta range for data collection 4.88 to 73.03° Index ranges -15<=h<=15, -8<=k<=8, -15<=l<=15 Reflections collected 12001 Independent reflections 2063 [R(int) = 0.0275] Absorption correction multi-scan Max. and min. transmission 0.8450 and 0.6890 Structure solution technique direct methods Structure solution program SHELXTL XT 2014/4 (Bruker AXS, 2014) Refinement method Full-matrix least-squares on F 2 Refinement program SHELXL-2014/7 (Sheldrick, 2014) Function minimized Σ w(F o 2 - F c 2 ) 2 Data / restraints / parameters 2063 / 10 / 177 Goodness-of-fit on F 2 1.057 Final R indices 1894 data; I>2σ(I) R1 = 0.0324, wR2 = 0.0883 all data R1 = 0.0354, wR2 = 0.0911 Weighting scheme w=1/[σ 2 (F o 2 )+(0.0552P) 2 +0.4286P] where P=(F o 2 +2F c 2 )/3 Largest diff. peak and hole 0.261 and -0.389 eÅ -3 R.M.S. deviation from mean 0.053 eÅ -3 270 A5.6 References (1) SAINT+ V8.34A, Bruker AXS Madison, WI. (2) SADABS V2014/4, Bruker AXS Madison, WI. (3) SHELXTL V2014/6, Bruker AXS Madison, WI. (4) Sheldrick, G. Acta Crystallogr. Sect. A: Found. Crystallogr. 2008, 64, 112. (5) Sheldrick, G. Acta Crystallogr. Sect. A: Found. Crystallogr. 2015, 71, 3. (6) Sheldrick, G. Acta Crystallogr. Sect. C: Cryst. Struct. Commun. 2015, 71, 3. (7) Farrugia, L. J. Appl. Crystallogr. 2012, 45, 849. (8) CCSD 2014. (9) Luk'yanov, O. A.; Anikin, O. V.; Gorelik, V. P.; Tartakovsky, V. A. Russ. Chem. Bull. 1994, 43, 1457. (10) Lingam, H. K.; Chen, X.; Zhao, J.-C.; Shore, S. G. Chem. - Eur. J. 2012, 18, 3490. (11) Grimme, S. J. Chem. Phys. 2006, 124, 034108. (12) Weigend, F.; Ahlrichs, R. PCCP 2005, 7, 3297. (13) Neese, F. Wiley Interdisciplinary Reviews: Computational Molecular Science 2012, 2, 73. 271 APPENDIX 6: FURTHER INFORMATION ON THE EXISTENCE OF DINITRAMIDOBORATES (CHAPTER 7) A6.1 Experimental Details Unless otherwise specified, all the reactions were performed using standard Schlenk and high-vacuum line techniques.NMR spectra were recorded on a Bruker AMX-500 spectrometer. 11 B NMR and 14 N NMR spectra were referenced externally to BF 3 .OEt 2 in CDCl 3 and to neat MeNO 2 , respectively. NaDN was prepared according to literature methods. 1 KDN was purified by extraction in glyme to remove the insoluble nitrate and drying under vacuum. Some of the KDN was also kindly donated by EURENCO Bofors and purified by recrystallization from propanol solutions. It was later found that such materials contained significant amounts of moisture, as observed by a sizeable doublet at 27 ppm in the 11 B NMR spectra in HDN+BH 4 - reactions. NaBH 4 and TMABH 4 were used as supplied. Anhydrous HCl (Matheson) was used as supplied. Glyme was distilled over Na/benzophenone ketyl and stored over 3Å molecular sieves. Deuterated acetonitrile (Cambridge isotopes) was dried over 3Å molecular sieves and degassed by freeze-pump-thaw cycles. NaBH 4 + HDN. In a typical experiment, NaBH 4 (max. 0.1 mmol) was placed in a flame-sealable NMR tube in a dry-nitrogen glovebox. Glyme was condensed onto the borohydride at -196°C. In parallel, the appropriate amount of potassium or sodium dinitramide was placed into a glass reactor in a glovebox. Glyme was condensed onto the dinitramide salt and a stoichiometric amount of HCl, measured by PV=nRT in a calibrated volume of a vacuum line was condensed onto the frozen mixture. The mixture was thawed and stirred for an hour, after which time it was left to decant. The supernatant HDN solution in glyme was then transferred into the NaBH 4 /glyme frozen mixture by cannulation. The NMR tube containing HDN and NaBH 4 was evacuated at -196 °C and the tube was flame-sealed. The mixture was then thawed at ca -40 °C and analyzed by VT NMR between -40 and 20 °C. The process was repeated for <1:1, 2:1, 3:1 and 4:1 HDN: NaBH 4 ratios. TMABH 4 + HDN. A solution of HDN (1.2 mmol) was prepared as described previously from KDN and HCl. It was cannulated onto a frozen TMABH 4 (94 mg; 1.1 mmol) solution in glyme in a glass reactor. The mixture was thawed and slowly warmed to room temperature and gas evolution was observed. 1.3 mmol of non-condensable gases where measured, as expected. Once the gas evolution subsided, a significant amount of insoluble white material was observed (usually due to impurities in the borohydride 272 starting material). A small NMR sample was cannulated from the reaction mixture. The NMR analysis revealed that [BH 3 DN] - (1) formed as the major species, with significant amounts of [BH 2 (DN) 2 ] - (2), BH 3 . glyme and other unidentified species. The reaction mixture was then transferred into a Teflon-FEP reactor and the solvent evaporated in vacuo, yielding a yellowish white flaky solid. The NMR analysis of the compound revealed that it did not contain any 1 but instead several unidentified species, 2 and dinitramide as the major species. TMABH 4 + excess HDN. HDN (1.3 mmol) in glyme prepared in situ as described previously from KDN and HCl was cannulated onto a TMABH 4 (23 mg; 0.26 mmol) suspension in glyme frozen at -196 °C. The reactor was evacuated at -196 °C and the mixture was thawed at -64 °C. The reactor was slowly warmed up. When it reached -15 °C, the mixture was frozen at -196 C° and non-condensable gases where measured by PV=nRT on a calibrated glass vacuum line (0.5 mmol of H 2 ), indicating the formation of [BH 2 (DN) 2 ] - had reached completion. A part of the reaction mixture was cannulated into a glass flame- sealable NMR tube while the solution was being kept at ca -20 °C. The solution was analyzed by NMR and was shown to contain mostly [BH 2 (DN) 2 ] - along with a few unidentified species. The NMR sample was then exposed to the same conditions as the main reactor and the reaction was thus followed between - 20 and +60 °C. At room temperature, species tentatively assigned to BH(DN) 2 and [BH(DN) 3 ] - (3 and 4) started to form along with several more unidentified species. After several days, boron oxides started to appear as well, indicating decomposition. The reaction mixture was then heated between 50-60 °C for 1-2 days. The analysis of the mixture then showed that almost all of the 3 and 4 species had been consumed, leaving only boron oxides and a new species with a 11 B NMR signal at 22.4 ppm. When the isolatation of the final mixture of products was attempted, the Teflon-FEP reactor in which the solution had been transferred exploded. This is known 2 to occur during HDN isolation attempts, suggesting that the excess HDN remaining exploded upon concentration. PPh 4 DN + 2 HDN. A cold solution of HDN (0.3 mmol) in glyme prepared from KDN as described previously was pipetted onto PPh 4 BH 4 (60 mg; 0.17 mmol) in a nitrogen-filled J. Young tube. After the vigorous effervescence subsided, copious amounts of solid were found in the reaction mixture. The analysis of the supernatant by NMR indicated that PPh 4 [BH 2 (DN) 2 ] formed predominantly, along with several unidentified species at 2.7 (t), 6 and -5.6 ppm in the 11 B NMR spectrum and at -27.8, -33.8 and - 32.7 ppm in the 14 N NMR spectrum. The precipitated solid was dissolved in acetone and proved to be PPh 4 DN. When the supernatant was dried in vacuo overnight, the resulting white solid became increasingly yellow. Most of the isolated solid did not dissolve in deuterated acetonitrile and the analysis of that solution indicated that very small amounts of PPh 4 [BH 2 (DN) 2 ] remained in the isolated materials, 273 along with unidentified species and significant amounts of nitrate, dinitramide and N 2 O, indicating that some of the solid decomposed upon dissolving in acetonitrile. A6.2 Additional Remarks Reaction of dinitramides with boron trihalides. The outcome of the reaction between boron trihalides and dinitramide salts is summarized in Figure A6.1. When condensing BCl 3 (in various proportions) on top of a frozen mixture of sodium dinitramide and glyme, a vigorous gas evolution was observed as the mixture thawed. Nitrous oxide (N 2 O) and significant amounts of NOCl were detected in the volatile products by IR spectroscopy, while sodium nitrate was detected as a major reaction product by Raman spectroscopy. Boron oxides were detected by 11 B NMR spectroscopy. Since the complex formed was apparently thermally unstable, the observation of the formation of a transient dinitramido-borate species by NMR spectroscopy was attempted. BF 3 was chosen as a starting material, since it was assumed that the Lewis acidity of BCl 3 was likely facilitating the decomposition of the dinitramide moiety. When frozen mixtures containing ca 1:1 ratio of BF 3 and sodium or potassium dinitramide in glyme were thawed in an NMR spectrometer at -30 °C, the formation of N 2 O as a major product was observed. However, the formation of a transient species at ca -28 ppm in 14 N NMR was also observed. This species disappeared quickly at -10 °C, leaving behind a new species with a corresponding 14 N NMR resonance at -31 ppm. Such signal was identical to the one observed when BF 3 was reacted with NaNO 3 in similar conditions, indicating that the transient species consists of small amounts of undecomposed dinitramidoborate, which eliminates N 2 O to form the parent nitratoborate. The boron trihalide/nitrate reaction system has been studied before 3 and the relatively rapid exchange of the halide and nitrate ligands was noticed. It is most likely what happened in this case, both for the dinitramidoborate and the nitratoborate, since BF 4 - was detected as a major end-product by 11 B and 19 F NMR spectroscopy. It is however likely that the transient dinitramidoborate observed by 14 N NMR is [BF 3 N(NO 2 ) 2 ] - , and that the putative B[N(NO 2 ) 2 ] 4 - resulting from ligand scrambling decomposes too quickly to be observed. 274 Figure A6.1: Attempts at isolating and observing dinitramidoborates through halide elimination from boron trihalides. HDN/DN - equilibrium. As more than one equivalent of HDN was reacted with 1, the NMR resonances became increasingly difficult to assign. Indeed, mixtures containing HDN and DN - tend to have variable and unpredictable shifts (between -8 and -40 ppm) depending on the ratios of the two species. The 11 B spectrum also became increasingly complex, with at least two major species for each stoichiometry studied. Through the correlation of several experiments run in various conditions, some confidence in the assignments described herein was gained. Reaction of HDN with borane. The study of the reaction of HDN with BH 3 complexes might help confirming the assignment of several additional species as the neutral BH n (DN) 3-n , although preliminary studies with strong borane complexes, such as BH 3 . SMe 2 , did not display any reactivity at all. Last steps of the reaction of excess HDN with TMABH 4 . Upon heating the mixture containing the putative 3 and 4 at 60 °C for a few days, 11 B signals at -4.6, 19.5 and 22.6 were observed, the latter two being strongly indicative of tricoordinate boron oxides (Figure A6.2). A single resonance at -31 ppm in 14 N NMR was observed, however, suggesting that no decomposition of the dinitramide moiety had yet occurred. It was thus concluded that B(N 3 O 4 ) 3 (5) might have formed, although the experimental conditions did not allow us to confirm such an assignment. Attempts to isolate this last species lead to an explosion which destroyed the Teflon-FEP reactor in which the solution was concentrated under vacuum. It is not clear if it was the excess HDN, the HDN sensitized by the boron product in solution or the unknown boron-dinitramide species which triggered the explosion. This compound was not pursued any further. 275 Figure A6.2: 11 B NMR TMABH 4 + excess HDN in glyme heated for two days at 50 °C, showing an unusual signal at ca 23 ppm which could correspond to 5. A6.3 References (1) Christe, K. O.; Wilson, W. W.; Petrie, M. A.; Michels, H. H.; Bottaro, J. C.; Gilardi, R. Inorg. Chem. 1996, 35, 5068. (2) Luk'yanov, O. A.; Anikin, O. V.; Gorelik, V. P.; Tartakovsky, V. A. Russ. Chem. Bull. 1994, 43, 1457. (3) Titova, K. V. Russ. J. Inorg. Chem. 2002, 47, 1121. 276 APPENDIX 7: ADDITIONAL INFORMATION ON SYNTHESIS OF NITRYL CYANIDE, NCNO 2 (CHAPTER 8) A7.1 Experimental Details Caution! Many of the materials described in this work are toxic, as well as energetic and likely to be explosive, especially in the condensed phase. They should only be handled on a small scale while taking appropriate safety measures, such as wearing ear plugs, face shield, leather gloves and protective clothing, and working in a well-ventilated environment. Also, care should be exercised when using nitromethane as a solvent, particularly in a closed system, because it can result in violent explosions when ignited by a suitable initiator. Instruments. NMR measurements were performed on a Bruker AMX 500 and on a Varian NMRS-600 spectrometers. Spectra were externally referenced to neat CH 3 NO 2 for 14 N, to TMS for 1 H and 13 C and to BF 3 . Et 2 O/CDCl 3 for 11 B NMR. Raman spectra were recorded in J-Young NMR tubes on a Bruker Equinox 55 FT-RA spectrophotometer using a Nd:YAG laser at 1064 nm. IR spectra were recorded on a Midac, M series, FT-IR spectrometer, using an AgCl-windowed cell. All manipulations were performed on Pyrex or passivated stainless steel high-vacuum lines. Fractional condensations were performed under a dynamic vacuum. Materials. CFCl 3 (≥98%, Aldrich) and CD 3 CN (99.8%, Cambridge Isotope Labs) were degassed and stored over 3 Å molecular sieves. CH 3 CN (anhydrous, ≥99.8%, Sigma-Aldrich)) was distilled over P 2 O 5 (99.6%, J. T. Baker), and stored over 3 Å molecular sieves. 2-H-heptafluoropropane (98%, Synquest Labs) was distilled over P 2 O 5 and stored in a passivated steel cylinder. TMSCN (97%, Matrix scientific), was degassed and stored in an ampule equipped with a Teflon valve. Nitromethane was pre-dried with MgSO 4 , then dried over 3 Å molecular sieves. Dry solvents were stored over 3 Å molecular sieves in flasks sealed with Teflon valves. TMSCN was purified prior to use by fractional condensation through - 26 °C and -196°C traps. t-BuMe 2 SiCN (Aldrich) was purified by pumping through a -20 °C trap in order to remove the more volatile Me 3 SiCN impurity. NO 2 BF 4 (≥95.5%, Aldrich) and NO 2 SbF 6 (Sigma- Aldrich) were stored in an inert atmosphere and both used as received, as well as exposed to a dynamic vacuum overnight to remove any volatile impurities. Stearic acid (≥95%, Sigma-Aldrich) was dried by heating at 60 °C under vacuum for 2 hours prior to use. All solid reagents were dried under vacuum prior to use. 277 FNO 2 was prepared by condensing F 2 (89 mmol) and N 2 O 4 (76 mmol) into a steel cylinder at -196 °C and then letting the mixture slowly warm up to room temperature over 12 h. The cylinder was then cooled down to -196 °C and the excess F 2 was removed under vacuum and passed through a NaCl/soda lime column. The crude product contains variable amounts of FNO, which were removed by adding an amount of ClF 3 corresponding to the estimated amount of FNO and passing the cooled gases through a -142 °C trap. The pure FNO 2 was collected in a -196 °C trap. IR: ṽ = 3567 (w), 3080 (m), 2614 (w), 1875 (m), 1793 (vs), 1644 (w), 1484 (w), 1394 (m), 1308 (vs), 1140 (m), 823 (vs), 741 (s), 571 (s) cm -1 . 14 N NMR (2H-heptafluoropropane, unlocked, -60 °C) δ/ppm: -92.5 (d, 1 J( 14 N- 19 F) = 117 Hz). The synthesis of HCN was inspired by a previously reported synthesis of HN 3 . 1 Finely crushed KCN (422 mg, 10.8 mmol) was added to stearic acid (15.3 g, 53.8 mmol). The reaction vessel was heated to 80 °C for 3 hours while the formed HCN was transferred under a dynamic vacuum and collected in a trap cooled by liquid nitrogen. Subsequent purification by fractional condensation (traps at -20, -90 and -196 °C) resulted in pure HCN (by IR) in the -90 °C trap. IR: ṽ = 3310 (s), 1411 (w), 712 (vs) cm -1 . 14 N NMR (Freon-11, unlocked,-20 °C) δ/ppm: -118.5 (s, broad). AgCN was prepared according to a literature procedure. 2 ClNO 2 was prepared mostly following literature procedures. 3 Notable exceptions are the use of a U-trap connected to a halocarbon oil bubbler instead of a condenser/collection vessel. The material was strongly colored (orange-brown) and contained NOCl, NO 2 and likely Cl 2 impurities. While the NOCl and NO 2 content was low by IR spectroscopy, vapor pressure measurements suggest a Cl 2 content between 10 and 15 mol%. The purity was sufficient for the screening experiments that were performed. The material was freshly refractionated prior to use. Me 3 SnCN was prepared according to literature methods. 4 K 3 PO 4 was dried at 110 °C overnight while under dynamic vacuum. TMAF was prepared according to a literature method. 5 Synthesis of NCNO 2 (1). The importance of purifying the commercially available t-BuMe 2 SiCN (Aldrich) must be stressed, since it typically contains a few percent of Me 3 SiCN, which upon reaction with NO 2 BF 4 forms Me 3 SiF, which is practically impossible to separate from NCNO 2 by any of the numerous chemical and physical purification methods explored throughout this work. Fractional condensation of t-BuMe 2 SiCN through -20 °C and -196 °C traps, collecting the contents of the -20 °C trap, has proven effective. Similarly, since commercially available NO 2 BF 4 may contain hydrolysis products, depending on the handling by the manufacturer, it is advisable to finely grind the reagent and to subject it to a dynamic vacuum overnight to remove any adventitious acidic hydrolysis products, which will inevitably lead to the formation of HCN when exposed to most cyanide sources, thus creating further purification challenges. 278 NO 2 BF 4 (404 mg; 3.0 mmol) was placed in a glass vessel equipped with a Teflon-FEP valve. Nitromethane (2-5 mL) was condensed at -196 °C into the reactor. The mixture was thawed and stirred, then frozen at -196 °C. t-BuMe 2 SiCN (217 mg;1.5 mmol) was then condensed above the frozen mixture at -196 °C. The frozen mixture was thawed at ca -30 °C. Once the thawing was completed, the cold liquid was swirled to dissolve small portions of the t-BuMe 2 SiCN frozen on the reactor's walls. This was repeated until all of the t-BuMe 2 SiCN had reacted. This ensures that the silicon compound is always in the presence of an excess of oxidizer and typically prevents the appearance of a green-blue color, usually indicative of significant decomposition. Effervescence followed each silyl cyanide addition to the suspension. The complete addition of the cyanide starting material was carried out within ca 5 minutes after thawing the frozen mixture, after which time the yellow mixture was frozen at -196 °C and fractionally condensed through -80 and -196 °C traps. The contents of the -196 °C trap were then fractionated three times through -96 -112 and -196 °C traps, NCNO 2 accumulating mostly in the -112 °C trap (0.19 mmol, ca 30 mol%, total yield; 12 mol% isolated yield based on t-BuMe 2 SiCN). The yields were estimated by measuring the pressure of the different gaseous fractions in calibrated volumes on the glass vacuum line. The purity of the isolated materials was estimated by a combination of IR and NMR measurements. The only major impurity found was a few mol% of N 2 O 4 . NCNO 2 : 14 N NMR (500 MHz, SO 2 , unlocked, -33 °C) /ppm: -62.6 (s, sharp, NCNO 2 ), -175 (s, broad, NCNO 2 ). 13 C NMR (600MHz, SO 2 , unlocked,-30 °C) /ppm: 106.6 (1:1:1 t, 1 J( 13 C- 14 N)= 25 Hz, NCNO 2 ); IR (11 torrs): ṽ = 2869 (vw), 2593 (vw), 2463 (vw), 2239 (w), 2186 (vw), 1580 (s), 1300 (m), 890 (m), 802 (vw) 720 (w), 576 (vw) cm -1 . Raman (-90 °C): ῦ/cm -1 = 2242 (10), 1577 (broad, 1.2), 1304 (4.9), 896 (4.1), 721 (1.4), 583 (1.4), 272 (broad, 1.0), 216 (4.7) cm -1 . In another synthesis, tBuMe 2 SiCN (425 mg; 3.0 mmol) was reacted with NO 2 BF 4 (835 mg; 6.3 mmol) in nitromethane (5 mL) as described above. The crude reaction mixture was fractionated through - 80, -96, -126 and -196 °C traps and the contents of the -126 trap were refractionated twice through -96, - 112 and -196 °C traps, then once through -112 and -196 traps. This afforded ca 1.4 mmol of NCNO 2 (purity >95mol% by IR spectroscopy; 45 mol% isolated yield based on t-BuMe 2 SiCN), demonstrating how sensitive the yield of NCNO 2 is to conditions, both during the synthesis and separation steps. A7.1.1 Measurement of the Physical Properties of NCNO 2 (1) Temperatures of cold baths were measured with a thermocouple digital thermometer precise to 0.1 °C. The accuracy of the digital thermometer was found to be well within 1 °C when slush baths of well- known temperatures were measured. Pressures were measured with a Heise Bourdon pressure gauge with 1 torr graduations. The NCNO 2 used for the measurements was prepared as described above. 279 Vapor pressure curve. ca 1.5 mmol of NCNO 2 was condensed at the bottom of a glass U-trap on a high vacuum line. The bottom of the trap was immersed in slush baths of different temperatures. Pressures were measured with the gas phase at 24 °C. The different temperatures were approached from above and below, and for each temperature the pressure was the same within the precision of the pressure gauge. Vapor pressures were measured in the range from -113 to -24 °C. The values obtained are listed in Table A7.1. Table A7.1: Vapor pressures of NCNO 2 measured at various temperatures Temperature (°C) Pressure (torr) -112.6 0 -95.6 0.5-1 -90.3 1.5 -89.6 1 -63.8 13 -63.5 12 -57.0 24 -56.3 23 -53.0 29 -31.4 117.5 -24.8 166 When the logarithm of the vapor pressure is plotted against the inverse temperature (in Kelvin) (for temperatures between -95.6 and -24.8 °C in Table A7.1), a straight line is obtained, which can be described by the equation y =-1459.1x + 8.0908 by linear regression (R 2 = 0.989). From that equation, a boiling point (P = 760 torrs) of 7 °C was extrapolated for NCNO 2 . For the latent heat of vaporization determination, points between -63.8 and -24.8 °C (Table A7.1) were selected since temperatures below - 80 °C were too close to the melting point of NCNO 2 . When the natural logarithm of the pressure (in Pa) was plotted against the inverse temperature (in Kelvin), a straight line described by the equation - 3439.4x+23.881 (R 2 = 0.9967) was obtained by linear regression. Following the Clausius-Clapeyron relation, the slope of such curve corresponds to ΔH evap /R (where R is the gas constant 1.987x10 -3 kcal K - 1 mol -1 ), from which a value of ΔH vap = 6.8 ± 0.2 kcal/mol was obtained. Gas density. The reported value is an average of two measurements. A tared evacuated glass vessel of known volume (148.0 mL) equipped with a Teflon valve was filled with NCNO 2 gas at a measured pressure at 23.4 °C. The vessel containing the gas was weighed and from the mass of the gas, assuming an ideal gas, from PV = nRT, a molecular weight of 70 ± 2 g/mol (calcd MW = 72) was obtained. 280 Density of the liquid. ca 1.5 mmol of NCNO 2 were condensed in a tared graduated vessel equipped with a Teflon valve. The volume (0.079 mL) was measured at -79.0 °C, at which temperature the amount of NCNO 2 in the gas phase was negligible. The volumes in the graduated vessel had been calibrated beforehand by accurately weighing volumes of water in it. The vessel was then warmed to ambient temperature to weigh the NCNO 2 (98.4 mg). The resulting density was 1.24 ± 0.08 g/mL at -79.0 °C. Melting point. ca 1.5 mmol of NCNO 2 was condensed into a narrow Pyrex vessel. The compound was frozen at -196 °C and placed in cold baths of different temperatures. By observing melting (or absence of melting) of NCNO 2 at increasingly close temperature intervals, the melting point of NCNO 2 was narrowed down to a range between -84.5 and -85.9 °C. A7.1.2 Computational Details Energies and structure: Geometry optimization and frequency analysis of 1, 2 and 3 in the gas phase were performed using the second order perturbation corrected “double hybrid” density functional B2PLYP of Grimme 6 together with the resolution of identity (RI) approximation 7 , and Ahlrich’s Def2-TZVPP 8 with its corresponding auxiliary basis set, using ORCA 3.0. 9 B2PLYP provide highly accurate energetics, for example, the mean absolute deviation for the G3/05 test set is 2.5 kcal/mol with a polarized QZV basis. 10 Species 1-7 were also calculated using the hybrid meta exchange-correlation functional M06-2X and the cc-pVTZ basis set, using Gaussian 09, rev A02. 11 M06-2X is a reliable general-purpose density functional theory (DFT) functional for main-group chemistry, with a mean absolute deviation of 2.2 kcal/mol, as demonstrated by several benchmarks. 12-15 The CBS-QB3 16,17 composite method was employed to treat various smaller structures in the gas-phase. CBS-QB3 is based on CCSD(T) energies extrapolated to the basis set limit using MP2 and MP4 calculations together with empirical corrections, and is expected to be highly reliable for thermochemistry. 16-19 Its mean absolute deviation in the G2 test set is reported to be 0.87 kcal/mol. 16 For reactions between NCNO 2 and the NO 2 radical, the CBS-QB3 composite method was modified to use geometries and thermal corrections from M06-2X/cc-pVTZ calculations, which should provide more reliable structures and frequencies than the B3LYP/6-311(2d,2p) default. Relative energies in the gas phase were calculated for 298 K and 1 atm. Spin-unrestricted and geometry relaxed scans of the NC … NO 2 potential energy surface was performed by broken symmetry DFT calculations at the M06- 2X/6-31+G(d) level of theory. The standard polarizable continuum model (PCM) of Gaussian 09 was employed for implicit treatment of the surrounding acetonitrile solution. Heats of formation. The heats of formation of 1 in the condensed phase, ΔH 0 f(l) , could be estimated after first calculating ΔH 0 f(gas) in the gas phase (50.7 kcal/mol) as an average of three reactions, 281 NCNO 2 (1) → CN (+104.0) + NO 2 (+7.9) (A7.1) (ΔH 0 r,CBS-QB3 = +62.0 kcal/mol, ΔH 0 f,1 = +49.9 kcal/mol) NCNO 2 (1) + C 2 H 6 (-20.0) → CH 3 CN (+17.7) + CH 3 NO 2 (-19.4) (A7.2) (ΔH 0 r,CBS-QB3 = -32.4 kcal/mol , ΔH 0 f,1 = +50.8 kcal/mol ) NCNO 2 (1) → CO 2 (-94.1) + N 2 (0.0) (A7.3) (ΔH 0 r,CBS-QB3 = -145.6 kcal/mol, ΔH 0 f,1 = +51.5 kcal/mol) where ΔH 0 f for the individual compounds are shown in kcal/mol within parentheses. The variation of the electrostatic surface potential (ESP) has been shown to correlate with a range of macroscopic properties relating to intermolecular interactions. 20,21 Heats of vaporization and sublimation, ΔH vap and ΔH sub , were here estimated from the electrostatic surface potential (ESP) calculated at the B3LYP/6-31G(d) level, using a parameterized relationship 22 and the HS95-v9 program. 23 The heat of vaporization (ΔH 1,vap ≈ 6.4 kcal/mol) and sublimation (ΔH 1,sub ≈ 7.0 kcal/mol) were estimated. Based on the low value and comparisons made elsewhere, 24 the compound can be expected to behave similarly to ClCN, and exhibit a significant vapor pressure at ambient conditions. This was later confirmed by experiments. The predicted heat of formation for 1 in the liquid phase is obtained by subtracting ΔH 1,vap from ΔH 0 f(1,gas) , i.e. ΔH 0 f(1,l) = 50.7 – 6.4 = 44.3±5 kcal/mol. Using the identical approach used for 1, values for 3 were estimated: ΔH 0 f(3,gas) = 62.6 ± 2 kcal/mol, ΔH 3,vap ≈ 13.7 kcal/mol, ΔH 3,sub ≈ 18.0 kcal/mol and ΔH 0 f(3,s) = 44.6 ± 5 kcal/mol. Vibrational spectroscopy and NMR estimates. Nuclear magnetic shielding tensors for 1 were calculated in acetonitrile at the GIAO-PCM-B3LYP/6-311++ (3df,3pd) level, and referenced to those of nitromethane. Such calculations have in previous work shown good agreement with experiment for nitro- groups. 24 In order to achieve higher theoretical accuracy, an empirical fit to experiment, described below, was constructed. Frequency analyses were performed at the M06-2X/cc-pVTZ (analytic), B3LYP/aug-cc-pVTZ (analytic), RMP2(fc)/aug-cc-pVTZ (analytic) using Gaussian 09, and RI-B2PLYP/Def2-TZVPP, RI-SCS- MP2/Def2-TZVPP (numerical) using ORCA 2.9. RIJCOSX-mPW2PLYP/Def2-TZVPP calculations were performed with ORCA 3.0, together with the auxiliary def2-TZVPP/J and def2-TZVPP/C basis sets for Coulomb and semi-numeric exchange integration. All numerical gradients were calculated with 0.001 bohr increments, DFT grid 5 and tightly converged SCF wave functions and geometries. Unscaled 282 vibrational spectra at all levels of theory are shown in Table A7.2 and Table A7.3. Raman intensities were obtained at the B3LYP/aug-cc-pVTZ and MP2(fc)/aug-cc-pVTZ levels. Excited states and UV-VIS spectroscopy. The lowest eight singlet and triplet UV/VIS transitions of 1 were estimated within the frameworks of equation-of-motion EOM-CCSD 25,26 and time-dependent DFT, at the EOM-CCSD/Def2-TZVPP and TD-ωB97X-D/aug-cc-pVTZ levels of theory, 27 respectively. Geometry optimizations of excited states were done at the TD-ωB97X-D/aug-cc-pVTZ level. Rocket propellant and explosive performance. The specific impulse (Isp, s) of a propellant mixture describes performance per mass of propellant, and is the most common measure or rocket propellant performance. I sp(vac) calculations were performed using RPA v.1.2.8.0 assuming a chamber pressure of 7 MPa, a nozzle expansion ratio of 70, and expansion to vacuum. The RPA code is based on known equations of states for thousands of known compounds and the I sp for 1 could be performed after adding the calculated condensed phase heat of formation (43.9 kcal/mol) to the program’s database. The ideal case of exhaust gases in thermal equilibrium was considered here. The actual performance of a real rocket motor powered by nitryl cyanide will be slightly lower than what is presented. The detonation pressure and velocity of 3 was estimated using the thermodynamic code Cheetah 7.0 28 assuming a solid-state density of 1.99 g/cm -3 , as discussed in the main article. A7.2 Further Computational Results and Vibrational Analysis Figure A7.1: Selected tetrameric cyclo-oligomerization products of 1. Calculated geometries (Å) and relative enthalpies and Gibbs free energies are calculated at the M06-2X/cc-pVTZ level (kcal/mol, 1 atm, 298 K). Assignment of the combination bands and overtones; indirect measurement of the δ rock NO 2 mode. Table A7.4 shows all of the IR absorption bands observed for NCNO 2 in the gas phase, including 283 the very weak ones, which were assigned to the appropriate overtones or combination bands and the Raman bands observed for the liquid. The very weak IR band observed at 802 cm -1 matches best with a combination of the δ rock NO 2 mode (predicted to be 600 cm -1 at the B2PLYP/Def2-TVZPP level and having zero IR and Ra intensities, Table A7.2) and of the Raman-observed δ in-plane NCN mode (216 cm -1 ). Subtracting the experimental value of 216 cm -1 from the observed value of 802 cm -1 , an estimated value of ca 590 cm -1 was obtained for the δ rock NO 2 mode (Table A7.4). Table A7.2: Unscaled vibrational frequencies (cm -1 ) for C 2v nitryl cyanide (1) at four levels of theory, together with experimentally observed values. a Vibrational mode and symmetry: B3LYP/ aug-cc-pVTZ MP2(fc)/ aug-cc-pVTZ B2PLYP/ Def2-TVZPP SCS-MP2/ Def2-TVZPP mPW2PLYP/ Def2-TVZPP IR [b] (exp) Raman [c] (exp) δ ip NCN (B 2 ) 215 ( 2:4) 202 ( 8:5) 211 (2) 204 (1) 233 (4) x 216(4.7) δ oop NCN (B 1 ) δ scissor NO 2 (A 1 ) δ rock NO 2 (B 2 ) δ wag NO 2 (B 1 ) ν C-N (A 1 ) ν sym NO 2 (A 1 ) ν asym NO 2 (B 2 ) ν C≡N (A 1 ) 275 ( 5:1) 571 ( 0:3) 627 ( 0:0) 763 ( 2:1) 907 ( 15:7) 1343( 40:27) 1614(100:10) 2336( 5:100) 275 ( 24:0) 575 ( 7:6) 596 ( 0:0) 734 ( 4:1) 901 ( 44:16) 1320( 92:20) 1787(100:100) 2156( 28:64) 281 (5) 577 (1) 607 (0) 747 (2) 901 (10) 1315(33) 1603(100) 2252(15) 282 (3) 575 (1) 600 (0) 735 (1) 906 (7) 1331(22) 1778(100) 2200(1) 316 (5) 584 (1) 647 (4) 764 (2) 918 (11) 1343(32) 1638(100) 2288(4) x 576 (vw) (≈590) [d] 720(vw) 890(m) 1300(s) 1580(vs) 2238(w) 272(1.0) 583(1.4) - 721(1.4) 896(4.1) 1304(4.9) 1577(1.2) 2242(10) a Calculated IR (km/mol) and Raman (Å 4 /amu) intensities are referenced to the strongest mode and given in % within parenthesis (IR:Raman). b 25 °C, gas-phase, ca 5 torrs. “x” denotes out of range of instrument. w = weak; vw = very weak; m = medium; s = strong; vs = very strong. c Liquid, ca -90 °C. Intensities expressed relative to the strongest band being 10. d Estimated from IR-observed combination band (vide infra). Table A7.3: Unscaled vibrational frequencies (cm -1 ) for C 2v nitryl isocyanide (2) at two levels of theory showing a poor match with the observed spectra attributed to 1. a Vibrational mode and symmetry: B3LYP/ aug-cc-pVTZ B2PLYP/ Def2-TVZPP IR [b] (exp) Raman [c] (exp) δ oop NCN (B 1) 65 ( 0:2) 59 (0) x 216(4.7) δ ip NCN (B 2) δ scissor NO 2 (A 1) δ rock NO 2 (B 2) δ wag NO 2 (B 1) ν C-N (A 1) ν sym NO 2 (A 1) ν asym NO 2 (B 2) ν C≡N (A 1) 117 ( 0:7) 515 ( 10:27) 622 ( 1:1) 767 ( 4:2) 836 ( 48:10) 1288 ( 64:42) 1694 (100:15) 2117 ( 93:100) 124 (0) 511 (8) 620 (1) 753 (4) 829 (42) 1262 (55) 1702 (100) 2081 (85) x 576 (vw) - 720(vw) 890(m) 1300(s) 1580(vs) 2238(w) 272(1.0) 583(1.4) - 721(1.4) 896(4.1) 1304(4.9) 1577(1.2) 2242(10) a Calculated IR (km/mol) and Raman (Å 4 /amu) intensities are referenced to the strongest mode and given in % within parenthesis (IR:Raman). b 25 °C gas-phase, ca 5 torr. Observation attributed to 1. “x” denotes out of range of instrument. w = weak; vw = very weak; m = medium; s = strong; vs = very strong. c Liquid, -90 °C. Intensities expressed relative to the strongest band being 10. 284 Table A7.4: Complete assignment of the observed vibrational frequencies of C 2v nitryl cyanide (1) Vibrational mode(s) and symmetry: IR [a] Raman [b] (int.) [c] δ ip NCN (B 2) 216 (4.7) δ oop NCN (B 1) 272 (broad, 1.0) δ scissor NO 2 (A 1) 576 (vw) 583 (1.4) δ wag NO 2 (B 1) 720 (vw) 721 (1.4) [δ rock NO 2 [d] + δ ip NCN] (A 1) 802 (vw) - ν C-N (A 1) 890 (m) 896 (4.1) ν sym NO 2 (A 1) 1300 (s) 1304 (4.9) ν asym NO 2 (B 2) 1580 (vs) 1577 (broad, 1.2) [ν C-N + ν sym NO 2] (A 1) 2186 (vw) - ν C≡N (A 1) 2238 (w) 2242 (10) [ν C-N + ν asym NO 2] (B 2) 2463 (vw) - 2 . ν sym NO 2 (A 1) 2593 (vw) - [ν sym NO 2 + ν asym NO 2] (B 2) 2869 (vw) a Gas-phase, 11 torr. w = weak; vw = very weak; m = medium; s = strong; vs = very strong. b Liquid, ca -90 °C. c Intensities expressed relative to the strongest band (=10). d the calculated IR- and Raman-intensities for this mode are zero and it was not directly observed (Table A7.2). Metathetical reactions, and the formation of isomers 8 and 9. The metathetical union of NO 2 + and CN - to 1 is greatly exothermic (~ -60 kcal/mol in CH 3 CN, Figure A7.2). It is likely that limitations in heat dissipation cause the decomposition of both 1 and 2 shortly after their formation, thus explaining the inability to conclusively detect either even at low temperature using this preparatory approach. The kinetic ambivalence of the cyanide anion with respect to C- or N-attack is addressed in Figure A7.2, which illustrates the electrostatics favoring the N-attack (forming 2), whereas the HOMO orbital favors the thermodynamically preferred C-attack (forming 1). Because both pathways are barrierless, a statistical distribution close to 1:1 can reasonably be expected from randomly colliding species. However, the situation is further complicated by two additional possibilities where the cyanide anion can attach to either oxygen of NO 2 + (which have a statistical advantage over nitrogen), by either N- or C- attack. Whereas the resulting [CNO … NO] complex 8 lies 4.7 kcal/mol above 2, the latter [NCO … NO] complex 9 lies 18.8 kcal/mol below 1 at the CBS-QB3 level (Figure A7.2). 285 Figure A7.2: Relaxed broken symmetry scans at the PCM-UM06-2X/6-31+G(d) level hint at the reaction dynamics in acetonitrile solution. The ambivalence of CN - is illustrated by a carbon-dominated HOMO orbital and a negative electrostatic potential that favors nitrogen. MOs and ESPs are plotted on constant 0.01 and 0.001 a.u. isosurfaces, respectively. One isomerization transition state was identified (TS-2, Figure A7.3) between 2 and 8, which corresponds to a free energy barrier of 93 kcal/mol relative to 2. One possible route to the thermodynamically favored 9 is by initial barrierless formation of 1, followed by a 50.7 kcal/mol transition state barrier (TS-1, Figure A7.3). Spin unrestricted broken symmetry potential energy surface scans at the PCM-UM06-2X/6-31+G(d) level revealed that the reaction of NO 2 + and CN - into the trans form of 8 proceeds following a sizable transition state barrier, likely arising from an avoided crossing due to a change in electron configuration. For an accurate treatment of this process a multiconfigurational method with explicit consideration of electron correlation, such as MR-CI with a very large active space, would be required. Because the broken symmetry DFT scans indicate that the crossing occurs at ~3.3 Å, and that the barrier is close to 100 kcal/mol, this was deemed needless. In searching for a transition state corresponding to the formation of the cis conformer of 8 from free ions, none was found. However, TS-3 was identified, which corresponds to the isomerization of 8 into 9. This structure lies below the isolated NO 2 + and CN - anions, and only 10 kcal/mol over 8. Therefore, it is possible that if a proximate PES structure allows for the direct formation of 8 from free ions, the thermodynamically favored 9 could form rapidly. Such a pathway would offer yet another explanation for the inability to use direct metathesis, as a synthetic route to reach 1. 286 Figure A7.3: Isomerization transition states between 1 and 9, 2 and 8, 9 and 8, calculated at the PCM- M06-2X/6-31+G(d) level of theory. All energies in kcal/mol, and are shown relative to 1. Intermolecular reactions of NCNO 2 , and reactions with NO 2 . Based on both previous literature studies, as well as the extensive computational and experimental investigations presented herein, NCNO 2 is inherently kinetically stable with respect to unimolecular decomposition in the gas-phase at ambient conditions. One likely cause for its observed slow decomposition in the liquid phase at room temperature are intermolecular reactions. NO 2 is one of the main side-products in the synthesis and decomposition of NCNO 2 , and trace amounts are always present, even after careful purification procedures. A multitude of transition states were identified that allow for the reaction between NCNO 2 and NO 2 . Details of all these calculations will not be disclosed here, save that all but one are of high energy (ΔG ‡ >40 kcal/mol). The one identified exception is shown in Figure A7.4. This transition state corresponds to a free energy barrier of 28.6 kcal/mol at the CBS-QB3/M06-2X/cc-pVTZ level of theory, which is sufficiently low to allow for slow decomposition close to room temperature. It describes the attachment of an NO 2 radical to 287 one of the oxygen atoms in NCNO 2 , and the formation of a kinetically unstable NCNO(O)-NO 2 radical intermediate (not shown). The conformational space and reactivity of this intermediate has not been studied in detail, but it is easy to see how various movement of the bound NO 2 group can facilitate in its decomposition. For instance, dissociation of NO 3 would leave NCNO, a known unstable molecule. PCM- TD-ωB97X-D/aug-cc-pVTZ calculations were performed on NOCN, which predict that it has an allowed electronic transition at ~770nm, corresponding to a blue color. It is thus possible that the blue color observed in most samples of partially decomposed liquid samples of NCNO 2 are in part due to traces of dissolved NCNO. In the presence of small amounts of NO 2 , the described pathway (Figure A7.4) effectively decreases the barrier to decomposition of NCNO 2 by 26 kcal/mol, corresponding to a large difference in the decomposition rate. Figure A7.4: Transition state corresponding to the reaction of NO 2 with NCNO 2 . Energies (relative NCNO 2 + NO 2 ) and geometries are at the CBS-QB3//M06-2X/cc-pVTZ level of theory (kcal/mol, 1 atm, 298K). Bond lengths are in Å. NMR shift calculations. For more accurate NMR shift estimations, the calculated PCM-GIAO- B3LYP/6-311++G(3df,2p) values were corrected according to a fit based on a range of experimental values in acetonitrile. In most cases, the GIAO method consistently and increasingly underestimates the nuclear shielding. Spectra have been acquired between -40 C and room temperature. Temperature effects are assumed to be small and largely cancel in the fit. The external reference used for all nitrogen signals, experimental and theoretical, is nitromethane. The fit is shown in Figure A7.5 and shows an excellent correlation between experimental and theory (r 2 = 0.999). The root mean square deviation from experiment is 3.7 ppm, and the fitting equation reads as, 288 Estimate (ppm) = 0.915 * calculated + 5.966 (A7.4) Figure A7.5: Theoretical GIAO-PCM-B3LYP 14 N NMR shifts versus experimental values in acetonitrile. Table A7.5: Experimental and theoretical NMR data used to construct fitting equation in Figure A7.5. Species Theoretical (ppm) Experimental (ppm) Fitted value (ppm) NCNH 2 -403.6 -366 -363 NH 4 + N(NO 2 ) 2 - -399.5 -360 -360 NNO -255.6 -231 -228 NCNH 2 -211.6 -188 -188 NNO -163.7 -147 -144 CH 3 CN -162.2 -136 -142 NO 2 + BF 4 - -150.2 -132 -131 HCN -149.4 -127 -131 BrCN -139.9 -122 -122 FNO 2 -99.8 -91 -85 TMSCN -111.6 -87 -96 N 2 -89.5 -71 -76 NCNO 2 -66.4 -60 -55 NH 4 + (NO 2 ) 2 - -68.4 -55 -57 HNO 3 -52.0 -42 -42 (NO 2 ) 2 -25.7 -19 -18 NH 4 + N(NO 2 ) 2 - -16.9 -9 -9 Ag + NO 3 - -8.4 -3 -2 PPh 4 + NO 2 - +255.8 +238 +240 289 Theoretical UV/VIS spectrum. EOM-CCSD/Def2-TZVPP and TD-DFT calculations at the TD-ωB97X- D/aug-cc-pVTZ level suggest that nitryl cyanide (1) does not have any spin-allowed electronic transition states in the visible spectrum, and only weakly allowed transitions in the UV region (vide infra). Optimization of the 1 B 2 and 1 A 1 excited states at the TD-ωB97X-D/aug-cc-pVTZ level suggest that they are non-dissociative. EOM-CCSD/Def2-TZVPP transitions and oscillator strengths: Excited State 1: Singlet-A2 4.2405 eV 292.38 nm f=-0.0000 Excited State 2: Singlet-B1 4.7267 eV 262.30 nm f=0.0010 Excited State 3: Singlet-B2 6.7147 eV 184.64 nm f=0.1914 Excited State 4: Singlet-A2 7.4161 eV 167.18 nm f=0.0000 Excited State 5: Singlet-A1 7.6830 eV 161.38 nm f=0.0534 Excited State 6: Singlet-B1 8.2836 eV 149.67 nm f=0.0183 Excited State 7: Singlet-B2 8.7215 eV 142.16 nm f=0.0060 Excited State 8: Singlet-A2 9.1978 eV 134.80 nm f=-0.0000 TD- ω B 9 7X-D/aug-cc-pVTZ transitions and oscillator strengths: Excited State 1: Triplet-B2 2.6972 eV 459.67 nm f=0.0000 <S**2>=2.000 Excited State 2: Triplet-A2 3.4567 eV 358.68 nm f=0.0000 <S**2>=2.000 Excited State 3: Triplet-B1 3.9835 eV 311.25 nm f=0.0000 <S**2>=2.000 Excited State 4: Singlet-A2 4.0296 eV 307.68 nm f=0.0000 <S**2>=0.000 Excited State 5: Singlet-B1 4.5534 eV 272.29 nm f=0.0008 <S**2>=0.000 Excited State 6: Triplet-A1 5.1176 eV 242.27 nm f=0.0000 <S**2>=2.000 Excited State 7: Triplet-A2 5.9494 eV 208.40 nm f=0.0000 <S**2>=2.000 Excited State 8: Singlet-A2 6.3660 eV 194.76 nm f=0.0000 <S**2>=0.000 Excited State 9: Triplet-A1 6.6154 eV 187.42 nm f=0.0000 <S**2>=2.000 Excited State 10: Singlet-B2 6.6174 eV 187.36 nm f=0.1370 <S**2>=0.000 Excited State 11: Triplet-B1 6.7070 eV 184.86 nm f=0.0000 <S**2>=2.000 Excited State 12: Singlet-A1 6.7191 eV 184.52 nm f=0.0619 <S**2>=0.000 Excited State 13: Singlet-B1 7.6237 eV 162.63 nm f=0.0094 <S**2>=0.000 Excited State 14: Triplet-A2 7.8316 eV 158.31 nm f=0.0000 <S**2>=2.000 290 RIJCOSX-mPW2PLYP/Def2-TZVPP transitions and oscillator strengths: ----------------------------------------------------------------------------- ABSORPTION SPECTRUM VIA TRANSITION ELECTRIC DIPOLE MOMENTS ----------------------------------------------------------------------------- State Energy Wavelength fosc T2 TX TY TZ (cm-1) (nm) (au**2) (au) (au) (au) ----------------------------------------------------------------------------- 1 31154.3 321.0 0.000000000 0.00000 -0.00000 -0.00000 -0.00001 2 35604.9 280.9 0.000555705 0.00514 -0.07168 0.00000 0.00000 3 53214.5 187.9 0.000000000 0.00000 -0.00000 0.00001 -0.00000 4 56063.6 178.4 0.068957743 0.40493 0.00000 -0.00003 0.63634 5 55579.8 179.9 0.225883411 1.33796 0.00000 1.15670 0.00001 6 69967.3 142.9 0.000000000 0.00000 0.00000 0.00001 -0.00001 7 60109.1 166.4 0.018212317 0.09975 -0.31583 0.00000 -0.00000 8 69542.0 143.8 0.004265888 0.02019 0.00002 -0.14211 -0.00019 9 21083.1 474.3 spin forbidden (mult=3) 10 28466.0 351.3 spin forbidden (mult=3) 11 32600.0 306.7 spin forbidden (mult=3) 12 40667.2 245.9 spin forbidden (mult=3) 13 46885.6 213.3 spin forbidden (mult=3) 14 51926.6 192.6 spin forbidden (mult=3) 15 56218.9 177.9 spin forbidden (mult=3) 16 60612.8 165.0 spin forbidden (mult=3) 291 ----------------------------------------------------------------------------- ABSORPTION SPECTRUM VIA TRANSITION VELOCITY DIPOLE MOMENTS ----------------------------------------------------------------------------- State Energy Wavelength fosc P2 PX PY PZ (cm-1) (nm) (au**2) (au) (au) (au) ----------------------------------------------------------------------------- 1 31154.3 321.0 0.000000000 0.00000 -0.00000 0.00000 0.00000 2 35604.9 280.9 0.000549274 0.00013 0.01156 -0.00000 0.00000 3 53214.5 187.9 0.000000000 0.00000 -0.00000 0.00000 -0.00000 4 56063.6 178.4 0.002965027 0.00114 0.00000 -0.00000 -0.03371 5 55579.8 179.9 0.013426767 0.00510 0.00000 0.07142 -0.00000 6 69967.3 142.9 0.000000000 0.00000 -0.00000 0.00000 0.00000 7 60109.1 166.4 0.000000891 0.00000 0.00061 0.00000 0.00000 8 69542.0 143.8 0.003502460 0.00166 -0.00000 0.04080 0.00003 9 21083.1 474.3 spin forbidden (mult=3) 10 28466.0 351.3 spin forbidden (mult=3) 11 32600.0 306.7 spin forbidden (mult=3) 12 40667.2 245.9 spin forbidden (mult=3) 13 46885.6 213.3 spin forbidden (mult=3) 14 51926.6 192.6 spin forbidden (mult=3) 15 56218.9 177.9 spin forbidden (mult=3) 16 60612.8 165.0 spin forbidden (mult=3) A7.3 Other Synthetic Approaches to Prepare NCNO 2 (1) The following describes some of the numerous synthetic approaches used in this work to prepare NCNO 2 . Complementary NMR and Raman spectra are given below. A7.3.1 TMSCN + NO 2 BF 4 CH 3 CN (~ 10mL) was condensed over NO 2 BF 4 (1.412 g, 10.6 mmol), followed by TMSCN (211 mg, 2.1 mmol), in a 200 mL Pyrex glass reactor. The reaction was allowed to proceed for 1 h at -30 °C. Repeated fractional condensations through -63, -95 and -196 °C traps removed CH 3 CN and most NO 2 (bulk of 1 stops at -95 °C). Subsequent fractionations through -84, -95, -126 and -196 °C traps enabled the isolation 292 of 1 and TMSF in the -126 °C trap. 14 N NMR (SO 2 , -30 °C) δ/ppm -62.6 (NCNO 2 ), -175 (NCNO 2 ). 13 C NMR (SO 2 , -35 °C) δ/ppm 106.8 (1:1:1 t 1 J( 13 C- 14 N)=25 Hz, NCNO 2 ). IR: ṽ = 2238 (w), 1580 (s), 1300 (m), 890 (w) cm -1 . Raman: ṽ = 2239 (s), 1298 (w) cm -1 . Figure A7.6: Raman spectrum of a partially purified reaction mixture at -60 °C. Spectral features of 1 are seen together with HCN and one CN-stretching mode of an unidentified compound. The spectrum of TMSF has been subtracted. A7.3.2 t-BuMe 2 SiCN + NO 2 BF 4 in Various Solvents. In acetonitrile. NO 2 BF 4 (158 mg; 1.19 mmol) was placed into a glass vessel equipped with a Teflon valve. Acetonitrile (5-10 mL) was condensed into the reactor. The mixture was quickly thawed, then refrozen to -196 °C. t-BuMe 2 SiCN (126 mg; 0.891 mmol) was then condensed above the frozen solvent. The solvent was thawed between -30 and -35 °C. Once the NO 2 BF 4 suspension had all thawed, the cold liquid was swirled to dissolve the t-BuMe 2 SiCN frozen on the reactor's walls. This ensured that the silicon compound was always in the presence of an excess of oxidizer and typically prevented the appearance of a greenish blue color, indicative of some decomposition and formation of NO 2 . As soon as all the t- BuMe 2 SiCN had dissolved (roughly 5 minutes after all the solvent had thawed), the mixture was frozen to -196 °C and separated by fractional condensation through -80, -96, -112, -126, -196 °C traps. The isolated yields were obtained by measuring the different gas fractions by PVT. Reasonably pure NCNO 2 was found in the -112 and -126 °C traps (0.79 mmol containing about 30 mol% acetyl fluoride, estimated by NMR; 60% yield). The -196 °C trap typically contained CO 2 and N 2 O, while the -80 °C trap contained acetonitrile and most of the t-BuMe 2 SiF. The -96 °C trap contained t-BuMe 2 SiF, NO 2 and sometimes (highly dependent upon the choice of warmer traps and upon the composition of the crude mixture) some NCNO 2 . Refractionation of the -112 °C trap through -96, -112 °C and -196 °C traps (collecting the -112 293 °C trap) allowed the removal of the last traces of cyanogen, t-BuMe 2 SiF and adventitious HCN. 14 N NMR (500 MHz, SO 2 , unlocked, -43 °C) /ppm: -62.6 (s, sharp, NCNO 2 ), -175 (s, broad, NCNO 2 ). 13 C NMR (600MHz, SO 2 , unlocked, -30 °C) (acetyl fluoride impurity omitted) /ppm: 106.6 (1:1:1 t, 1 J( 13 C- 14 N)= 25Hz, NCNO 2 ); Raman (-90 °C, 100 mw) (crude sample, impurities omitted): ῦ/cm -1 = 2242 (10), 1575 (broad, 1.3), 1304 (5.6), 894 (4.8), 720 (1.5), 582 (1.5), 267 (broad, 1.4), 216 (4.7) cm -1 . A relatively pure sample of 1 by IR spectroscopy was obtained through several refractionations of a -96 °C fraction collected from a reaction mixture that had been passed through -60, -96, -112,-126 and -196 °C traps, which contained some NCNO 2 along with t-BuMe 2 SiF, NO 2 and small amounts of acetyl fluoride. IR (gas phase, ca 5 torrs): ṽ = 2869 (vw), 2601 (vw), 2240 (w), 2186 (vw), 1581 (s), 1299 (m), 891 (m), 720 (w), 576 (vw) cm -1 . It should be noted that acetyl fluoride has a very similar vapor pressure as NCNO 2 and it is therefore difficult to remove it by fractional condensation. Given the high reactivity of NCNO 2 , chemical separation is also unlikely to be a viable purification method. With an excess of t-BuMe 2 SiCN. In another attempt, an excess of t-BuMe 2 SiCN was used, in hopes of avoiding the formation of acetyl fluoride by reducing the availability of BF 4 - . Upon complete dissolution of the t-BuMe 2 SiCN, a colorless solution was obtained, which was frozen, then slowly warmed up and passed in a dynamic vacuum through -80, -96, -196 °C traps. Upon thawing, the solution abruptly turned green, and blue material accumulated in the -196 °C trap. The analysis of the traps by IR spectroscopy indicated a somewhat lower acetyl fluoride content than in most other syntheses. The -196 °C trap was refractionated through -96, -112, -126, -196 °C traps. Based on the amount of gas in the -112 and -126 °C traps, the crude yield of NCNO 2 was roughly 40 mol% based on t-BuMe 2 SiCN. However, a purity analysis by IR revealed a significant cyanogen content, which means that the actual NCNO 2 yield was likely closer to 10-20%. The deep blue material found in the -196 °C trap proved to be nitrosyl cyanide, NCNO, a known, unstable compound previously described. 29 This reaction strongly suggests that NCNO 2 reacts with the silyl cyanide starting material. PCM-TD-ωB97X-D/aug-cc-pVTZ calculations were performed on NOCN, which predict that it has an allowed electronic transition at ~770 nm, agreeing with the observed blue color. It is then possible that the blue color observed in most samples of NCNO 2 could be partially due to traces of dissolved NCNO. This example illustrates the highly reactive character of NCNO 2 , as well as the sensitivity of its formation to reaction conditions. In other solvents. All attempts to use other solvents which are less reactive than acetonitrile (fluoroform, DCM, DCM/nitromethane) lead to poor or no yields and impractically low reaction rates, while causing 294 further separation problems. One reaction was performed in propionitrile in hopes of obtaining the less volatile propionyl fluoride as a by-product. However, the reaction times were longer in propionitrile than they were in acetonitrile and the substantially lower melting point of propionitrile created other separation issues. Since nitromethane does not appear to react extensively with NCNO 2 , the propionitrile system was not explored any further. NO 2 BF 4 + CH 3 CN. In order to investigate the cause for the acetyl fluoride formation, acetonitrile was reacted with NO 2 BF 4 . Acetonitrile (1.1 g; 26.9 mmol) was condensed on top of finely ground NO 2 BF 4 (85 mg; 0.64 mmol). The mixture was sonicated for a few minutes at room temperature and then fractionally condensed through -60 and -196 °C traps. The contents of the -196 °C trap were analyzed by gas-phase IR spectroscopy, which confirmed the formation of acetyl fluoride (characteristic absorptions at 1868 and 1187 cm -1 ). However, the amounts formed in the given reaction time (traces observed by IR) do not agree with the significant amounts of acetyl fluoride observed for similar reaction times at lower temperature (0 °C to -35 °C) when NCNO 2 was present. This suggests that NCNO 2 either synergistically enhances the formation of acetyl fluoride from NO 2 BF 4 and acetonitrile or, more likely, reacts very quickly with acetonitrile in the presence of BF 4 - or BF 3 . acetonitrile to form acetyl fluoride. Other reactions involving oxidizer salts of BF 4 - in acetonitrile have been shown to lead to the formation of acetyl fluoride. 30 A7.3.3 Metathesis of CN - and NO 2 - Several attempts at the direct metathesis of cyanide and nitryl ions were attempted (Equation A7.5). These were followed from -40 to 25 °C by 14 N NMR. In some instances trace amounts of 1 were indicated at room temperature by a fleeting signal at -60.3 ppm in suspensions of mostly unreacted KCN in CD 3 CN. This chemical shift is identical to that observed in experiments with FNO 2 and TMSCN in the same solvent. However, due to poor signal-to-noise ratios and reproducibility, the formation of 1 through this approach could not be conclusively claimed. Attempts to increase the solubility of the cyanide anion through experiments with 18-crown-6/KCN resulted in rapid decomposition at low temperature, likely due to exothermic reactions between the crown ether and NO 2 + . Reactions of n-NBu 4 CN with NO 2 BF 4 in DCM resulted in the formation of NO 3 - (-2.0 ppm) and N 2 O at 0 °C. NO 3 - was confirmed by the addition of HCl, which shifted the -2.0 ppm peak to -38 ppm, characteristic for HNO 3 . MCN + NO 2 + BF 4 - (M = K, n-NBu 4 ) → NCNO 2 (1) + M + BF 4 - (A7.5) 295 In a typical reaction 1 mL solvent (CD 3 CN, DCM, CFCl 3 or SO 2 ) was condensed over 0.1 mmol MCN and 0.1mmol NO 2 BF 4 or NO 2 SbF 6 in a Young NMR-tube. The reaction was monitored from -40 to +25 °C by 14 N NMR. The insolubility of KCN in CH 3 CN prompted the use of 18-crown-6, which was dried by sublimation and added (0.1 mmol) under identical reaction conditions (at -20 °C). This resulted in the exothermic decomposition of the crown ether. The use of NBu 4 CN, which is soluble in CD 3 CN, as a source of cyanide, yielded NO 3 - , as well as ample amounts of N 2 O as observed by 14 N NMR. The poor solubility of NO 2 BF 4 in DCM resulted in no reaction. nBu 4 NCN + NO 2 BF 4 in DCM. 0.1mmol (13.3 mg) of NO 2 BF 4 and 0.1mmol (26.9 mg) n-Bu 4 NCN were put together in a J-Young NMR tube, and kept cold. ~1.5 mL of DCM was condensed over the solids, and kept below -40 °C. The reaction was followed by NMR between -40 and 20 °C. At 0 °C, significant amounts of N 2 O and very weak signals at -10.4 and -32.7 ppm were observed by 14 N NMR. All cyanide was consumed at 20 °C, yielding a bright red solution. NO 2 SbF 6 + KCN in SO 2 . 0.1 mmol (28 mg) of NO 2 SbF 6 was added to 0.1mmol (6.5 mg) of KCN in a J- Young NMR tube. The solids were kept at -196 °C. ~2ml of SO 2 were condensed in, and the reaction mixture was kept at -50 °C, before it was monitored between -30 and 20 °C by 14 N NMR. Sufficient solubility was obtained around 0 °C, but significant reactivity was observed only at room temperature. After 24 h, NCNO 2 was observed as the major product. The commercially available NO 2 + salt was found to contain significant amounts of HNO 3 , which in all likelihood catalyzed the reaction by forming HCN, which, contrary to KCN, does not form a strong complex with SO 2 . This was confirmed by a later experiment in which HCN replaced KCN as the cyanide source, which exhibited enhanced reactivity compared to the reaction with KCN. NH 4 + was detected by 1 H NMR, probably through reaction of HNO 3 with HCN. NMe 4 CN + NO 2 BF 4 in CD 3 CN. 0.15 mmol (20 mg) of NO 2 BF 4 and 0.15 mmol (15 mg) of NMe 4 CN were placed into a into a J-Young NMR tube. 1 mL of CD 3 CN was condensed over the solids, and the reaction was followed by 14 N NMR. At -40 °C, N 2 O was already the major product. At -30 °C, after a few minutes, traces of NCNO 2 were detected and remained in the mixture even at room temperature. Since small amounts of HNO 3 and HCN were detected in the solution (likely from acidic impurities in the 296 commercially available NO 2 BF 4 ), it is likely that the traces of NCNO 2 formed originated from the less exothermic HCN + NO 2 BF 4 reaction, and the direct CN - + NO 2 + reaction resulted in decomposition. A7.3.4 Nitration of HCN These reactions were inspired by Klapötke et al.'s reported observation of the CNNO 2 isomer from the nitration of HCN, 31 which was demonstrated to actually be NCNO 2 . Most of the systems explored allowed the observation of sizeable amounts of NCNO 2 , but the conversion of HCN was always quite low, which lead to purification problems. Reactions run in SO 2 with NO 2 SbF 6 allowed faster reactions than those in Freon-11 but added the problem of separating NCNO 2 from SO 2 . In addition, in all cases the reaction rates were substantial only close to ambient temperature, which favored decomposition and the formation of side-products. HCN + NO 2 BF 4 in CH 2 Cl 2 . NO 2 BF 4 (0.1 mmol, 13 mg) was placed into a J-Young NMR tube and DCM (~1.5 mL) was condensed over the solids. HCN (0.1 mmol) was condensed over the solution. The temperature was kept at -30 °C before the reaction was followed by 14 N NMR. No significant reaction was observed, even after several weeks at room temperature. HCN + NO 2 BF 4 in CFCl 3 and CD 3 CN. In a typical NMR-scale experiment, HCN (0.18 mmol) and Freon-11 (CFCl 3 ) were condensed on top of NO 2 BF 4 (used as received, 20.6 mg; 0.16 mmol) in a J- Young NMR tube. The temperature was then ramped slowly from -60 °C to 25 °C while in the NMR instrument. After several days at room temperature, significant amounts of a compound with -64.2 and - 171 ppm 14 N NMR signals were detected. When a similar experiment was performed over a few hours in CD 3 CN, no reaction took place below room temperature. After a few days at room temperature, only N 2 O and the starting materials could be detected in the reaction mixture. In a scaled-up experiment NO 2 BF 4 (664 mg, 5 mmol) was added to a Pyrex glass reactor, followed by ~10 mL of CFCl 3 , and HCN (5 mmol). The reaction was left at 25 °C for 48 h, after which time most of the NO 2 BF 4 appeared unreacted. Fractional condensation (traps at -63, -94, -126 and -196 °C) allowed the isolation of 1 and CFCl 3 in the -126 °C trap. 14 N NMR (CFCl 3 , unlocked) /ppm: -64.2 (s, sharp, NCNO 2 ), -171.9 (s, broad, NCNO 2 ). IR: ṽ = 2238 (w), 1580 (s), 1300 (m), 890 (w) cm -1 . Raman: ṽ = 2239 (s), 1298 (w) cm -1 . When a partially purified reaction mixture containing 1, unreacted HCN and CFCl 3 solvent was left at room temperature for a few hours and analyzed by 14 N NMR spectroscopy, 1 disappeared while a 297 new broad signal at -139.8 ppm was observed with increasing intensity. This could indicate the formation of higher oligomeric structures. Figure A7.7: 14 N-NMR (25 °C) of the reaction between HCN (-117.7 ppm) and NO 2 BF 4 in CFCl 3 . NCNO 2 is observed at -64.2 ppm and NCNO 2 at -171.9 ppm. The minor peak at -64.7 ppm is likely ClNO 2 (see section below on reaction with ClNO 2 ). Figure A7.8: Calculated and observed infrared spectra of 1 in the gas-phase; CFCl 3 background subtracted. HCN + NO 2 SbF 6 in SO 2 . NO 2 SbF 6 (0.1 mmol, 28 mg) and KCN (0.1 mmol, 6.5 mg) were placed into a J-Young NMR tube and SO 2 (~2 mL) was condensed over the solids. The tube was kept at -50 °C before the reaction was followed between -20 and 20 °C by 14 N NMR. Significant reaction rates were observed only at room temperature. After 24 h, NCNO 2 was observed as a major species. An unidentified signal appeared at δ -249 ppm (t, J = 75 Hz) in 14 N NMR, suggesting an -NH 2 or =NH 2 + species. The ammonium 298 cation was detected by 1 H and 14 N NMR. The sample was condensed onto K 3 PO 4 in an attempt to chemically separate unreacted HCN. This led to the decomposition of NCNO 2 . HCN + NO 2 SbF 6 in SO 2 . NO 2 SbF 6 (1.5 mmol, 423 mg) was placed into a glass reactor, and 1 mL of SO 2 was condensed over the solid. HCN (1 mmol) was condensed over the frozen solvent at -196 °C. The reaction was left at -30 °C for 3 h, then at room temperature overnight. Fractional condensation through - 84, -95, -117, -126 and -196 °C traps showed only CO 2 and N 2 O in the last trap, and only HCN and SO 2 in the -117 °C trap. The content of the -117 °C trap was also examined by NMR and showed mostly the HCN starting material. It is likely that the amount of solvent was insufficient, making the reaction too exothermic, thus explaining the difference compared to the NMR scale experiment described above. In a modification of the previous reaction NO 2 SbF 6 (1.6 mmol, 450 mg) was placed into a glass reactor, and ~3 mL of SO 2 was condensed over the solid. HCN (0.5 mmol) was condensed over the frozen solvent at -196 °C. The reaction was left at -30 °C for 3 h without stirring, then at 0 °C for six d. The IR analysis of the crude reaction mixture in a CsI windowed cell allowed the detection of NCNO 2 , N 2 O, and CO 2 . A7.3.5 Reactions with FNO 2 The reaction of TMSCN with FNO 2 in acetonitrile was quite exothermic and only traces of NCNO 2 were formed. The reactions were also plagued by the difficult removal of TMSF and by problems related to obtaining good reaction rates in solvents other than acetonitrile. FNO 2 did not appear to nitrate HCN, as NO 2 BF 4 does, to form NCNO 2 in significant amounts. FNO 2 + KCN in a steel cylinder. KCN (610 mg; 9.38 mmol) was placed into a ClF 3 -passivated steel cylinder. FNO 2 (1.1 mmol) was condensed in and the cylinder was kept at -80 °C for 15 d. The weight gain was measured after evacuation of the volatiles and found to be negligible. Only FNO 2 and negligible amounts of NO 2 were observed in the fractionated volatiles. All of the FNO 2 was recondensed onto KCN in the cylinder and the reaction was followed at -45 °C, -20 °C, 0 °C and room temperature. Throughout the monitoring, only negligible amounts of CO 2 and N 2 O as well as unidentified products could be detected by IR, attesting to the expectedly slow gas-solid reaction. No NCNO 2 was detected in any of the experiments. In another cylinder, FNO 2 (8 mmol) was condensed on top of KCN (78 mg; 1.2 mmol). The cylinder was kept at the same temperatures as the previous one, and the reaction systems showed no 299 significant difference to the first ones. Small amounts of potassium nitrate were found in the orange solid residue by Raman spectroscopy. FNO 2 + TMSCN in CD 3 CN. FNO 2 (0.11 mmol) and TMSCN (12 mg; 0.12 mmol) and CD 3 CN were condensed into a sealed 5 mm Teflon-FEP NMR tube and sealed. The reaction was monitored between - 40 and -30 °C. A relatively sharp 14 N NMR signal at δ -60 ppm was detected among other side-products (N 2 O 4 , N 2 O) in the resulting blue solution. The presence of a doublet in the 1 H NMR at δ -0.39 (J = 7 Hz) ppm confirmed the expected elimination of TMSF. The short lifetime of the species corresponding to the -60 ppm 14 N signal did not allow the observation of the expectedly broad CN signal. Signals at +406 ppm (very broad, presumably a nitroso or nitrite species) and -43 ppm (HNO 3 ) were also found by 14 N NMR. FNO 2 + TMSCN, in CFCl 3 . In a typical experiment, TMSCN (492 mg; 4.96 mmol) was weighed into a Teflon-FEP reactor. CFCl 3 was condensed in on the glass line, and the reactor was transferred to the metal line. FNO 2 (5.17 mmol) was then condensed into the reactor at -196 °C, and the mixture was warmed to -60 °C during 1 h. After 40 min at -60 °C, the mixture was fractionated through -20, -64, -96, - 126 and -196 °C traps. The contents of the -120 °C trap were condensed into a J-Young NMR tube. 14 N NMR (CFCl 3 , -40 °C) δ/ppm: -64.4 (s), -173.0 (bs), along with other signals at -18.5 (N 2 O 4 ), -71.5 (N 2 ), - 116.4 (cyanogen), -147.6 (N 2 O), -231.8 (N 2 O). FNO 2 + Freon-11. FNO 2 was condensed into CFCl 3 . A signal at -66.3 ppm was observed by 14 N NMR at -20 °C, consistent with the formation of ClNO 2 , which suggests that nitrating agents are likely to form variable amounts of ClNO 2 in Freon-11, which might explain the fact that, in one occasion, more than one signal around -66 ppm were observed when HCN was reacted with NO 2 BF 4 (Figure A7.7). FNO 2 + TMSCN in CH 3 CN. FNO 2 (2.9 mmol) was condensed on top of frozen TMSCN (293 mg; 2.95 mmol) and acetonitrile (5 mL). Upon thawing at -20 °C, the reaction turned violently exothermic with evolution of gas. No NCNO 2 was found in the fractionated products, but evidence for the formation of methyl-containing derivatives other than TMSF were found. Mostly N 2 O, FNO 2 , TMSF and CO 2 were identified in the reaction products. Although this reaction was shown to work by NMR, careful heat control is absolutely critical for scale-up reactions. FNO 2 + TMSCN in 2H-heptafluoropropane. The previous reaction was repeated using 2H- heptafluoropropane as the solvent. The reaction proceeded at a surprisingly low rate at -20 °C. Even after several days at that temperature, significant amounts of unreacted FNO 2 could be found in solution. It is interesting to note that even at very low temperatures, the decomposition products (N 2 O 4 and N 2 O) are formed the fastest. After several days at -20 °C, only minute amounts of NCNO 2 could be found by 14 N 300 NMR at -64.6 ppm, while N 2 O 4 and N 2 O were observed in large amounts. Signals at +467 (R-NO?), - 61.1, -126.6, -137.8 and -149.9 ppm were also observed by 14 N NMR. Given the slow rate of the reaction and the large amounts of side-products formed, this system afforded no advantage compared to the ones previously investigated. FNO 2 + tBuMe 2 SiCN in DCM. FNO 2 (0.05 mmol) was condensed on top of a t-BuMe 2 SiCN (15 mg; 0.11 mmol) solution in DCM in a 5 mm Teflon-FEP tube. The tube was heat sealed under vacuum and warmed to -80, then -50, then -31 °C. The sample was analyzed by VT NMR, which showed no significant reaction between the two starting materials. After a few hours, some Si-F species could be observed in small quantities by 1 H NMR. This indicates that the solvent has a dramatic effect on the reaction between FNO 2 and R 3 SiCN. FNO 2 + HCN in 2H-heptafluoropropane. FNO 2 (0.1 mmol) and HCN (0.1 mmol) were condensed into a Teflon-FEP tube along with 2H-heptafluoropropane. The reaction was monitored between -40 and -20 °C. Even after several weeks at -20 °C, mostly starting materials and a few small unknown peaks could be observed by 14 N NMR (-72.9; -75), and what were perhaps small amounts of NCNO 2 (-65.7 ppm). FNO 2 + HCN in CD 3 CN. FNO 2 (0.07 mmol) and HCN (0.07 mmol) were condensed into a Teflon-FEP tube along with CD 3 CN. The reaction was monitored between -40 and 20 °C. Very broad 14 N NMR signals at +428 and +351 ppm were observed. A small signal -64 ppm in the 14 N NMR was indicative of NCNO 2 . Several 1 H NMR signals were observed, indicative of complex oligomerization reactions of HCN. At higher temperatures, decomposition products were observed. A7.3.6 Reactions with ClNO 2 In hopes of facilitating the purification of NCNO 2 by generating less volatile side-products, the reactivity of ClNO 2 was explored as a substitute for FNO 2 and NO 2 BF 4 . In no case did ClNO 2 behave as a source of "NO 2 + " and no NCNO 2 was observed in any of the systems explored. ClNO 2 + AgCN. ClNO 2 (3.3 mmol) was condensed on top of AgCN (434 mg; 3.24 mmol) in a Teflon- FEP reactor. At -96 °C, the ClNO 2 was adsorbed to the solid AgCN. At -50 °C, the solid abruptly turned orange. The products were separated by fractional condensation. Nearly all of the ClNO 2 was consumed in the reaction. The main reaction products were NO 2 and ClCN by IR spectroscopy. Small amounts of starting materials (ClNO 2 and most likely its NOCl impurity) and CO 2 were detected. The solid residue contained still significant amounts of AgCN by Raman spectroscopy, which suggests that 1 mole of AgCN reacts with 2 moles of ClNO 2 through a radical pathway to yield AgCl, ClCN and N 2 O 4 . This 301 contrasts with the reaction between AgOCN and ClNO 2 , observed by Klapötke et al., in which evidence was obtained for the transient formation of OCNNO 2 . 32 ClNO 2 + AgCN in CD 3 CN. ClNO 2 (0.13 mmol) was condensed over a CD 3 CN suspension of AgCN (17.1 mg; 0.13 mmol) in a J-Young NMR tube. The reaction was monitored between -30 °C and room temperature. Evidence for ClCN was found by NMR, together with traces of HCN, N 2 and NO 2 . A grey solid was obtained as the final residue (which suggests the formation of photosensitive AgCl), and CO 2 , N 2 O and NO 2 were detected by IR. ClNO 2 + Me 3 SnCN. ClNO 2 (0.11mmol) was condensed onto Me 3 SnCN (21.8 mg; 0.12 mmol) in CD 3 CN in a J-Young NMR tube. The reaction was monitored between -30 and 0 °C. A blue solution was obtained at -30 °C, with the formation of a -17.7 ppm signal in 14 N NMR (see data for the NO 2 BF 4 + Me 3 SnCN reaction) as well as a broad signal at +50 ppm, which suggests XNO or RONO species. At -10 °C, a signal at -116.1 ppm (likely cyanogen) and the disappearance of the +50 signal were noted in the 14 N NMR spectrum. One major signal at 0.78 ppm in the 1 H NMR was observed (deshielded with respect to the Me 3 SnCN starting material), suggesting the formation of a Me 3 Sn + or of a NO 2 -containing species. Upon re-cooling to -30 °C, the broad 14 N NMR signal at +50 ppm was again observable. No evidence for the formation of NCNO 2 was found. ClNO 2 + TMSCN in CD 3 CN. TMSCN (12 mg; 0.12 mmol) and ClNO 2 (0.12 mmol) were condensed into a J-Young NMR tube along with CD 3 CN. The reaction was monitored between -30 °C and room temperature. During the 2-3 h of reaction time, the solution turned yellow, then greenish blue. In the 1 H NMR, the gradual appearance of a signal (0.25 ppm), shielded with respect to Me 3 SiCN (0.53 ppm), indicated the formation of TMS-O-TMS as a major product, while the 14 N NMR indicated a significant increase in NOCl content. At room temperature, ClCN, N 2 O and NOCl were observed by 14 N NMR. N 2 O, CO 2 , NOCl and NO 2 were observed by IR. This indicates that ClNO 2 and TMSCN do not react in a simple, predictable "metathetical" way to form NCNO 2 and TMSCl. A7.3.7 Nitration of Me 3 SnCN In hopes of facilitating the isolation of NCNO 2 by generating the heavier Me 3 SnF side-product instead of Me 3 SiF, the nitration of the tin analogue of trimethylsilyl cyanide was explored. The reaction proved to be difficult to control, similarly to a direct metathesis between NO 2 + and CN - , and only decomposition products were observed. 302 Me 3 SnCN + NO 2 BF 4 in CD 3 CN. Me 3 SnCN (29 mg; 0.15 mmol) and NO 2 BF 4 (20 mg; 0.15 mmol) were placed into a J-Young NMR tube, and CD 3 CN was condensed over the solids. When left to react at -35 °C, the mixture evolved gas, and the solids changed color to light green. Intense bubbling followed, as the reaction proceeded for 30 min. The tube was then frozen at liquid nitrogen temperature, and all volatiles were pumped off to relieve pressure. The green reaction mixture was stored over night at -78 °C. The reaction was then followed between -30 °C and ambient temperature. At -30 °C, one major 1 H species was observed at lower field than the starting material. N 2 O, NO 2 and a new species at -17.5 ppm (possibly a Sn-NO 2 species) were observed by 14 N NMR. The species at -17.5 ppm appeared to be stable for a few days at room temperature. BF 4 - was detected by 19 F and 11 B NMR. 11 B NMR showed signals for other minor boron products (-0.4 ppm and three other signals in the same region). No evidence for the formation of NCNO 2 was obtained. A7.3.8 Nitration of Cyanamide None of the attempts involving the reaction of cyanamide with oxidizers yielded NCNO 2. A7.3.9 Reactions of BrCN and NO 2 - In an attempt to reduce the exothermicity problem associated with direct metathesis (discussed above), an approach of using BrCN together with nitrite salts was attempted at low temperature (Equation A7.6). After slowly warming up the reaction mixtures towards ambient temperature, neither of the reactions resulted in the desired product. Instead various other products were observed. BrCN + M + NO 2 - (M = Ag, Na, PPh 4 ) → NCNO 2 (1) + M + Br - (A7.6) PPh 4 NO 2 + BrCN in DCM. BrCN (0.1 mmol, 11 mg) and PPh 4 NO 2 (0.1 mmol, 39mg) were placed into a J-Young NMR tube and DCM was condensed over the solids. The reaction was monitored between -40 and 20 °C by 14 N NMR spectroscopy. N 2 , N 2 O and nitrate (-2.5 ppm) and a signal at -300 (broad, small) were detected at -40 °C. At 0 °C, most of the starting materials had been consumed. BrCN + PPh 4 NO2 in CD 3 CN. BrCN (0.1 mmol, 11 mg) and PPh 4 NO 2 (0.1 mmol, 39 mg) were placed into a J-Young NMR tube and CD 3 CN was condensed over the solids. The solution was analyzed by 14 N 303 NMR spectroscopy at -30 °C. Signals corresponding to N 2 O, N 2 and nitrite were identified, as well as a broad singlet at δ -301.1 ppm and a sharp singlet at δ -1.0 ppm. BrCN + PPh 4 NO 2 in 1,2-dimethoxyethane (glyme). BrCN (0.1 mmol, 11 mg) and PPh 4 NO 2 (0.1 mmol, 39mg) were placed into a J-Young NMR tube and glyme was condensed over the solids. The reaction bubbled intensely at -40 °C, and had to be frozen and pumped clear of volatiles to avoid over-pressurizing the tube. The solution was analyzed at -40 °C by 14 N NMR. Signals for N 2 and N 2 O could be observed. Warming up to -10 °C did not lead to the appearance of new signals. At room temperature, the BrCN starting material was observed in addition to the other signals previously observed by NMR. PPh 4 NO 2 + BrCN in CD 3 OD. BrCN (0.4 mmol, 42 mg) and PPh 4 NO 2 (0.4 mmol, 154 mg) were placed into the reactor under air, and CD 3 OD was added to the solids at room temperature. The solution was analyzed by 14 N NMR spectroscopy. On the first day of reaction, N 2 O and N 2 and unreacted nitrite were detected in solution, along with signals at 181.7 (perhaps a RONO -type compound), -3.1(most likely nitrate, as it shifts to -27 when HCl is added), -311.2, and -366.6 ppm. After a day, the nitrite starting material was completely consumed. After four days, the signal at -311.2 ppm had disappeared. Crystals were obtained from the solution, and identified as PPh 4 Br 3 by X-ray crystallography. NaNO 2 + BrCN in 1,2-dimethoxyethane (glyme). NaNO 2 (0.15 mmol, 10 mg) and BrCN (0.15 mmol, 16 mg) were placed into a J-Young NMR tube, and glyme was condensed over the solids. After three weeks at room temperature, N 2 and small amounts of N 2 O were detected by 14 N NMR. AgNO 2 + BrCN in CD 3 CN. AgNO 2 (0.2 mmol, 31 mg) and BrCN (0.2 mmol, 21 mg) were placed into a J-Young NMR tube and CD 3 CN was condensed over the solids. The reaction was monitored by 14 N NMR between -30 and 20 °C. Even at -30 °C, the formation of N 2 O, nitrate (-1.7 ppm) and of what is likely to be a RONO (+195 ppm) species was observed. The volatiles were transferred to a second tube and were shown to contain the same RONO-type species, N 2 O and HNO 3 (-40.1 ppm). The non-volatile residue was shown to contain nitrate (-1.7 ppm) and an unidentified species in the amine/ammonium region (-350 ppm). AgNO 2 + BrCN in 1,2-dimethoxyethane (glyme). AgNO 2 (0.1 mmol, 15 mg) and BrCN (0.1 mmol, 11 mg) were placed into a J-Young NMR tube and glyme was condensed over the solids. No reaction products could be detected by 14 N NMR . AgNO 2 + BrCN in D 2 O. AgNO 2 (0.4 mmol, 62 mg) and BrCN (0.4 mmol, 42 mg) were placed into an NMR tube and D 2 O was put over the solids. BrCN appeared to be only sparingly soluble, whereas AgNO 2 was not soluble at all. Early on, a species at 123.9 ppm in 13 C NMR could be detected (likely a nitrile 304 species). Signals at 195.3 ppm (broad, likely a RONO or a soluble nitrite), -5.0 ppm (most likely nitrate), -133.2 ppm, -362 ppm (nonet, J = 7 Hz, ND 4 + ) were observed by 14 N NMR. After one day, only ND 4 NO 3 remained. AgNO 2 + BrCN in CD 3 OD. AgNO 2 (0.4 mmol, 62 mg) and BrCN (0.4 mmol, 42 mg) were placed into an open vial and CD 3 OD was put over the solids. The reaction was left to stir overnight. Major signals were seen at 184.2 ppm (likely a nitrite species) and -120.6 ppm (nitrile or cyanogen derivative) by 14 N NMR. Other minor signals at -0.7 ppm, -1.8 ppm, (possibly nitrates) and -37.5 ppm (likely HNO 3 ) were observed. When the sample was cooled to -10 °C, an additional signal at -24.6 ppm (N 2 O 4 ) was found. In 1 H NMR, a major signal was detected at 6.49 ppm, consistent with the presence of the ammonium cation. AgNO 2 + BrCN (scale-up). AgNO 2 (1 mmol, 154 mg) and H 2 O (deionized, 2 mL) were stirred in a glass reactor under air, which was covered to keep the mixture in the dark. AgNO 2 is only slightly soluble. BrCN (1 mmol, 106 mg) was added, and the reaction was left standing for 1 h. The solution/solid mix was then filtered through a glass filter to remove yellow AgBr, and washed with diethyl ether. The ether phase was dried with MgSO 4 and filtered. The aqueous phase was left in a crystallization dish to evaporate, and yielded transparent needle crystals. The crystals, which are soluble in CD 3 CN and H 2 O, but not soluble in chlorobenzene, EtOH, or acetone, were identified as [NH 4 ][Ag(NO 3 ) 2 ] by X-ray crystallography. A7.3.10 Attempts at Chemical Purification of NCNO 2 and Reactivity of NCNO 2 These reactions illustrate the highly reactive character of NCNO 2 and the challenges that its purification entails. Removal of NO 2 /N 2 O 4 with Hg. Following the reaction of TMSCN and NO 2 BF 4 in CH 3 CN (described above), a fraction (~0.38 mmol of gas) predominantly consisting of TMSF, NCNO 2 and traces of NO 2 /CO 2 was obtained by fractional condensation (-95 °C trap). The fraction was condensed over Hg (2.2g, 11 mmol), and kept at -20 °C while shaken for 10 min. A black precipitate formed immediately on the surface of the metal. IR of the volatiles showed that the NO 2 content had decreased relative to NCNO 2 , but that both had decreased significantly relative to TMSF. After 30 additional minutes, all NO 2 and all NCNO 2 had reacted. Thus, the method is not viable for the separation of NCNO 2 from NO 2 . The black precipitate was not analyzed. 305 NCNO 2 /HCN + NEt 3 . Following the reaction of HCN and NO 2 BF 4 in CFCl 3 (described above), all volatiles were condensed over an excess of NEt 3 , and kept at -20 °C for 3 h. Subsequent fractional condensation and IR analysis revealed only N 2 O, CO 2 , CFCl 3 , NEt 3 and small amounts of HCN. It was concluded that triethylamine is not a strong enough base for efficient removal of HCN from the reaction mixtures. In addition, it was showed to decompose NCNO 2 . NCNO 2 /HCN/SO 2 + K 3 PO 4 . Following a reaction of NO 2 SbF 6 (0.1 mmol, 28mg) and HCN (0.07 mmol) in SO 2 in a J-Young NMR tube, strong signals for NCNO 2 were seen by 14 N NMR. All volatiles were transferred over 0.7 mmol (>10 eqv.) K 3 PO 4 , to neutralize and capture all acidic species, and kept briefly at -15 °C. The solution turned slightly yellow. After 10 min the colorless volatiles were transferred to a new tube. 14 N NMR showed only traces of NCNO 2 and N 2 O and HCN as the major species. NCNO 2 /TMSF + molecular sieves. Following the reaction of TMSCN and NO 2 BF 4 in CH 3 CN (described above), a fraction predominantly consisting of TMSF, NCNO 2 and traces of NO 2 /CO 2 was obtained by fractional condensation, and condensed over 3 Å molecular sieves. Nothing could be retrieved from the sieves. Reaction of NCNO 2 /SO 2 with DABCO and CsF. Following a reaction of NO 2 SbF 6 (1.6 mmol, 450 mg) and HCN (0.5 mmol) in SO 2 (~3 mL, ~68 mmol), NCNO 2 was observed among the various volatile compounds of the reaction mixture after 6 days at 0 °C (using a CsI-windowed IR-cell). N 2 O and CO 2 were also detected. The entire reaction mixture was expanded into the vacuum-line, after which half was condensed on top of CsF (5.2g, 34 mmol), and half over DABCO (3.8 g, 34 mmol). Both samples were left at -55 °C for four hours, followed by fractional condensation (with the samples kept at -55 °C). Only small amounts of CO 2 , N 2 O and SO 2 could be recovered from the DABCO-sample. CsF did not absorb SO 2 even at -55 °C when subjected to a dynamic vacuum. Both compounds appear to cause the decomposition of NCNO 2 , and are thus not viable methods for the chemical separation of SO 2 from the reaction mixture. NCNO 2 /TMSF + B(C 6 F 5 ) 3 . About 0.26 mol of the NCNO 2 /TMSF gas mixture was condensed onto B(C 6 F 5 ) 3 (69.7 mg; 0.14 mmol) in DCM in a J-Young NMR tube. The reaction was monitored between - 30 and 20 °C. No 14 N NMR signal could be observed and only very broad signals were obtained by 11 B NMR. A bright red-orange solid, then solution, were obtained. 11 B NMR showed a signal at high field, suggesting the formation of an adduct with the borane. 1 H NMR showed a small signal around 5.5 ppm, suggesting that some HCN might have formed a complex with the borane. The red solid did not evolve any volatile product for several hours at RT. The formation of a brightly colored solid that did not have any 14 N NMR signal suggests the formation of a radical species. 306 B(C 6 F 5 ) 3 + HCN. HCN (ca 0.05 mmol) was condensed over B(C 6 F 5 ) 3 (21 mg; 0.04 mmol) in DCM in a J.Young NMR tube, and the reaction was monitored by NMR. 1 H NMR showed a signal at 5.51 ppm, 11 B NMR showed a signal at -8.5 ppm, and 14 N NMR showed a signal at -190 ppm, all indicative of the formation of a complex. No color change was observed, indicating that the red color observed in the same reaction with NCNO 2 was not due to HCN. NCNO 2 + TMAF in CD 3 CN. About 0.26 mmol of a gas mixture containing TMSF/NCNO 2 was condensed, along with CD 3 CN on top of tetramethylammonium fluroride (TMAF) (14.2 mg; 0.15 mmol) in a J-Young NMR tube. The reaction was monitored at -30 °C by NMR. Upon thawing, the mixture was already bright orange. 14 N NMR indicated the formation of N 2 O and of a nitrate salt or unknown nitro compound at -1.8 ppm. No more NCNO 2 could be found. 1 H NMR indicated the formation of bifluoride, and a broadening of the TMSF signal at 0.392 ppm, which suggests minimal interaction of TMAF with TMSF and that most of it reacted with HCN/NCNO 2 . These results suggest that the presence of TMAF or at the very best, of TMAF and a trace of moisture, is highly detrimental to NCNO 2 . NCNO 2 /TMSF + H 2 O in DCM. About 0.5 mmol of a TMSF/NCNO 2 mixture was condensed into a J- Young NMR tube, along with DCM. Degassed water was condensed on top, and the reaction was monitored by VT NMR. At 0 °C, water had some solubility in DCM and a new species could be observed by 14 N NMR at -358 ppm. The DCM-phase was discarded from the two-phase mixture and the aqueous phase diluted with some additional water. Ammonium nitrate was identified by 1 H and 14 N NMR spectroscopy as the main reaction product. A7.4 References (1) Christe, K. O.; Wilson, W. W.; Sheehy, J. A.; Boatz, J. A. Angew. Chem. Int. Ed. 1999, 38, 2004. (2) Bryce, D. L.; Wasylishen, R. E. Inorg. Chem. 2002, 41, 4131. (3) Kaplan, R.; Shechter, H.; Castorina, T. C.; Tomlinson, W. R. In Inorg. Synth.; John Wiley & Sons, Inc.: 2007, p 52. (4) Seyferth, D.; Kahlen, N. J. Org. Chem. 1960, 25, 809. (5) Christe, K. O.; Wilson, W. W.; Wilson, R. D.; Bau, R.; Feng, J. A. J. Am. Chem. Soc. 1990, 112, 7619. (6) Grimme, S. J. Chem. Phys. 2006, 124, 034108. (7) Eichkorn, K.; Treutler, O.; Öhm, H.; Häser, M.; Ahlrichs, R. Chem. Phys. Lett. 1995, 240, 283. (8) Weigend, F.; Ahlrichs, R. PCCP 2005, 7, 3297. (9) Neese, F.; Schwabe, T.; Grimme, S. J. Chem. Phys. 2007, 126, 124115. 307 (10) Schwabe, T.; Grimme, S. PCCP 2006, 8, 4398. (11) Frisch, M. J. et al., Gaussian 09, Revision A.02, Gaussian, Inc., Wallingford CT, 2009. (12) Valero, R.; Gomes, J. R. B.; Truhlar, D. G.; Illas, F. J. Chem. Phys. 2008, 129, 124710. (13) Zhao, Y.; Truhlar, D. G. Theor. Chem. Acc. 2008, 120, 215. (14) Goerigk, L.; Grimme, S. PCCP 2011, 13, 6670. (15) Zhao, Y.; Truhlar, D. G. J. Chem. Theory Comput. 2011, 7, 669. (16) Montgomery, J. J. A.; Frisch, M. J.; Ochterski, J. W.; Petersson, G. A. J. Chem. Phys. 1999, 110, 2822. (17) Montgomery, J. J. A.; Frisch, M. J.; Ochterski, J. W.; Petersson, G. A. J. Chem. Phys. 2000, 112, 6532. (18) Guner, V.; Khuong, K. S.; Leach, A. G.; Lee, P. S.; Bartberger, M. D.; Houk, K. N. J. Phys. Chem. A 2003, 107, 11445. (19) Ess, D. H.; Houk, K. N. J. Phys. Chem. A 2005, 109, 9542. (20) Murray, J. S.; Brinck, T.; Politzer, P. Chem. Phys. 1996, 204, 289. (21) Politzer, P.; Murray, J. S. J. Phys. Chem. A 1998, 102, 1018. (22) Rice, B. M.; Pai, S. V.; Hare, J. Combust. Flame 1999, 118, 445. (23) Brinck, T. Hardsurf program. (HS95v09), 2009. (24) Rahm, M.; Dvinskikh, S. V.; Furó, I.; Brinck, T. Angew. Chem. Int. Ed. 2011, 50, 1145. (25) Stanton, J. F.; Bartlett, R. J. J. Chem. Phys. 1993, 98, 7029. (26) Kállay, M.; Gauss, J. J. Chem. Phys. 2004, 121, 9257. (27) Chai, J.-D.; Head-Gordon, M. PCCP 2008, 10, 6615. (28) Sorin, B.; Laurence, E. F.; Kurt, R. G.; Howard, W. M.; I-Feng, W. K.; Souers, P. C.; Vitello P. A., Cheetah 7.0, rev2280, Lawrence Livermore National Laboratory, 2012. (29) Bak, B.; Nicolaisen, F. M.; Nielsen, O. J.; Skaarup, S. J. Mol. Struct. 1979, 51, 17. (30) Christe, K. O.; Wilson, W. W.; Bélanger-Chabot, G.; Haiges, R.; Boatz, J. A.; Rahm, M.; Prakash, G. K. S.; Saal, T.; Hopfinger, M. Angew. Chem. Int. Ed. 2015, 54, 1316. (31) Klapötke, T. M.; McIntyre, G.; Schulz, A. J. Chem. Soc., Dalton Trans. 1996, 3237. (32) Klapötke, T. M.; Schulz, A. Inorg. Chem. 1996, 35, 7897. 308 APPENDIX 8: ADDITIONAL INFORMATION ON SYNTHESIS OF FLUORODINITRAMINE (CHAPTER 9) A8.1 Experimental Details Caution! Anhydrous HF can cause severe burns and contact with the skin must be avoided. Many of the materials described in this work are energetic and should be handled on a small scale while taking appropriate safety measures, such as wearing face shields, leather gloves and protective clothing, and working in a well-ventilated environment. All volatile materials were handled in either a stainless-steel/Teflon-FEP 1 or Pyrex-glass vacuum line with greaseless Teflon stopcocks. Solids were handled in the dry Ar or N 2 atmosphere of a glove box. HF was dried by storage over BiF 5 or TaF 5 . 2 Acetonitrile was dried by storage over P 4 O 10 and Linde 3Å molecular sieves and distilled prior to use. A literature method was used for the preparation of NF 4 SbF 6 3 and the sample of KN(NO 2 ) 2 was kindly donated by EURENCO Bofors. NMR spectra were recorded on a Bruker AMX 500 ( 14 N,  0 = 36.13 MHz) and on a Varian-400 spectrometer. Spectra were externally referenced to neat CH 3 NO 2 (  0 = 0.00 ppm). Raman spectra were recorded in 3 mm Pyrex tubes on a Cary Model 83 using the 4880 Å excitation line of an Ar-ion laser. Preparation of FN(NO 2 ) 2 . In a typical experiment, NF 4 SbF 6 (2.00 mmol) and KN(NO 2 ) 2 (2.00 mmol) were loaded in the drybox into a passivated ¼“ o.d. Teflon-FEP ampule closed by a stainless steel valve. The solvent (SO 2 or CH 3 CN) (2-5 mL) was added at -196 °C on the vacuum line and the mixture was warmed to the melting point of the solvent at which point NF 3 evolution began. The volatile products were separated by repeated fractional condensations through a series of -64, -80, -95, and -196 °C traps in a dynamic vacuum. The bulk of the desired FN(NO 2 ) 2 product was found in the -78 and -95 °C traps. The purity of the isolated materials was estimated by Raman and NMR measurements. Computational Details. Structure optimizations of NF 3 , F 2 N(NO 2 ), FN(NO 2 ) 2 , and N(NO 2 ) 3 in the gas phase were calculated using the hybrid meta exchange-correlation density functional M06-2X, the aug-cc-pVTZ basis set, and Gaussian 09, rev A02. 4 M06-2X 5 is a reliable general-purpose density functional theory (DFT) functional for main-group chemistry, with a mean absolute deviation of 2.2 kcal/mol, as demonstrated by several benchmarks. 6,7 The CBS-QB3 8 composite method was employed for calculating adiabatic bond dissociation energies. CBS-QB3 is based on CCSD(T) energies extrapolated to the basis set limit using MP2 and MP4 calculations together with empirical corrections, and is expected to 309 be highly reliable for thermochemistry. 8-10 Its mean absolute deviation in the G2 test set is reported to be 0.87 kcal/mol. 11,12 Harmonic frequencies were calculated at the mPW2PLYP/Def2-TZVPP level 11,12 of theory using the ORCA 3.0 code, with implicit consideration of CH 3 CN solution, as treated by the COSMO method. Raman intensities were calculated at the B3LYP/aug-cc-pVTZ level of theory, using Gaussian09’s standard implementation of the polarizable continuum model (PCM). HELP analyses 13 were performed using DGRID 4.6. 14 A8.2 Computational Method Description The CBS-QB3 8,15 composite method was employed for accurate gas-phase energies. CBS-QB3 is based on CCSD(T) energies extrapolated to the basis set limit using MP2 and MP4 calculations together with empirical corrections, and is expected to be highly reliable for thermochemistry. 8-10,15 For the vibrational analyses at the PCM-B3LYP/aug-cc-pVTZ level, numerical integration of all modes was done to obtain anharmonic frequencies, using Gaussian 09. 4 Harmonic frequencies were also calculated at the RIJCOSX- mPW2PLYP/Def2-TZVPP 11,12 level of theory using the ORCA 3.0 code, together with the auxiliary def2- TZVPP/J and def2-TZVPP/C basis sets for coulomb and semi-numeric exchange integration, and implicit consideration of CH3CN solution by the COSMO method. All numerical gradients were calculated with 0.001 bohr increments, DFT grid 5 and tightly converged SCF wave functions and geometries. HELP 13 and QTAIM 16 analyses of all species were done in gas-phase. Generation and analysis of the electron localization function, η(r), and the electron density from M06-2X/aug-cc-pVTZ wave functions were performed over a 0.026 Å grid in DGRID 4.6, 14 where η(r) was calculated using the spin- polarized definition of Kohout and Savin. 17 Natural Bond Orbital (NBO) analyses and Wiberg bond indices were calculated using NBO version 3.0 18-20 in Gaussian 09. A8.3 Bond Dissociation Energies Previous studies have clearly demonstrated that N-N bond breakage in nitramines (N-NO 2 ) preferentially occurs through homolytic bond dissociation. 21-25 When concerned with the free energy of activation in endothermic homolytic bond dissociation it is necessary to realize that the entropy which arises due to the additional rotational and translational degrees of freedom associated with a full dissociation, is largely gained after the dissociation barrier has been overcome. This has the practical consequence of making the free energy of dissociation smaller than the free energy of activation. Because of this, the bond 310 dissociation enthalpy can be considered a more reasonable upper-approximation to the factual free energy barrier. This is usually a good approximation when comparing calculated barriers with experimental observations. 24-27 The CBS-QB3 gas-phase estimates show a distinct difference in bond dissociation energies, in which N(NO 2 ) 3 > FN(NO 2 ) 2 > F 2 NNO 2 (Table A8.1). As expected, the reverse trend is true for bond lengths. Table A8.1: N-N bond distances (M06-2X/aug-cc-pVTZ) and bond dissociation enthalpies and free energies (CBS-QB3) for N(NO 2 ) 3 , FN(NO 2 ) 2 and F 2 NNO 2 . N-N distance (Å) ΔH 0 ΔG 0 N(NO2) 3 1.501 28.2 14.2 FN(NO2) 2 1.547 22.1 9.4 F 2 NNO 2 1.559 14.2 2.3 311 A8.4 Vibrational Analysis Table A8.2: Summary of calculated frequencies and Raman intensities for FN(NO 2 ) 2 , together with experimental solid-state results. Vibrational assign. in C s symmetry, Approx. mode description Obsd. freq., rel. Ra int. a PCM-B3LYP /aug-cc-pVTZ harmonic (analytic) a PCM-B3LYP /aug-cc-pVTZ anharmonic (numeric) COSMO- mPW2PLYP /Def2-TZVPP harmonic (numeric) COSMO- mPW2PLYP /Def2-TZVPP individually scaled c A’ ν 1 ν as NO 2 ip b ν 2 ν sym NO 2 ip 1743 [1] 1351 [3], 1335 [2], 1332 [14] 1743 [16] 1377 [100] 1696 1345 1762 1360 1715 1329 ν 3 ν NF ν 4 δ sciss NO 2 ip ν 5 ν sym N 3 ν 6 δ rock NF ν 7 δ sciss N 3 ν 8 δ rock NO 2 + δ rock NF ip ν 9 δ wag NO 2 ip ν 10  NO 2 ip A’’ ν 11 ν as NO 2 oop b ν 12 ν sym NO 2 oop ν 13 δ sciss NO 2 oop ν 14 ν asym N 3 ν 15 δ wag NF ν 16 δ rock NO 2 + δ rock NF oop ν 17 ν sym NO 2 oop ν 18  NO 2 oop lattice vibrations 1064 [1], 1054 [.5] 842 [2],825 [31] 794 [22] 615 [4], 606 [9] 430 sh, 425 [88] 329 [70] 207 [26], 200 [15] 77 [30] 1682 sh, 1680 [4], 1668[.5] 1254 [1], 1249 [3] 758 [3] 680 sh, 675 [5] 593 [9], 589 [9] 337 [10] 324 [100 Not obsd. 165 [20], 99 [20], 87 [8] 1066 [8] 831 [63] 807 [14] 606 [15] 417 [86] 310 [24] 187 [16] 46 [9] 1720 [44] 1284 [7] 749 [0] 686 [3] 604 [17] 332 [3] 273 [50] 46 [5] 1041 825 785 591 406 302 185 66 1677 1256 742 676 591 315 286 6i 1089 836 821 623 439 327 201 74 1737 1272 750 702 618 345 291 34 1064 831 798 607 428 318 198 59 1693 1244 743 692 604 328 306 33 a Frequencies in cm -1 ; uncorrected intensities based on peak heights in percent based on the most intense band being 100. b ip and oop stand for in phase and out of phase, respectively. c Each frequency was corrected using an individual scaling factor obtained from harmonic and anharmonic B3LYP calculations. A8.5 Bonding, Structure and Electron Density Analysis For straightforward comparison of the entire family of compounds, N(NO 2 ) 3 , FN(NO 2 ) 2 , F 2 NNO 2 and NF 3 , selected properties have been summarized in Table A8.3. For estimation of atomic charges Quantum Theory of Atoms in Molecules (QTAIM) approach was used, which allows for direct topological analysis of the electron density. The trend in QTAIM charges is mirrored by the Natural Population Analysis (NPA) charges, and predicts an increasingly positive value on the center nitrogen with fluorine substitution, as expected. To provide estimates to the localization and size of the sterically active lone pair domain on the center nitrogen, the High-ELF Localization domain Population (HELP) was utilized. HELP works as to provide a number of electrons that, according to a certain criterion (electron localization function, η(r) > ½), 13 can be considered localized. HELV is the corresponding 312 volume of this collection of electrons. Similar to the trend in N-N bond length, the largest change in the lone pair domain happens with the first fluorine substitution, which causes the lone pair domain to expand appreciably (1.73e in 2.8 Å 3 → 1.79e in 3.5 Å 3 ). The diffusivity of the lone pair domains can be approximated as HELP/HELV, which follows reasonably well the observed trend in pyramidilization angle and N-N bond lengths. Finally, the extent of covalent character of the nitrogen-nitrogen bonds was analyzed using Wiberg bond indices, 28 which were calculated from natural atomic orbitals (NAOs). As expected, the bond indices point to a gradual decrease of atomic orbital overlap between nitrogen atoms with increasing fluorine substitution. This can be explained as a direct consequence of longer nitrogen- nitrogen bonds (or vice versa). Table A8.3: Summary of calculated properties for N(NO 2 ) 3 , FN(NO 2 ) 2 , F 2 NNO 2 and NF 3 N(NO 2) 3 FN(NO 2) 2 F 2NNO 2 NF 3 N-N bond energy (ΔH, kcal-mol) 28.2 22.1 14.2 - N-N distance (Å) 1.501 1.547 1.559 - pyramidalization angle (°) a 107.9 104.4 103.2 101.9 QTAIM charge: center nitrogen -0.11 0.27 0.62 0.94 QTAIM charge: fluorine - -0.31 -0.32 -0.31 NPA charge center nitrogen -0.13 0.15 0.42 0.66 NPA charge fluorine - -0.19 -0.21 -0.22 HELP (e) nitrogen lone pair 1.73 1.79 1.73 1.69 HELV (Å 3 ) nitrogen lone pair 2.8 3.5 3.7 3.6 Wiberg bond index (N-N) 0.826 0.773 0.759 - a Given as average of the three bond angles around the triply bound nitrogen center. A8.6 References (1) Christe, K. O.; Wilson, W. W.; Schack, C. J.; Wilson, R. D.; Bougon, R. In Inorg. Synth.; John Wiley & Sons, Inc.: 2007, p 39. (2) Christe, K. O.; Wilson, W. W.; Schack, C. J. J. Fluorine Chem. 1978, 11, 71. 313 (3) Christe, K. O.; Schack, C. J.; Wilson, R. D. J. Fluorine Chem. 1976, 8, 541. (4) M. J. Frisch et al., Gaussian 09, Revision A.02, Gaussian, Inc., Wallingford CT, 2009. (5) Zhao, Y.; Truhlar, D. G. Theor. Chem. Acc. 2008, 120, 215. (6) Zhao, Y.; Truhlar, D. G. J. Chem. Theory Comput. 2011, 7, 669. (7) Goerigk, L.; Grimme, S. PCCP 2011, 13, 6670. (8) Montgomery, J. J. A.; Frisch, M. J.; Ochterski, J. W.; Petersson, G. A. J. Chem. Phys. 1999, 110, 2822. (9) Guner, V.; Khuong, K. S.; Leach, A. G.; Lee, P. S.; Bartberger, M. D.; Houk, K. N. J. Phys. Chem. A 2003, 107, 11445. (10) Ess, D. H.; Houk, K. N. J. Phys. Chem. A 2005, 109, 9542. (11) Weigend, F.; Ahlrichs, R. PCCP 2005, 7, 3297. (12) Schwabe, T.; Grimme, S. PCCP 2006, 8, 4398. (13) Rahm, M.; Christe, K. O. ChemPhysChem 2013, 14, 3714. (14) M. Kohout, Dgrid, version 4.6, 2011. (15) Montgomery, J. J. A.; Frisch, M. J.; Ochterski, J. W.; Petersson, G. A. J. Chem. Phys. 2000, 112, 6532. (16) Bader, R. F. W. Atoms in Molecules: A Quantum Theory; Clarendon Press: Oxford, 1990. (17) Kohout, M.; Savin, A. Int. J. Quantum Chem 1996, 60, 875. (18) Glendenig, E. D.; Landis, C. R.; Weinhold, F. Wiley Interdiscip. Rev.: Comput. Mol. Sci. 2012, 2, 1. (19) Reed, A. E.; Curtiss, L. A.; Weinhold, F. Chem. Rev. 1988, 88, 899. (20) Reed, A. E.; Weinstock, R. B.; Weinhold, F. J. Chem. Phys. 1985, 83, 735. (21) Rahm, M.; Brinck, T. Chem. Phys. 2008, 348, 53. (22) Rahm, M.; Brinck, T. Chem. Eur. J. 2010, 16, 6590. (23) Rahm, M.; Brinck, T. Chem. Commun. 2009, 2896. (24) Rahm, M.; Brinck, T. J. Phys. Chem. A 2010, 114, 2845. (25) Rahm, M.; Dvinskikh, S. V.; Furó, I.; Brinck, T. Angew. Chem. Int. Ed. 2011, 50, 1145. (26) Bélanger-Chabot, G.; Rahm, M.; Haiges, R.; Christe, K. O. Angew. Chem. Int. Ed. 2013, 52, 11002. (27) Rahm, M.; Bélanger-Chabot, G.; Haiges, R.; Christe, K. O. Angew. Chem. Int. Ed. 2014, 53, 6893. (28) Wiberg, K. B. Tetrahedron 1968, 24, 1083.

Abstract (if available)

Abstract There is an increasingly pressing need for new, high-performance, environmentally benign energetic materials. Indeed, the state-of-the-art explosives and propellants are causing growing concern over their detrimental impact on the environment and human health. It is extremely challenging to find alternatives to the compounds currently in use, which are high performing, relatively insensitive and inexpensive to synthesize. This challenge requires the synthesis of new types of compounds. This dissertation summarizes my contribution to this effort. The approach of the Christe-Haiges research group to the synthesis of new energetic materials involves the targeting of groundbreaking, novel, highly challenging species, the development of efficient synthetic pathways and the characterization of these species. ❧ Chapters 2-3 deal with derivatives of 3,5-dinitro-1,2,4-triazole, an energetic building block with suitable properties for potential applications. In CHAPTER 2 the improved synthesis of 3,5-dinitro-1H-1,2,4-triazole (HDNT) and the first structural characterizations of that compound, which had been used in situ for decades, are described. This chapter also discusses the identifcation and structural characterization of common impurities formed when literature synthetic methods are followed. Some of these impurities were found to be dangerously more sensitive than the targetted HDNT. CHAPTER 3 describes the synthesis and structural characterization of over twenty 3,5-dinitro-1,2,4-triazolate (DNT) salts and polymorphs thereof, most of which are highly thermally stable and insensitive. ❧ CHAPTER 4-CHAPTER 7 describe the syntheses of hydro‐borane and ‐borate compounds with energetic ligands. Halogen‐free boron‐based energetic materials are promising compounds because of their environmentally benign combustion products and because of the high heat of formation of boron oxide. CHAPTER 4 describes the synthesis of salts of nitroazolate‐trihydroborate complex anions using, among several nitroazolates, some of the precursors described in CHAPTER 2 and CHAPTER 3. CHAPTER 5 describes the synthesis and full characterization of the first room temperature‐stable solid salts of trinitromethyl‐trihydroborate. The compounds demonstrate that boron hydrides can form stable derivatives even with highly oxidizing moieties. CHAPTER 6 describes the synthesis of the dinitramide‐substituted ammonia‐borane NH₃.BH₂[N(NO₂)₂], another example of a surprisingly stable energetic boron compound and the first structurally characterized Group 13‐dinitramide compound. CHAPTER 7 describes the observation of several dinitramide‐hydroborates, which demonstrates the potential use of dinitramide as a weakly coordinating anion. ❧ CHAPTER 8 describes the synthesis and characterization of nitryl cyanide, a small molecule which had long eluded synthesis. This species is another spectacular example of the combination of a strong oxidizer and a reducing group. The energy density of this molecule is extremely high and the compound could eventually serve as a replacement for hydrazine as a monopropellant. ❧ CHAPTER 9 describes the synthesis of fluorodinitramide, a particularly reactive and marginally stable compound which was characterized by low‐temperature NMR and vibrational spectroscopy. The study of the compound provided insight into the stability of the intriguing fluoronitroamine family.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

New synthetic methods for organonitrogen compounds

PDF

Towards selective fluoroalkylating reactions: synthesis and mechanistic studies

PDF

Hydrogen fluoride addition reactions and a fluorinated cathode catalyst support material for fuel cells

PDF

Molecular dynamics study of energetic materials

PDF

Synthesis of multifunctional heterocycles, amino phosphontes using boronic acids

PDF

Novel fluoroalkylation reactions and microwave-assisted methodologies

PDF

Analytical investigation of the proteasome inhibitor Bortezomib and the total synthesis of specialized pro-resolving lipid mediators

PDF

Synthesis, characterization and reaction chemistry of polyazides and cyanometallates

PDF

Carbocations and superelectrophilic intermediates

PDF

On the pursuit of pushing the limits: from the isolation of highly reactive intermediates to the synthesis of energetic materials

PDF

Development of sulfone-based nucleophilic fluoromethylating reagents and related methodologies

PDF

Stereoselective nucleophilic fluoromethylations: from methodology to mechanistic studies

PDF

New bifunctional catalysts for ammonia-borane dehydrogenation, nitrile reduction, formic acid dehydrogenation, lactic acid synthesis, and carbon dioxide reduction

PDF

Influence of polymer structure on electronic properties and performance as donor materials in bulk heterojunction solar cells

PDF

Total synthesis of specialized pro-resolving lipid mediators and their analogs

PDF

Improving the sustainability of conjugated polymer synthesis via direct arylation polymerization

PDF

Investigations in cooperative catalysis: synthesis, reactivity and metal-ligand bonding

PDF

Development of new bifunctional iridium complexes for hydrogenation and dehydrogenation reactions

PDF

Harnessing fluorinated C1 nucleophilic reagents for the direct fluoroalkylation of ubiquitous C(sp2)-X and C(sp)-H centers

PDF

Design and synthesis of helicene-based macromolecules

Asset Metadata

Creator Bélanger-Chabot, Guillaume (author)

Core Title Towards groundbreaking green energetic materials

School College of Letters, Arts and Sciences

Degree Doctor of Philosophy

Degree Program Chemistry

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag boron chemistry,dinitramide,energetic materials,nitryl cyanide,OAI-PMH Harvest,tetrazoles,triazoles,trinitromethanide

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Prakash, G. K. Surya (committee chair), Christe, Karl O. (committee member), Vashishta, Priya (committee member)

Creator Email gbchabot@gmail.com,gbelange@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c3-589668

Unique identifier UC11300955

Identifier etd-BlangerCha-3566.pdf (filename),usctheses-c3-589668 (legacy record id)

Legacy Identifier etd-BlangerCha-3566.pdf

Dmrecord 589668

Document Type Dissertation

Format application/pdf (imt)

Rights Bélanger-Chabot, Guillaume; Belanger-Chabot, Guillaume

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA