University of Southern California Dissertations and Theses

The photocycloaddition of acetone to ketenimines and the syn-anti isomerization of the B-adducts

(USC Thesis Other)

The photocycloaddition of acetone to ketenimines and the syn-anti isomerization of the B-adducts

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content THE PHOTOCYCLOADDITION OF ACETONE TO KETENIMINES i t AND THE SYN-ANTI ISOMERIZATION OF THE g-ADDUCTS by Roger Lief Knutsen in A Thesis Presented to the FACULTY OF THE GRADUATE SCHOOL UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment ©f the Requirements for the Degree MASTER OF SCIENCE (Chemistry) August 1970 UMI Number: EP41652 All rights reserved INFORMATION TO ALL USERS The quality of this reproduction is dependent upon the quality of the copy submitted. In the unlikely event that the author did not send a complete manuscript and there are missing pages, these will be noted. Also, if material had to be removed, a note will indicate the deletion. Dissertation Publishing UMI EP41652 Published by ProQuest LLC (2014). Copyright in the Dissertation held by the Author. Microform Edition © ProQuest LLC. All rights reserved. This work is protected against unauthorized copying under Title 17, United States Code ProQuest LLC. 789 East Eisenhower Parkway P.O. Box 1346 Ann Arbor, Ml 4 8 1 0 6 - 1346 U N IVE R SITY O F S O U TH E R N C A L IF O R N IA TH E G RA DU ATE SC HO O L U N IV E R S IT Y PARK LOS A N G ELES. C A L IF O R N IA 9 0 0 0 7 £ I 7-2-1 C T ’ A i ’j thesis, written by Roger Lief Knutsen under the direction of h.xs— Thesis Committee, and approved by all its members, has been pre sented to and accepted by the Dean of The Graduate School, in partial fulfillm ent of the requirements fo r the degree of Master of Science Dean THESIS COMMITTEE ACKNOWLEDGEMENTS I wish to express my deep appreciation for the encouragement and guidance given by Dr. Lawrence A. Singer, who initiated and directed this study. I also wish to thank Dr. Kenneth L. Servis for his patient instruction and direction in the variable temperature nmr portion of this work. A traineeship for the 1968-1970 academic years provided by the National Science Foundation is gratefully acknowledged. TABLE OF CONTENTS Page ACKNOWLEDGEMENTS .............................. ii LIST OF TABLES . . .. .. ... ... ... ... . iv LIST OF FIGURES .......... • • -vi INTRODUCTION ........... ........ .... 1 RESULTS .. .. ........... ............. 13 Preparation of Ketenimines Photbcycloaddition Reactions Other Imines Photolysis of Acetone Variable Temperature NMR Study DISCUSSION . . .; . . . ..... .. .. .. ... . 32 Ketenimines Photoreactions Preparation of Imines Variable Temperature NMR Study EXPERIMENTAL ......................... . . 48 Analytical Instruments Reagents Preparation of Ketenimines Preparation of Imines Photoreactions Variable Temperature NMR SUMMARY . ......................... 68 FOOTNOTES ...... 69 iii LIST OF TABLES I Table I. Barriers to inversion in selected amines | ! Table II. The barrier to inversion about nitrogen | in benzylmethylmethoxyamine, as a function of solvent dielectric constant. I I Table III. Spectral data for dimethyl-N-(phenyl)- ketenimine 3 and dimethyl-N-(cyelohexyl)- i ketenimine 4. AV : Table IV. Yields and boiling points for dimethyl- N-(phenyl)ketenimine g and dimethyl-N- (cyelohexyl)ketenimine 4. ; Table V. Spectral and physical data for photoadducts j I Table VI. Material balance in the photocycloaddition I of acetone to 3 and 4. /w ft i j Table VII. Physical data for the compound 11 I recovered from the reaction of acetoneT with ! ketenimine 3. J ^ j Table VIII. Spectral data for imine 12 and acetone I anil perchlorate 13. ! i Table IX. Spectral data of the crude mixture from | the photolysis of acetone. j Table X. Summary of the nmr data for simple ketones j and diols which may result from acetone photolysis. j Table XI. Summary of the retention times in the vpe ! analysis of the acetone photolysis mixture. 1 i Table XII. The data used for matching observed and theoretical spectra, and for determining for ; each temperature, in the case of 2,.2,4,4-tetra- ! methyl-N-(phenyl)iminoxetane 9. ■ Table XIII. Thermodynamic data for the nitrogen inversion in imines 9, 10, and 12. /*v iv Page 7 12 15 15 17 18 19 21 23 24 25 30 31 Page Table XIV. Barriers to inversion in selected imines. 46 v LIST OF FIGURES Page Figure I. Observed and calculated nmr spectra for imine % in p-dichlorobenzene at various temperatures (0C)f and with the corresponding mean lifetimes Or) for inversion. 27 Figure II. Observed nmr spectra for imine 10 in p-dichlorobenzene at various temperatures (°C). 29 Figure III. A plot of the log 1/r vs 1/T for the inversion about nitrogen in imine 9. 43 INTRODUCTION The purpose ©f this project has been two-fold. First, the photocycl©addition reactions of acetone with dimethyl- N-(phenyl)ketenimine and dimethyl-N-(cyclohexyl)ketenimine have been investigated. Second, the svn-anti isomerization about the imine bond in the B-iminoxetanes formed have been analyzed. The following discussion is directed, then, I i to the subject of photoeycloaddition reactions as well as | the question of isomerization of the geometrical config uration about nitrogen in amines and imines. 1 2 Since Patemo and Btlchi first described the light- eatalyzed addition of aldehydes and ketones to olefins to form oxetanes 1, photoeycloaddition reactions have been of R, R C=0 + r6 R5 + / lb / . increasing interest and have been extended to include a 3 wide range of reactions. Two considerations have been i i shown to be important in determining the reactivities in j these reactions. First, the nature of the reacting species j j 2 is dependent on the ketone or aldehyde involved. Thus the reaction of aliphatic ketones with olefins has been shown 4 to proceed via the n,rr* singlet state of the ketone, while reactive aromatic ketones have been shown to react 5 6 via the n,n* triplet state. * Furthermore, ketones with low-lying tt,tt* triplet states have been found to be much less reactive than those with n,TT* triplet states,^ paralleling the reactivities found in the photoreduction of the ketones.^ Second, if the triplet energy level of the carbonyl compound is above that of the olefin, energy transfer from the triplet carbonyl to the ground state olefin will occur with the result that there is no oxetane formation.^ As an extension of the Patemo-Btichi reaction Singer g and Bartlett demonstrated that aldehydes and ketones would add to ketenimines to give a and 3 iminoxetanes 2. R-, R + X R3^ ^C=0 + O O N - R 5 > R R2 R4 2a 23 9,10 Singer and Davis have extended this reaction to 3 include, in the cases of benzophenone and fluorenone, reactions with a number of aliphatically and aromatically substituted ketenimines. The two considerations which were found to be important in the reactions of ketones and aldehydes with olefins were also found to be important in these reactions. First, ketones with low-lying n,TT* triplet states were more reactive than those with t t , t t * triplet states, as demonstrated by the decreased reactiv ity of 2-acetonaphthone and 1-naphthaldehyde toward keten- g imines. Second, energy transfer from the triplet ketone . to the triplet energy level of the ketenimine was compet- , itive with cycloaddition when the triplet energy level of the ketenimine was below that of the ketone. That energy transfer was, in fact, important was demonstrated by the i ketenimine quenching of the benzophenone photoreduction 9 reaction. This energy transfer process was increasingly important with increasing aryl substitution on the keten imine which resulted in the lowering of the triplet energy level of the ketenimine. Finally, while these reactions were originally thought to proceed via the triplet state of the carbonyl, more recently, evidence has been found for the formation of a cycloadduct via a fluorenone singlet-ketenimine complex in the reaction of fluorenone with dimethyl-N - ( cyclohexyl )ketenimine. * * 4 Throughout the studies of the photocycloadditions of ketones to ketenimines the structural question as to svn- anti isomerization about the imine bond had not been settled. This question has been examined by Curtin and his 12 13 co-workers * in the case of open-chain imines. Using both IR and nuclear magnetic resonance (nmr) data, they concluded that svn and anti isomers of triarylimines^ and 13 N-alkylimines of benzophenones were present in solution at room temperature. The activation energy barriers to inversion in the triarylimines were on the order of 17-20 kcal/mole while for the N-alkylimines the barriers were on the order of 25-27 kcal/mole. In addition, the svn and anti isomers of N-haloimines and oxime ethers were iso lated and found to be quite stable with respect to isomerization about the imine bond. These results suggested two tentative generalizations regarding inversion about the imine bond. First, the heteroatom had a marked effect in stabilizing the config- * ■ uration about nitrogen, permitting the isolation of svn and anti isomers. Second, the N-aryl imines underwent isomerization at a much faster rate than N-alkyl imines. The reasons for these two observations will be discussed later. 14 More recently, Woman and Schmidt have reported the ^.geometrical isomerization of bisimines of 2,2,4,4-tetra- 5 methyleyclobutadione. Coalescence temperatures of 100- 165**C were observed in these compounds, indicating higher than average activation barriers to inversion. Staab and co-worker s*~* have reported a barrier of - 21 kcal/mole for acetone anil in quinoline, with the coalescence temper ature being lAO'C. This, too, is a relatively high barrier to inversion when compared to the compounds discussed below. In related work, the question of the rapid inversion of the geometrical configuration about nitrogen in ammonia, amines, and related compounds has also received a great deal of interest in recent years. Due to the very rapid inversion in ammonia and most amines it has been impossible to investigate the kinetic parameters of the inversion process. Thus a great deal of attention has been directed to finding means by which the nitrogen pyramid could be stabilized, or at least the inversion process slowed considerably. Brois has had notable success in isolating stable pyramids, finding that substitution of an -NHL, group^ or halogens^ on the nitrogen in aziridines stabilized the configuration about the nitrogen. Felix and Eschenmoser18 have prepared stable s^n and anti isomers of T-chloro-T-azabieyclo/2 ?. 1. OJheptane. Recently a wide variety of compounds have been found in which the rates of inversion are intermittent between the very rapid 6 and the completely rigid, and which lend themselves to a study of the thermodynamic parameters for the inversion ; process. p A variety of methods have been tried in order to determine rates of inversion about nitrogen. Techniques 13 19 such as IR or pH studies have not had more than very limited usei largely due to the fact that results were not very quantitative. Variable temperature nmr techniques are by far the most commonly used, even though much of the work which has been done has been of limited benefit when only the rates of inversion, at a wide range of tempera tures, were reported and no thermodynamic data deter- • 20 mined. A further limitation on the use of nmr has been the fact that, for one reason or another, some compounds have not lent themselves to easy or precise analysis, and thus only qualitative estimates of the energy barriers to 21 inversion have been made. 22 Using techniques outlined by Roberts, namely, line shape analyses of nmr spectra over a wide range of temper atures, several groups have been able to report the thermodynamic data for the nitrogen inversion process in a variety of compounds. A sampling of some of this work is presented in Table I, and will serve to illustrate several of the important factors which influence the rate of inversion about nitrogen. !Table I. Barriers to inversion in selected amines. Compound Solvent c •c ag: kcal/mole kcal/mole Reference i (0ch2)2nch3 N-CH, CH2CHC1 HCC1F. -146 -125 6.G + 0.5 23 24 III N-CHL HCC1F, -100 8 24 IV N-CH< FCC1, -98 8.85 25 ;v N-Cl FCC1, vi (0ch2)ch3noch3 hcci3 -54 -26 11.5 11.7 + 0.4 12.2 25 26 Table I. Continued Compound Solvent e *C a c kcal/mole kcal/mole Reference VII VIII 1J5-CH3 f CE3 IX C13CSN(CH20)2 X CH CH XI N-CH, Cl-N CH2C12 CH2C12 HCC1, CC1, 42 28 13.7 + 0.5 15.6 + 0.5 15.3 + 0.2 DCC1, 67.5 32.50+1.31 18.9 3 — 23.5 27 27 28 29 30 a. Calculated by this author. oo The effect of the strongly electronegative heteroatom, | when attached to nitrogen, in reducing the rate of j i inversion is seen when amines IV and V are compared. The substitution of the chlorine atom for the methyl group clearly raises the barrier to inversion. Amine XI provides • the extreme example of a very high barrier to inversion, ; ' 31 This effect is believed to be due to the fact that a ( strongly electronegative substituent on nitrogen tends to increase the s-character of the lone pair of electrons on j ' > ‘ nitrogen. Since the lone pair must transfer to a p-orbital i ; in the planer transition state required for inversion, this ■ greater s-character of the electron pair would inhibit inversion. : A second effect caused by a heteroatom is illustrated ; by comparing amines I and IX. In the planer transition ; state the nonbonding electron pairs on sulpher will be repulsed by the electron pair on nitrogen, more so than in ' i I the ground state. This greater electrostatic repulsion in i 1 • the transition state would reduce the rate of inversion, | as is observed. Howeveri the state of the art is not currently such that this electrostatic effect can be separated from electronegative effects. Thus, while some 32 have found some evidence for this electrostatic effect, just how important this effect is on the inversion process - is not .clearly understood at this time. The role steric effects play in governing the rate of inversion is illustrated by a comparison of amines II and III, as well as amines VII and VIII. In going from the five-membered rings to larger rings the transition is from a condition of less to more interaction in the ground state of the N-methyl group with the ring protons. Thus the greater steric crowding in the larger ring compounds contributes to slightly increased rates of inversion and consequent lower barriers to inversion. This steric influence on rates of inversion has been discussed by 31 Brois, who finds a high rate of inversion for N-t-butyl- aziridine compared to other aziridines. However, the accurate measurements of inversion rates leading to the thermodynamic data necessary for a definitive discussion of the subtle differences in steric requirements between these sorts of compounds are not yet in hand. A final influence on rates of inversion which is not ! illustrated by these compounds but which has been discussed 31 33 by Brois and others is the conjugative effect. This involves the interaction of the nitrogen lone pair and adjacent n-electronsi' such as in a benzene ring. Since this sort of overlap would be greatest in the planer transition state, this effect would serve to increase the rate of inversion. Again, the problem is that accurate thermo dynamic data are not available to enable one to make """11 1 quantitative statements regarding the extent of the effect this conjugative factor may have on rates of inversion. There is an important consideration to be kept in mind when comparing barriers to inversion and this is the effect of the solventa problem observed in several instances.^*^*^ The results of Griffith and Roberts,^ reproduced in Table II, serve to illustrate the importance of the effect of the solvent on the rate of inversion. They found that the activation barrier to inversion was inversely proportional to the solvent dielectric constant, i.e., the more polar the solvent, the lower the barrier to inversion. The point to be made is that when comparisons of barriers to inversion are to be made the same or very similar solvents must have been used in order that valid comparisons can be made. In this project, then, interest has been directed to the photoeycloaddition reactions of acetone, as an ali phatic ketone, with dimethyl-N-(phenyl)ketenimine and dimethyl-N-(cyclohexyl)ketenimine. In addition, the svn- anti isomerization about the imine bond in the P-iminoxetanes have been analyzed using the line shape analysis of the variable temperature spectra. The results of these investigations will now be discussed. Table II. The barrier to inversion about nitrogen in benzylmethylmethoxyamine, as a function of solvent 26 dielectric constant. Solvent ca T c °C Ea kcal/mole Log A n-Hexane 1.89 -16 12.9 + 0.3 12.8 + 0.3 cs2 2.64 -27 12.4 + 0.5 12.7 + 0.4 hcci3 4.81 -26 11.7 + 0.4 11.8 + 0.4 h2cci2 9.08 -34 9.4 + 0,4 10.0 + 0.4 Acetone 20.7 b • • • Methanol 32.6 b • • • a. Solvent dielectric constant. b. Coalescence not reached above -70°C. RESULTS Preparation of Ketenimines. The dimethyl-N-(phenyl)- i I s J and dimethyl-N-(cyelohexyl)ketenimines were prepared by | the dehydration of the corresponding amides using a | phosphorus pentoxide-Florisil slurry in triethylaminef I under an inert atmosphere, and at reflux temperatures. The i j amides were obtained by the reaction of a two-fold excess | of either aniline or cyclohexylamine with iso-butyryl- ■ chloride. The acid chloride had in turn been prepared by the treatment of iso-butyric acid with an excess of thionyl chloride. This reaction sequence is summarized j as follows; CH, XCHC0oH / 2 CH, S0C1, CH, \ CH, / CHCGC1 H2NR CH, \ / CHCONHR CH, PzOs/Florisil CH, \ C=C=N-R I The spectral data for the ketenimines are listed in f i ' j j Table Illi while the yields and boiling points are. | summarized in Table IV. i I Photoeycloaddition Reactions. The photoeycloaddition j f • ' i ; ! reactions of acetone with the ketenimines were carried out i i . ; ■ in sealedi degassed Vycor tubes using a 450 w Hanovia lamp ! i ' i I as the light source. The reactions may be summarized as i j j follows: | | 3: R=C6H5 5: R=C6H5 9s R=C6H5 j 4 s R=C^H11 6: R=C<cH1 n 10 s R-O.H., ! * • 6 11 V- 6 11 6 11 ’ ■j I Ketenimine 3 reacted much more slowly than ketenimine 4, ! : - ■ | | requiring 30 hours of irradiation before the ketenimine i • • .: j was consumed. Ketenimine 4 was usually consumed within j | 10-12 hours of irradiation. ; ! The a-adduct 6 was not isolated from the product ! ! i ! . I ! mixture. Insteadi the a-adduct from the acetone-4 reaction : j had apparently rearranged to form the unsaturated amide 8, as follows: ! 15 Table III. Spectral data for dimethyl-N-(phenyl)ketenimine 3 and dimethy1-N-(cyclohexyl)ketenimine 4. (CCl^) Ketenimine IR (cm” ) NMR (PPM 6) Ketenimine Band Methyl H Aromatic H Cyclohexyl H 3 2026 1.18 7.34 .... 4 2035 " 1.58 .... 0•83-2•17 Table IV. Yields and boiling points for dimethyl-N- (phenyl)ketenimine 3 and dimethyl-N-(cyclohexyl)- ketenimine 4. A* Ketenimine Yield3 Boiling Point 79 pC/mm 3 M 18 35.5/0.2 4 24.4 58/1.65 a. Yields calculated from the corresponding amides. The spectral evidence for the unsaturated amide 8 is /r/ included in Table V, along with the spectral and physical data for all the photoadduets. Neither the a-adduet 5, nor the unsaturated amide 7 were found in the product mixture /w from the acetone-3 reaction. The material balance for these reactions is summarized in Table VI. In the reaction of acetone with ketenimine 3 another /W compound 11 was recovered from the reaction mixture, and its physical characteristics are summarized in Table VII. Since the data indicated that this compound was the keten imine trimer, attempts were made to prepare it from the ketenimine 3 and its dimer. These efforts were not successful. Other Imines. As part of our investigation of the inversion about the imine nitrogen, the syntheses of other Table V. Spectral and physical data for photoadducts. (CCl^) Compound IR? cm” NMRb Chemical Shift (PPM 6) Relative Abundance Melting Point °C 8 A * / 1672(s)(C=0) 1.20s»1.70s (methyl) 6.3 71-76 1633(w)(C=C) 0.75-1.93 (cyclohexyl) 11.7 4.73-4.87m (vinyl) 2 3.55m (N-H) 1.25 9 0 / 1740(s)(C=N) 1.24s,1.49s (methyl) c i i • 6.55-7.33m (aromatic) 10 A M / 1728(s)(C=N) 1.35s,1.42s (methyl) 54-56 0.70-1.98m (cyclohexyl) a. Notation used: (s), strong; (w), weak. b. Tetraraethylsilane used as the internal standard. Notation used: ra, multiplet; s, singlet. c. Liquid at room temperature. ; Table VI. Material balance in the photocycloaddition of ! | acetone to 3 and 4. System Photolysis Times Mole Reaction % in Productsa Total Ketenimine Accounted for 3 5 7 2 / W f+S Acetone-3 30 hr 15.0 0 0 41.1 56.1 % 4 6 8 10 Acetone-4 10 hr 33.3 0 27.8 29.4 90.5 % ' a. Based on starting ketenimine. Table VII. Physical data for the compound 11 recovered from the reaction of acetone with ketenimine 3. (CC1,) M.W. Melting Point r?i cm NMR (PPM 6)a °C Methyl H Aromatic H 435 163-165 1600,1675 1.44 6.72-7.23 ; ' a. Tetramethylsilane used as the internal standard. 20 imines were carried out. 2,2,4»4-Tetramethyl-N-(phenyl)- 3-iminocyclobutanone 12 was prepared by the Schiff*s-base reaction of aniline with tetramethylcyclobuta-1,3-dione in 23 % yield. The spectral data for imine 12 are summarized in Table VIII. An attempt was made to prepare acetone anil by the treatment of the acetone anil perchlorate 13 with triethyl- amine. The crude product, upon examination by nmr, contained two methyl hydrogen absorptions and aromatic hydrogens, as well as amine protons. The integrated spectrum indicated that a yield of the desired imine of approximately 64 % was obtained. Purification of the i product by distillation resulted in decomposition of much of the material, with no improvement in the degree of purity of the distillate. A chromatography on neutral , alumina also was not successful in enhancing the purity of the adduct. Spectral data of the pure adduct were therefore 35 not obtained. The acetone anil perchlorate had been prepared by the reaction of acetone with aniline perchlor ate, which in turn had been prepared by the treatment of aniline with perchloric acid. The spectral data for the acetone anil perchlorate 13 are recorded in Table VIII. A V Photolysis of Acetone. In an effort to understand better some of the non-ketenimine side products from the photoreactions, the photolysis of pure acetone was investi- Table VIII. Spectral data for imine 12 and acetone anil perchlorate 13. Compound cm” NMRb PPM 6 Relative Abundance 12° 1680(s)(C=N) 1.11sj1.38s (methyl) 6.3:6.3 . 1800(s)(C=0) 6.58-7.43m (aromatic) 5 13d /vV 1665(s)(C=N) 2.25s,2.58s (methyl) 3:3 7.30-7.55m (aromat i e) 5 1.94s 0.5 a. Notation used: (s), strong. b. Tetramethylsilane used as the internal standard. Notation useds s, singlet? m, multiplet. c. Spectra in CCl^. d. Spectra in B^CC^. 22 gated. The photolysis was carried out in sealed, degassed Vyeor tubes, using a 450 w Hanovia lamp, for thirty hours. The spectral data for the crude product mixture are recorded in Table IX. In Table X are recorded the nrar data . i I for several compounds which have been reported to result o/: from the photolysis of acetone. The crude mixture was analyzed by vapor-phase chromatography. The retention times for known compounds and the crude product mixture are all recorded in Table XI. Variable Temperature NMR Study. The nmr spectra of imines 9, 10, and 12 were recorded over a wide range of f * r * * /Vv ' ” temperatures. Sample experimental curves for imines 9 and 10 are shown in Figures I and II, respectively. Theoret- ical curves were then calculated by means of a computer 37 program and these were matched with the observed curves, 22 using the line shape analysis technique. The theoretical curves for imine 9, are included in Figure I. Representative data used in the matching of these theoretical and observed curves in the case of imine j? are given in Table XII. In the case of imine 10 it was not possible to match theoret- i ical curves with the observed curves due to the overlap of the cyelohexyl absorption with the methyl absorptions. The thermodynamic data for the inversion about the imine bond in the three imines are summarized in Table XIII, I Table IX. Spectral data for the crude mixture from the photolysis of acetone. (CCl^) *5? cm A NMRb PPM 6 • 3425(b) 1.15s 4.2q 0.95-1.6° 1725(s) 1.22s 1.24s 1.35s 2.08s 2.57s 2.63s a. Notation used: (b), broad; (s), strong. ~ b. Notation used: s, singletj q, quartet. Tetramethyl- silane used as the internal standard, c. Broad shoulder. ■ Table X. Summary of the nmr data for simple ketones and diols which may result from i. ■ | acetone photolysis. ;Compound H3C-C -ch2- -CH- 0- (PPM 6)a -0-H H-C= Acetone* 3 2.17 Acetone0 2.05 Biaeetyl* 3 2.33 2,4Pentadione & Enolc d 2.12, 1.97 2,5Hexadione° 2.03 iso-Propanolc 1.07 & 1.16 2,3-Butadiol 1.13 3.58 3.58 3.58-4.15 ca 3.8 4.98&5.05 3.05 5.50 a. In CCl^, with tetramethylsilane as an internal standard. b. Varian Associates NMR Spectra Catalog. c. This author's results. d. Chemical shift of the methyl group attached to the enolic carbon. I & | Table XI. Summary of the retention times in the vpe analysis of the acetone photolysis mixture. Ketones RT Ketones RTb Photolysis Mixture Acetone 2:05 2:20s Biacetyl 2 , : . 40 4:20m 4: 50w 7: 50w 2,4-Pentadione 9i25 10:OOw 10:50w 13:OOw 14:20w 16:45w 17iOOw 2,5-Hexadione 18:00 18:00s a. The time is listed in minutes and seconds. b. Notation used! s, strong; m, medium; w, weak. Figure I. Observed and calculated nmr spectra for imine 9 in p-dichlorobenzene at various temperatures (°C), and with the corresponding mean lifetimes Or) for inversion. 26 T» II5*C T (SECS.) 000414 T«97»C 0.01410 Q0I930 T»89® 0.02693 T* 60* C 0.2321 Figure II. Observed nmr spectra for imine 10 in p-dichloro- benzene at various temperatures (°C). 28 | t -------------- V a 0 0 1 --- i 0 >09 *1 0 o02l >1 ' 6Z 30 , Table XII. The data used for matching observed and theoret ical spectra# and for determining nr for each temperature, in the case of 2,2,4,4-tetramethyl-N^(phenyl)iminoxetane 9. ' Calculated Observed _______ r(sec) Peak/Valley 1/2-width T<°C) Peak/Valley 1/2-width 0.2321 81.499 60 81.5 0.04807 4.408 80 4.403 0.0344 2.432 85 2.434 0.0299 1.925 87 1.925 0.02693 1.634 89 1.631 0.0219 1.231 90 1.218 0.02125 1.188 91 1.189 0.01961 1.095 92 1.101 0.0193 1.080 93 1.085 0.01535 25.72 95 25.75 0.0141 23.24 97 23.22 0.01023 15.02 100 15.04 0.00732 9.81 105 9.825 0.00414 5.25 115 5.26 f 31 | Table XIII. Thermodynamic data for the nitrogen inversion j ; in imines 9, 10 f and 12. /A I CH CH CH CH CH CH CH CH CH CH 9 r * 10 / V W ' 12 AA/ Imine m£l e °c Ea kcal/moleb AH* kcal/mole r4 = a Gc kcal/mole c e.u. 9 94 19.0 + 0.3 18.3 ± 0.3 18.7 + 0.3 -1.1 + 0.35 10 /W 141 21.3 21.8 (-i.i)c 12 114 22.8 + 2.7 22.0 + 2.6 19.8 + 2.3 5.68 a. Coalescence temperature. b. The uncertainties are root-mean-square deviations, e. Assumed to be the same as for imine 9. i DISCUSSION | i I Ketenimines. Modest improvements in the yields of the ; | 9 1 ketenimines 3 and 4k over those reported by prior workers | were realized. This was due to an improved technique in | I preparing the phosphorus pentoxide-Florisil dehydrating slurry, that is, the phosphorus pentoxide was added to the triethylamine slurry after the Florisil, instead of mixing the two solids prior to addition to the solution. The benefit of this improved dehydrating slurry was that it made possible a drastic reduction in reaction times which l resulted in improved yields, since long reaction times lead to very low yields. Photoreactions. Photoeycloaddition reactions involving! ketenimines have now been extended to include the reaction of an aliphatic ketone with ketenimines. While it has been beyond the scope of this investigation to determine the I exact nature of the reacting species in these cycloaddition ■ i reactions, there is some important recent work whieh ! relates to this point. As discussed earlier, Singer and co-workers^ have noted the reaction of a ketone singlet ! ■ . ■ • i with a ketenimine, in the ease of the fluorenone-dimethyl- j N-(cyclohexyl)ketenimine reaction, providing a precedent ; ! '33 1 I I ifor the reaction of ketones, which have long-lived singlets i . ■ ■ . ! » . |with ketenimines by either singlet or triplet pathways. But' |more pertinent to this discussion is the work done by Turro: !and his group on the cyeloadditions of acetone to j I 38 ' j1,2-dicyanoethylene (DCE) and to 1 -methoxy-1-butene | 39 ! {(MB). j ; I | In the reaction of acetone with the electron-deficient' | olefin DCE it was proposed that the reaction proceeded by aj ! I i "quasi" concerted nucleophilic attack of singlet acetone, j | Quenching studies indicated that there was no triplet j | reaction. Gn the other hand, the reaction of acetone with J | the electron-rich olefin MB was found to proceed by the j ; reaction of both singlet and triplet acetone, as indicated > I by the observations that (a) in the absence of quenchers . the reaction was non-stereospecific, i.e., the stereo chemistry about the olefin was not preserved upon cyclo- ! i ■ j addition; and (b) in the presence of triplet quenchers the j j cyeloaddition reaction was repressed only partially and, | ! . i I | further, the stereochemistry about the olefinic bond was I ' I ; preserved. | I Thus, in the reaction of acetone with MB the reaction I ; was proceeding by both the singlet and triplet states of j ; acetone, and the chemistry of each intermediate was j i different. These observations strongly suggest that the ! reaction of acetone with ketenimine 3 and 4 may well he a* * proceeding by both the triplet and singlet pathways. The observation that the reaction of acetone with ketenimine 4 requires a much shorter reaction time than the reaction with ketenimine 3 points to one factor governing J these reactions. A very simple scheme (Scheme 1) describing ■ - I Scheme 1. A + K hy-^ k A* Product (1) (2) + K kr K* + A (3) K* k . K + energy (4) this reaction involves (1) the excitation of acetone to some unspecified reactive statet (2) reaction of the excited acetone with the ketenimine to form products, and (3) transfer of energy from the excited acetone to the ground state ketenimine, followed by (4) return of the ketenimine to the ground state through some deactivation process. For purposes of simplicity side reactions and other deactivation modes of acetone have been omitted. If j i one considers this scheme, the reactivity difference j ! between ketenimines 3 and 4 is seen, to be due to step (3) j becoming more important relative to step (2) as the triplet energy of the ketenimine is decreased with aryl substi tution on the ketenimine. The factors governing the direction of addition of ketones to ketenimines are not clearly understood. Singer 9 10 and Davis * have suggested that dipolar orienting influences may be important in these photocycloadditions• As the data in Table VI indicates* the reaction of acetone with ketenimine 4 proceeds with nearly equal distribution between the two potential directions of addition. Since ; nearly half of ketenimine 3 reacts to give the 3-adduct 9 it is quite possible that the acetone-3 reaction proceeds with nearly equal production of the a and 3 adducts, with subsequent loss of the a-adduct 5. The observed rear rangement of the a-adduct 6, as shown in the following equation* HN CH CH O % CH, _ CH " \ J c=ch9 / 2 CH3 6! R=C6Hh 8. R=C6Hll suggests that over the long reaction time for ketenimine 3, 36 the a-adduct 5 may rearrange and react further, which would explain the fact that neither the a-adduct 5 not its rear- ranged unsaturated amide 7 were found in the product mixture. In fact, the photochemical cleavage of fluorenone i a-adducts has been observed.^® The appearance of trace amounts of the ketenimine 3 trimer in the product mixture of the acetone-3 reaction is not clearly understood. The nmr, IR, and melting point data listed in Table VII were not consistent with the ketenimine 40 T _ dimer data. In fact, the mass spectral analysis indicated that the molecular weight of the compound was 435, while a j very important fragment had a mass of 145 . Since the molec- ; ular weight of ketenimine 3 is 145 a.m.u., this evidence suggested the ketenimine trimer. In order to see whether j the trimer had been formed during the course of the j acetone-3 reaction a sample of ketenimine 3 and its dimer was photolysed and then analyzed by nmr. There was no evidence of the formation of the ketenimine trimer after six hours of irradiation, nor was there any trimer formed when the sample was heated at 100®C for six hours. These results indicated that the ketenimine trimer had not been formed by a reaction of the ketenimine and its dimer during the course of the acetone-3 reaction® nor had it been formed in a thermal reaction at 100°C. It is possible 37 that the trimer was present in the starting ketenimine 3. Nonetheless, just where it is coming from is an open, and . intriguing, question. The overriding difficulty in these reactions was the problem of the appearance of the photolysis products of 36 acetone. It has been reported that the photolysis of acetone results in the formation of biacetyl, 2,4-penta- dione, 2,5-hexadione, iso-propanol, and pinacol. In order to better understand what was occurring under the conditions of the cycloaddition reactions, samples of acetone were irradiated in a manner duplicating the conditions of the photoreactions. A very complex mixture of products resulted from this photolysis of the acetone. The IR data (Table IX) suggest the presence of not only a ketone functionality but also one or more hydroxyl groups. , The nmr data of the crude mixture (Table IX) suggest that there are several compounds present if one considers the variety of aliphatic protons present. A comparison of these spectral data with that for some of the proposed adducts from the photolysis of acetone (Table X) leads to few conclusions as to the content of the mixture of adducts. An analysis of the crude mixture by vapor-phase chromatography was then attempted. The retention times for some known compounds as well as those for the crude product mixture are summarized in Table IX. As indicated, there were two major adducts in the photolysis mixture with at least eight minor adducts. The only compound which has been i identified is 2,5-hexadione. Unfortunately, this left a large number of unanswered questions with respect to what may be happening to the , acetone when the photocycloaddition reactions are carried out. Quite apparently there are a large number of side ' reactions occurring which may be very important when the photocycloaddition reaction is carried out over a long period of time, as was the case in the acetone-3 reaction. / > / The side reactions may, of course, involve the ketenimines and account for the loss of a large portion of ketenimine 3, in its cycloaddition reaction. ; ! Preparation of Imines. In order to broaden the scope of the variable temperature nmr study it was our desire to synthesize other imines. An ordinary Schiff's-base prepa ration of 2,2,4,4-tetramethyl-N-(phenyl)-3-iminocyelo- butanone 10 was quite successful in preparing that imine. However, Schiff's-base reactions using HC1 or zinc chloride as catalysts were not successful in preparing the anils of acetone or 2,2,4,4-tetramethylcyclobutanone and thus an alternative preparative path was sought. Leonard and 35 Paukstelis had reported the preparation of secondary amine perchlorate salts which were then reacted with ketones to obtain the iminium perchlorate salts. The i 39 ! i ! ■ i i i i < adaptation of this procedure tb prepare acetone anil j perchlorate was apparently quite successful, as evidenced j i by the spectral data in Table VIII. The imine band showed i j ' ' | j up in the IR, and in the nmr spectra the methyl doublet j j and aromatic protons were both clearly present and in the i proper relative abundance. J Several attempts were then made to release the anil byj i neutralizing the acid salt. However, any technique ! ! ! j involving the use of water resulted only in conversion of r j the perchlorate salt to aniline and aeetone, with little or ! J no anil present. An exchange reaction using triethylamine j was then tried and this appeared successful. Nonetheless ! • ' ' ; | there was still aniline present in the product mixture. j Since distillation led to decomposition of much of the product, a chromatography on neutral alumina was tried. However, complete separation was not effected, and we were i unable to prepare samples suitable for a variable | temperature nmr study. j Nonetheless this preparative approach holds a great i ' ' I deal of promise, particularly since the perchlorate salts j i I ; can be prepared in quantity. Two things must be accom- ! piished for this preparation to succeed. First, water must ; | be rigorously excluded from the product anil in order to \ i ; ' prevent reconversion to starting materials. Second, a J suitable technique for purifying the ketone anil must : be found. t I Variable Temperature NMR Study. The nmr spectra of | imines 9, 10, and 12 change dramatically with ehanges in r t I J I temperature as shown in Figures 1 and II. This results from; I - • i I the increasing rate of inversion about the imine bond with j i ! | increasing temperature. The doublet observed at lower ! | temperatures collapses at the coalescence temperature to j t ! ; give a single absorption for the methyl groups of the imine! I where the rate of inversion is faster than the time reso- ! | i | lution of the nmr. Theoretical spectra were calculated by | i i j utilizing the chemical shift and line width data obtained ! ' i at low temperatures, but with different values of r, the j | mean lifetime before inversion occurs. | Comparison of the theoretical and observed curves thus! permitted an evaluation of r as a function of temperature, j as is illustrated in Figure I for imine 9. From this I information the activation parameters for the inversion ; process were readily available. The logarithmic form of the Arrhenius equation, relating the rate of inversion, k = 1/r, to the activation energy, E_, may be written as follows: d -E log 1/r = f 1 + log A, 2.303R T where T is the temperature in ®K, and R is the gas j constant, 1.987 cal/deg-mole. If one prepares a plot of log 1/r vs 1/1* the slope is then -E /2.303R, from which SI the energy of activation can be calculated. In practice ! this data was analyzed by the method of least squares. A I i representative plot which was derived in the case of imine j 9 is shown in Figure III. j ! The Eyring equation, relating the free energy of j activation, &GC, to the mean lifetime for inversion at the | coalescence temperature is as follows: i G+ log 1/r = log KkT - c . h 2.303RT where R is the gas constant and k and h are physical constants. Since, in the unimolecular inversion process, K = 1, this can be rearranged to give: •i - The equations giving the other thermodynamic parameters at AG* = 2.303RT(10.32 + log T - log 1/r). the coalescence temperature are as follows: AH* = E_ - RT, and v- cl AS* = (aH*-AG*)/T. In the ease of imine 10 the overlap of the eyelohexyl I 1 i ■ i I absorption with the methyl absorption made the matching of j : theoretical curves impossible. In this case the Gutowsky ! 41 : ; equation was used to determine the lifetime for inversion; I ‘ - i at the coalescence temperature. This equation is written \ Figure III. A plot of the log 1/r vs 1/T for the inversion about nitrogen in imine 9. 42 L o g (l/T ) 43 2.4 2.2 2.0 / - COALESCENCE TEMPERATURE 1 . 6 1 . 4 1 . 0 0 . 8 0.6 2.8 3.0 3 . 1 2.6 2 . 7 2.9 (l/T) x I03 as follows!. . ‘C 1 . 1 j V = 1 - 1 2 ! ^ O D 2 tt2 t 2 ( 6 v ) 2 Q D i I f 1 I ! where 6, , is the chemical shift difference at low temper- | ® , ; atures, i.e., at the slow exchange limit. Since the i i I I chemical shift difference 6^ = 0 at the coalescence | temperature, this equation can be rearranged and simplified so that ■ i 1/r = s I2 tt6v m. | [ I j The Eyring equation was then used to calculate the free j energy barrier to inversion at the coalescence temperature. In order that the enthalpy change during inversion for imine 10 might be calculated, the entropy changes during inversion were assumed to be very much the same in imines x 9 and 10, and thus a value of aST = -1.1 e.u. was assumed a e for imine 10. The thermodynamic parameters for the inversion about the imine bond in imines 9, 10, and 12 are summarized in A A/ /W Table XIII. When these results are compared with the free energy barriers to inversion listed in Table I, one sees that these imines have relatively high barriers to inversion when compared to amines. In fact, only in the case of amine XI, which has a halogen attached to the 45 nitrogen, is there a greater barrier to inversion, with the 1,2,2-trimethylaziridine X having a barrier to inversion comparable to that for imine 9^. One of the original reasons for interest in aziridines I was the observation that in strained systems the rate of 31 inversion about nitrogen was slowed. Thus some workers turned to aziridines as a potential source of stable nitrogen pyramids. It is now seen that imines as a group may have even higher barriers to inversion about nitrogen, 'in fact, extrapolation back to room temperature, using the ■Eyring equation, gives a lifetime for inversion of about ,13 minutes for imine 9. Thus inversion in solution at room A * temperature is still relatively fast, and suggests that the isolation of svn and anti isomers may not be possible even in the cases of unsymmetrical p-photoadducts. Of course, in the case of the symmetrical P-adducts prepared here there is no opportunity of svn and anti isomers. In order to compare the results of this work with I results obtained for other imines the free energy barriers to inversion, aGc, were calculated, using the Eyring equation, for several of the imines discussed by Curtin, 12 Grubbs, and McCarty, and these values are listed in Table XIV. It must be pointed out that since these results were obtained using different solvents, and not all are nmr studies, these figures must only be considered as relative ol Table XIV. Barriers to inversion in selected imines. Compound °C k Ea kcal/mole kcal/mole Solvent (CF3)2C=NCF(CF3)2 32 io3 13 ± 3 14 None (p-ch3oc6h4)2c=n-c6h4(cg2ch2ch3)-p 29.8 12.4 16 cci4 (p-CH3OC6H4)2C=N-C6H5 62.2 10.9 17-20 18 CC14 (p-CH3OG6H4) 2C=N-C6H4(CH3) -p 69.5 T l oo 18 c CC14 (CH3)2C=N-C6H5 140 10 • • • 21 Quinoline (CH3)2C=N-CH2C6H5 170 • • • • • • 23° Quinoline P-C1C6H4(C6H5)C=N-CH3 40 7xl0‘6 25 26 Cyclohexane a. Reference 12, and references cited therein. ! - P * ■ ' O N Lb. Calculated by this author. c. Reference 15. ___________________________________ and for purposes of comparison only. It is immediately seen that imines 9. 10, and 12 fit right in the middle of these /vV /vA ywV other imines. The important trend in the data in Table XIV is that substitution of an alkyl group for an aryl group raises the barrier to inversion, and that this effect is most important when the substitution is made on the nitrogen. This is, of course, due to the conjugative effect. Comparison of the barriers to inversion for imines 9^ and 10 very clearly demonstrates the importance of the conjugative effect in influencing the rate of inversion in imines. In changing from the cyclohexyl ring in imine 10 to the benzene ring in imine 9, there is a marked reduction in the barrier to inversion. Clearly the opportunity for overlap between the electron pair on nitrogen and the tt- electrons of the benzene ring, in the transition state, has enhanced the rate of inversion. Finally, the data for imine 12 must by considered as less than adequate. The values obtained for rates of inversion as a function of temperature did not produce a satisfactory plot for determining the activation energy with precision. The data was treated by the method of least squares and the results reported here for comparison with the other imines. EXPERIMENTAL Analytical Instruments. Infrared spectra were recorded: on a Perkin-Elmer IR-337 infrared spectrophotometer. Cali- -1 bration was done against the 1601.4 cm band of a poly styrene film. All nuclear magnetic resonance (nmr) spectra,j ■ * with the exception of the variable temperature spectra, j | were taken on a Varian Associates A-60 Analytical NMR J Spectrometer with tetramethylsilane (TMS) as the internal I standard. Chemical shifts (6) are reported in parts per i million (PPM) downfield from TMS. The variable temperature ! ! nmr spectra were recorded on a Varian Associates HA-100. The vapor-phase chromatographic analysis was made on a Varian Aerograph model 90-P using a SE-52 (Silicon Gum Rubber, Phenyl) column (20%) with helium as the nonsta- tionary phase. Melting points were determined with an i electrothermal melting point apparatus and were uncorrected. Reagents i I ! I i Acetone, from Mallinekrodt Chemical Works, was j distilled and then further dried over Linde Molecular j ■ Sieve, type 4A. \ | 49 : I ; 1 i p-Dichlorobenzene. from Mallinckrodt Chemical Works, j i j was purified by fractional sublimation at 25 mm pressure. | Florisil, 60-100 Mesh, from Fisher Scientific Company,I I was dried overnight at 120°C prior to use. Petroleum Ether. 30-60. from Mallinckrodt Chemical Works, was distilled and the fraction boiling at 30-50°C collected. j Triethvlamine. from Matheson Coleman and Bell, was ! dried over potassium hydroxide. All other reagents and solvents were used without further purification and were reagent grade. Preparation of Ketenimines Iso-butvrvlchiofide. To a 100 g sample of iso-butyric j acid in a one-necked flask was added 340 g of thionyl j I chloride dropwise while stirring. The mixture was then set ! i ! | up on a steam bath with a reflux condenser and drying tube ; attached, and allowed to react for five hours. At the end of this period the excess thionyl chloride was removed by fractional distillation, the thionyl chloride coming off as> ; ■ . i a forerun. The product was then collected at atmospheric i > pressure at 89-91°C, in approximately 40% yield. ! Isobutvrvl-N-phenvlamide. A 1000 ml, three-necked , flask was fitted with a mechanical stirrer, condenser, and i dropping funnel. To a solution of 25 g of iso-butyryl- ! chloride in 450 ml of ether in the flask was added, drop- wise, with stirringi a solution of 49 g of aniline in 100 ml of ether. An extra 100-150 ml of ether were then added to facilitate the stirring. The reaction was complete within an hour. The ether layer was then transferred to a | separatory funnel, the flask rinsed thoroughly with ether and water, and these fractions added to the funnel. Just i enough water was added so that any remaining solid was : dissolved. The water layer was then discarded, and the ether layer washed twice with water. After drying the ether layer over magnesium sulfate the ether was stripped ! off . .in vacuo so that approximately 100 ml of 'solution remained. To this was added 200 ml of petroleum ether, and i the product allowed to crystallize over ice. The amide was . an off-white crystalline material, m.p. 103-107®C, ’ collected in 75-85% yields. ' Iso-butvrvl-N-cvclohexvlamide. This preparation was identical to that for iso-butyryl-N-phenylamide. In this case 30 g of iso-butyrylchloride and 61.5 g of cyclo- hexylamine were used. The product consisted of white , crystals, m.p. 112-117°C, and was collected in 85-90% yields• 51 Dimethyl-N- (phenyl )ketenimine. The following is a modification of the 'procedure of Singer and Davisj To a ; 1000 ml, three-necked flask, fitted with a condenser, a : strong electric stirrer, and an inlet tube for nitrogen, was added 400 ml of triethylamine and 20 g of iso-butyryl- 1 N-phenylamide. The system was flushed with nitrogen while ■ the amide was being taken up in solution. To the mixture was added 55 g of the oven-dried Florisil and then 55 g of phosphorus pentoxide. The reaction was allowed to proceed under reflux for 5-6 hours, and then cooled to room temperature. The organic layer was collected, and the solids washed with ether and these fractions added to the ; product mixture. After the solvent had been removed in vacuo, a few small crystals of hydroquinone were added as a free-radical inhibitor, and the product vacuum distilled to give the product as a colorless liquid in 11-18% yields, b.p. 35.5°C/G.2 mm. The adduct showed the characteristic ketenimine band in the IR at 2026 cm"* (CCI4). The nmr (CCl^) of the ketenimine showed a sharp methyl singlet at l.l8 PPM (6), as well as the phenyl absorption at 7.34 1 PPM (6). Dimethyl -N-Ccyclohexyl)ketenimine. This preparation was identical with that for dimethyl-N-(phenyl)ketenimine, with the exception that the reaction time was 15 hours. The ketenimine was collected at 58°C/1.65 mm, in 24.4% yield. | The ketenimine band in the IR (CC1.) appeared at 2035 cm”*. i . ' The nmr (CCl^) exibited a broad cyclohexane absorption j from 0.83 to 2.17 PPM (6), with the methyl peak appearing | at 1.58 PPM (6). | i ; i Photocycloadditions ! | 1 Acetone and Dimethvl-N-(phenyl)ketenimine. A solution j I of 1.54 g of dimethyl-N-(phenyl)ketenimine in 150 ml of thej I ' j I distilled, dried acetone was prepared and placed in Vycor j ~ . . . 1 tubes. These were capped with red, sleeve-type serum caps j and degassed (freeze-thaw cycle) by means of syringe j needles through the serum caps. The vacuum was always I maintained when the needle was extracted. These were taped to a Quartz immersion well fitted with a Vyeor sleeve and containing a 450 w Hanovia lamp, No. 679A-36, and irra- ! diated for 30 hours at l8-20°C. j The excess solvent was removed in vacuo giving 3.1 g j of crude products. The IR and nmr of this mixture indicated I this to be a very complex mixture. Of this crude yield, j ; t 11.69 g were analyzed as follows: The fraction was treated j ! . i with petroleum ether, precipitating 0.35 g of dark tarry ; material, which was 20.7% of the crude material. A ehroma- | ,tography on Florisil, using petroleum ether as the elutant,! 1 [was carried out on the remaining 1.34 g. i The first fraction consisted of a trace amount of a ;white solid, m.p. 163-165°C. The IR (CCl^) showed sharp |absorptions in the carbonyl region, at 1600 and 1675 cm”^. The nmr (CCl^) showed phenyl protons at 6.72 to 7.23 PPM j(6) as well as a single sharp peak at 1.44 PPM (6). The ;mass spectrum contained a parent peak at 435, and a very j strong peak at 145, which suggested that this was the ;ketenimine trimer (M.W. = 145 for monomer). | The second fraction collected was a colorless liquid, 0.48 g, 28.4% of the crude material. The IR (CCl^) showed :a sharp imine absorption at 1740 cm , while the nmr (CCl^) exibited a doublet at 1.24 and 1.49 PPM (6), as well as a phenyl absorption from 6.55 to 7.33 PPM (6), indicating |this to be the 3-photocycloadduct of acetone and dimethyl- i JN-(phenyl)ketenimine. This accounted for 41,1% of the starting ketenimine. I The third fraction was a white solid, 0.14 g, or 8.3% Sof the crude materiali with a m.p. of 85-94°C (unrecrys- j : tallized). This physical evidence suggested that this was i i iso-butyryl-N-phenylamide, the result of ketenimine |remaining in the crude product mixture reacting with water, i ;and accounts for 15% of the starting ketenimine. Later fractions (0.50 g total) taken from the column were not readily analyzable. The nmr spectra (CCl^) of these fractions exibited broad humps from 0.85 to 1.60 PPM 54 (6), suggesting that polymeric material was present. There were phenyl absorptions in these spectra, suggesting the fate of the remaining ketenimine. In total 1.12 g, or 83.6% of the material on the column was recovered. Acetone and Dimethyl-N-(cvclohexvl)ketenimine. A solution c£ 1.63 g of dimethyl-N-(cyclohexyl)ketenimine in 150 ml of the distilled, dried acetone was prepared and placed in Vycor tubes. These were capped with red, sleeve- type serum eaps and degassed (freeze-thaw cycle) by means of syringe needles through the serum caps. The vacuum was always maintained when the needle was extracted. These were taped to a Quartz immersion well fitted with a Vycor sleeve and containing a 450 w Hanovia lamp, No. 679A-36, and irradiated for 10 hours at lo“20°C. The excess solvent was removed in vacuo giving 7.57 g of crude products. The IR and nmr spectra (CCI4) of this crude mixture Indicated that this was a highly complex mixture consisting largely of adducts resulting from the photolysis of acetone. These spectra also indicated that the reaction was incomplete. Of this crude mixture 2.00 g s |were analyzed as follows s A chromatographic separation 1 • i !on Florisil, using petroleum ether as the eluent, was only j jslightly successful due to overloading of the eolumn. A i first fraction (0.76 g) was found, upon analysis by nmr, to I 55 I j contain the desired a and 3 adduets as well as acetone !photolysis adducts. Later fractions contained only photol ysis adduets of acetone. The first fraction was rechro matographed on a larger column. A first fraction (0.175 g) of a white solid, m.p. 54-56°C, was obtained. The IR (CCI4) showed a sharp imine absorption at 1728 cm~^. The nmr (CCI4) showed a, doublet at 1.35 and 1.42 PPM (6) , as well as the broad cyclohexyl absorption from 0.70 to 1«98 PPM (6). These data indicated that this was the 3-photo- eycloadduct of acetone and dimethyl-N-(cyclohexyl) / ' ketenimine. This accounted for 29.4% of the starting ketenimine. A second fraction (0.165 g) consisting of a white solid, m.p. 71-76°C, was collected. The IR (CCI4) showed a sharp amide band at 1672 em"^, a less intense but sharp olefin band at 1633 cm“* , and a sharp N-H band at 3405 cm"*^. The nmr (CCI4) showed olefinic protons at 4.73 to 4.87 PPM (6), two types of methyl protons at 1.20 and 1.70 PPM (6), in a 2s1 ratio, respectively, as well as the cyclohexyl absorption from 0.75 to 1.93 PPM (6). These I I spectra suggested that this compound was 2,2,3-trimethyl- but-3-en-N-(cyclohexyl)amide, which resulted from the i _ I rearrangement of the a-photocyeloadduct of acetone 56 i I and dimethyl-N-(cyclohexyl)ketenimine. This accounted for 27.8% of the starting ketenimine. A third white solid (0.16 g) was collected. The melting point of this compound, 107-110°C, confirmed that this was iso-butyryl-N-cyclohexylamide, the result of the unreacted ketenimine in the crude product mixture reacting with water. This accounted for a further 33.3% of the starting ketenimine. Thus 90,5% of the starting ketenimine was accounted | for. Of the ketenimine which did react, 44.1% formed the ! 3 -photocycloadduct, while 41.7% formed the a-photocyclo- adduct. The direction of addition of acetone to dimethyl- i N-(cyclohexyl)ketenimine was therefore nearly equally divided between the two possibilities. Preparation of Imines 2,2,4,4-Tetramethyl-N-(phenyl)-3-iminocyclobutanone. A solution of 28 g of tetramethyl-1,3-cyclobutadione in 300 ml of benzene was prepared in a 500 ml flask. To this | was added 22.4 g of aniline and 1 ml of concentrated HC1. i | This mixture was attached to an extractor and heated under i- I j ; j reflux conditions overnight. The reaction was then quenched t ( ■ • S I in water and ether extracted. After the organic layer had ! " i 1 been washed several times with .water_and.then, dried, over J j magnesium sulfate, the solvent was stripped off. A total of J 10 g (23%) of the monoimine was recovered from this and subsequent concentrations, while 10 g (36%) of the starting diketones were recovered. A chromatographic separation of the crude product on alumina using 10% ethyl ether in petroleum ether as the elutant gave a light yellow liquid as a first fraction. The IR (CCI4) showed two strong absorptions of importance, at 1800 and 1680 cm"1, corresponding to the C=0 and C=N stretching frequencies, respectively. The nmr (CCI4) showed a doublet at 1.11 and 1.38 PPM (6), and a methyl/phenyl ratio, upon integration, of 12.7/5 (Anticipated=12/5). The slight shoulder at 1.25 PPM (6) indicated that the purity of the sample was about 95%. Aniline Perchlorate. In a 250 ml Erlenmeyer flask a solution of 18.6 g of 'aniline in 50 ml of ether was prepared and chilled in an ice bath. While stirring, 28.7g i of perchloric acid solution (70% in ethanol) was added dropwise. At the end of the addition the mixture was cooled and the white, crystalline perchlorate was collected by filtration. The compound melted at 285-290®C, after j darkening at 265-275°C. The yield was 12-14 g, 62-72%. j i [ j Acetone Anil Perchlorate. In a 100 ml round-bottom j j flask was prepared a solution of 19.36 g of aniline j 58 perchlorate in 15 ml of distilled, dried acetone. After stirring for one-half hour 30 ml of benzene were added. Approximately half of the solvent was removed in vacuo, and the mixture allowed to stand overnight in the refrigerator. Collection of the white crystals gave 14.18 g of adduct (61%) m.p. 75.5-79°C. Removal of more' of the solvent yielded 6.24 g of a second fraction whose melting point characteristics were found to be the same as the starting material, accounting for 32% of the aniline perchlorate. The IR (H2CCI2) showed an imine band at 1665 cm-*. The nmr . (H2CCI2) of the compound showed two sharp methyl singlets at 2.25 and 2.58 PPM (6), a sharp band at 1.94 PPM (6), and a phenyl absorption from 7.30 to 7.55 PPM (6). The integra tion showed these to be in a ratio of 3s3j0.5j5.0. Acetone anil. In a separatory funnel 10.0 g of acetone ! • * anil perchlorate were dissolved in as little dry acetone as , possible. To this was added 50 ml of dry ether and then 4.34 g of distilled, dried triethylamine. After vigorous shaking the lower fraction was removed and the ether layer dried over magnesium sulfate for one hour. Removal of the solvent in vacuo yielded 4.3 g of crude adduct. The nmr of ; this crude mixture indicated that about 15% of the material was aniline and the remainder acetone anil, suggesting a 64% yield of the desired anil. Purification by 59 distillation led to 'decomposition of the material. A chromatography on neutral alumina using petroleum ether as j I I the eluent resulted in a mixture of aniline and acetone j anil in the ratio of about 1*5. The nmr contained the two methyl peaks ffom the acetone anil, the N-H peak of aniline, and a complex phenyl absorption. \ Photoreactions j Dimethyl-N-(phenyl)ketenimine and Tetramethyldi-N- (phenyl)cvclobutadiimine. A solution of two compounds (0.5 M in each) in benzene was prepared, placed in an nmr tube, degassed (freeze"-thaw cycle) and sealed. The sample was taped to a Quartz immersion well containing a 450 w Hanovia lamp, No. 679A-36, and irradiated for a total time of six hours. A Vycor filter was used for the first two hours and then removed for the remainder of the irra diation. At one hour intervals the nmr spectrum of the sample was recorded in order to detect any reaction. After the full period of irradiation there was no detectable ! • • I' reaction. The sample was then heated in boiling water for | a further six hoursJ but again no reaction was found to j | have occurred. j | j Photolysis of Acetone. Distilled and dried acetone ! 60 ] ! (150’ ml) was placed in Vycor tubes, capped with the red, sleeve-type serum caps and degassed (freeze-thaw cycle) by means of a syringe needle through the serum caps. The samples were taped to a Quartz immersion well fitted with a Vycor sleeve and irradiated for thirty hours using a 450 w Hanovia lamp, No. 679A-36, The excess solvent was removed in vacuo leaving a colorless liquid. The IR (CCl^) of the crude product showed a very broad band centered around 3425 cm"*. The nmr (CCI4) had a quartet centered at 4.2 PPM (6), as well as sharp peaks at 2.63, 2.57, 2.08, 1.35, 1.24, 1.22V and 1.15 PPM (6). In addition, there was a broad shoulder from 0.95 to 1.6 PPM (6), which was superimposed on the last four peaks. The VPC analysis of this mixture was carried out on a Varian Aerograph model 90-P, using an SE-52 column (20%). All samples injected ^ere 1 pi, and consisted of the followings (1) a mixture of acetone, biacetyl, 2,4-penta- dionej and 2,5-hexadione, (2) the acetone photolysis mixturei and (3) various combinations of one of the j compounds in (1) and the acetone photolysis mixture. The retention times for the components of these mixtures were ! determined and comparisons between the known compounds and I ■ i components of the crude reaction mixture made in order to j I • ! ! cheek on the possible presence of those compounds listed iri [ ■! 61 I (1) in the photolysis mixture. The retention times are i i summarized in Table XI. Variable Temperature NMR i ! ! . ! Preparation of the Samples for NMR Analysis. 2.2,4,4- ! ' ! 1 Tetramethyl-N-Cphenyl)iminoxetane. Prior to the preparation j i | of the nmr tube the imine was further purified by chroma- | i tography on Florisil, a center cut being taken. Into an ! ? ' i i nmr tube were then weighed 0.0320 g of the imine and \ i ! ► • I ! 0.5176 g of p-dichlorobenzene. The solution at 80.3®C was | i therefore 5.8% in the imine, or 0.46 M. A neck was drawn in the nmr tube and the sample degassed (freeze-thaw ! i i | cycles) and sealed. Some difficulty was encountered in j | that under the full vacuum gases escaping from the melting j j solid caused vigorous bumping. 2.2.4,4-Tetramethvl-N- j I (cvclohexvl)iminoxetane. Prior to the preparation of the : ! ; j nmr tube the imine was further purified by chromatography I I ' • | i on Florisil, a center cut being taken. Into the hmr tube ! . ■ j ; were then weighed 0.1018 g of the imine and 0.8965 g of I • - • : '! ! p-dichlorobenzene, and a neck drawn in the nmr tube. The I ! ; solution at 80.3°C was 10.2% in the imine, or 0.83 M. For I • i. the degassing a slightly different procedure was employed, i For the first several freeze-thaw cycles, the vacuum was i partially replaced using nitrogen. This eliminated any 62 vigorous bumping caused by gases escaping from the melting solid. After several cycles, increasing the vacuum each time, it was possible to melt the sample under full vacuum without bumping, and the sample was sealed. 2.2,4,4-Tetra- methvl-N-(phenyl)-3-iminocvclobutanone. Prior to the preparation of the nmr tube the imine was further purified by a chromatography on Florisil, a center cut being taken. Into the nmr tube were then weighed 0.0349 g of the imine and 0.4796 g of p-dichlorobenzene. The solution at 80.3°C was 6.8% in the imine, or 0.51 M. A neck was drawn in the nmr tube and the sample degassed (several freeze-thaw cycles) and sealed. Recording the NMR Spectra. Before recording the spectra the temperature control was calibrated in the 42 manner described in the instruction manual, using ethylene glycol as the standard. The spectra of ethylene glycol were recorded over the range of 60 to 120°C, and after calibration of the chart paper, the chemical shift between the methylene and hydroxyl protons determined. These chemical shifts were then compared to a graph of chemical shifts vs temperature to determine the actual temperature of the sample probe. From this information a simple plot of Dial Setting vs Actual Temperature was made and this used to determine the temperature at which each 63 spectra was recorded. The variation in temperature was felt to be less than ± 2“C. The sample tubes were then inserted in the probe and, as the sample melted, the instrument was locked on the proton signal from the p-dichlorobenzene. Scanning upfield from this signal the expanded spectra were recorded through the Jegion of the absorption of the methyl groups of the imines, the scan width being about 75-100 cycles. Spectra were recorded over a range of temperatures from 60 to 1506C, giving several spectra above and below the coalescence temperature. It was necessary to wait at least twenty minutes between spectra in order to insure that temperature equilibration had been achieved. Figures I and II show representative curves for imines 9 and 10. There were two types of curves observed over this range of temperatures. At low temperatures the curves had two peaks, while at high temperatures there was but one. The point at which the two merged and became one was the coalescence temperature, corresponding to the rate of inversion about nitrogen at which it is no longer possible for the instrument to distinguish between the two types of methyl groups in the imines. Computer Calculation of Theoretical Curves. Theoret- j I . | ieal curves which were calculated by means of a computer \ L ' program were -used to determine the mean lifetime, for inversion about nitrogen at the temperatures the spectra were recorded. This program utilized the following input for these calculations: 1. Titlej which was for the convenience of the j f | operator and could be anything. j i 2. 6, Chemical shift, which was the separation of the ! i two methyl peaks at low temperature (ground state).! This was a constant of the calculations, and was measured from the spectra taken at 60 C. 3. Jcc» the coupling constant between the protons of the methyl groups which were on opposite sides of of the imine molecules. This was actually zero, but, in order that there might be some number to be used in the calculations the computer made, the value of 0.1 cps was chosen as being suitably small. Values of 0.2 and 0.05 cycles were also tried with no effect on the spectrum. 4. v, the lifetime in seconds for inversion about nitrogen, was the value chosen for each calculation and was the variable from one calculation to the j ! next. In order to pick a first suitable value to be f used in the calculations the Gutowsky equation was j used, where: I where 6^ is the 6 given in the second paragraph above, and &Ur is the peak separation at some temperature other than the ground state. By- varying from this value suitable curves for matching with the recorded curves were calculated. A 5. T21 the natural line width, was the half-width of the one methyl peak in the nmr spectrum observed at 60*0. B 6. T2» the natural line width, was the half-width of the other methyl peak in the nmr spectrum observed at 60°C. 7. AX, was the width of the calculated spectrum. In all eases 75 cycles was chosen as being suitably wide to calculate the curve desired. 8. N, was Jin arbitrary number which would give a sufficient number of points along the curve for plotting. In all cases 299 was chosen. (N must be j odd and less than 402.) ! The computer then printed out two sets of data. One jwas therdistance, in cycles, from the center of the curve ! ; |each point lay. The second was a value for the height of |the curve at each particular point. These two sets of data were then used to plot a theoretical nmr curve. The computer also plotted a curve alongside the print-out data, giving a rough indication of what the curve would look like. Hatching Of Observed and Calculated Curves. There were two criteria for the matching of the observed with the computed spectra. Below the coalescence temperature the peak/valley ratio was used. This was the ratio of the maximum curve height to the height of the curve in the valley between the two peaks, and was always greater than unity. Above the coalescence temperature the peak half- width was used. The data for the calculated curves, taken directly from the computer print-out, was used to determine these two parameters for the theoretical spectra, while they were measured directly from the observed spectra. With the observed and calculated peak/valley ratios and line half-widths in hand one could match curves. A curve calculated using a given 7* was found which was the same as a curve observed at some particular temperature. Ir. this manner the lifetime for inversion was known for each temperature. The data used in matching theoretical to observed curves in the case of 2,2,4,4-tetramethyl-N- ! (phenyl)iminoxetane is listed in Table XII. j • f ,j i i 67 I I j j Finally, treatment of these data by the method of i least squares led to the thermodynamic data for the ! nitrogen inversion proeess, which are summarized in Table j XIII. A typical log 1/r vs l/T plot, for 2,2,4,4-tetra- j methyl-N-(phenyl)iminoxetane is recorded in Figure III. | SUMMARY The photoeycloaddition reactions of acetone with dimethyl-N -(phenyl)ketenimine 3 and dimethyl-N-(cyclo- hexyl)ketenimine 4 have been carried out and the products analyzed. In the case of the acetone-3 reaction the pt ... 3-adduct 2,2,4,4-tetramethyl-N-(phenyl)iminoxetane 9 was A' formed, but no evidence for an a-adduct was found. In the acetone-4 reaction both the a-adduct 2,2,3,3-tetramethyl- A * N-Ccyel©hexyl)iminoxetane 6 and the 3-adduct 2,2,4,4- tetramethyl-N-(cyclohexyl)iminoxetane 10 were formed in about equal amounts. The a-adduct 6 was found to rearrange to 2,2,3-trimethyl-3-butene-N-(cyclohexyl)amide 8. The barriers to inversion about the imine bond in imines 9 and 10 were found to be = 18.7 + 0.3 and 21.8 kcal/mole, respectively. These barriers to inversion are higher than those found for most amines, but are very much in the middle of barriers to inversion in other imines. The lower barrier to inversion in the case of imine 9 is attributed to the tact that the conjugative /1A effect is enhancing the rate of inversion in imine 9^. 68 FOOTNOTES 1. E. Patemo and G. Chieffi, Gazz. Chim. Ital., 39, 341 (1909). 2. G. Btiehi, C. G. Inman, and E. S. Lipinsky, J. Am. Chem. Soe., 76, 4327 (1954). 3. See, for examples a. G. Bttchi, J. T. Kofron, E. Koller, and D. Rosenthal, ibid.. 78. §76 (1956); b. R. Srinivasan, ibid., 82, 775 (I960); e. J. F. Harris, Jr. and D. D. Coffman, ibid. . 84, 1-553 (1962); d. M. Kawanisi, K. Karaogawa, T. Okada, and H. Nozaki, Tet., 24, 6557 (1968); e. Y. Shigemitsu, H. Nakai, and Y. Odaira, Tet.. 25, 3039 (1969). 4. a. N. J. Turro, et al., J. Am. Chem. Soe., 89, 3950 (1967); b. J. A. Barltrop and H. A. J. Carless, Tet. Let.. 390 (lr 5. N. C. Yang, M. Nussim, M. J. Jorgenson, and S. Murov, ibid..3657 (1964). 6. D. R. Arnold, R. L. Hinman, and A. H. Gliek, ibid., 1425(1964). 7. G. S. Hammond and P. A. Leermakers, J. Am. Chem♦ Soc.. ‘ - , 207 (1962). 8. L. A. Singer and P. D. Bartlett, Tet. Let.. (1964). 9. L. A. Singer and G. A. Davis, J. Am. Chem. Soc.. 89, 598 (1967). 10. L. A. Singer and G. A. Davis, ibid.. 89. 941 (1967). 11. L. A. Singer, G. A. Davis, and V. P. Muralidharan, ibid.. 91, 897 (1969). 12. D. Y. Curtin, E. J. Grubbs, and C. G. McCarty, ibid., “8, 2775 (1966). 13. D. Y. Curtin and J. W. Hauser, ibid., 83, 3474 (1961). 69 70 ' 14. J. J. Worman and E. A. Schmidt, J. Ora. Chem.. 35, 2463 (1970). 15. H. A. Staab, F. Vttgtle, and A. Mannschreck, Tet. Let., 697 (1965). 16. S. J. Brois, ibid., 5997 (1968). 17. S. J. Brois, ibid.. 506 and 508 (1968). 18. D. Felix and A. Eschenmoser, Anaew. Chem.. 80. 197 (1968). 19. J. J. Delpuech and M. N. Descamps, Chem. Commun., 870(1969). 20. a. A. T. Bottoni and J. D. Roberts, J. Am. Chem. Soc., 80. 5203 (1958); b. A. Loewenstein, J. F. Newmer, and J. D. Roberts, ibid.. 82, 3599 (1960); c. F. A. L. Anet and J. M. Osyany, ibid.. 89. 352 (1967); F. A. L. Anet, R. D. Treoka, and D. J. Cram, ibid., o9, 357 (1967). 21. See, for example; a. Q. R. Boggs and J. T. Gerig, J. Ora. Chem.. 34, 1484 (1969); b. S. K. Brauman and M. E. Hill, J. Chem. Soc.. B, 1091 (1969); c. G. W. Gribble, N. R. Easton, Jr., and J. T. Eaton, Tet. Let.. 1075 (1970). ; 22. J. D. Roberts, Chem. Brit., 2, 529 (1966). 23. C. H. Bushweller and J. W. O'Neil, J. Am. Chem. Soc.. 92, 2159 (1970). 24. J. B. Lambert and W. L. Oliver, Jr., ibid., 91, 7774 (1969). 25. J. M. Lehn and J. Wagner, Chem. Commun.. 148 (1968). 26. D. L. Griffith and J. D. Roberts, J. Am. Chem. Soc.. 87, 4087 (1965). 27. F. G. Riddell, J. M. Lehn, and J. Waaner. Chem. Commun.. 1403 (1968). 28. M. Raban, ibid., 1017 (1967). 71 29. M. Jautelat and J. D. Roberts, J. Am. Chem. Soc.. 91, 642 (1969). 30. V. Rautenstraueh, Chem. Commun.. 1122 (1969). This was an equilibrium study done at 23°C. 31. S. J. Brois, Trans. N. Y. Acad. Sci., 31. 931 (1969). 32. See further Reference 20d. 33. See further References 20c and 20d. 34. J. E. Anderson, Chem. Commun.. 284 (1968). 35. N. J. Leonard and J. V. Paukstelis, J. Org. Chem., 28, 3021 (1963). 36. a. K. Pfordte and G. Leuschner, Ann.. 646, 23 (1961)? b. K. Shima and S. Tsutsumi, Kogyo Kagaku Zasshi, 64, 460 (1961), as reported in CA, 57, 2088d (1962). 37. H. S. Gutowsky, R. L. Void, and E. J. Wells, J. Chem. Phvs.. 43, 4107 (1965). A part of this program was developed at the Computer Sciences Laboratory at the University of Southern California. 38. a. J. C. Dalton, P. A. Wriede, and N. J. Turro, J. Am. Chem. Soc., 92, 1318 (1970); b. N. J. Turro, P. A. Wriede, and J. C. Dalton, ibid.. 90. 3274 (1968). 39. a. N. J. Turro and P. A. Wriede, ibid., 92, 320 (1970); b. N. J. Turro and P. A. Wriede, ibid.. 90, 6863 (1968). 40. V. P. Muralidharan, Unpublished Results. 41. H. S. Gutowsky and C. H. Holm, J. Chem. Phvs.. 25, 1228 (1956). 42. "Variable Temperature System" under the tab marked Reference Information in the NMR Spectrometer System Manual, p. 5-1.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

The acid catalyzed isomerization and dimerization of styrene oxide

PDF

Medium effects on the excited states of 1-pyrenecarboxaldehyde and photocycloaddition of fluorenone to ketenimine.

PDF

Kinetics of the reductive elimination reaction of trans-bromohydroxytetracyanoplatinate ion with thiosulfate and tetrathionate ions

PDF

Synthesis and reactivity studies of cyclopentadienylmanganese diphosphine complexes

PDF

The rate of aquation of the sulfito ligands in disulfitotetracyanocobaltate (III) ion

PDF

Tests and properties of crude petroleum emulsions

PDF

The reaction of phenylmagnesium bromide with methyl B-benzoylpropionate and with diethyl succinate

PDF

The purification and study of inulin

PDF

Studies of 2-nitrobenzenesulfenyl bromide and 2,4-dinitrobenzenesulfenyl bromide

PDF

The chemistry of 3-methyl-3-nitro-1, 2-butanedicarboxylic acid

PDF

The catalytic conversion of parahydrogen by paramagnetic ions in aqueous solution

PDF

Development of a flow NMR system

PDF

Kinetics and mechanism of the substitution of water in aquapentacyanocobaltate (III) by pyridine

PDF

The stereochemical course of the nucleophilic displacement of cyanide from alpha-ferrocenyl-alpha-aminonitriles, and model experiments for the synthesis of optically active alpha-ferrocenyl alkyl...

PDF

Syntheses in the thianaphthene series

PDF

I. Intramolecular reaction of alkenyl substituted silylenes. II. Determination of the relative migratory aptitudes of a trimethylsilyl group versus a hydrogen to a carbenoid center

PDF

The reaction H+ClCN to CN+HCl

PDF

The preparation and properties of some derivatives of amino-borine

PDF

Some properties of the tetrakis (ethylenediamine)-u-dioldichromium (III,III) and - Dicobalt (III,III) complex ions

PDF

The synthesis of bis (triphenylphosphine) iminium salts of mononuclear and polynuclear iron carbonyl anions and the crystal structure of the hydridotetracarbonyl iron (II) anion, HFe (CO)4

Asset Metadata

Creator Knutsen, Roger Lief (author)

Core Title The photocycloaddition of acetone to ketenimines and the syn-anti isomerization of the B-adducts

Contributor Digitized by ProQuest (provenance)

Degree Master of Science

Degree Program Chemistry

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

Tag chemistry, organic,OAI-PMH Harvest

Language English

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c17-796068

Unique identifier UC11347825

Identifier EP41652.pdf (filename),usctheses-c17-796068 (legacy record id)

Legacy Identifier EP41652.pdf

Dmrecord 796068

Document Type Thesis

Rights Knutsen, Roger Lief

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 au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA