University of Southern California Dissertations and Theses

Molecular orientation and structure of hydrogen bonds at water interfaces

(USC Thesis Other)

Molecular orientation and structure of hydrogen bonds at water interfaces

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content MOLECULAR ORIENTATION AND STRUCTURE OF HYDROGEN BONDS AT WATER INTERFACES by Chayan Dutta 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) December 2017 Copyright 2017 Chayan Dutta To all my Teachers ii Abstract The unique ability of water to form extended hydrogen bond network in the bulk en- vironment is responsible for many unusual physical properties of liquid water. This extended hydrogen bond network is terminated at the interface producing an inherent inhomogeneity in structure, orientation and distribution of hydrogen bonds at the inter- face. Because of these intrinsic properties interfacial water plays a significant role in a variety of systems, e.g., biological membranes, environmental interfaces, electrochemi- cal interfaces, heterogeneous catalysis etc. Understanding the basic interactions of water at the interface is crucial. We have used surface sensitive spectroscopic techniques to investigate into the nature of hydrogen bonds and molecular orientations at the inter- face. This thesis is mainly focused on assigning the water bending mode at the air/water interface, establishing the bend mode as a complementary probe of hydrogen bonding and investigating molecular orientation of different hydrogen bonded species at neutral interfaces, chemically and electrochemically charged surfaces and in micro-emulsion confinement. iii Acknowledgements Graduate school at USC has been a great learning experience in my carrier and a true defining chapter in my life. It goes without saying that this journey would have been different without the guidance and constant support of several people and it gives me great pleasure to express my gratitude to them. I will be indebted forever to my graduate advisor, Dr. Alexander V . Benderskii, for giving me an opportunity to work in his laboratory. He has always given me the necessary freedom for critical thinking and encouraged to pursue my research ideas. His passion and enthusiasm for science and research is contagious and his simple outlook towards life is exemplary. I thank my dissertation committee members, Prof. Jahan Dawlaty and Prof. Aiichiro Nakano for their valuable time and helpful discussions and comments. I also thank two other members of my qualifying examination committee, Prof. Susumu Takahashi and Prof. Sri R. Narayan for their gracious involvement in my screening and qualifying examinations. I greatly appreciate the Department of Chemistry and Graduate School at University of Southern California for this opportunity and the financial support provided throughout my stay at the graduate school. iv I am fortunate to have met and learned from some awesome people in the Bender- skii lab. Dr. Sergey Malyk was kind enough to tolerate me during my early days at USC. The lessons and tricks of ultrafast spectroscopy and optics I learned from Sergey prepared me for the later part of my graduate life after Sergey left USC and started his career in Industry. I wish him and his family success and happiness. I would also thank other members of the Benderskii group, Purnim Dhar, Angelo Montenegro, Dhritiman Bhattacharya, Muhammet Mammetkuliyev and Ariel Nessl. Purnim was not only a wonderful senior and mentor but also a great friend. Angelo was a great help in the laboratory. We had a fun time and great learning experience through mutual failure. I will dearly cherish Angelo and his wife Kim’s friendship. I am thankful to Muhammet and Dhritiman for proofreading some of the chapters of my thesis. Muhammet shared some of his Matlab codes which were helpful in some of my works. Our latest addition to the group, Ariel surely is a great candidate to “Carry the LASERS”. I had the good fortune to work with two summer undergraduate researchers, Anton Svirida and Vadim Trepalin. I wish them great success. Part of my thesis work would have been incomplete without the contributions from our collaborators, Dr. Stephen Cronin’s group at USC and Dr. Francesco Paesani’s group at the University of California, San Diego. Bingya Hou, Haotian Shi and Lang Shen from the Cronin group did an excellent job in keeping up with our ever-growing need for new graphene samples. We collaborated with the Paesani group on the water orientation project. I am grateful to Dr. Daniel Moberg and Shelbey Straight for sharing their results. I am thankful to the ultrafast community at USC. I will miss our joint group meet- ings on Wednesday evenings with the Bradforth group and the Dawlaty group. These meetings were great opportunity to present our current research works and sharing ideas. v I am happy to acknowledge our Physical Chemistry community at SSC for making graduate school a memorable journey by cheering our success together and by being there in time of crisis. I would also like to thank our Indian community at SSC for being a great support over the years. Amit Samanta, Parichita Majumdar, Saptaparna Das, Anirban Roy, Atanu Acharya, Subodh Tiwari, and Piyush Deokar were great seniors and friends who made sure we feel at home after coming to the US for the first time and lend a helping hand whenever needed. Amit, Saptaparna and Atanu were great help during my screening and qualifying preparations. Cheers to Atanu for teaching me how to do some simple calculations for my qualifying project and sharing his L A T E Xtemplete. I am glad I met Sougata Pal and his family who welcomed us to their home with homemade Indian food and pure “Adda”. This list won’t be complete without mentioning Haipeng, Subhasish, Satya, Ankit, Gaurav (aka. FNU) and Deepak, my amazing roommates or apartment mates over the years. I will cherish the time we spent together and wish them a great future ahead. I would like to thank my Parents, Sister and Brother for their love, support and sacrifices. I find great source of joy and love in my nieces, Sonai and Monai, who will always have a spacial place in my heart. I also appreciate the love and constant support from my In-laws, sister-in-law (Nilanjana) and her husband (Achin). Finally I would like to acknowledge my wonderful wife, Arijita for putting up with me from our college days. She has always believed in me and encouraged me to persue my dreams. She deserves a special “Thanks” for carefully proofreading my thesis. Its been an eventful journey and I am furtunate to have her by my side. I am thankful to her for all the love, support and togetherness. vi Table of Contents Dedication ii Abstract iii Acknowledgements iv List of Tables x List of Figures xii Chapter 1: Introduction 1 1.1 Water: a simple yet fascinating molecule . . . . . . . . . . . . . . . . . 1 1.2 Spectroscopic Observation of Water . . . . . . . . . . . . . . . . . . . 3 1.3 Surfaces: The Devil of Mother Nature . . . . . . . . . . . . . . . . . . 6 1.4 Surface Specific Techniques . . . . . . . . . . . . . . . . . . . . . . . 8 1.5 Thesis Outline . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 Chapter 1 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 Chapter 2: Vibrational Sum Frequency Generation Spectroscopy 23 2.1 Sum Frequency Generation . . . . . . . . . . . . . . . . . . . . . . . . 23 2.1.1 Experimental Setup . . . . . . . . . . . . . . . . . . . . . . . . 24 2.2 Orientational analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 2.3 Appendix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 2.3.1 Single-Shot Autocorrelator . . . . . . . . . . . . . . . . . . . . 35 Chapter 2 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 Chapter 3: Water bending Vibration: a Complementary Probe of Hydrogen Bonding 39 3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 3.2 Assignment of the bend spectra . . . . . . . . . . . . . . . . . . . . . . 41 3.2.1 Previous Studies . . . . . . . . . . . . . . . . . . . . . . . . . 41 3.2.2 Experimental Results and Discussion . . . . . . . . . . . . . . 43 3.3 Appendix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 vii 3.3.1 Tantative construction of the bend spectra from the stretch spectra 49 3.3.2 SFG in D 2 O . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51 Chapter 3 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 Chapter 4: Comprehensive Molecular Picture of the Air/Water Interface 56 4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56 4.2 Experimental results and discussions . . . . . . . . . . . . . . . . . . . 58 4.3 Orientational Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . 61 4.3.1 Orientation from theoretical calculations . . . . . . . . . . . . . 67 4.3.2 Comprehensive picture of molecular orientation at the air/water interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68 4.4 Appendix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71 4.4.1 MB-MD Simulations and SFG spectra calculation . . . . . . . . 71 4.4.2 Normal Mode Analysis and Orientational Analysis from MD Trajectories . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72 Chapter 4 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76 Chapter 5: Refined Picture of Molecular Orientation at Charged Interfaces 79 5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79 5.2 Results and discusssions . . . . . . . . . . . . . . . . . . . . . . . . . 82 5.2.1 Comparison of water bend SFG spectra in SDS and CTAB solu- tions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82 5.2.2 Change in water bend SFG spectra with CTAB concentration . . 85 5.3 Orientation at the Interface . . . . . . . . . . . . . . . . . . . . . . . . 88 5.4 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92 5.5 Appendix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93 5.5.1 Experimental details and spectral fitting . . . . . . . . . . . . . 93 5.5.2 Comparison of SFG signals from pure water, 0.4 mM CTAB and 1 mM SDS solution . . . . . . . . . . . . . . . . . . . . . . . . 94 Chapter 5 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96 Chapter 6: Surface Propensity of Cations and Anions 99 6.1 Solvated H + and OH at the air/water Interface . . . . . . . . . . . . . 99 6.1.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99 6.1.2 Low Concentration Spectra . . . . . . . . . . . . . . . . . . . . 100 6.1.3 High Concentration Spectra . . . . . . . . . . . . . . . . . . . 101 6.1.4 Orientational Changes in Acid and Base Solutions . . . . . . . . 105 6.2 SFG spectra of salt solutions . . . . . . . . . . . . . . . . . . . . . . . 108 6.2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108 6.2.2 Effect of anions in the VSFG spectra of sodium salts . . . . . . 110 6.2.3 Effect of cations in the VSFG spectra . . . . . . . . . . . . . . 112 6.3 Appendix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117 viii 6.3.1 FTIR measurements . . . . . . . . . . . . . . . . . . . . . . . . 117 Chapter 6 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120 Chapter 7: SFG Spectroscopy at Graphene Electrode Interface 123 7.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123 7.2 Result and Discussions . . . . . . . . . . . . . . . . . . . . . . . . . . 126 7.2.1 CaF 2 /D 2 O Interface . . . . . . . . . . . . . . . . . . . . . . . . 127 7.2.2 Al 2 O 3 /D 2 O Interface . . . . . . . . . . . . . . . . . . . . . . . 128 7.2.3 Graphene/D 2 O Interface . . . . . . . . . . . . . . . . . . . . . 129 7.3 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134 7.4 Appendix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135 7.4.1 Experimental Details . . . . . . . . . . . . . . . . . . . . . . . 135 Chapter 7 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142 Chapter 8: Water Structure in micro-emulsion confinement 146 8.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146 8.2 Experimental Details . . . . . . . . . . . . . . . . . . . . . . . . . . . 150 8.2.1 Preparation of the reverse microemulsion and determination of the particle size . . . . . . . . . . . . . . . . . . . . . . . . . . 150 8.2.2 V olumetric Analysis . . . . . . . . . . . . . . . . . . . . . . . 152 8.2.3 SFG Experiment . . . . . . . . . . . . . . . . . . . . . . . . . 153 8.3 Results and Discussions . . . . . . . . . . . . . . . . . . . . . . . . . . 155 8.3.1 FTIR results . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155 8.3.2 SFG results . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158 8.3.3 Isotopic dilution studies . . . . . . . . . . . . . . . . . . . . . . 161 8.4 Orientational analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . 162 8.5 Appendix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165 Chapter 8 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 168 ix List of Tables 2.1 Character table for C 2v point group. . . . . . . . . . . . . . . . . . . . 33 3.1 Fitting parameters of the vibrational SFG spectra of the water bend mea- sured for SSP and PPP polarizations at the air/water and SDS/water in- terface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 4.1 Average orientation angle for different H-bonded species. . . . . . . . 69 4.2 Fitting parameters of the vibrational SFG spectra of the water bend mea- sured for SSP, PPP and SPS polarizations at the air/water. All spectra were normalized by the maximum intensity of the SSP spectrum. . . . 74 4.3 ‘c’ and ‘d’ parameters for the water bending mode in three polarization combinations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75 4.4 ijk values used for sensitivity analysis. . . . . . . . . . . . . . . . . . 75 5.1 Surface charge density of surfactants (CTAB) for varying bulk surfactant concentrations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87 5.2 Change in tilt angle of the C 2V axis for the free-OH (up oriented) and H-bonded (down oriented) water molecules as a function of CTAB con- centration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92 5.3 Fitting parameters of the vibrational SFG spectra of the water bend mea- sured for SSP, SPS and PPP polarizations at the and CTAB/water inter- face in 0.01, 0.03 and 0.07 mMol concentrations. . . . . . . . . . . . . 95 6.1 Average Orientation of the C 2V axis of the free -OH and the H- bonded water species at the acid and base solution interface . . . . . . . . . . . 106 6.2 Average Orientation of the C 2V axis of the free -OH and the H- bonded water species at the iodide salt solution interface . . . . . . . . . . . . 115 6.3 Fitting parameters of the vibrational SFG spectra of the water bend mea- sured for SSP and PPP polarizations at the 2M acid and base solution interface. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118 6.4 Fitting parameters of the vibrational SFG spectra of the water bend mea- sured for SSP and PPP polarizations at the Iodide salt solution interface. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119 x 7.1 Fitting parameters for the VSFG spectra of the graphene/D 2 O and Al 2 O 3 /D 2 O interfaces in the absence and presence of 1M NaCl. . . . . . . . . . . . 139 7.2 Fitting parameters for the VSFG spectra of the graphene/D 2 O with neg- ative and positive surface potential on the graphene. . . . . . . . . . . 140 7.3 Fitting parameters for the VSFG spectra of the graphene/D 2 O in pres- ence of 1M NaCl with negative and positive surface potential on the graphene. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141 8.1 Dynamic light scattering results and the calculated radius of water pool inside the reverse micelles . . . . . . . . . . . . . . . . . . . . . . . . 152 8.2 Fitting parameters for VSFG spectra in the water stretching region at AOT/ water and AOT/ HOD interfaces measured in SSP and PPP polar- izations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165 8.3 ’C’ and ’D’ parameters used for orientational analysis. aac = bbc =0.32, ccc =1 for 3700 cm 1 and 3560 cm 1 peak and aac = bbc =0.40, ccc =1 for 2616 cm-1 peak. . . . . . . . . . . . . . . . . . . . . . . . . . . . 167 xi List of Figures 1.1 Donor and acceptor hydrogen bonds around a water molecule . . . . . . 2 1.2 Infrared spectra of liquid water at various temperature. . . . . . . . . . 4 1.3 Structure of the air/water interface. . . . . . . . . . . . . . . . . . . . . 7 1.4 Energy level scheme for SFG process with one incoming beam resonant with the vibrational lebels of the molecules. . . . . . . . . . . . . . . . 11 2.1 Schematic of the homodyne detected VSFG setup. . . . . . . . . . . . . 25 2.2 Spectra of the output from the Ti:sapphire oscillator (red) and the com- pressed pulse output for the cavity (blue). . . . . . . . . . . . . . . . . 26 2.3 (a) Spectrum of the 800nm pulse before the 4f stretcher and after the 4f stretcher, measured using an ocean optics spectrometer. (b) Visible pulse spectrum at the sample measured in CCD. . . . . . . . . . . . . . 27 2.4 (a) IR spectrum in the CH stretch region , measured using the MCT detector. (b) SFG spectrum at the gold surface in the CH stretch region. 28 2.5 (a) Frequency-resolved cross-correlation of femtosecond IR pulse and picosecond visible pulse (b) Temporal profile of the visible pulse passed through the 4f- stretcher. . . . . . . . . . . . . . . . . . . . . . . . . . 29 2.6 A three-layer model for SFG in reflection geometry with molecularly thin interface layer. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 2.7 Special intensity distribution of the SFG signal from the BBO crystal captured using a VGA camera for the output (800nm) of the internal compressor (a) and the external compressor (b) . . . . . . . . . . . . . 35 3.1 Illustrations for the different H-bonded classes present at the air/water interface. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 3.2 Top panel: Vibrational SFG spectra of the water bend mode at the air/water interface for PPP (red) and SSP (blue) polarization combinations of SFG, visible, and IR. Black solid lines show fit described in the text. Bottom panel: Two principal resonant Lorentzians (positive and nega- tive) used in the fit (left axis). The background signal fit is shown on top (right axis). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 xii 3.3 The OH-stretch SFG spectrum of the air/water interface in absence (red) and presence (blue) of a monolayer of SDS (1 mM SDS solution)., for both SSP and PPP polarizations. . . . . . . . . . . . . . . . . . . . . . 46 3.4 Comparison of the water bend SFG spectra of the air/water interface in absence (top left) and presence (top right) of an SDS monolayer, for SSP (blue) and PPP (red) polarizations. Solid black lines represent fit to the model described in text. Bottom panels show the two main Lorentzians used in the fitting. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 3.5 Tentative assignment of the water bend SFG spectrum by analogy with the OH-stretch spectrum. . . . . . . . . . . . . . . . . . . . . . . . . . 50 3.6 SFG spectra at air/water (H 2 O) , air/D 2 O , SDS/water (H 2 O) and SDS/D 2 O interface in SSP and PPP polarization combinations. Legends are col- ored according to the corresponding spectra. . . . . . . . . . . . . . . . 52 4.1 Experimentally measured SFG spectra, two main resonant components of the spectra and the calculated SFG spectra of the water bend at the air/water interface. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60 4.2 Experimentally measured SFG spectra taken from Literature and the cal- culated SFG spectra of the water stretch at the air/water interface. . . . . 62 4.3 Calculated SFG amplitude (2) eff as a function of the tilt angle of the HOH bend dipole with respect to the surface normal in SSP, PPP and SPS polarization combinations for SFG-Visible-infrared laser beams . . 63 4.4 Sensitivity analysis of the SFG amplitudes of the water bending mode with the change of hyperpolarizability ratios. . . . . . . . . . . . . . . . 65 4.5 Estimation of possible orientational distribution of H-bonded and free- OH species. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66 4.6 (a) A snapshot from the MB-MD trajectory for the air/water interface. (b) Histogram of up (red) and down (blue) oriented water molecules, with the SSP SFG spectrum (black trace) overlaid. The up oriented wa- ters have a maximum around 1660 cm 1 and the down oriented water molecules have a maximum near 1680 cm 1 . . . . . . . . . . . . . . . 67 4.7 Model illustration of orientation of the water molecules at the interface. 69 4.8 Pictorial description for the angular distribution of orientation of water molecules at the air/water interface. . . . . . . . . . . . . . . . . . . . 70 5.1 Schematic of the water H-bond reorientation for different types of water molecules. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81 5.2 Vibrational SFG spectra of the water bend mode at the air/CTAB solu- tion (Left) and air/SDS solution interface (right) for PPP (red) and SSP (blue) polarization combinations of SFG, visible, and IR. Bulk concen- tration for CTAB and SDS was 0.4 mM and 1.0 mM respectively. . . . . 84 xiii 5.3 Vibrational SFG spectra of the water bend mode at the CTAB/water in- terface with increasing CTAB concentration from 0.01 mM to 0.1 mM for SSP (top), SPS (middle) and PPP (bottom) polarization combina- tions of SFG, visible, and IR. Black solid lines show fit described in the text. PPP spectra in 0.07 mMol has been offset by -0.03 for clarity. . . . 86 5.4 Variation of amplitudes for free-OH molecules (Left) and hydrogen bonded -OH molecules (right) from spectral fitting of the water bending spectra in SSP (blue), SPS (green) and PPP (red) polarization from CTAB/water interface as a function of bulk CTAB concentration. B1 is the amplitude at 1630 cm-1 for molecules with free-OH and B2 is the amplitude at 1662 cm-1 for hydrogen bonded -OH molecules. . . . . . . . . . . . . . 88 5.5 Defined coordinate system for the orientation analysis of the free-OH and H-bonded -OH species. represents the tilt angle of the C 2 V axis of the corresponding water species with respect to the surface normal. . 89 5.6 Calculated SFG amplitude (2 eff as a function of the tilt angle of the water C 2V axis with respect to the surface normal in SSP, PPP and SPS polar- ization combinations for SFG-Visible-infrared laser beams. Change in tilt angle for the free-OH species (a) and Hydrogen bonded species (b) as a function of the surface charge density and the corresponding molec- ular picture at the interface ((c) and (d) respectively). Vertical lines are color coded according to bulk CTAB concentration, purple (0.01 mMol), blue (0.03 mMol) and red (0.07 mMol) respectively . . . . . . . . . . . 91 5.7 Comparative picture of molecular orientation at the SDS/water and CTAB /water interface. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92 5.8 SFG spectra of water bending mode at air/water interface, negatively charged SDS (1mMol) /water interface and positively charged CTAB (0.4 mMol)/water interface. . . . . . . . . . . . . . . . . . . . . . . . 94 6.1 SFG spectra of HCl solution (pH 1.7) and NaOH solution (pH 13.0) at the water bending region . . . . . . . . . . . . . . . . . . . . . . . . . 100 6.2 SFG spectra of 2M solutions of HI, HBr, HCl, NaOH and NaCl at the water bending region . . . . . . . . . . . . . . . . . . . . . . . . . . . 102 6.3 SFG amplitudes of the free-OH (B 1 / 1 ) and the H-bonded (B 2 / 2 ) water bend peak in 2M solutions of HI, HBr, HCl and NaOH . . . . . . . . . 105 6.4 Illustrations of the plausible Hydrogen bond structure at the aqueous halide solution interface . . . . . . . . . . . . . . . . . . . . . . . . . . 107 6.5 SFG spectra of 2M solutions of sodium salts (NaI, NaBr, NaCl, NaOH) at the water bending region . . . . . . . . . . . . . . . . . . . . . . . . 111 6.6 SFG spectra of 2M solutions of iodides (HI, LiI, NaI and KI) at the water bending region . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112 6.7 SFG amplitudes of the free-OH (B 1 / 1 ) and the H-bonded (B 2 / 2 ) water bend peak in 2M solutions of LiI, NaI, KI and HI . . . . . . . . . . . . 113 xiv 6.8 SFG spectra of 2M solutions of bromides (HBr, LiBr, NaBr) at the water bending region . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114 6.9 SFG spectra of 2M solutions of chlorides (HCl, LiCl, NaCl) at the water bending region . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114 6.10 Illustrations for a cartoon picture of the plausible Hydrogen bond struc- ture at the salt solution interface . . . . . . . . . . . . . . . . . . . . . 116 6.11 FTIR spectra of 2M solution of salts . . . . . . . . . . . . . . . . . . . 117 7.1 Schematic of Single layer graphene/water interface. . . . . . . . . . . . 125 7.2 VSFG spectra of D 2 O in the -OD stretch region in the absence and pres- ence of 1M NaCl solution at the CaF 2 /D 2 O interface. . . . . . . . . . . 127 7.3 VSFG spectra of D 2 O in the -OD stretch region in the absence and pres- ence of 1M NaCl solution at the Al 2 O 3 /D 2 O interface. The black traces are fits to the raw spectra. . . . . . . . . . . . . . . . . . . . . . . . . . 128 7.4 VSFG spectrum of D 2 O at graphene and Al 2 O 3 interfaces in presence and absence of 1M NaCl. . . . . . . . . . . . . . . . . . . . . . . . . . 130 7.5 VSFG spectra of D 2 O in the -OD stretch region at the graphene/D 2 O interface at various surface charges in presence (a) and absence (b) of 1M NaCl. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132 7.6 Frequency shift of the free-OD with surface charge . . . . . . . . . . . 134 7.7 Schematic of the sample cell used in our SFG experiments (Left). Single layer graphene on CaF 2 window (right). . . . . . . . . . . . . . . . . . 135 7.8 Sample characterization before and after SFG experiments . . . . . . . 136 7.9 Frequency shift of the G-band with external potential using Raman spec- troscopy. Carrier concentrations at the corresponding voltages were also calculated and ploted in the right axis. . . . . . . . . . . . . . . . . . . 137 8.1 Interfacial and bulk water in reverse micelles (top left) and surfactant monolayer/water interfaces (top right) used in this study. . . . . . . . . 147 8.2 (a)Structure of sodium 1,4-bis-2-ethylhexylsulfosuccinate (AOT), bot- tom left and Polyoxyethylene(4)lauryl ether (Brij L-4), bottom right. (b) Reverse micelle structure showing the total radius (Rh), radius (R) of the water pool inside the reverse micelle and the length (Ltail) of the surfactant tail group . . . . . . . . . . . . . . . . . . . . . . . . . . . 151 xv 8.3 SFG spectra of OH-stretching band at the air/water interface for ssp and ppp polarization combinations of the SFG, visible and the IR pulses measured at different center frequency for the IR beam and stitched to- gether afterwards during the data analysis. The right panel shows the full spectra at the water stretching region where the sharp peak around 2750 cm-1 is due the dangling Oh molecules at the interface and the red shifted broad peaks are due to hydrogen bonded water molecules. All other SFG spectra presented in the paper have been analyzed by this method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154 8.4 SFG spectra of OH-stretching band at the air/water interface for ssp and ppp polarization combinations of the SFG, visible and the IR pulses measured at different center frequency for the IR beam and stitched to- gether afterwards during the data analysis. The right panel shows the full spectra at the water stretching region where the sharp peak around 2750 cm-1 is due the dangling Oh molecules at the interface and the red shifted broad peaks are due to hydrogen bonded water molecules. All other SFG spectra presented in the paper have been analyzed by this method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156 8.5 Separated spectral responses of bulk (dotted) and interfacial (solid) wa- ter regions for AOT (blue) and Brij (red) reverse micelle determined from the fits of A/zN versus r, where r is the radius of different micro- emulsion solutions. Solid black line shows the pure bulk water IR ab- sorption spectra. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157 8.6 (a) SFG spectra of OH-stretching band in AOT/water interface (left panel) and Brij/water interface (right panel) for ssp and ppp polarization com- binations of the SFG, visible and the IR pulses. Interface spectral re- sponses separated from FTIR spectra of the AOT and Brij reverse mi- celle is plotted in the same graphs for comparison. (b)Comparison of the SFG spectra of air/water, AOT/water and Brij/water interfaces for SSP (left) and PPP (right) polarization combinations of the SFG, visible and the IR pulses. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159 8.7 Isotopic dilution study at the HOD/AOT interface. Homodyne detected SFG spectra of OD stretch at HOD/AOT interface. D 2 O and H 2 O were mixed at 1:3 ratio so that D 2 O/HOD/H 2 O is 1:6:9. SSP and PPP polar- ization spectra are plotted in the left and right panel respectively. Spectra were fit with three resonant lorentzians with center frequencies at 2540 cm 1 , 2616 cm 1 and 2670 cm 1 . . . . . . . . . . . . . . . . . . . . . 161 8.8 Simulated SFG intensity of the 3560 cm 1 /2616 cm 1 bond at different tilt angle with distribution for SSP and PPP polarization at the air/H 2 O (Left) and at the air/HOD (Right) interface respectively. The gray line represents the experimental intensity ratio in PPP and SSP polarization (I PPP /I SSP ) values which are 2.46 and 1.82 . . . . . . . . . . . . . . . 164 xvi 8.9 4 lorentzian fitting for the SFG spectra in SSP polarization (a) and in PPP polarization (b) at the H 2 O/AOT interface and the contribution of each peak towards the overall spectra. Vertical colored lines are placed as a guide to the center frequencies for each of the fitted lorentzians. . . 166 xvii Chapter 1: Introduction 1.1 Water: a simple yet fascinating molecule Water is the most abundant liquid on Earth’s surface (more than 70%) that is essential to sustain life. The entire animal and plant kingdom depend on water to maintain most of their physiological activities. It is simple and most stable liquid that the living world use in many ways everyday. As Nobel Laureate A. Szent-Gyorgy once aptly said ”Water is life’s matter and matrix, mother and medium. There is no life without water”. Yet the pecularity of water lies in its simplicity. Water is a triatomic, polar molecule and a liquid at room temperature with a high dipole moment of 1.83 Debye (1 Debye = 3..34 x 10 30 C m) directed along the C 2V axis of symmetry. Even though water is a stable molecule, the high electronegativity difference between the oxygen and the hydrogen atoms renders the oxygen atom a partial negative and the hydrogen atoms a partial positive charge. This dipolar nature of water molecules play a significant role in determining various physical properties of liquid water, e.g., unusually high specific heat and surface tension, high melting and boiling points and its unique ability to form extensive ”hydrogen bond” (this will be referred to as ’H-bond’ throughout this thesis) network. The electronic structure of water consists of two bonds with two hydrogen atoms and two non-bonding orbitals on oxygen atoms, each occupied by two lone pairs of 1 electrons. The particular tetrahedral arrangement of these four orbitals around the oxy- gen atom facilitates the formation of two donor H-bonds (through two hydrogen atoms on the water molecule) and two acceptor H-bonds (accepting through two lone pairs of electron on the oxygen atom), as shown in Figure 1.1. Due to low average energy (3-7 kcal/mol) of OHO bond of liquid water at room temperature, the network of hydrogen bonding is very dynamic. In other words, the hydrogen bonds are being made and broken constantly on sub-pico second (< 1ps) timescale 1 and the average number of hydrogen bonds around a water molecule changes. This leads to a very random orien- tation of hydrogen bonds in the bulk water. This isotropic picture changes dramatically when one approaches from bulk to the interface of any aqueous system. The dynamic hydrogen bonding network of water is resposible for many key processes in chemistry, biology and atmospheric science. One practical example would be the role of hydro- gen bonds in vapor condensation at sufficient pressures and the interactions between atmospheric aerosol particles and water vapor which leads to the formation of clouds. Water being ’The Solvent of life’, plays a crucial role in many biological, chemical and physical processes and many of these processes occur at the interface. Figure 1.1: Donor and acceptor hydrogen bonds around a water molecule. 2 1.2 Spectroscopic Observation of Water As discussed earlier, water plays central role in determining various key features of many different systems. Due to the size and nature of these systems and the fleeting, labile nature of hydrogen bond configurations in liquid water it is almost impossible to apply a particular technique to all such systems. X-ray scattering, neutron scattering, NMR spctroscopy and microwave spectroscopy are some of the common experimental techniques employed to study H-bonded systems. However, each of these techniques are unique and limited in their information content and only applicable to a particu- lar type of system. For example, X-ray scattering is a powerful method to determine structure of ordered crystalline systems which makes it very useful in studies of ice and crystalline hydrates in mineral chemistry or geology. However, in disordered sys- tems or semi-crsytalline large macromolecular systems are not best understood by this technique mainly due to its limitation in precisely observing less ordered or more la- bile systems. Neutron scattering is complementary to X-ray scattering for observing H-atoms or D-atoms in Hydrogen bonded clusters. Even though these techniques are highly sensitive and precise in determining the intermoleculer distances between water molecules (through intermolecular pair correlation function of a system, 2, 3 e.g, O-O cor- relations or O-H/O-D correlations etc.) these experiments are limited by the availability of a scattering source and extreme experimental conditions. Vibrational spectroscopy is a widespread technique that provides direct molecular information in the condensed phase. It offers an observation window into the structure and dynamics of aqueous hydrogen bonding network. 4, 5 Particularly, Infrared (IR) spec- troscopy is hypersensitive to H-bonds, the number of H-bond around water molecules, 3 and intra- and inter moleculer coupling between neighbouring water molecules. Wa- ter is a triatomic molecule with 3 vibrational normal mode, symmetric stretching ( s or 1 , 3657 cm 1 ), assymetric stretching ( a or 3 , 3756 cm 1 ) and bending ( b or 2 , 1594.6 cm 1 ) 6, 7 mode. Frequencies of these vibrations are sensitive to their environment and number of hydrogen bonding partners in liquid phase. This is pre- cisely the reason that the IR spectrum of liquid water is markedly different from the gas phase spectrum. In liquid water the stretching band of water consists of a broad peak (3100-3700 cm 1 ) red shifted from that of gas phase frequencies (Figure 1.2). H- Figure 1.2: Infrared spectra of liquid water at various temperature.Reproduced from Ref. 8 bonding between water molecules weakens the OH bond in a water molecule which in turn shifts the vibrational frequencies to the red side of the spectrum. The red shift of the OH-stretch vibration is dependent on the degree of H-bonding and different water moieties with different degree of H-bonds overlap in frequency. Much of the research has focused on steady-state and time-resolved spectroscopy of the OH-stretch mode to study both bulk 4, 5, 9 and interfacial water. 10–17 4 Vibrational frequency of the water bend is blue shifted relative to liquid water 7, 18–21 ( 1645 cm 1 ) and amorphous ice 22 ( 1670 cm 1 ). The blue shift of the bending frequency in H-bonded system can be qualitatively understood by the fact that, due to H-bonding more energy is required for the bending motion to overcome the H-bonding energy barrier. Hence, the stronger the H-bonding environment the more the vibrational frequency of the water bend is shifted towards the blue side of the spectrum. There is an empirical linear correlation (see equation 3.1) between the blue-shift of the bend and the red-shift of the OH stretch frequency. 6 The water bend mode can also be used as a spectroscopic probe of the aqueous H-bonds. However, water bend spectroscopy has not received as much attention as the O-H stretch, in part due to its weaker transition dipole 23 and weaker frequency dependence on the H-bond strength. The blue-shift of the bend is a factor of 4 smaller than the red-shift of the OH-stretch. 6 We will come back to this discussion about water bend spectroscopy and how the complemetary nature of water bending to that of water stretching can be exploited for understanding of aqueous interfaces in Chapter 3. Another interesting feature in liquid water IR spectra appears in the low frequency region (0-700 cm 1 , Figure 1.2), know as the libration modes. IR spectra of water vapour shows rotational substructure in the far-IR region due to free rotations of iso- lated H 2 O molecules. These rotations are mixed with internal vibrations which gives rise to the “rotational structure” in the spectrum. In liquid water, the free rotation of isolated water molecules is restricted and controlled by H-bonds formed between water molecules. 5 1.3 Surfaces: The Devil of Mother Nature Surfaces or interfaces are integral parts of heterogeneous systems. It is the outermost layer of molecules or atoms in a medium or in between two media. Although atomi- cally thin, the physical, electronic and optical properties of interfaces or molecules at the interface are markedly different than that of the bulk medium. Because of this, many interesting and important chemical and physiological processes occur at surfaces or interfaces. Ions, enzymes or protein transport through cell membrane, adsorption des- orption of oxygen and carbon dioxide in the respiratory system, absorption-desorption of atmospheric gases and particles in clouds, heterogeneous catalysis are only a few of the examples where surfaces play a key role. Many of these processes occur at the in- terface because of the inherent asymetry present at the interface. For example, in bulk liquid water the orientation of different molecules are random, i.e, molecules in the bulk experience similar symmetric environments as its neighbours. In going form bulk to the surface, this symmetry is broken which makes the surface heterogeneous and more reactive than the bulk. Due to this different behaviour, surfaces are an intriguing subject of research. However, in the past, due to the unavalability of surface specific techniques it was relatively unexplored and our understanding of the surface was somewhat blurry. Recent advancement in experimental and theoretical methods in the past few decades have stirred our interest in the study of surfaces and interfaces. Nobel Laureate Wolfgang Pauli once said: ”God created the bulk; the devil invented surfaces”. It is indeed true, considering the differences between the bulk and the sur- faces and the level of complexity associated with the understanding of surface processes. Understanding the true nature of surfaces, especially, the aqueous interfaces is critical. 6 Figure 1.3: Structure of the air/water interface.Reproduced from Ref. 24 The H-bonding structure and dynamic motion of water molecules at the air/water in- terface is the focus of many experimental and theoretical studies not only because of its complex nature but also due to its applicability in many physical, chemical and bi- ological processes. 25 The extended H-bonding network of bulk water is terminated at the interface producing an inherent inhomogeneity in structure, orientation and distri- bution of H-bonds which leads to many unusual physical properties of surface water. Figure 1.3 shows a molecular picture of the air/water interface and various types of hydrogen bonded species present at the interface. One of the most interesting feature of the aqueous interface is the presence of ’non-hydrogen bonded’ or ’Dangling-OH’ molecules. 26, 27 While the presence of such species in the bulk is still debatable, these constitute almost 30% of the interfacial layer at the air/water interface, 16, 24 solid min- eral interfaces, 28, 29 surfactant inetrfaces 30–34 and at electrochemical interfaces. 35–39 The reduced dimensionality of hydrogen bonds are expected to govern the chemical and physical processes at various types of aqueous interfaces, however, their role in specific processes are yet to be completely understood. 7 1.4 Surface Specific Techniques Understanding processes happening at the atomically thin interface requires state-of-the art techniques that have extreme surface specificity with minimum perturbation to the system under study and that can be performed under ambient conditions. Many exist- ing techniques, such as X-ray scattering, 40 high resolution electron energy loss spec- troscopy (HREELS), 41 scanning tunneling spectroscopy 42 etc. depend on the interac- tion of electrons or x-rays with the surface under ultra high vaccum (UHV) conditions. Scanning probe techniques, such as scanning tunneling microscopy (STM), atomic force microscopy (AFM) etc. and optical spectroscopic techniques, such as, near field optical scanning microscopy (NSOM), 43 surface enhanced Raman spectroscopy (SERS), 44 tip enhanced Raman spectroscopy (TERS) 45 etc. provide some degree of surface specificity and are invaluable in characterizing surfaces. However, many of these are restricted in their application to particular types of systems and/or surface signals are overwhelmed by the bulk response. Even-order nonlinear spectroscopic techniques are proven to be efficiently applica- ble to the studies of liquid interfaces or specifically aqueous interfaces. Under electric- dipole approximation these processes are forbidden in centrosymmetric medium. Due to the non-centrosymmetric character of surfaces or interfaces (as discussed earlier), these processes are very selective towards generating a strong surface signal which is essential to obtain molecular level information from the interface. Many reseach groups in the last 30 years have contributed towards the tremendous developement of these techniques. Specifically, second harmonic generation (SHG) and sum frequency generation (SFG) spectroscopy have been proven to be indispensable in studing steady-state (molecular 8 orientation, vibrational or electronic states) and time-resolved (pico- and femto second dynamics) measurements of liquid interfaces. 10, 11, 34, 46–51 Second Order nonlinear Spectroscopy When an intense optical field interacts with a material, the induced polarization can be written as a power series in the field strength E j as 52, 53 P i = (1) jk E j + (2) ijk E j E k + (3) ijkl E j E k E l +::: (1.1) =P (1) +P (2) +P (3) +::: (1.2) The first term, P (1) , is linear with the optical field which is the basis of linear spectro- scopic techniques such as optical reflection or absorption. (1) jk is the linear susceptibility. All other terms in equation 1.1 (P (n) ; n2) represents non-linear optical processes. A detailed description of these different processes can be found in these references. 52–54 Nonlinear spectroscopies reveal much information about complex molecular systems which are otherwise not achievable by linear techniques. Even-order nonlinear processes, specially 2 nd order spectroscopies are very useful in providing molecular level information about surfaces and interfaces as they are sur- face selective. In a second order process two optical beams interact at the surface of the material to generate a signal in the phase-matched direction which is proportional to (2) ijk . Second order susceptibility which describes the material response under the influence of two incoming optical beams. (2) ijk is a third-rank tensor with 27 elements and is connected to the molecular hyperpolarizability ( abc ). It is possible to detect this second order signal from the surface or interface of a material and gain molecular level 9 information (such as molecular orientation) through detailed analysis of the signal. This will be discussed in detail in Chapter 2. The surface selectivity of the second order process can be understood from symmetry considerations. From equation 1.1 the second order polarization can be written as P (2) i = (2) ijk :E j E k (1.3) Under the inversion operator, the sign of the electric fields will change and the sign of the second order polarization will also change. (2) ijk , on the other hand, is a material property and will remain unchanged under inversion operation for a centrosymmetric material. This gives us P (2) i = (2) ijk :E j E k (1.4) Both of these equations 1.3 and 1.4 can be true only when, (2) ijk = 0. Hence, in a cen- trosymmetric environment, such as in bulk material, the second order response is zero. On the contrary, at the surface or interface where there exist no inversion symmetry, the second order response is non-zero. A non-zero second order (2) ijk can also be under- stood by considering the properties of the third-rank tensor under inversion symmetry. For centrosymmetric environments all directions are equivalent, so the value of (2) ijk for two opposing directions must be equal, i.e, (2) ijk = (2) ijk (1.5) However, a sign change in the subscript of the third-rank tensor is equivalent to a reversal in the axis system. thus (2) ijk = (2) ijk (1.6) 10 Again, to satisfy both equation 1.5 and 1.6, (2) ijk must be equal to zero. Sum frequency generation spectroscopy Sum frequency generation (SFG) is a second order non-linear technique where two input optical beams (with frequency! 1 and! 2 ) interacts at the surface or interface to generate Figure 1.4: Energy level scheme for SFG process with one incoming beam resonant with the vibrational lebels of the molecules. a signal in the phase-matched direction with an energy equal to the two incident optical fields. In case of vibrational sum frequency generation (VSFG), one of the input beams is resonant with the vibrational energy (in the IR region) levels of the molecules at the surface and the other input beams is usually very high in frequency (usually in the visible region), so the signal beam frequency comes out to be ! SF =! IR +! vis (1.7) Figure 1.4 depicts the energy level scheme for the broadband SFG (BB-SFG) process details of which will be discussed in the next chapter. It should be noted here that a gross selection rule of the SFG process can be deduced from the energy level scheme 11 depicted here. Since the IR beam has to be resonant with the vibrational energy levels of the molecules to resonantly enhance the SFG signal, the molecule has to be IR active, i.e, Q 6= 0, where, is the dipole moment of the molecule. The next step involving the upconversion with visible frequency followed by the generation of the SFG beam is equivalent to the anti-Stoke Raman process. Thus, the molecules at the surface has to be Raman active, i.e, Q 6= 0, where, is the Raman polarizability of the molecules. Hence, in order to be SFG active, the molecules have to be both IR and Raman active. VSFG spectroscopy at the air/water interface VSFG spectrum of the air/water interface in the water stretching region was first re- ported by Shen and co-workers. 55, 56 Following these pioneering work, many other re- search groups have studied the water interface using improved analysis procedures and advanced laser systems to unravel various problems related to structure and orientation of molecules at the interface. 11, 51, 57, 58 In general, the water stretch spectra at the in- terface consists of a broad H-bonded region in the red side of the spectrum (3000-3600 cm 1 ) and a sharp blue shifted peak3700 cm 1 . Different types of H-bonded species with different degree of H-bond contribute to the overall spectra which overlap to give this broad feature. In addition to that, inter- and intra- molecular coupling complicate the analysis of the spectra. However, the water bending vibrational mode at the interface has not been explored much which in principle could produce an wealth of information about the structure and orientation of H-bonds, inter-molecular coupling mechanisms and energy relaxation from the stretch to the librational modes via the bending mode. We have studied the interfacial water using the water bend mode as a vibrational probe to understand and develop this relatively unexplored topic. We have also used the water 12 stretching mode for some of the specific projects presented in this thesis, all of these projects are summarized in the next section. 1.5 Thesis Outline The aim of this thesis is to understand the molecular orientation and structure of dy- namic hydrogen bonding network of water at aqueous interfaces using surface specific vibrational spectroscopic technique. We began our discussion by summarizing the ubiq- uitous nature of liquid water and various experimental and theoretical methods used to study liquid water. Then we had discussed the differences between bulk and interfa- cial water and provided an overview of some of the scientific problems associated with interfacial water. In Chapter 2, we will discuss the basics of non-linear spectroscopic techniques, in particular the surface sensitive broadband vibrational sum frequency generation (SFG) spectroscopy. The influence of optical electric field polarization on the SFG spectra and the analysis of polarization dependent SFG spectra to extract the orientation information of specific vibrational modes at the interface will be reviewed. We will make use of these analyses throughout the course of this thesis. In Chapter 3, the SFG spectrum of the water bend vibrational mode ( 2 ) at the air/water interface measured using vibrational SFG technique, for the first time in our laboratory, is presented. 59 Experimental evidence to aid the assignment of the 2 spec- tral features to H-bonded classes of interfacial water has been presented, 60 which is in general agreement with two recent independently published theoretical studies. 61, 62 The dispersive line shape shows an apparent frequency shift between SSP vs. PPP po- larization combinations (SFG-visible-IR). This is naturally explained as an interference 13 effect between the negative (1630 cm 1 ) and positive (1662 cm 1 ) peaks correspond- ing, respectively, to free-OH and H-bonded species, which have different orientations and thus different amplitudes in SSP vs. PPP spectra. Surfactant monolayer of sodium dodecyl sulfate (SDS) was used to suppress the free OH species at the surface, and the corresponding SFG spectral changes indicate that these water molecules with one of the hydrogens pointing up into the air phase, contribute to the negative peak at 1630 cm 1 . In Chapter 4, a detailed quantitative analysis of the water orientation at the air/water interface based on polarization sensitive SFG spectra of water bending vibrational mode at the air/water interface will be provided. Spectral lineshapes for water bending is mostly dispersive with the resonant features sitting on top of a broad non-resonant back- ground. As water bending is affected by both the OH groups on a single water molecule, orientational analysis using water bending vibration on the free-OH species or the H- bonded species provides more localized information as complementary to that of water stretching spectroscopy. Our orientational analysis using the polarization dependent SFG spectra reveals an average orientation of 80 5 for the water bending dipole from the surface normal for a free-OH species on the surface, and an average of 107 5 for the H-bonded species. In Chapter 5, a refined picture of molecular orientation of water molecules at charged surfactant aqueous interfaces will be presented. Spectral line shape of the water bend SFG spectra at the positively charged surfactant cetyltrimethylammonium bromide (CTAB) /water interface is polarization dependent and different from that of negatively charged surfactant sodium dodecyl sulfate (SDS)/water interface. Orientational analysis using water bending as a vibrational probe at these different surfaces reveals structural differences between them and provides orientational changes of differently hydrogen bonded species around the surfactant head groups as the surface charge density changes. 14 We have reported the systematic sign flipping of the PPP bend spectra at CTAB/water in- terface in a concentration dependent study. We believe that careful analysis of the CTAB concentration dependence of the water bend spectra in different polarization combina- tion provides quantitative information on the surface charge required for a complete re- versal of water structure at the interface. In general, the C 2V axis of the water molecules with free-OH bonds moves away from the bulk water at a positively charged interface, whereas water molecules with higher degree of hydrogen bonding orient themselves to- wards the bulk water mainly because of charge dipole interaction with the surfactant head groups. In Chapter 6, the VSFG spectrum of the water interface was studied in presence of electrolytes. Molecular dynamics (MD) simulations based on polarizable force field models suggest that hydrated protons have an affinity to appear at the interface while the hydroxyl ions do not prefer to stay at the surface layer probably due to very differ- ent hydration structure of these ions. Inorganic ions, specially large anions, also have a tendency to enrich at the interface as predicted by recent theoretical simulations while the surface tension measurements predict that the air/ water interface repels such ions. Nonetheless, the presence of any ions at the interface will have severe implications on the surface structure and orientation of H-bonds through direct perturbation at the sur- face. Also, due to the presence of ions at the interface there exists an interfacial electric field which enhances the SFG signal through an electric field induced third order pro- cess. We have studied these effects in aqueous halide acids, hydroxide base and at the alkali halide salt solution interface using VSFG in the water bending region. In Chapter 7, We present evidence for the presence of non-hydrogen bonded (i.e., dangling) -OD groups at the graphene electrode interface, for the first time, by using 15 optical spectroscopic techniques under electrochemical conditions. The unique proper- ties of monolayer graphene provide a high conductivity, transparent electrode that can be used for a variety of applications. The hydrophobic nature of graphene material can affect the hydrogen bonding structures of water molecules in close contact with the elec- trode surface. Understanding the nature of these surface water structures is key to unrav- eling many of the mechanistic details of aqueous phase electrochemistry. One of the key features of water structure at hydrophobic interfaces is the presence of weakly interact- ing or free water molecules with one -OH (-OD) groups pointing towards the graphene surface. Despite numerous theoretical predictions, direct spectroscopic evidence for the presence of such species at the graphene interface has been elusive until now. Our VSFG spectra exhibit a strong narrow peak at 2680 cm 1 present at the graphene/D 2 O interface in the absence and presence of 1M sodium chloride (NaCl) due to the free-OD stretch mode of water molecules. In Chapter 8, We present a comparative study of water/surfactant interactions using two different spectroscopic approaches: water at planar surfactant monolayers by sum frequency generation (SFG) spectroscopy, and interfacial water confined in reverse mi- celles formed by the same surfactants using infrared (IR) absorption spectroscopy. We report spectral features in the OH stretching region (3200-3700 cm 1 ) that are observed in both IR and SFG spectra, albeit with different relative amplitudes, for ionic surfactant sodium 1,4 bis 2 ethylhexylsulfosuccinate (AOT) and nonionic surfactant polyoxyethy- lene 4-lauryl ether (Brij L-4) reverse micelles in hexane and corresponding monolayers at the air/water interface. A prominent feature in the SFG spectra of the OH-stretch at 3560 cm 1 is attributed to water molecules that have one donor hydrogen bond to a nearby water molecule and another weak donor hydrogen bond to the surfactant head 16 group. The same feature is observed in IR spectra of reverse micelles after deconvolut- ing the interfacial vs. bulk spectral contributions. We performed orientational analysis on these water molecules utilizing the polarization dependent SFG, which showed the average tilt angle of the OH-stretch of surfactant bound water molecules 155 5 with respect to the surface normal. 17 Chapter 1 References [1] Eaves, JD; Loparo, JJ; Fecko, Ch J; Roberts, ST; Tokmakoff, A; Geissler, PL Hy- drogen bonds in liquid water are broken only fleetingly Proc. Nat. Acad. Sci. 2005, 102, 13019–13022. [2] Lovesey, Stephen W Theory of neutron scattering from condensed matter 1984. [3] Amann-Winkel, Katrin; Bellissent-Funel, Marie-Claire; Bove, Livia E; Loerting, Thomas; Nilsson, Anders; Paciaroni, Alessandro; Schlesinger, Daniel; Skinner, Lawrie X-ray and neutron scattering of water Chem. Rev. 2016, 116, 7570–7589. [4] Fecko, C. J.; Eaves, J. D.; Loparo, J. J.; Tokmakoff, A.; Geissler, P. L. Ultrafast hydrogen-bond dynamics in the infrared spectroscopy of water Science 2003, 301, 1698–1702. [5] Bakker, H. J.; Skinner, J. L. Vibrational spectroscopy as a probe of structure and dynamics in liquid water Chem. Rev. 2010, 110, 1498–1517. [6] Falk, Michael The frequency of the hoh bending fundamental in solids and liquids Spectrochim. Acta, Part A 1984, 40, 43–48. [7] Bertie, John E; Ahmed, M Khalique; Eysel, Hans H Infrared intensities of liq- uids. 5. optical and dielectric constants, integrated intensities, and dipole moment derivatives of water and water-d2 at 22. degree. c J. Phys. Chem. 1989, 93, 2210– 2218. [8] Mar´ echal, Y From physics to biology: A journey through science accompanying the hydrogen bond and the water molecule J. Mol. Struct. 2008, 880, 38–43. [9] Asbury, J. B.; Steinel, T.; Stromberg, C.; Corcelli, S. A.; Lawrence, C. P.; Skinner, J. L.; Fayer, M. D. Water dynamics: Vibrational echo correlation spectroscopy and comparison to molecular dynamics simulations J. Phys. Chem. A 2004, 108, 1107–1119. [10] Du, Q.; Superfine, R.; Freysz, E.; Shen, Y . R. Vibrational spectroscopy of water at the vapor water interface Phys. Rev. Lett. 1993, 70, 2313–2316. [11] Richmond, GL Molecular bonding and interactions at aqueous surfaces as probed by vibrational sum frequency spectroscopy Chem. Rev. 2002, 102, 2693–2724. 18 [12] Raymond, E. A.; Tarbuck, T. L.; Brown, M. G.; Richmond, G. L. Hydrogen- bonding interactions at the vapor/water interface investigated by vibrational sum- frequency spectroscopy of hod/h2o/d2o mixtures and molecular dynamics simula- tions J Phys Chem B 2003, 107, 546–556. [13] Tian, Chuan-Shan; Shen, Y Ron Isotopic dilution study of the water/vapor inter- face by phase-sensitive sum-frequency vibrational spectroscopy J. Am. Chem. Soc. 2009, 131, 2790–2791. [14] Auer, B. M.; Skinner, J. L. Vibrational sum-frequency spectroscopy of the water liquid/vapor interface J. Phys. Chem. B 2009, 113, 4125–4130. [15] Nihonyanagi, Satoshi; Ishiyama, Tatsuya; Lee, Touk-kwan; Yamaguchi, Shoichi; Bonn, Mischa; Morita, Akihiro; Tahara, Tahei Unified molecular view of the air/water interface based on experimental and theoretical (2) spectra of an iso- topically diluted water surface J. Am. Chem. Soc. 2011, 133, 16875–16880. [16] Stiopkin, Igor V .; Weeraman, Champika; Pieniazek, Piotr A.; Shalhout, Fadel Y .; Skinner, James L.; Benderskii, Alexander V . Hydrogen bonding at the water sur- face revealed by isotopic dilution spectroscopy Nature 2011, 474, 192–195. [17] Morita, Akihiro; Hynes, James T. A theoretical analysis of the sum frequency generation spectrum of the water surface Chem. Phys. 2000, 258, 371–390. [18] Max, Jean-Joseph; Chapados, Camille Isotope effects in liquid water by infrared spectroscopy. iii. h 2 o and d 2 o spectra from 6000 to 0 cm- 1 J. Chem. Phys. 2009, 131, 184505. [19] Bertie, John E; Lan, Zhida Infrared intensities of liquids xx: The intensity of the oh stretching band of liquid water revisited, and the best current values of the optical constants of h2o (l) at 25 c between 15,000 and 1 cm- 1 Appl. Spectrosc. 1996, 50, 1047–1057. [20] Draegert, David A; Stone, NWB; Curnutte, Basil; Williams, Dudley Far-infrared spectrum of liquid water J. Opt. Soc. Am. 1966, 56, 64–69. [21] Walrafen, GE; Blatz, LA Weak raman bands from water J. Chem. Phys. 1973, 59, 2646–2650. [22] Devlin, J Paul; Sadlej, Joanna; Buch, Victoria Infrared spectra of large h2o clus- ters: New understanding of the elusive bending mode of ice J. Phys. Chem. A 2001, 105, 974–983. [23] Martins Filho, Harley P Approximate molecular orbital anharmonic parameters for the infrared spectrum of h 2 o Spectrochim. Acta Part A 2001, 57, 83–93. 19 [24] Jungwirth, Pavel Physical chemistry: Water’s wafer-thin surface Nature 2011, 474, 168–169. [25] Bjorneholm, Olle; Hansen, Martin H; Hodgson, Andrew; Liu, Li-Min; Limmer, David T; Michaelides, Angelos; Pedevilla, Philipp; Rossmeisl, Jan; Shen, Huaze; Tocci, Gabriele et al. Water at interfaces Chem. Rev. 2016, 116, 7698–7726. [26] Maier, Sabine; Salmeron, Miquel How does water wet a surface? Acc. Chem. Res. 2015, 48, 2783–2790. [27] Verdaguer, A; Sacha, GM; Bluhm, H; Salmeron, M Molecular structure of water at interfaces: Wetting at the nanometer scale Chem. Rev. 2006, 106, 1478–1510. [28] Zhang, Luning; Tian, Chuanshan; Waychunas, Glenn A; Shen, Y Ron Structures and charging of-alumina (0001)/water interfaces studied by sum-frequency vi- brational spectroscopy J. Am. Chem. Soc. 2008, 130, 7686–7694. [29] Braunschweig, Bj¨ orn; Eissner, Stephan; Daum, Winfried Molecular structure of a mineral/water interface: effects of surface nanoroughness of -al2o3 (0001) J. Phys. Chem. C 2008, 112, 1751–1754. [30] Ma, Gang; Chen, Xiangke; Allen, Heather C Dangling od confined in a langmuir monolayer J. Am. Chem. Soc. 2007, 129, 14053–14057. [31] Perera, PN; Fega, KR; Lawrence, C; Sundstrom, EJ; Tomlinson-Phillips, J; Ben- Amotz, Dor Observation of water dangling oh bonds around dissolved nonpolar groups Proc. Nat. Acad. Sci. 2009, 106, 12230–12234. [32] Dutta, Chayan; Svirida, Anton; Mammetkuliyev, Muhammet; Rukhadze, Marina; Benderskii, Alexander V Insight into water structure at the surfactant surfaces and in microemulsion confinement J. Phys. Chem. B 2017, 121, 7447–7454. [33] Gragson, DE; Richmond, GL Potential dependent alignment and hydrogen bond- ing of water molecules at charged air/water and ccl4/water interfaces J. Am. Chem. Soc. 1998, 120, 366–375. [34] Scatena, LF; Brown, MG; Richmond, GL Water at hydrophobic surfaces: weak hydrogen bonding and strong orientation effects Science 2001, 292, 908–912. [35] Gordillo, MC; Marti, J Water on graphene surfaces J. Phys.: Condensed Matter 2010, 22, 284111. [36] Mart´ ı, J; Gordillo, MC Effects of confinement on the vibrational spectra of liquid water adsorbed in carbon nanotubes Phys. Rev. B 2001, 63, 165430. 20 [37] Kolesnikov, Alexander I; Zanotti, Jean-Marc; Loong, Chun-Keung; Thiyagara- jan, Pappannan; Moravsky, Alexander P; Loutfy, Raouf O; Burnham, Christian J Anomalously soft dynamics of water in a nanotube: a revelation of nanoscale con- finement Phys. Rev. Lett. 2004, 93, 035503. [38] Politano, Antonio; Marino, Antonio Raimondo; Formoso, Vincenzo; Chiarello, Gennaro Hydrogen bonding at the water/quasi-freestanding graphene interface Carbon 2011, 49, 5180–5184. [39] Dutta, C et. al. Dangling water at the graphene electrode interface Submitted. [40] Schwartz, DK; Schlossman, ML; Kawamoto, EH; Kellogg, GJ; Pershan, Peter S; Ocko, BM Thermal diffuse x-ray-scattering studies of the water-vapor interface Phys. Rev. A 1990, 41, 5687. [41] Ibach, H Comparison of cross sections in high resolution electron energy loss spec- troscopy and infrared reflection spectroscopy Surf. Sci. 1977, 66, 56–66. [42] Bayman, A; Hansma, PK; Kaska, WC; Dubois, LH Inelastic electron tunneling spectroscopic study of acetylene chemisorbed on alumina supported palladium particles Appl. Surf. Sci. 1983, 14, 194–208. [43] Betzig, EPLJS; Finn, PL; Weiner, JS Combined shear force and near-field scanning optical microscopy Appl. Phys. Lett. 1992, 60, 2484–2486. [44] Menzel, Henning; Mowery, Mark D; Cai, Mei; Evans, Christine E et al. Surface- confined nanoparticles as substrates for photopolymerizable self-assembled mono- layers Adv. Mater. 1999, 11, 131–134. [45] Pettinger, Bruno; Ren, Bin; Picardi, Gennaro; Schuster, Rolf; Ertl, Gerhard Nanoscale probing of adsorbed species by tip-enhanced raman spectroscopy Phys. Rev. Lett. 2004, 92, 096101. [46] Chen, CK; De Castro, ARB; Shen, YR Surface-enhanced second-harmonic gener- ation Phys. Rev. Lett. 1981, 46, 145. [47] Du, Quan; Freysz, Eric; Shen, Y Ron Vibrational spectra of water molecules at quartz/water interfaces Phys. Rev. Lett. 1994, 72, 238. [48] Shen, Yuen Ron; Ostroverkhov, Victor Sum-frequency vibrational spectroscopy on water interfaces: polar orientation of water molecules at interfaces Chem. Rev. 2006, 106, 1140–1154. [49] Eisenthal, KB Liquid interfaces probed by second-harmonic and sum-frequency spectroscopy Chem. Rev. 1996, 96, 1343–1360. 21 [50] Eisenthal, Kenneth B Second harmonic spectroscopy of aqueous nano-and mi- croparticle interfaces Chem. Rev. 2006, 106, 1462–1477. [51] Richmond, GL Structure and bonding of molecules at aqueous surfaces Annu. Rev. Phys. Chem. 2001, 52, 357–389. [52] Shen, Yuen-Ron The principles of nonlinear optics New York, Wiley-Interscience, 1984, 575 p. 1984. [53] Boyd, Robert W Nonlinear optics; Academic press, 2003. [54] Lambert, Alex G; Davies, Paul B; Neivandt, David J Implementing the theory of sum frequency generation vibrational spectroscopy: a tutorial review Appl. Spec- tros. Rev. 2005, 40, 103–145. [55] Du, Q; Superfine, R; Freysz, E; Shen, YR Vibrational spectroscopy of water at the vapor/water interface Phys. Rev. Lett. 1993, 70, 2313. [56] Du, Quan; Freysz, Eric; Shen, Y Ron et al. Surface vibrational spectroscopic stud- ies of hydrogen bonding and hydrophobicity Science 1994, 264, 826–827. [57] Shen, Yuen Ron; Ostroverkhov, Victor Sum-frequency vibrational spectroscopy on water interfaces: Polar orientation of water molecules at interfaces Chem. Rev. 2006, 106, 1140–1154. [58] Raymond, EA; Tarbuck, TL; Richmond, GL Isotopic dilution studies of the va- por/water interface as investigated by vibrational sum-frequency spectroscopy J. Phys. Chem. B 2002, 106, 2817–2820. [59] Vinaykin, Mikhail; Benderskii, Alexander V Vibrational sum-frequency spectrum of the water bend at the air/water interface J. Phys. Chem. Lett. 2012, 3, 3348– 3352. [60] Dutta, Chayan; Benderskii, Alexander V On the assignment of the vibrational spectrum of the water bend at the air/water interface J. Phys. Chem. Lett. 2017, 8, 801–804. [61] Nagata, Yuki; Hsieh, Cho-Shuen; Hasegawa, Taisuke; V oll, Judith; Backus, Ellen H. G.; Bonn, Mischa Water bending mode at the watervapor interface probed by sum-frequency generation spectroscopy: A combined molecular dynamics simu- lation and experimental study J. Phys. Chem. Lett. 2013, 4, 1872–1877. [62] Ni, Yicun; Skinner, JL Ir and sfg vibrational spectroscopy of the water bend in the bulk liquid and at the liquid-vapor interface, respectively J. Chem. Phys. 2015, 143, 014502. 22 Chapter 2: Vibrational Sum Frequency Generation Spectroscopy I have performed linear and non-linear spectroscopic experiments in this thesis. Linear spectroscopic techniques (such as FTIR and Raman ) are well established techniques with commercially available instruments which we have mostly used for sample char- acterization. Detailed working principle for these experiments are available elsewhere and are not needed in the context of this thesis. In this chapter, the theoretical back- ground, selection rules, and experimental setup for vibrational sum frequency gener- ation (VSFG) spectroscopic technique are discussed in detailed. Another part of this chapter is dedicated to the discussions of orientational analysis from the polarization depended SFG data. 2.1 Sum Frequency Generation SFG is a non-linear spectroscopic technique which involves two laser pulses, spatially and temporally overlapped onto the sample surface to genearate a signal at the sum of two incoming laser frequencies. The broad-band SFG (BB-SFG) scheme was first in- truduced in the late 1990’s. 1, 2 In BB-SFG, a broadband femtosecond pulse in the IR region (for vibrational spectroscopy 1–5 ) exites all the vibrational resonances (within the available bandwidth) to create a first order polarization. This gets upconverted by a 23 stretched (picosecond) narrowband (in frequency) nonresonant visible pulse. The BB- SFG spectra represents a convolution of the molecular response function with the up- conversion pulse (visible pulse). For a frequency domain detection, the SFG signal is Fourier analysed by a monochromator and it can be written as 4, 6 E SFG (!;)/ Z +1 1 P (2) (t;)dt = Z +1 1 P (1) E Vis (t)e i!t dt = Z +1 1 [S(t)E IR (t)]E Vis (t)e i!t dt = [S(!)E IR (!)]E Vis (!;) (2.1) Where, P (2) is the second-order polarization created by the interaction of P (1) and the visible pulse (E Vis (t)). P (1) is the first-order polarization in the sample created by the IR pulse (E IR (t)). S(t) is the molecular response in the time domain picture and ‘’ denotes convolution. S(!) is the spectral response function in the frequency domain. Hence, the SFG signal in frequency domain is a product of the spectral response S(!) with the resonant IR pulse spectrum E IR (!), which is convoluted with the visible pulse spectrum E Vis (!). The frequency resolution of the experiment will depend on the spectral width of the upconversion pulse (the visible pulse) and the spectral width of the broad-band IR pulse will determine the spectral range covered by the measurements. 2.1.1 Experimental Setup The workhorse of our experimental setup (shown in figure 2.1) is a femtosecond Ti:sapphire laser (Coherent Micra) centered at800nm with full width half maxium 50nm, which acts as seed pulse for our dual stage amplifier system (Coherent Lagend Elite-Duo) pumped with a Nd:YLF laser (Evolution 30). Our broad-band mid-infrared 24 Figure 2.1: Schematic of the homodyne detected VSFG setup. 25 pulses (bandwidth 350 cm 1 ) were generated by an optical parametric amplifier (OPA) followed by a difference frequency generator (DFG) with a GaSe crystal. The OPA (TOPAS-C, Light Conversion) was pumped with one portion of the Ti: sapphire amplifier output ( 50fs, 1.8mJ/pulse) operating at a 5 kHz repetition rate. The signal and idler pulses ( = 1.1-2.6m) produced from the TOPAS are mixed in DFG (nDFG, Light Conversion) to yield tunable infrared (IR) pulses (1400-3600 cm 1 ). Another por- Figure 2.2: Spectra of the output from the Ti:sapphire oscillator (red) and the compressed pulse output for the cavity (blue). tion of the amplifier output ( 1.2 mJ/pulse) was directed to an external compressor and the compressed pulse was passed through a 4f stretcher. At the focal plane of the 4f- stretcher a narrow band region ( 20 cm 1 ) was selected spatially using a mechanical slit. The width of the slit determines the power and bandwidth (spectral resolution) of the visible pulse. Figure 2.3a compares the visible pulse spectrum before and after the 4f- stretcher measured using an Ocean Optics spectrometer. Visible pulse spectrum (figure 2.3b) at the sample was measured using the CCD detector. Both the visible and the IR pulses were focused on the sample surface to a spot size of 160m using a spherical 26 mirror. The laser power at the sample surface was1.0 J and 12.0 J per pulse for the infrared ( 1650 cm 1 for water bending region) and the visible (800nm), respec- tively. The angle of incidence from the surface normal was 67 for the visible and 62 for the infrared beams. The time delay between the two pulses was varied by a com- puter controlled translation stage. The polarizations of the IR, visible, and SFG pulses are controlled by waveplates and polarizers (zero-order quartz half-wave plate, 800 nm, CVI Melles Griot for visible beam; zero-order CdSe half-wave plate, 1000-19000 nm, 5 mm thick, Alphalas for IR beam; zero-order quartz half-wave plate, 670 nm, CVI Melles Griot for SFG beam). (a) (b) Figure 2.3: (a) Spectrum of the 800nm pulse before the 4f stretcher and after the 4f stretcher, measured using an occean optics spectrometer. (b) Visible pulse spec- trum at the sample measured in CCD. The VSFG signal was recollimated, spatially and spectrally filtered (a Raman notch lter and a 800nm high reector is used) before focusing it onto the entrance slit of a monochromator (Princeton Instruments Acton SP2500 or SpectraPro 500i monochro- mator) and liquid-nitrogen-cooled CCD detector (Roeper Scientific, Spec-10:100B, 1340 X 100 pixels). For reference spectrum, the narrowband visible and broad-band 27 IR pulses are overlapped on the surface of a gold or GaAs sample to generate a nonres- onant SFG spectra. Collected VSFG spectra for all samples are normalized with respect to this nonresonant signal measured following each sample which partially deconvo- lutes the IR pulse spectrum. For each acquisition, a background spectrum is collected separately by blocking the IR beam and is subtracted from each VSFG spectrum to re- move contributions from scatter and dark counts to the measured signals. In order to get rid of the contributions from the atmospheric CO 2 and H 2 O in the measured IR spec- trum, plexi-glass boxes were built around the NDFG, IR delay, and the SFG setup with an option to continuously purge them with dry air. The spectra of the IR pulses were (a) (b) Figure 2.4: (a) IR spectrum in the CH stretch region , measured using the MCT detector. (b) SFG spectrum at the gold surface in the CH stretch region. also measured using an IR grating blazed at 5m and a liquid nitrogen-cooled mer- cury cadmium telluride (MCT) detector (IR Associates) while alligning the OPA/DFG 28 combination before an SFG experiment. Figure 2.4 compares the IR spectrum in the CH-stretching region measured in the MCT detector (2.4a) and the nonresonant SFG signal from the gold surface (2.4b) in the same region. The temporal profile of the narrow-band picosecond visible pulse after the 4f- stretcher was measured by SFG cross-correlation on the gold substrate. Figure 2.5a (a) (b) Figure 2.5: (a) Frequency-resolved cross-correlation of femtosecond IR pulse and picosecond visible pulse (b) Temporal profile of the visible pulse passed through the 4f- stretcher. shows the cross-correlation plot and the corresponding temporal profile of the visible pulse is plotted in figure 2.5b. To obtain clean water for the samples, we used 15 M water from our Millipore sys- tem and distilled it through a sealed distillation apparatus cleaned with piranha solution before any measurement. 29 2.2 Orientational analysis VSFG technique has been routinely used to investigate structure and orientation of water molecules at liquid interfaces. SFG is a unique surface specific technique where the sig- nal intensity is highly polarization dependent. A quantitative analysis of the polarization dependent SFG signal provides detailed informations regarding the complex molecular structure and orientations at the interface. 7–11 SFG intensity from an interface detected Figure 2.6: A three-layer model for SFG in reflection geometry with molecularly thin interface layer. in the reflection geometry can be written as 8, 12 I(! SF ) = 8 3 ! 2 SF sec 2 SF c 3 n 1 (! SF )n 1 (! vis )n 1 (! IR ) j (2) eff j 2 I(! vis )I(! IR ) (2.2) Where! SF ,! vis and! IR are the frequencies of the SFG, visible and the infrared laser beams respectively,n 1 (! i ) are the refractive indices at different frequencies, SF is the 30 reflection angle of the sum frequency beam from the surface normal, I(! i ) are the inten- sities of the three laser beams, (2) eff is the effective second order susceptibility. (2) eff is a rank-3 tensor with 27 elements, however, for an isotropic surface with a C 1v symmetry there are only four independent nonzero (2) eff elements, namely xxz = yyz , xzx = yzy , zxx = zyy and zzz . Here the lab coordinate system has been chosen such that the z axis is the surface normal and the lasers beams are in the xz plane. Different compo- nents of this (2) eff can be measured by carefully choosing appropriate polarizations of the incoming and the outgoing beams according to (2);ssp eff =L yy (! SF )L yy (! vis )L zz (! IR ) sin IR yyz (2.3) (2);sps eff =L yy (! SF )L zz (! vis )L yy (! IR ) sin vis yzy (2.4) (2);ppp eff =L xx (! SF )L xx (! vis )L zz (! IR ) cos SF cos vis sin IR xxz L xx (! SF )L zz (! vis )L xx (! IR ) cos SF sin vis cos IR xzx +L zz (! SF )L xx (! vis )L xx (! IR ) sin SF cos vis cos IR zxx +L zz (! SF )L zz (! vis )L zz (! IR ) sin SF sin vis sin IR zzz (2.5) Where, ‘p’ and ‘s’ stands for polarizations of the laser beams parallel or perpendicular to the plane of incidence respectively. L (! i ) are the Fresnel factors as calculated by a three layer model 8 (Figure 2.6) and can be written as L xx (! i ) = 2n 1 (! i ) cos SF n 1 (! i ) cos SF +n 2 (! i ) cos SF (2.6) L yy (! i ) = 2n 1 (! i ) cos SF n 1 (! i ) cos SF +n 2 (! i ) cos SF (2.7) L zz (! i ) = 2n 2 (! i ) cos SF n 1 (! i ) cos SF +n 2 (! i ) cos SF n 1 (! i ) n 0 (! i ) 2 (2.8) 31 SF is the refractive angle for the sum frequency beam as calculated from n 1 (! SF ) sin SF =n 2 (! SF ) sin SF (2.9) The experimentally measured (2) ijk tensors in the lab-frame are connected to the molec- ular hyperpolarizability tensor ( (2) i 0 j 0 k 0 ) by rotational transformation matrix R(;; ). 13 (2) ijk =N s X i 0 j 0 k 0 hR ii 0R jj 0R kk 0i (2) i 0 j 0 k 0 (2.10) N s is the number density of molecules at the interface. \hi 00 represents orientational average of the corresponding property over the orientational distribution function as defined by Gan et. al. 9 We have assumed a Gaussian distribution function of the form f() = 1 p 2 2 exp ( 0 )=2 2 (2.11) we assumed that the water orientation is centered around the tilt angle 0 and the stan- dard deviation of the angular distribution is. Molecular hyperpolarizabity ( (2) i 0 j 0 k 0 ) is a frequency dependent quantity and can be written as a sum of resonant and nonresonant quantities. The resonant part contains the vibrational information of the molecules at the interface and the nonresonant part generally provides a broad background over the entire spectral region. (2) i 0 j 0 k 0 =A NR exp(i) + X q b m ! IR ! m +i m (2.12) 32 All these parameters were defined previously. b m in equation 2.12 is related to the in- frared dipole derivative m Qm and the raman polarizability (1) ab Qm of the m th vibrational mode according to equation 2.13. 14, 15 (b m ) i 0 j 0 k 0 = 1 2 0 ! m (1) i 0 j 0 Q m k 0 Q m (2.13) Molecular symmetry group of the vibrational mode determines the nonzero tensor el- ements of the molecular hyperpolarizability tensor (2) i 0 j 0 k 0 . Water bending is a totally symmetric mode with C 2v symmetry and the non-zero tensor elements can be deduced from the character table (see Table 2.1) of the C 2v point group. Since water bending mode belongs to the irreducible representation A 1 , there will be three independent com- Table 2.1: Character table for C 2v point group. C 2v E C 2 v (xz) 0 v (yz) A 1 1 1 1 1 z x 2 ;y 2 ;z 2 aac ; bbc ; ccc A 2 1 1 -1 -1 xy B 1 1 -1 1 -1 x xz aca B 2 1 -1 -1 1 y yz bcb ponents of the hyperpolarizability tensor, aac ; bbc ; ccc . Effective (2) tensor element for the C 2v mode for different polarization combinations can be calculated from equation 2.10. For example the (2) yyz would be (2) yyz =N s hR ya R ya R zc i (2) aac +hR yb R yb R zc i (2) bbc +hR yc R yc R zc i (2) ccc (2.14) This can be simplified in the form of the following expression (equation 2.15) for all polarization combinations (2) eff =N s d cos c cos 3 (2.15) 33 c and d parameters for the three polarization combinations (SSP, SPS, PPP) can be rig- orously deduced after transforming the molecular coordinate fixed hyperpolarizability tensors into the laboratory fixed axis and can be expressed in a simplified form as: C ssp = L K ;C sps = 1;C ppp = ALBL +CL 2DL AK +BLCL + 2DM (2.16) d ssp = L yy (! SF )L yy (! vis )L zz (! IR ) sin IR L 4 (2.17) d sps = L yy (! SF )L yy (! vis )L zz (! IR ) sin vis L 4 (2.18) d ppp = AK +BLCL + 2DM 4 (2.19) Where A, B, C, D, K, L and M are A =L xx (! SF )L xx (! vis )L zz (! IR ) cos SF cos vis sin IR ; B =L xx (! SF )L zz (! vis )L xx (! IR ) cos SF sin vis cos IR ; C =L zz (! SF )L xx (! vis )L xx (! IR ) sin SF cos vis cos IR ; D =L zz (! SF )L zz (! vis )L zz (! IR ) sin SF sin vis sin IR ; K = aac + bbc + 2 ccc L = aac + bbc 2 ccc M = aac + bbc (2.20) 34 2.3 Appendix 2.3.1 Single-Shot Autocorrelator A homebuilt single-shot autocorrelator was used to measure the visible pulse duration (t p ) in our SFG experiments. A detailed description of the optical layout can be found in ref. 16 In general, two identical input femtosecond pulses are specially and temporally overlapped onto a non-linear crystal (type-I BBO, 1mm thick) to produce a SFG signal. A delay stage between the two pulses was introduced for changing the temporal delay (). (a) (b) Figure 2.7: Special intensity distribution of the SFG signal from the BBO crystal captured using a VGA camera for the output (800nm) of the internal compressor (a) and the external compressor (b) The intensity of the SFG signal depends on as the autocorrelation function given by A() = Z 1 1 I(t)I(t)dt (2.21) 35 When =0, intensity will be maximum. When these two identical pulses cross each other inside the BBO crystal with a very small angle (2) between them, a SFG signal will be generated from each point of crossing between the two beams. The generated signal was projected onto a white screen and a screenshot of the signal was captured by a VGA camera. Spatial intensity distribution (x) of the SFG signal was measured by an image processing program in Igor Pro. The laser pulse duration (t p ) was calculated from the special intensity distribution of the SFG spectrum generated at the BBO crystal by the following relation given by 17 t p = (k xsin) c (2.22) where, ‘c’ is the velocity of light, and ‘k’ is a numerical factor depended on the laser pulse shape. For a Gaussian pulse shape, k = p 2. 36 Chapter 2 References [1] Van Der Ham, EWM; Vrehen, QHF; Eliel, ER Self-dispersive sum-frequency gen- eration at interfaces Opt. Lett. 1996, 21, 1448–1450. [2] Richter, Lee J; Petralli-Mallow, Teresa P; Stephenson, John C Vibrationally re- solved sum-frequency generation with broad-bandwidth infrared pulses Opt. Lett. 1998, 23, 1594–1596. [3] Roke, Sylvie; Kleyn, Aart W; Bonn, Mischa Time-vs. frequency-domain femtosec- ond surface sum frequency generation Chem. Phys. Lett. 2003, 370, 227–232. [4] Bordenyuk, Andrey N; Jayathilake, Himali; Benderskii, Alexander V Coherent vibrational quantum beats as a probe of langmuir- blodgett monolayers J. Phys. Chem. B 2005, 109, 15941–15949. [5] Carter, Jeffrey A; Wang, Zhaohui; Dlott, Dana D Ultrafast nonlinear coherent vi- brational sum-frequency spectroscopy methods to study thermal conductance of molecules at interfaces Acc. Chem. Res. 2009, 42, 1343–1351. [6] Stiopkin, Igor V; Jayathilake, Himali D; Weeraman, Champika; Benderskii, Alexander V Temporal effects on spectroscopic line shapes, resolution, and sen- sitivity of the broad-band sum frequency generation J. Chem. Phys. 2010, 132, 234503. [7] Wei, Xing; Shen, YR Motional effect in surface sum-frequency vibrational spec- troscopy Phys. Rev. Lett. 2001, 86, 4799. [8] Zhuang, X; Miranda, PB; Kim, D; Shen, YR Mapping molecular orientation and conformation at interfaces by surface nonlinear optics Phys. Rev. B 1999, 59, 12632. [9] Gan, Wei; Wu, Dan; Zhang, Zhen; Feng, Ran-ran; Wang, Hong-fei Polarization and experimental configuration analyses of sum frequency generation vibrational spectra, structure, and orientational motion of the air/water interface J. Chem. Phys. 2006, 124, 114705. [10] Gan, Wei; Wu, Dan; Zhang, Zhen; Guo, Yuan; Wang, Hong-fei Orientation and motion of water molecules at air/water interface Chinese J. Chem. Phys. 2006, 19, 20–24. 37 [11] Wang*, Hong-Fei; Gan, Wei; Lu , Rong; Rao, Yi; Wu, Bao-Hua Quantitative spectral and orientational analysis in surface sum frequency generation vibrational spectroscopy (sfg-vs) Int. Rev. Phys. Chem. 2005, 24, 191–256. [12] Wei, Xing; Hong, Seok-Cheol; Zhuang, Xiaowei; Goto, Tomohisa; Shen, YR Non- linear optical studies of liquid crystal alignment on a rubbed polyvinyl alcohol surface Phys. Rev. E 2000, 62, 5160. [13] Hirose, Chiaki; Akamatsu, Naotoshi; Domen, Kazunari Formulas for the analysis of the surface sfg spectrum and transformation coefficients of cartesian sfg tensor components Appl. Spectrosc. 1992, 46, 1051–1072. [14] Superfine, Richard; Huang, Ji Y; Shen, YR Experimental determination of the sign of molecular dipole moment derivatives: an infraredvisible sum frequency generation absolute phase measurement study Chem. Phys. Lett. 1990, 172, 303– 306. [15] Zhu, XD; Suhr, Hajo; Shen, YR Surface vibrational spectroscopy by infrared- visible sum frequency generation Phys. Rev. B 1987, 35, 3047. [16] Shalhout, Fadel Y Vibrational sum frequency spectroscopy of molecules on metal, semiconductor, and aqueous surfaces; University of Southern California, 2013. [17] Raghuramaiah, M; Sharma, AK; Naik, PA; Gupta, PD; Ganeev, RA A second- order autocorrelator for single-shot measurement of femtosecond laser pulse dura- tions Sadhana 2001, 26, 603–611. 38 Chapter 3: Water bending Vibration: a Complementary Probe of Hydrogen Bonding In this chapter homodyne detected SFG was implemented to study the water bending ( 2 ) vibrational mode at the air/water interface. 1, 2 Experimental measurements were followed by spectral fitting and analysis which validated the previous theoretical assign- ments of different H-bonded water species. This will establish water bending vibrational mode as a complemetary probe of hydrogen bonding at the air/water interface which should be considered as a background for the next three chapters. 3.1 Introduction The dynamic hydrogen bonding network of liquid water is the microscopic underpin- ning of most, if not all, of its physical, chemical, and biological properties, including in- terfacial phenomena such as surface tension and hydrophobic/hydrophilic interactions. Vibrational spectroscopy, predominantly in the OH-stretching region of the spectrum, has been applied extensively to study the H-bond structures, distribution, and ultrafast dynamics in bulk 3–5 and interfacial water. 6–14 Recently, several groups started exploring 39 the water bend vibrational mode ( 2 ) for the studies of the aqueous H-bonding at inter- faces. 1, 15–17 Water bend spectroscopy may provide molecular information complemen- tary to that available from the OH-stretch spectra. 16 For example, while the OH-stretch frequency is predominantly affected by a single donor H-bond, the water bend mode, which involves both hydrogens, is sensitive to at least two donor H-bonds (in addition to possibly being affected by the acceptor H-bonds through the two lone pairs). In- tramolecular coupling between the two local OH-stretch modes on each water molecule, as well as intermolecular coupling between neighboring molecules significantly compli- cate the spectral and orientational analysis of the vibrational SFG measurements in the OH-stretch region because, in general, the direction of the transition dipole does not represent the orientation of any single OH-bond. 4 OH-stretch in neat water can be delo- calized over as many as 10 molecules. 4 To circumvent the effect of this delocalization or the inter-molecular dipole-dipole coupling effects, isotopic dilution experiements are particularly useful for studying bulk and interfacial water. 18, 19 Additional complexity in OH-stretch spectra arises due to strong non-Condon effects, i.e, IR transition dipole is strongly dependent on the local environment 20, 21 and increases by a factor of5 from the blue to the red side of the OH band (from 3700 cm 1 to 3200 cm 1 ). 20 In contrast, the water bend SFG spectroscopy should be free of many of these com- plications since there is only one mode per molecule, and the intermolecular coupling is weaker due to the smaller transition dipole. 16 Unlike the delocalized OH-stretch mode in neat water, the bend mode is expected to be a local probe of H-bonding. The water bend mode is much more harmonic than the stretch mode and it is weakly coupled to the H-bond degrees of freedom. This makes the non-Condon effects almost negligible for the bend mode. It should also be noted that the frequency shift of the water bend 40 mode (due to H-bonding) is almost4 times smaller than for the stretch and the sepa- ration between the water bend peaks for the free-OH and H-bonded OH species is much smaller than for the OH-stretch spectrum (80-100 cm 1 for the bend versus 250-450 cm 1 for the stretch). The bend mode is also a doorway state for vibrational relaxation in water, situated right in the middle of the 3000 cm 1 band gap between the OH-stretch modes and the librational, rotational, and translational modes. Of considerable interest is therefore the mechanism of water bend mode coupling to the librational overtones, which provide a broad background on which the 2 spectral peak is observed in liquid water 22 and at the air/water interface. 1 3.2 Assignment of the bend spectra 3.2.1 Previous Studies Vinaykin et al. reported the first surface-selective vibrational sum-frequency generation (VSFG) spectrum of the water bend at the air/water interface. 1 They offered an Occams razor interpretation of the spectrum in terms of the minimum number of Lorentzians sufficient to fit the observed line shape within the available signal-to-noise. Briefly, al- though the observed line shape (Fig. 3.2) appears to have both positive and negative peaks, it rides on top of a strong background signal which presumably is due to the librational overtones in this spectral region. Thus, the band shape can be satisfactorily approximated by an interference of a single main Lorentzian peak with a spectrally-flat background. Shortly thereafter, Nagata and Bonn 15 suggested an assignment of the 2 line shape based on molecular dynamics (MD) simulations. Analysis of the MD trajec- tories revealed that the VSFG spectrum has indeed two distinct contributions of opposite 41 signs: the negative peak (less blue-shifted from the gas-phase 2 frequency) due to wa- ter molecules with 0 or 1 donor H-bond predominantly, the free-OH species with one of the hydrogens not participating in H-bonding, and the positive peak (stronger blue- shifted) due to species with an average of 2 donor H-bonds. Recently, Ni and Skinner 16 presented a mixed quantum-classical calculation by first computing the spectroscopic maps of the water bend that correlate to the transition frequency, dipole, and couplings with the local electric field, and then applying them to the MD trajectories to obtain the 2 spectrum of both bulk liquid and interfacial water . As with the previously published OH-stretch calculations, 23–25 analysis of the MD trajectories allowed them to separate out spectral signatures of different H-bonding classes. While the contributions from each H-bonding class are broad and overlap one another, a clear general trend indicates negative contributions from the free-OH species (1 N , 2 S , and 3 S H-bonded classes in their notation) on the red side of the spectrum and a positive blue-shifted contributions from H-bonded structures (3 D , 4 D H-bonded classes presented in figure 3.1). We have validated this spectral assignment of the the free-OH and hydrogen bonded -OH species in the bending region using vibrational sum frequency generation spectroscopy. 2 The theoretical calculations focused on the SSP spectrum (obtained using P- polarized infrared light resonant with the ( 2 ) mode, S-polarized nonresonant visible light, and detecting S-polarized component of the SFG signal), which contains a single tensor element of the nonlinear susceptibility (2) xxz . In contrast, PPP polarized SFG spectra contains contributions from multiple tensor elements ( (2) zzz , (2) yyz , (2) yzy , (2) zyy ) and therefore provide different orientational averaging of the chromophores. Thus compar- ing SSP vs. PPP spectra can be useful in situations where distinct species with different orientations are present, e.g. different H-bonded classes proposed by theoretical simu- lations 15, 16 described above. 42 Figure 3.1: Illustrations for the different H-bonded classes present at the air/water interface. Some of the possible classes with major contribution towards the overall spectrum have been presented only. 3.2.2 Experimental Results and Discussion Water bend spectra at the air/water interface Figure 3.2 shows experimentally measured SFG spectra of the water bend at the air/water interface obtained using SSP and PPP polarization combinations. The main spectral feature appears to be red-shifted by 15 cm 1 in the PPP spectrum relative to the SSP spectrum. In light of the proposed assignment of this dispersive spectral feature to the free-OH species contributing a negative peak at lower frequency and H-bonded species contributing a positive peak at higher frequency, the apparent PPPvs: SSP spec- tral shift has a natural explanation. The two interfering peaks of opposite sign have different relative amplitudes in the PPPvs: SSP spectra, and thus the overall spectrum appears shifted even though the component peaks have the same center frequencies. 43 This is illustrated in Figure 3.2 using two Lorentzians of opposite sign, negative cen- tered at 1630.5 cm 1 and positive at 1662.5 cm 1 , to fit the main spectral feature. The Figure 3.2: Top panel: Vibrational SFG spectra of the water bend mode at the air/water interface for PPP (red) and SSP (blue) polarization combinations of SFG, visible, and IR. Black solid lines show fit described in the text. Bottom panel: Two principal resonant Lorentzians (positive and negative) used in the fit (left axis). The background signal fit is shown on top (right axis). Reproduced from Ref. 2. different amplitudes of the negative and positive components are indeed expected since molecules belonging to different H-bonding classes have different orientation relative to the surface normal. The overall fit (solid black lines in Figure 3.2, upper panel) also in- cludes two other Lorentzians and a constant background term with a phase factor which 44 collectively describe the pedestal underlying the resonant water bend peaks. The inten- sity spectrum of this background pedestal is also shown in the bottom panel of Figure 3.2. We note that this is not a nonresonant background usually used in SFG spectral fits, since there are known librational overtone modes in this spectral region. 1, 22 The frequencies of the two main Lorentzians (! 1 &! 2 ) were constrained to be the same for SSP vs: PPP spectra, within the 1 cm 1 experimental uncertainty, while their ampli- tudes were varied. The line width parameters 1 and 2 were also constrained within 45 cm 1 to 60 cm 1 range. The fitting parameters describing the background pedestal were allowed to vary. The fitting parameters are summarized in Table 3.1 . We emphasize that the two-Lorentzian fit of the main spectral feature is an oversim- plification used here only to illustrate our argument. Several H-bonded classes likely contribute to each peak, 16 each having its unique spectral shape and average molecu- lar orientation. We therefore do not attempt to fit both SSP and PPP spectra using the deconstructed H-bonding class contributions presented by Ni and Skinner, 16 since they were calculated only for the SSP spectrum. Experimental validation for the proposed spectral assignment It is well known that the free-OH species observed at the air/water interfaces and oil/water interfaces are greatly suppressed when water is in contact with a hydrophilic substance. In particular, spreading a monolayer of an anionic amphiphilic surfactant on the water surface effectively eliminates the free-OH species, presumably by H-bonding the free-OH groups to the surfactant headgroups. 26–29 This is demonstrated in Figure Figure 3.3 which shows the SFG spectrum of the free OH-stretch mode at the air/water interface of pure water in comparison with the 1mM solution of sodium dodecyl sulfate (SDS), which forms a monolayer of SDS at the water surface. The free OH peak at 45 3700 cm 1 is almost entirely suppressed in the presence of SDS, in both SSP and PPP spectra. We therefore expect the corresponding changes in the water bend SFG spectra to point to the contribution of the free-OH species. Figure 3.3: The OH-stretch SFG spectrum of the air/water interface in absence (red) and presence (blue) of a monolayer of SDS (1 mM SDS solution)., for both SSP and PPP polarizations. Reproduced from Ref. 2. Figure 3.4 shows the comparative SFG spectra of the water bend at the air/water interface of pure water and 1 mM SDS solution. It is evident that the negative peak at 1630 cm 1 is significantly diminished in the presence of the SDS monolayer for both SSP and PPP polarizations. Also, the frequency shift between SSP and PPP spectra is no longer apparent, since the interference effect is now diminished due to lower amplitude of the negative peak. This solidifies the assignment of the negative 1630 cm 1 feature in the water bend SFG spectrum to the free OH species (1 N , 2 S , 3 S H-bonded classes). We note that the SDS molecule itself does not give any SFG signal in this frequency range. This was verified experimentally by recording spectra of D 2 O/SDS mixtures under the identical experimental conditions (See Figure 3.6). 46 Figure 3.4: Comparison of the water bend SFG spectra of the air/water interface in absence (top left) and presence (top right) of an SDS monolayer, for SSP (blue) and PPP (red) polarizations. Solid black lines represent fit to the model described in text. Bottom panels show the two main Lorentzians used in the fitting . Reproduced from Ref. 2. Figure 3.4 also shows the curve fitting to the same model as in Figure 3.2, with the main spectral feature described using two Lorentzians of opposite signs. The fitting pa- rameters can be found in Table 3.1. Interestingly, the fitting indicates that the negative peak at 1630 cm 1 , while diminished, does not entirely disappear in the presence of 1 mM SDS. This is in contrast with the free OH stretch peak that is completely suppressed (Figure 3.3). Clearly, there are contributions to the negative spectral amplitude at 1630 47 cm 1 other than the free OH species. Indeed, Ni and Skinner calculations suggest that the fully tetrahedrally-coordinated species (4 D H-bonded class) provide a negative am- plitude signal at the lower-frequency end of the spectrum as well as positive signal on the blue-side of the spectrum. Another possible source of the residual negative ampli- tude at 1630 cm 1 could be the librational overtone resonances, which are not included in the theoretical treatments and only crudely approximated by our fitting. Indeed, the constant term in our fitting equation has a phase relative to the resonant Lorentzian terms which is not zero, and further, is different for different polarizations (SSP vs. PPP) and also different in the absence vs. presence of SDS (Table 3.1). This indicates that this term does not simply describe a nonresonant background, but rather approximates a spectrally broad collection of resonant signals. The assignment of the broad blue-shifted peak observed at 1750 cm 1 in the PPP spectrum of the pure air/water interface remains an open question. Although it is clearly present in the PPP spectrum, the SSP spectrum shows a much weaker blue-shifted fea- ture at 1700 cm 1 (due to interference effects and the significant line width, frequency assignment of these peaks is somewhat ill-determined). MD simulations of the SSP spectrum have not found appreciable spectral density in this frequency range, 15, 16 but no attempt has been made so far to calculate the PPP spectrum. It is possible that this feature is due to the librational overtones of water which were not explicitly included in the calculations, or H-bonded structures with extreme blue-shift of the bend frequency which contribute to susceptibility tensor elements other than, or perhaps a combina- tion band of the water bend ( 2 ) and low-frequency collective modes. We also cannot completely rule out the possibility of an impurity contributing to this spectral feature. In summary, the experimental evidence presented here strongly supports the recent theoretical assignment of the water bend SFG spectrum of the air/water interface in 48 terms of the hydrogen-bonded classes. The main resonant spectral feature of the water bend at the surface has a dispersive line shape which arises due to interference between a lower-frequency negative peak and higher frequency positive peak. The negative red- shifted peak at 1630 cm 1 is mostly associated with the free OH species with zero or one donor H-bond, although there is also a discernible contribution from the fully hydrogen- bonded species. The positive blue-side peak at 1662 cm 1 is mostly due to the structures where both hydrogens are tied up in donor H-bonds. The nearly quantitative agreement between the experimentally measured and theoretically calculated SSP spectra of the pure air/water interface is both an encouraging validation of the spectroscopic maps and MD simulations and a valuable input for future spectroscopic studies of aqueous inter- faces. In particular, since different hydrogen-bonded species are spectrally separated, the calculated spectroscopic parameters can be used to independently characterize the molecular orientation of water molecules at the surface involved in different H-bonding motifs. 3.3 Appendix 3.3.1 Tantative construction of the bend spectra from the stretch spectra The blue shift of the observerd water bend peak frequencies with respect to the 1595 cm 1 value of the gas-phase water molecule indicates the presence of hydrogen bonded species at the air/water interface. Frequency of the free-OH or the weakly hydrogen bonded species and the strongly hydrogen bonded species can be tentatively assigned by analogy with the OH-stretch SFG spectum of the air/water interface. 6, 11, 30 Both water bend and OH-stretch vibrations are spectroscopic probes of aqueous H-bonding 49 network. Hence, these two modes can be seen as representations of the distribution of H-bond configurations present at the interface. OH-stretch SFG spectra at the air/water interface consists of a sharp blue shifted peak ( 3700 cm 1 ) corresponding to free-OH bonds and a red shifted broad feature 3100-3500 cm 1 for H-bonded structures. 11 M. Falk 22 established an emphirical correlation between the stretch and bend frequencies as given by equation 3.1 2 = 1590 0:26(3705 OH ) (3.1) Thus, we expect the least blue-shifted feature in the water bend spectrum to correspond Figure 3.5: Tentative assignment of the water bend SFG spectrum by analogy with the OH-stretch spectrum. to the least red-shifted feature in the OH-stretch spectrum. Also, the frequency shift of the bend is 4 times smaller than the stretch, hence the seperation between water 50 bend peaks between free-OH species and the H-bonded species is expected to be smaller than in OH-stretch spectrum. Figure 3.5 shows this tentative assignment, where start- ing from the OH-stretch SFG spectrum (A), the frequency shift is inverted (step 1) and scaled down by a factor of 4 (step 2). The sign of the free -OH peak is negative as de- scribed earlier. Next, the amplitude of the H-bonded feature (red) relative to the free-OH is reduced (step 3) as the non-Condon effects are weak or non-existent in the bend spec- trum. The resulted predicted spectrum is in qualitative agreement with experimentally measured water bend spectrum in Figure 3.2 3.3.2 SFG in D 2 O We have also measured the SFG spectrum of D 2 O and SDS dissolved in D 2 O in the water bending region (1500-1800 cm 1 ) to make sure that this region is spectrally flat and no other spectral contamination is coming due to the SDS molecules. SFG spectra at air/water (H 2 O) , air/D 2 O , SDS/water (H 2 O) and SDS/D 2 O interface in SSP and PPP polarization combinations are plotted in Figure 3.6. 51 Figure 3.6: SFG spectra at air/water (H 2 O) , air/D 2 O , SDS/water (H 2 O) and SDS/D 2 O interface in SSP and PPP polarization combinations. Legends are col- ored according to the corresponding spectra . Reproduced from Ref. 2. Table 3.1: Fitting parameters of the vibrational SFG spectra of the water bend measured for SSP and PPP polarizations at the air/water and SDS/water interface Air/Water SDS/Water SSP PPP SSP PPP ! 1 , cm 1 1630.5 1.0 1630.5 1.0 1630.5 1.0 1630.5 1.0 ! 2 , cm 1 1662.5 1.0 1662.5 1.0 1662.5 1.0 1662.5 1.0 ! 3 , cm 1 1457 20.0 1605 2.0 1423 5.0 1432 5.0 ! 4 , cm 1 1705 10.0 1751 5.0 1700 5.0 1729 10.0 B 1 -12.7 1.0 -19.8 0.5 -20 1.0 -25 2.0 B 2 8.1 0.5 10.5 0.5 17.4 0.5 30 0.5 B 3 8.5 1.5 1.7 0.5 40 2.0 40 5.0 B 3 20.1 1.0 5.8 1.0 12.6 1.0 33 1.0 1 , cm 1 46.4 2.0 56 2.0 48 5.0 60 5.0 2 , cm 1 46.3 2.0 40 2.0 35 2.0 44 5.0 3 , cm 1 80 20.0 27 5.0 30 10.0 30 10.0 4 , cm 1 81 2.0 71 5.0 59 5.0 99 5.0 A NR 0.710.01 0.870.002 0.650.01 0.49 0.05 , rad -1.41 0.10 -2.05 0.1 -2.63 0.1 -2.74 0.05 52 Chapter 3 References [1] Vinaykin, Mikhail; Benderskii, Alexander V . Vibrational sum-frequency spectrum of the water bend at the air/water interface J. Phys. Chem. Lett. 2012, 3, 3348– 3352. [2] Dutta, Chayan; Benderskii, Alexander V On the assignment of the vibrational spectrum of the water bend at the air/water interface J. Phys. Chem. Lett. 2017, 8, 801–804. [3] Fecko, C. J.; Eaves, J. D.; Loparo, J. J.; Tokmakoff, A.; Geissler, P. L. Ultrafast hydrogen-bond dynamics in the infrared spectroscopy of water Science 2003, 301, 1698–1702. [4] Bakker, H. J.; Skinner, J. L. Vibrational spectroscopy as a probe of structure and dynamics in liquid water Chem. Rev. 2010, 110, 1498–1517. [5] Asbury, J. B.; Steinel, T.; Stromberg, C.; Corcelli, S. A.; Lawrence, C. P.; Skinner, J. L.; Fayer, M. D. Water dynamics: Vibrational echo correlation spectroscopy and comparison to molecular dynamics simulations J. Phys. Chem. A 2004, 108, 1107–1119. [6] Du, Q.; Superfine, R.; Freysz, E.; Shen, Y . R. Vibrational spectroscopy of water at the vapor water interface Phys. Rev. Lett. 1993, 70, 2313–2316. [7] Raymond, E. A.; Tarbuck, T. L.; Brown, M. G.; Richmond, G. L. Hydrogen- bonding interactions at the vapor/water interface investigated by vibrational sum- frequency spectroscopy of hod/h2o/d2o mixtures and molecular dynamics simula- tions J Phys Chem B 2003, 107, 546–556. [8] Richmond, G. L. Molecular bonding and interactions at aqueous surfaces as probed by vibrational sum frequency spectroscopy Chem. Rev. 2002, 102, 2693–2724. [9] Tian, C. S.; Shen, Y . R. Isotopic dilution study of the water/vapor interface by phase-sensitive sum-frequency vibrational spectroscopy J. Am. Chem. Soc. 2009, 131, 2790. [10] Auer, B. M.; Skinner, J. L. Vibrational sum-frequency spectroscopy of the water liquid/vapor interface J. Phys. Chem. B 2009, 113, 4125–4130. [11] Skinner, J. L.; Pieniazek, P. A.; Gruenbaum, S. M. Vibrational spectroscopy of water at interfaces Acc. Chem. Res. 2012, 45, 93–100. 53 [12] Stiopkin, Igor V .; Weeraman, Champika; Pieniazek, Piotr A.; Shalhout, Fadel Y .; Skinner, James L.; Benderskii, Alexander V . Hydrogen bonding at the water sur- face revealed by isotopic dilution spectroscopy Nature 2011, 474, 192–195. [13] Nihonyanagi, Satoshi; Ishiyama, Tatsuya; Lee, Touk-kwan; Yamaguchi, Shoichi; Bonn, Mischa; Morita, Akihiro; Tahara, Tahei Unified molecular view of the air/water interface based on experimental and theoretical (2) spectra of an iso- topically diluted water surface J. Am. Chem. Soc. 2011, 133, 16875–16880. [14] Morita, Akihiro; Hynes, James T. A theoretical analysis of the sum frequency generation spectrum of the water surface Chem. Phys. 2000, 258, 371–390. [15] Nagata, Yuki; Hsieh, Cho-Shuen; Hasegawa, Taisuke; V oll, Judith; Backus, Ellen H. G.; Bonn, Mischa Water bending mode at the watervapor interface probed by sum-frequency generation spectroscopy: A combined molecular dynamics simu- lation and experimental study J. Phys. Chem. Lett. 2013, 4, 1872–1877. [16] Ni, Yicun; Skinner, JL Ir and sfg vibrational spectroscopy of the water bend in the bulk liquid and at the liquid-vapor interface, respectively J. Chem. Phys. 2015, 143, 014502. [17] Kundu, Achintya; Tanaka, Shogo; Ishiyama, Tatsuya; Ahmed, Mohammed; Inoue, Ken-ichi; Nihonyanagi, Satoshi; Sawai, Hiromi; Yamaguchi, Shoichi; Morita, Ak- ihiro; Tahara, Tahei Bend vibration of surface water investigated by heterodyne- detected sum frequency generation and theoretical study: Dominant role of quadrupole J. Phys. Chem. Lett. 2016, 7, 2597–2601. [18] Ingrosso, Francesca; Rey, Rossend; Elsaesser, Thomas; Hynes, James T Ultrafast energy transfer from the intramolecular bending vibration to librations in liquid water J. Phys. Chem. A 2009, 113, 6657–6665. [19] Bastida, Adolfo; Z´ u˜ niga, Jos´ e; Requena, Alberto; Miguel, Beatriz Hybrid quan- tum/classical simulation of the vibrational relaxation of the bend fundamental in liquid water J. Chem. Phys. 2009, 131, 204505. [20] Loparo, Joseph J; Roberts, Sean T; Nicodemus, Rebecca A; Tokmakoff, Andrei Variation of the transition dipole moment across the oh stretching band of water Chem. Phys. 2007, 341, 218–229. [21] Corcelli, SA; Skinner, JL Infrared and raman line shapes of dilute hod in liquid h2o and d2o from 10 to 90 c J. Phys. Chem. A 2005, 109, 6154–6165. [22] Falk, Michael The frequency of the hoh bending fundamental in solids and liquids Spectrochim. Acta, Part A 1984, 40, 43–48. 54 [23] Auer, BM; Skinner, JL Vibrational sum-frequency spectroscopy of the liquid/vapor interface for dilute hod in d2o J. Chem. Phys. 2008, 129, 214705. [24] Auer, BM; Skinner, JL Water: Hydrogen bonding and vibrational spectroscopy, in the bulk liquid and at the liquid/vapor interface Chem. Phys. Lett. 2009, 470, 13–20. [25] Pieniazek, P. A.; Tainter, C. J.; Skinner, J. L. Interpretation of the water surface vibrational sum-frequency spectrum J. Chem. Phys. 2011, 135, –. [26] Gragson, DE; McCarty, BM; Richmond, GL Surfactant/water interactions at the air/water interface probed by vibrational sum frequency generation J. Phys.Chem. 1996, 100, 14272–14275. [27] Gragson, D. E.; McCarty, B. M.; Richmond, G. L. Ordering of interfacial water molecules at the charged air/water interface observed by vibrational sum frequency generation J. Am. Chem. Soc. 1997, 119, 6144–6152. [28] Ma, Gang; Chen, Xiangke; Allen, Heather C Dangling od confined in a langmuir monolayer J. Am. Chem. Soc. 2007, 129, 14053–14057. [29] Sovago, Maria; Vartiainen, Erik; Bonn, Mischa Observation of buried water molecules in phospholipid membranes by surface sum-frequency generation spec- troscopy J. Chem. Phys. 2009, 131, 161107. [30] Richmond, GL Molecular bonding and interactions at aqueous surfaces as probed by vibrational sum frequency spectroscopy Chem. Rev. 2002, 102, 2693–2724. 55 Chapter 4: Comprehensive Molecular Picture of the Air/Water Interface This chapter provides a comprehensive molecular picture of molecular orientation at the air/water interface. SFG spectra of the water bend at the air/water interface was mea- sured in three polarization combinations (SSP, PPP and SPS) and orientational analy- sis was performed. The experimental measurements were combined with many-body molecular dynamics (MB-MD) simulations 1–3 of the SFG spectra using the MB-pol po- tential energy function for water model. 4–7 These MB-MD simulation were performed by our collaborators, Daniel R. Moberg, Shelbey C. Straight and Francesco Paesani, at the University of California, San Diego. Muhammet Mammetkuliyev wrote the Matlab script for orientational analysis. 4.1 Introduction The structure and dynamic motion of water molecules at the air/water interface is the fo- cus of many experimental and theoretical studies not only because of its complex nature but also due to its applicability in many physical, chemical and biological processes. The extended hydrogen bond structure of bulk water is terminated at the interface pro- ducing an inherent inhomogeneity in structure, orientation and distribution of hydrogen bonds which leads to many unusual physical properties of surface water including high 56 surface tension, surface wetting and hydrophobic effect among many others. A detailed knowledge of the surface orientation of water molecules is essential for understanding these different properties. Vibrational spectroscopy at the air/water interface in the wa- ter stretch/bend mode can serve as structural probes for extracting orientational informa- tion. Shen et. al. 8 reported the first vibrational sum frequency generation (SFG) spectra of the -OH stretch at the air/water interface and concluded that the free-OH (dangling -OH) groups at the air/water interface has a broad orientational distribution centered at the surface normal. 9 Gan et al. 10, 11 quantitatively analyzed the SFG spectra of the water stretch at the air/water interface in different polarization combinations and with different experimental geometries to conclude that the average tilt angle of the free-OH is 30 from the surface normal at the interface with a small distribution width. However, orientational analysis of the -OH stretching mode is complicated due to intermolecu- lar coupling between neighboring water molecules 12, 13 and the fact that the direction of transition dipole does not represent any single-OH bond. In contrast, water bend mode is least affected by intra and intermolecular couplings due to its smaller transition dipole. As there is only one HOH bend mode per molecule which is along the C 2v axis of the molecule, orientational analysis on this mode will be a direct visualization of the water orientation and structure of differently hydrogen bonded water molecules at the inter- face. Here, we present a quantitative orientational analysis of the water molecules based on our experimentally measured SFG spectra at the air/water interface using the water bending mode as the molecular probe. The experimental measurements are combined with many-body molecular dynamics (MB-MD) simulations 1–3 of the SFG spectra us- ing the MB-pol potential energy function for water. 4–7 Orientational analysis based on polarization sensitive SFG spectra along with theoretical calculations provide a com- prehensive molecular picture of average orientations of water molecules at the air/water 57 interface. Our combined experimental and theoretical studies suggest up-oriented water molecules where C 2v axis of the water molecules with free-OH or dangling -OH bonds are oriented around 80 with respect to the surface normal and down-oriented water molecules where the average orientation of the C 2v axis is around 110 from the surface normal. 4.2 Experimental results and discussions Conventional homodyne detected sum frequency generation (SFG) spectra of the water bending mode at the air/water interface was first reported by Vinaykin and Benderskii. 14 The dispersive lineshape of the water bending was assigned based on qualitative consid- erations and analogies with known OH stretching spectra. Nagata and coworkers 15 have calculated the SFG spectra of the water bend mode using MD simulations and reported the experimental and theoreticalj (2) j 2 spectra at the air/water interface. Accord- ing to their calculations, the water bending region consists of a negative peak around 1645 cm 1 and a positive peak around 1730 cm 1 . Ni and Skinner 16 developed a mixed quantum classical approach to calculate the spectral signature of different types of H- bonding species at the air/water interface. They also predicted negative amplitudes of the free or weakly H-bonded water molecules (1 N , 2 S and 3 S hydrogen bond classes) in the less blue shifted (with respect to the gas phase 17 frequency of 1595 cm 1 ) region and a positive amplitude for strongly hydrogen bonded species (4 D , 3 D hydrogen bond classes) in the SSP polarization. We validated the theoretically calculated spectra and also provided a rationale for the apparent frequency shift between SSP and PPP polar- ization spectra in a recent letter. 18 We have shown that when the air/water interface is covered with surface active molecules (e.g., surfactants), there is evidence of decreased intensity for the free-OH molecules due to the change in the ratio of amplitudes for the 58 free-OH species and strongly hydrogen bonded species peaks. As the amplitudes of these different Lorentzians change, the overall spectra, which is an interference between the negative and the positive Lorentzians, shifts. Vibrational SFG spectra of the water bending mode at the air/water interface has been measured experimentally in SSP and PPP polarizations, however SPS spectra have not been reported so far. In SPS spectra the dipole components parallel to the interface can be probed. However, due to ran- dom orientation of the water molecules at the interface, the in-plane component of the dipole moment can be cancelled on average as qualitatively explained by Richmond et. al. 19 We present the VSFG spectrum of water bend mode at the air/water interface mea- sured at different polarization combinations (SSP, PPP and SPS for SFG, visible and infrared, respectively) and provide a detailed quantitative analysis of the water orienta- tion at the air/water interface based on our experimental data. Spectral lineshapes for water bending are mostly dispersive with the resonant features sitting on top of a broad non-resonant background. The overall spectra have a negative feature around 1630 cm 1 followed by a positive feature around 1660 cm 1 due to free-OH species and H-bonding species, respectively. As water bending is affected from both the OH groups on a sin- gle water molecule, orientational analysis using water bending vibration on the free-OH species or the H-bonded species might provide more detailed information as comple- mentary to that of water stretching spectroscopy. SFG spectra of all three polarization combinations (SSP, PPP and SPS) of the water bend vibration at the air/water inter- face has been shown in figure 4.1 (left panels: a, b, c). Data acquisition time for these three spectra were same and they are normalized by the maximum intensity of the SSP spectra. These intensity spectra were fitted using j (2) j 2 =jA NR e i NR + X j B j ! IR ! j +i j j 2 (4.1) 59 Figure 4.1: Left panels: Vibrational SFG spectra of the water bend mode at the air/water interface for SSP (blue), PPP (red) and SPS (green) polarization combi- nations of SFG, visible, and infrared. Black solid lines show fit described in the text. Middle panels (d,e,f): Two principal resonant lorentzians (positive and neg- ative) used in the fit (shown in left axis) for SSP, PPP and SPS polarization com- binations. b 1 and b 2 are the amplitudes of the dangling -OH and H-bonded OH respectively. The black line represents the background signal fits (right axis).Right Panels (g,h,i): Calculated SFG spectra at different polarization combinations using MB-Pol. Reproduced from 20 . where A NR is the non-resonant background, is the relative phase between the resonant and the non-resonant background, and b m ,! m and m are the amplitude, frequency and width of the mth resonant vibrational mode. These fitting parameters for each polariza- tion combination have been summarized in table 4.2 in the Appendix section. The 1630 cm 1 peak representing the free-OH species is negative for SSP and PPP polarizations 60 and the 1660 cm 1 peak for the H-bonded species is positive. In contrast, the SPS spec- trum has opposite signs for these two peaks. The two resonant Lorentzian fit might be an oversimplification of the fact that several H-bonding classes with different spectral shape and average orientation contribute to the overall spectra as predicted by theory. ?, 21 Due to smaller transition dipole and weaker coupling, the water bend lineshapes will be least effected by the neighboring H-bond networks. In other words, non-Condon effects on the water bend vibration are minimal. Therefore, it is reasonable to fit the overall spectra with two resonant Lorentzians to decrease fitting parameters in our multi-fitting scheme. The right panel in figure 4.1 shows the theoretically calculated SFG spectra in all three polarization combinations. SFG lineshapes calculated with MB-MD and the overall signal intensity are in quantitative agreement with the experimentally measured spectra, except for the PPP spectra where the calculated relative signal intensity is lower. Figure 4.2 compares the calculated SFG spectra in the water OH-stretching region with experimental spectra from Ref 22 with quantitative agreement between the two in terms of relative intensity and spectral lineshapes throughout the entire -OH stretch re- gion, which provide further evidence for the accuracy of MB-pol in accurately describ- ing the air/water interface. 3 4.3 Orientational Analysis SFG is a unique surface specific technique where the signal intensity is highly polar- ization dependent. Hence, a quantitative analysis of the polarization selective SFG data is needed in order to delve into the complex molecular structure and orientation 61 Figure 4.2: Left panels: Vibrational SFG spectra of the water stretch mode at the air/water interface for SSP (blue), PPP (red) and SPS (green) polarization combi- nations of SFG, visible, and infrared. Black solid lines show fit described in the text. Right Panel: Calculated SFG spectra at different polarization combinations using MB-Pol. Reproduced from 20, 22 . of molecules at the interface. 22–24 Rao et. al. 24 showed that (2) can be simplified in the form of the following expression, (2) eff =N S d(hcosichcos 3 i) (4.2) where the ’c’ and ’d’ parameters for the three polarization combinations (SSP, SPS, PPP) can be rigorously deduced after transforming the molecular coordinate fixed hy- perpolarizability tensors into the laboratory fixed axis. The ’c’ and ’d’ parameters for the 62 Figure 4.3: Calculated SFG amplitude (2) eff as a function of the tilt angle of the HOH bend dipole with respect to the surface normal in SSP, PPP and SPS polar- ization combinations for SFG-Visible-infrared laser beams. Reproduced from 20 . HOH bending mode in SSP, PPP and SPS polarizations are presented in Table 4.3. (2) can be thought of as the overall SFG amplitude since the intensity spectrum is propor- tional toj (2) eff j 2 . Hence, the variation of (2) as a function of average tilt angle of the HOH bend dipole with respect to the surface normal can be determined by equation 4.1. To calculate (2) for different polarizations, it is important to know the aac , bbc and ccc values for the bending ( 2 ) mode of water. Ni and Skinner 16 have calculated the dipole m Qm and polarizability derivative a ab Qm tensor of the gas-phase water molecule and used them to calculate the SFG spectra of water bending at the air/water interface. Computa- tional studies 16 reveal that the magnitude of the dipole derivative and the polarizability for the bend mode ( 2 ) of water is rather insensitive to the surrounding environment. In other words, non-Condon effects are weaker for the water bend mode which also 63 allows us to use these gas-phase values to calculate (2) eff . In view of the fact that the polarizability derivatives of the water bend mode in liquid water can be affected by its environment, we performed sensitivity analysis where we have changed the ratios ( aac ccc and bbc ccc ) to elucidate its effect on the overall SFG amplitude. This is discussed in detail in the next section. Figure 4.3 shows the amplitude variation of the SFG signal with the change of av- erage tilt angle. Experimental amplitude values for the resonant vibrations obtained by curve fitting can be compared for different polarizations and the corresponding average tilt angle for different H-bonded species can be obtained. Our experimental amplitude ratios (SSP:PPP:SPS) are 1:1.556:1.567 and 1:1.325:1.25, respectively (see Table [S1] in the Appendix) for the free-OH peak (b 1 ) and the H-bonded OH peak (b 2 ). Comparing the experimentally determined amplitude ratios to that of calculated SFG amplitudes, we obtain an average tilt angle for the bending dipole of 80 5 for the free-OH and 107 5 for the hydrogen bonded water molecules. We have used the gas-phase polarizability values from literature 16 assuming that non-Condon effects are minimal. Sensitivity analysis of SFG amplitudes with values We present the SFG amplitudes of the water bending mode at the air/water interface as a function of the average tilt angle of the HOH bending dipole with the change of ratios in figure 4.4. Our sensitivity analysis reveal that the overall shape of the SFG amplitude in all three polarization combinations remain similar and the corresponding average tilt angles for the HOH bending dipole (from experimentally determined values) are within 10 of the actual tilt angle for free-OH and Hydrogen bonded OH species for all the polarizations. Equation 4.2 provides a trivial solution for (= 90 ) and we can see that 64 (a) (b) (c) Figure 4.4: Sensitivity analysis of SFG amplitudes with the change of aac ccc and bbc ccc values. (a) aac ccc changed by50% from the gas-phase value (see Table 4.4. (b) bbc ccc changed by 50% from the gas-phase value. (c) both aac ccc and bbc ccc changed by 50%. Solid circles represent the calculated amplitudes with increased ratios and open circles represent the calculated amplitudes with decreased values, whereas the solid lines are for gas-phase values and same in all three graphs. Reproduced from 20 . the SFG amplitude changes sign around that angle for all three polarization combina- tions. However, it also gives a unique solution (cos = p ( 1 c ) ) that depends on the c parameter which is different in different polarizations, which in turn produces another 65 change of sign in the SFG amplitude around 45 and 135 for the PPP polarization. The angle at which the PPP amplitude crosses the zero line is quite insensitive to the change of ratios as the change in ‘c’ parameter is negligible. Figure 4.5: Change in PPP/SSP amplitude (Left Panel) and SPS/SSP amplitude (Right Panel) of the water bend with the change in standard deviation () of the orientational distribution of the HOH bend dipole. The dotted horizontal lines represent the experimentally measured ratios for the free-OH species (red) and the hydrogen bonded species (blue) in both graphs. The vertical solid lines are a guide to the experimentally measured average tilt angles for the respective species (red: free-OH and blue: H-bonded). The width of the solid lines shows the error estimated in the experimentally measured tilt angles. Reproduced from 20 . The present orientational analysis with the water bend mode predicts an average tilt angle of the HOH bisector to be around 80 5 , which agrees with the previous studies of water stretches 8–11 and provides complementary information about the orientational distribution of the bend mode. In this analysis, Gaussian distributions are assumed for all calculations of the SFG amplitudes as a function of the average tilt angle. How- ever, to obtain the angular range of distributions for different classes of H-bonded water molecules, the SFG amplitudes at different standard deviations have been calculated and compared with the experimentally measured amplitude ratios (Figure 4.5). Comparing these results with experimentally obtained ratios, it is found that the average angular distribution for the orientation of the C 2v axis of the water molecules at the air/water 66 interface are 12 5 and 20 5 for the free-OH and H-bonded water molecules, respectively. 4.3.1 Orientation from theoretical calculations A snapshot from the MB-MD trajectory for the air water interface has been presented in figure 4.6. Water molecules with free-OH bonds have been identified from the snap- shot and the corresponding bend frequency of that molecule was calulated. Criteria for (a) (b) Figure 4.6: (a) A snapshot from the MB-MD trajectory for the air/water interface. (b) Histogram of up (red) and down (blue) oriented water molecules, with the SSP SFG spectrum (black trace) overlaid. The up oriented waters have a maximum around 1660 cm 1 and the down oriented water molecules have a maximum near 1680 cm 1 . Reproduced from 20 . chosing a free-OH bond or the probing depth of the interfacial region has been defined according to standard procedures and presented in the appendix section. Histogram dis- tributions of up and down oriented water molecules have been calculated and plotted in figure 4.6. Calculations of the fraction of water molecules with free-OH bonds at the air/water interface reveal that about 29 % of the water molecules at the air/water inter- face are free-OH or non-hydrogen bonded species, which is in agreement with previous 67 studies. 25–27 It should also be noted that there is a 20 cm 1 frequency shift on average between water molecules oriented up and down with respect to the surface normal. It should also be noted that the calculated weighted average angles from MB-MD simu- lation at the air/water interface for up/down oriented water molecules are 60 and 118 respectively. The calculated tilt angle values for these two different species present at the top most layer of water is in great agreement with our experimental results. 4.3.2 Comprehensive picture of molecular orientation at the air/water interface OH stretch and HOH bend: A comparative study The molecular level information obtained from analysing the VSFG spectra of the OH- stretch mode at the air/water interface 22, 28 provides the average tilt angle for molecules with free-OH bond. Water molecules with one free-OH and onother -OH hydrogen bonded to its nearest neighbour are oriented with a 30 tilt angle (for the free-OH) with respect to the surface normal. Further analysis shows that the average orientation for this free-OH molecules has a very narrow distribution which forms a cone around the surface normal as shown in figure 4.7a. These studies can not predict or produce any orientational information for the other -OH bond in the molecule, as the overall stretch spectra in the H-bonded region is broad and all the different components overlap. On the other contrary, analysis of the bend spectra allows us to extract orientation of the C 2v axis of the water molecule. Since there is only one HOH bend mode per molecule, the orientation of the C 2v axis represents the average orientation for the whole molecule. On top of that, as the effect of inter-molecular coupling between water molecules to the bend mode is weaker, it is easier to differentiate between water molecules on the basis 68 (a) Free-OH stretch (b) Free-OH bend (c) H-bonded OH Bend Figure 4.7: Model illustration of orientation of the water molecules at the interface. (a) The water molecule at the interface (in the ‘xz’ plane) with a free-OH bond pointing towards the air phase, taken from 28 with permission. (b) Tilt angle of the C 2V axis for a molecule with a dangling-OH at the interface, mostly oriented upwards to the air phase and (b) Tilt angle of the C 2V axis for a molecule with a extended H-bond at the interface, mostly oriented downwards to the water phase. of their orientation (up/down from the surface). Figure 4.7b and 4.7c show a model representation for the average orientation of the water molecules orientated upwards and downwards respectively. On average, water molecules with free-OH bonds are oriented upwards and the average tilt angle for the C 2v axis is80 from the surface normal. This picture is in general agreement with the molecular picture obtained form the water stretch spectroscopy. We can also predict the average orientation for water molecules with higher degree of hydrogen bonding most of which will have a downward orientation from the surface layer. Average tilt angle for the C 2v axis for such extended H-bonded water molecules is110 from the surface normal. Table 4.1: Average orientation angle for different H-bonded species. Average tilt angle for Free-OH stretch 22, 28 Average tilt angle for C 2v axis for free-OH bend Average tilt angle for C 2v axis for H-bonded OH bend angle 30 80 5 107 5 width of distribution 15 or less 12 5 15 5 calculated 60 1 118 1 69 A comprehensive molecular picture for the possible orientational distribution of the water molecules at the interface is presented in Figure 4.8. A static picture for the free-OH molecules at the interface corresponds to that where the whole H 2 O molecule can have an orientational distribution in the ‘xz’ plane (Figure 4.8a) or out of the plane where one OH bond remains in the same position and the other OH bond lies out of the molecular plane (Figure 4.8b). For water molecules with two donor H-bonds, the average direction of the HOH bending dipole is oriented towards the bulk water and the average orientation for these molecules is in the ‘yz’ plane (Figure 4.8c). (a) ‘xz’ plane (b) away from‘xz’ plane (c) ‘yz’ plane Figure 4.8: Pictorial description for the angular distribution of orientation of water molecules at the air/water interface. (a) the whole molecule is in ‘xz’ plane and rotates in plane. (b) one -OH bond (free-OH) is in the ‘xz’ plane and the other -OH rotates away from the plane. (c) Both -OH are hydrogen bonded and the HOH dipole is pointed downwards towards the bulk and rotates in ‘yz’ plane. In conclusion, we have employed a quantitative orientational analysis at the air/water interface to determine hydrogen bonding structure and motion of water molecules us- ing water bending SFG spectra in different polarization combinations. Our orientation results from water bending spectroscopy along with other results obtained from water stretch spectroscopy 22, 27–29 provide a complete picture of the hydrogen bonding struc- ture at the air/water interface. As water bend spectra provides a more localized and complementary picture of water structure at the interface to that available from water 70 stretch spectra, it can be applied to study water orientation at charged interfaces as well as many other complex chemical and biological interfaces. 4.4 Appendix 4.4.1 MB-MD Simulations and SFG spectra calculation Molecular Dynamics simulations and theoretical modeling of the air/water interface were performed by the Paesani group (University of California, San Diego). Classical MB-MD simulations were performed using the MB-pol potential energy surface on a box with dimensions of 26 26 100 ˚ A in 3D periodic boundary con- ditions, containing 512 water molecules. This setup provides a slab of water molecules parallel to the xy plane. This system had previously been equilibrated at 298.15 K in the NVT ensemble using Nos e-Hoover chains of four thermostats with the equations of mo- tion propagated using the velocity-Verlet algorithm with a timestep of 0.02 fs. 30 These simulations provided 32 initial conditions for NVE simulations. Each of the 32 initial conditions were run for 250 ps resulting in8 ns of NVE trajectories which were used to calculate the real and imaginary parts of (2) in the PPP, SSP, and SPS polarizations. 3 For the calculation of sum frequency spectra from the correlation function, the truncated-cross correlation function (TCF) method was implemented as described in Ref. 31 In order to analyze only water molecules near the surface of the slab, and to 71 account for the double interface setup of the slab system, the screening function was used, g SC (z) =sign(z) 8 > > > > > > < > > > > > > : 0 ifjzjz c1 ; cos 2 ( (jzjz c2 ) 2(Z c1 z c2 ) ) ifz c1 <jzjz c2 ; 1 ifjzj>z c2 : (4.3) withz c1 =7 ˚ A andz c2 =8 ˚ A. The one-body and N-body induced contributions (1B+NB) approximation of the dipole moment surface and 1B approximation to the polarizabil- ity tensor were used in calculating the TCF. Only contributions from pairs of water molecules whose oxygen atoms were no more than 4 ˚ A were counted. For further de- tails on the implementation of the TCF method with classical MB-MD using MB-pol, see Ref. 3 4.4.2 Normal Mode Analysis and Orientational Analysis from MD Trajectories Individual frames were extracted from the 8 ns of NVE trajectories, then their con- figurations optimized. From these optimized configurations, the normal modes were calculated by diagonalization of the Hessian matrix of the simulation box in periodic boundary conditions. 32 The resulting frequencies and displacements for each normal mode from 8 independent frames were analyzed in the subsequent analysis to determine orientations of water molecules in the bending region and fractions of free-OH species. Water Bending Region The same screening function and values for z c1 and z c2 given in Equation 4.3 were used to select only water molecules near the surface normal. For each remaining water 72 molecule near the surface, the angle along each water molecule’s C 2V axis creates with the surface normal was calculated. To approximately identify which molecules con- tribute to the vibrational mode, the total displacement D i of the oxygen molecule was calculated to provide an approximate measure of the water molecule’s activity, D i = X j=x;y;z jd (i) j j 2 : (4.4) Here, d (i) j is the displacement of the oxygen atom of the i th water molecule along the j th cartesian axis in the k th normal mode. The D i values in each frame were normalized with respect to the largest value over all modes. As a computational cost saving method, all water molecules with a D i value less than 0.01 were not included in the follow- ing analysis. Note that the displacement is an approximate method of estimating each molecule’s contribution to the SFG signal. The average angle of the remaining water molecules was calculated using their corresponding D i values as weights. The average frequency of the sets of up and down oriented water molecules was also calculated. Free OH Calculation The following criteria were used to determine whether a water molecule contains a free- OH bond: 1. Normal modes with frequencies larger than 2800 cm 1 . 2. For this analysis a hard cutoff of 8 above the center of mass was used instead of the screening function g sc (z). 3. The two angles created between the surface normal and the two OH bond vectors in each water molecule were calculated. The smallest must be less than 90 . 73 4. The distance between the hydrogen of the smallest angle and the next nearest oxygen molecule must be greater than 2.5 ˚ A. If a molecule satisfies all four conditions, it is included and cross checked with the list of molecules from the bending region for each frame. From this cross checked list, the fraction of water molecules in the bending region with free-OH bonds was calculated to be 0.2930 0.0220. Table 4.2: Fitting parameters of the vibrational SFG spectra of the water bend measured for SSP, PPP and SPS polarizations at the air/water. All spectra were normalized by the maximum intensity of the SSP spectrum. Air/Water SSP PPP SPS ! 1 , cm 1 1630.5 1.0 1630.5 1.0 1630.5 1.0 ! 2 , cm 1 1662.5 1.0 1662.5 1.0 1662.5 1.0 ! 3 , cm 1 1457 20.0 1605 2.0 1541 10.0 ! 4 , cm 1 1705 10.0 1751 5.0 1750 10.0 B 1 -12.7 1.0 -19.8 0.5 19.9 0.5 B 2 8.0 1.0 10.5 0.5 -10 1.0.5 B 3 9.8 2.0 2.94 0.5 3.95 1.0 B 4 23.72.0 3.98 0.5 13.5 1.0 1 , cm 1 46.0 2.0 61.0 5.0 74 3.0 2 , cm 1 46.0 2.0 42 2.0 73 10.0 3 , cm 1 71 20.0 38 10 .0 36 20.0 4 , cm 1 90 2.0 62 10.0 98 5.0 A NR 0.6950.01 0.6870.002 0.1630.01 , rad -1.41 0.10 -1.95 0.1 -0.766 0.1 74 Table 4.3: ‘c’ and ‘d’ parameters for the water bending mode in three polarization combinations SSP PPP SPS c -1.5641 2.078 1.0 d -0.3682 0.7349 0.61 Table 4.4: ijk values used for sensitivity analysis. values calculated for single water molecule were used to calculated all the components and they are normalized by ccc . Actual value used in Figure 4.4a used in Figure 4.4b used in Figure 4.4c aac -6.614 -9.921 -4.409 -6.614 -6.614 -9.921 -4.409 bbc -2.468 -2.468 -2.468 -3.702 -1.645 -3.702 -1.645 ccc -2.468 1 1 1 1 1 1 75 Chapter 4 References [1] Medders, Gregory R; Paesani, Francesco Infrared and raman spectroscopy of liq- uid water through first-principles many-body molecular dynamics J. Chem. Theory Comput. 2015, 11, 1145–1154. [2] Straight, Shelby C; Paesani, Francesco Exploring electrostatic effects on the hy- drogen bond network of liquid water through many-body molecular dynamics J. Phys. Chem. B 2016, 120, 8539–8546. [3] Medders, Gregory R; Paesani, Francesco Dissecting the molecular structure of the air/water interface from quantum simulations of the sum-frequency generation spectrum J. Am. Chem. Soc. 2016, 138, 3912–3919. [4] Medders, Gregory R; Babin, V olodymyr; Paesani, Francesco Development of a first-principles water potential with flexible monomers. iii. liquid phase properties J. Chem. Theory Comput. 2014, 10, 2906–2910. [5] Babin, V olodymyr; Medders, Gregory R; Paesani, Francesco Development of a first principles water potential with flexible monomers. ii: Trimer potential energy surface, third virial coefficient, and small clusters J. Chem. Theory Comput. 2014, 10, 1599–1607. [6] Babin, V olodymyr; Leforestier, Claude; Paesani, Francesco Development of a first principles water potential with flexible monomers: Dimer potential energy sur- face, vrt spectrum, and second virial coefficient J. Chem. Theory Comput. 2013, 9, 5395–5403. [7] Reddy, Sandeep K; Straight, Shelby C; Bajaj, Pushp; Huy Pham, C; Riera, Marc; Moberg, Daniel R; Morales, Miguel A; Knight, Chris; Gtz, Andreas W; Paesani, Francesco On the accuracy of the mb-pol many-body potential for water: Interac- tion energies, vibrational frequencies, and classical thermodynamic and dynamical properties from clusters to liquid water and ice J. Chem. Phys. 2016, 145, 194504. [8] Du, Q; Superfine, Richard; Freysz, E; Shen, YR Vibrational spectroscopy of water at the vapor/water interface Phys. Rev. Lett. 1993, 70, 2313. [9] Wei, Xing; Shen, YR Motional effect in surface sum-frequency vibrational spec- troscopy Phys. Rev. Lett. 2001, 86, 4799. [10] Gan, Wei; Wu, Dan; Zhang, Zhen; Feng, Ran-ran; Wang, Hong-fei Polarization and experimental configuration analyses of sum frequency generation vibrational 76 spectra, structure, and orientational motion of the air/water interface J. Chem. Phys. 2006, 124, 114705. [11] Gan, Wei; Wu, Dan; Zhang, Zhen; Guo, Yuan; Wang, Hong-fei Orientation and motion of water molecules at air/water interface Chinese J. Chem. Phys. 2006, 19, 20–24. [12] Loparo, Joseph J; Roberts, Sean T; Nicodemus, Rebecca A; Tokmakoff, Andrei Variation of the transition dipole moment across the oh stretching band of water Chem. Phys. 2007, 341, 218–229. [13] Corcelli, SA; Skinner, JL Infrared and raman line shapes of dilute hod in liquid h2o and d2o from 10 to 90 c J. Phys. Chem. A 2005, 109, 6154–6165. [14] Vinaykin, Mikhail; Benderskii, Alexander V Vibrational sum-frequency spectrum of the water bend at the air/water interface J. Phys. Chem. Lett. 2012, 3, 3348– 3352. [15] Nagata, Yuki; Hsieh, Cho-Shuen; Hasegawa, Taisuke; V oll, Judith; Backus, Ellen HG; Bonn, Mischa Water bending mode at the watervapor interface probed by sum-frequency generation spectroscopy: A combined molecular dynamics sim- ulation and experimental study J. Phys. Chem. Lett. 2013, 4, 1872–1877. [16] Ni, Yicun; Skinner, JL Ir and sfg vibrational spectroscopy of the water bend in the bulk liquid and at the liquid-vapor interface, respectively J. Chem. Phys. 2015, 143, 014502. [17] Falk, Michael The frequency of the h o h bending fundamental in solids and liquids Spectrochim. Acta Part A 1984, 40, 43–48. [18] Dutta, Chayan; Benderskii, Alexander V On the assignment of the vibrational spectrum of the water bend at the air/water interface J. Phys. Chem. Lett. 2017, 8, 801–804. [19] Raymond, E. A.; Tarbuck, T. L.; Brown, M. G.; Richmond, G. L. Hydrogen- bonding interactions at the vapor/water interface investigated by vibrational sum- frequency spectroscopy of hod/h2o/d2o mixtures and molecular dynamics simula- tions J Phys Chem B 2003, 107, 546–556. [20] Dutta, C; Moberg, D. R.; Mammetkuliyev, M.; Straight, S.C.; Paesani, F.; Bender- skii, A.V . A comprehensive molecular picture of the air/water interface Submitted. [21] Nagata, Yuki; Hsieh, Cho-Shuen; Hasegawa, Taisuke; V oll, Judith; Backus, Ellen H. G.; Bonn, Mischa Water bending mode at the watervapor interface probed by sum-frequency generation spectroscopy: A combined molecular dynamics simu- lation and experimental study J. Phys. Chem. Lett. 2013, 4, 1872–1877. 77 [22] Wei, Xing; Shen, YR Motional effect in surface sum-frequency vibrational spec- troscopy Phys. Rev. Lett. 2001, 86, 4799. [23] Wang*, Hong-Fei; Gan, Wei; Lu , Rong; Rao, Yi; Wu, Bao-Hua Quantitative spectral and orientational analysis in surface sum frequency generation vibrational spectroscopy (sfg-vs) Int. Rev. Phys. Chem. 2005, 24, 191–256. [24] Rao, Yi; Tao, Yi-song; Wang, Hong-fei Quantitative analysis of orientational order in the molecular monolayer by surface second harmonic generation J. Chem. Phys. 2003, 119, 5226–5236. [25] Stiopkin, Igor V .; Weeraman, Champika; Pieniazek, Piotr A.; Shalhout, Fadel Y .; Skinner, James L.; Benderskii, Alexander V . Hydrogen bonding at the water sur- face revealed by isotopic dilution spectroscopy Nature 2011, 474, 192–195. [26] Morita, Akihiro; Hynes, James T. A theoretical analysis of the sum frequency generation spectrum of the water surface Chem. Phys. 2000, 258, 371–390. [27] Du, Q.; Superfine, R.; Freysz, E.; Shen, Y . R. Vibrational spectroscopy of water at the vapor water interface Phys. Rev. Lett. 1993, 70, 2313–2316. [28] Gan, Wei; Wu, Dan; Zhang, Zhen; Feng, Ran-ran; Wang, Hong-fei Polarization and experimental configuration analyses of sum frequency generation vibrational spectra, structure, and orientational motion of the air/water interface J. Chem. Phys. 2006, 124, 114705. [29] Gan, Wei; Wu, Dan; Zhang, Zhen; Guo, Yuan; Wang, Hong-fei Orientation and motion of water molecules at air/water interface Chinese J. Chem. Phys. 2006, 19, 20–24. [30] Martyna, Glenn J; Klein, Michael L; Tuckerman, Mark Noshoover chains: the canonical ensemble via continuous dynamics J. Chem. Phys. 1992, 97, 2635–2643. [31] Nagata, Yuki; Mukamel, Shaul Vibrational sum-frequency generation spec- troscopy at the water/lipid interface: molecular dynamics simulation study J. Am. Chem. Soc. 2010, 132, 6434–6442. [32] Moberg, Daniel R; Straight, Shelby C; Knight, Christopher; Paesani, Francesco Molecular origin of the vibrational structure of ice ih J. Phys. Chem. Lett. 2017. 78 Chapter 5: Refined Picture of Molecular Orientation at Charged Interfaces 5.1 Introduction Detailed understanding of the water hydrogen bonding network at the charged interfaces should provide a fundamental background for many chemical and biological processes. Despite various attempts 1–7 the true nature of interactions between water molecules and charged interfaces is yet to be completely recognized. Bulk water show unique prop- erties due to its ability to form extensive hydrogen bond networks, where most of the water molecules are tetrahedrally coordinated giving rise to maximum number of hy- drogen bonds possible around a single water molecule. This three dimensional network of hydrogen bonds is sharply terminated at the interface making it more heterogeneous than the bulk. 8, 9 Theoretical studies 10, 11 predicted the air/water interface to be very thin (about 3 ˚ A), however the direct experimental evidence is elusive till date. Theo- retical studies by Skinner and co-workers 10, 12–14 proved that this wafer-thin surface of water is comprised of several different types of water molecules with various degree of hydrogen bonding. This inherently diverse network of hydrogen bond undergoes 79 structural changes under various degree of charge on the surface. Direct charge-dipole interaction can be thought of as the major factor that determines the water structure at the charged interface. For a negatively charged interface the hydrogens of the water molecules should orient up towards the negative charge, whereas, in a positively charged interface the oxygen atom of a water molecule should orient towards the positive charge due to the negative charge of the oxygen atom. Nihonyanagi et al. 4 provided direct spectroscopic evidence for this orientational flip-flop of water molecules at charged in- terfaces by phase sensitive vibrational sum frequency generation spectroscopy (SFG). Their conclusion of water orientation at charged interfaces was based on the sign of the Im (2) SFG spectra at charged aqueous interfaces in the water stretching region which provides a qualitative (up/down) orientation information since stretching region of water is spectrally very broad and can be complicated due to intra and intermolecular coupling between neighboring water molecules. Water surface is very heterogeneous with a wide variety of water molecules with different degree of hydrogen bonding. Orientation of these different water molecules around the charged headgroup of surfactants requires ad- ditional attention. Also, the quantitative estimation of charge density required for a com- plete reversal of water orientation at interfaces is required for a detailed understanding of charged surfaces. Vibrational spectroscopy in the OH-stretching region is a powerful tool for studying the H-bond structures, distribution and ultrafast dynamics in bulk 15–17 and interfacial water. 8, 12, 18–20 However, water bend spectroscopy 21, 22 is believed to be free of many complexities (e.g., intra- and inter- molecular coupling) that make the wa- ter stretch spectroscopy difficult to interpret and it provides complementary information to the OH-stretch spectra. Since water bend mode is directly affected by both the donor hydrogen bonds in the molecule, it can be thought of as a local probe of the hydro- gen bond structure. Hence in a charged interface the water bend spectra could provide 80 Figure 5.1: Schematic of the water H-bond reorientation for different types of wa- ter molecules. valuable information regarding the structure of hydrogen bonds and orientation of water molecules at the interface. In this chapter, we present the water bend SFG spectrum in a positively charged surfactant cetyltrimethylammonium bromide (CTAB)/water inter- face in SSP, PPP and SPS polarization combination and provided average orientation of water molecules with free-OH bonds and fully hydrogen bonded water molecules. Our comparative lineshape analysis and orientation analysis of the water bend spectra from a negatively charged surfactant (sodium dodecyl sulfate, SDS) and positively charged surfactant interface shows structural and orientational dissimilarity between these two interfaces. Orientational changes as a function of surface charge density has also been recorded which provides a quantitative estimation of the surface charge density required for a complete flip-flop in geometry at the interface (Figure 5.1). 81 5.2 Results and discusssions 5.2.1 Comparison of water bend SFG spectra in SDS and CTAB solutions Previously we reported the first SFG spectrum of the water bend at the air/water inter- face. 21 In general the observed line shape consists of a lower frequency negative and a higher frequency positive feature for SSP and PPP polarizations 22 and a positive lower frequency followed by a higher frequency negative feature for SPS polarization, 23 that sits on top of a broad background which presumably comes due to librational overtones at this spectral region. Theoretical studies 14, 24 predict that the negative red-shifted peak belongs to the water molecules from 1 N , 2 S , 3 S H-bond classes as defined by Ni and skinner, 14 and the blue-shifted positive peak is attributed to molecules from 3 D , 4 D H- bonded classes in air/water interface for SSP polarization, where 2,3,4 represents total number of hydrogen bonds around a water molecule and N, S, D subscript signifies non-donor, single-donor and double-donor hydrogen bond respectively. For example, 2 S represents water molecules having 2 hydrogen bonds with single donor hydrogen, and 1 N type of water molecules having only one hydrogen bond without any donor hydrogen etc. Recent theoretical calculations by Paesani group 23 have confirmed the lineshapes of water bending vibration at the air/water interface in all three polarization combinations (SSP, PPP and SPS). We provided experimental evidence 22 in favor of this assignment of the bend spectrum and also explained that the apparent frequency shift in SSP and PPP polarization occur due to the interference of two peaks with opposite signs and different amplitudes. We presented the water bend spectra from SDS/water inter- face which proved the assignment of the 1630 cm 1 peak to free-OH species. Even 82 though free-OH species peak is suppressed 22 with respect to the hydrogen bonded peak in the negatively charged surfactant (SDS) interface, the overall line shape in the water bend region resembles that of an air/water interface with a negative peak and a positive peak in both SSP and PPP polarization combination. However, the SFG spectra form a positively charged surfactant interface in the water bend region is different as shown in Figure 5.2. In figure 5.2, SFG spectra of the water bend from a 0.4mM solution of CTAB and 1.0 mM solution of SDS has been shown for SSP and PPP polarization combination for SFG, Visible and IR pulses. ‘S’ stands for perpendicular polarization of light with respect to the surface normal and ‘P’ stands for parallel polarization with respect to the surface normal. It is quite clear that the SSP spectrum has similar line shape as pure air/water interface, having a negative peak at 1630 cm 1 and a positive peak at 1662.5 cm 1 , whereas the PPP spectrum is flipped with respect to the SSP spectra. Curve fit- ting of the SDS and CTAB spectra can be done with the two Lorentzian fitting model adopted for air/water interface keeping the resonant frequencies constant, but the signs of the two main resonant features have to be interchanged for the PPP spectrum. In SSP, the 1630 cm 1 peak is negative and the 1662.5 cm 1 peak has positive amplitude which is similar to SSP spectra of air/water interface. In contrast, PPP spectrum shows positive and negative amplitude for the 1630 cm 1 peak and the 1662.5 cm 1 peak re- spectively. As the instantaneous bend frequency in the water bending region is weakly dependent on the H-bond strength i.e., its hydrogen bonding environment, the blue shift of the bend is much smaller ( 4 times smaller) than the red-shift of the OH-stretch frequency. Because of this weaker environment dependence, bend frequencies from wa- ter molecules with various degree of hydrogen bonds are concentrated around smaller range of frequencies. Ni and Skinner 14 were able to separate the spectral contributions of different sub-classes of water molecules toward the overall water bend spectra at the 83 air/water interface by mixed quantum/classical calculations. Their calculations showed that the SFG spectra from different sub-classes of water molecules overlap one another with distinct negative intensity in the red side of the spectrum and positive intensity at the blue side of the spectrum. (a) (b) Figure 5.2: Vibrational SFG spectra of the water bend mode at the air/CTAB so- lution (Left) and air/SDS solution interface (right) for PPP (red) and SSP (blue) polarization combinations of SFG, visible, and IR. Bulk concentration for CTAB and SDS was 0.4 mM and 1.0 mM respectively. Reproduced from 25 Here, we would like to point out the fact that, positive and negative amplitudes in the fitting of the experimentally measured intensity spectra could contain contributions from both strongly hydrogen bonded species and weakly hydrogen bonded species. Hence, instead of trying to decipher the contribution of each species in the experimental SFG spectra if we try to understand the average orientation of different species (mainly for the negative and positive amplitude peaks), we should be able to get quantitative information about hydrogen bond structure and orientation at the air/water or charged interfaces. 84 5.2.2 Change in water bend SFG spectra with CTAB concentration It is not unreasonable to assume that a positive charge at the surface will orient the wa- ter molecules with their oxygen molecules pointing upwards to the surface. Since the PPP spectra from CTAB interface has a flipped spectral line shape as compared to the air/water interface, we should expect a gradual change in the water bend spectra with gradual increase in surface charge. In figure 5.3, SFG spectra from CTAB solutions in SSP, SPS and PPP polarization combination have been plotted for increasing CTAB concentration from 0.01 mM to 0.07 mM.. Surface density of surfactants can be calcu- lated from the bulk concentrations using Gibbs adsorption isotherm and are presented in Table 5.1. All spectra were fitted using the same two Lorentzian model keeping the resonant frequencies constant at 1630 cm 1 and 1662.5 cm 1 . Spectral lineshapes in SSP and SPS polarization combinations do not change with surface concentration of the surfactants. SFG spectra in SSP polarization in the water bending region has a negative and positive amplitude in the red side and the blue side of the spectrum respectively irrespective of the surface density of surfactants. On the other hand, SPS spectral line- shape has a positive and negative amplitude in the low to high frequency side of the spectrum respectively at all concentration. However, PPP spectra from the same solu- tions show different concentration-dependent behavior. At very low concentration ( 0.01 mM) spectral fitting of the line shape gives a negative and a positive resonant feature in PPP polarization. However, with increasing concentration the amplitude of the negative lorentzian at 1630 cm 1 become less negative and eventually become positive after 0.06 mM bulk concentration, whereas the amplitude of the positive lorentzian at 1662.5 cm 1 become negative at that concentration. Non-resonant background obtained from spectral fitting of CTAB solution at differ- ent concentrations and different polarization combinations is plotted in Figure 5.3d. As 85 (a) (b) (c) (d) Figure 5.3: Vibrational SFG spectra of the water bend mode at the CTAB/water interface with increasing CTAB concentration from 0.01 mM to 0.1 mM for SSP (a), SPS (b) and PPP (c) polarization combinations of SFG, visible, and IR. Black solid lines show fit described in the text. PPP spectra in 0.07 mMol has been offset by -0.03 for clarity. Variation of Non-resonant background (A NR ) as a function of bulk CTAB concentrationis plotted in (d) in SSP, PPP and SPS polarization. Reproduced from 25 . the concentration increases the non-resonant background also increases monotonically in all polarizations, which is an indication that the apparent sign change of SFG spectra in PPP polarization with increasing concentration is not an artifact of the sign change of the non-resonant background Figure 5.4 shows this variation of the amplitudes at 1630 cm 1 (B 1 in left panel) and at 1662.5 cm 1 (B 2 in right panel) with the change of bulk CTAB concentration. As 86 the surface charge density increases the B 1 amplitude decreases while the B 2 amplitude increases in SSP. In SPS, both B 1 and B 2 remains almost constant, i.e., B 1 remains positive and B 2 remains negative. However, the signs of B 1 and B 2 amplitude change as the surface charge density grows gradually from lower to higher value for the PPP spectra. The surface concentration of the CTAB molecules (molecules / ˚ A 2 ) has been calculated using Gibbs adsorption equilibrium and presented in Table 5.1. The surface density of CTAB molecule at around 0.06 mMol bulk concentration has been estimated to be around 10 13 (molecule /cm 2 ) which is roughly equal to a charge density of 0.001 charge/ ˚ A 2 at the surface. Table 5.1: : Surface charge density of surfactants (CTAB) for varying bulk surfac- tant concentrations Concentration (mMol) Charge Density (charge/ ˚ A 2 ) Molecules/cm 2 0.01 3 x 10 4 3 x 10 12 0.03 6 x 10 4 6 x 10 12 0.07 0.002 2x 10 13 Previous studies 26–28 of electro-kinetic measurements and measurements of diffuse electric double layer in aqueous solutions with increasing concentration of cationic sur- factant show charge reversal at these low concentrations. The diffuse electric layer po- tential changes from a negative value at the air/water interface to a positive value with increasing cationic surfactant concentration. Our quantitative measurement of the sur- face positive charge concentration at which the surface orientation of water molecules changes provides comparable charge density at which charge reversal occur as predicted previously by electro-kinetic measurements. 87 Figure 5.4: Variation of amplitudes for free-OH molecules (Left) and hydrogen bonded -OH molecules (right) from spectral fitting of the water bending spectra in SSP (blue), SPS (green) and PPP (red) polarization from CTAB/water interface as a function of bulk CTAB concentration. B1 is the amplitude at 1630 cm-1 for molecules with free-OH and B2 is the amplitude at 1662 cm-1 for hydrogen bonded -OH molecules. Reproduced from 25 . 5.3 Orientation at the Interface Orientation of the C 2V axis of the water molecules at the interface in presence of sur- factant molecules has been determined by orientational analysis of the polarization de- pended SFG spectra of the water bend mode. The experimentally measured (2) ijk tensors in the lab-frame are connected to the molecular hyperpolarizability tensor (2) i 0 j 0 k 0 by ro- tational transformation matrix R(,, ). 29 (2) ijk =N s X i 0 j 0 k 0 hR ii 0R jj 0Rkk 0 i (2) i 0 j 0 k 0 (5.1) where, N s is the number density of molecules at the interface. hi represents orienta- tional average of the corresponding property over the orientational distribution function. 88 Figure 5.5: Defined coordinate system for the orientation analysis of the free-OH and H-bonded -OH species. represents the tilt angle of the C 2 V axis of the corre- sponding water species with respect to the surface normal. Molecular hyperpolarizabity ( (2) i 0 j 0 k 0 ) is a frequency dependent quantity and can be writ- ten as a sum of resonant and nonresonant quantities. The resonant part contains the vibrational information of the molecules at the interface and the nonresonant part gener- ally provides a broad background over the entire spectral region. Molecular symmetry group of the vibrational mode determines the nonzero tensor elements of the molecular hyperpolarizability tensor (2) i 0 j 0 k 0 . Water bending is a totally symmetric mode with C 2V symmetry and there are three nonzero tensor elements which contributes to the symmet- ric bending mode; aac , bbc and ccc . Rao et. al. 30 showed that (2) eff can be simplified in the form of the following expression (2) eff =N S d(hcosichcos 3 i) (5.2) Where, is the tilt angle of the C 2V axis of the water molecule with respect to the surface normal. c and d parameters for the three polarization combinations (SSP, SPS, PPP) can be calculated after transforming the molecular coordinate fixed hyperpolarizability tensors into the laboratory fixed axis. A plot of (2) eff vs tilt angle () can be used to determine the average tilt angles of different hydrogen bonded species provided the SFG 89 amplitudes of the respective species in different polarizations are determined from the experimentally measured spectra. Figure 5.6 shows the variation of (2) eff as a function for SSP, PPP and SPS polarizations and the variation of the tilt angles for the free-OH and hydrogen bonded species as a function of bulk surfactant concentration. Since there is only one bend mode per water molecule which is weakly affected by intermolecular interactions, the direction of the C 2v axis for different hydrogen bonded species will provide a localized information of hydrogen bonded structure at the interface. Charged hydrophilic headgroups of the surfactant molecules are expected to be sub- merged into the water phase with their long chain hydrophobic tail groups hanging in the air phase. The head groups will be solvated at the subsurface layer such that there will be some water molecules above the headgroup. These water molecules above the head group at the top surface layer and the water molecules below the subsurface layer (layers within 3-5 ˚ A of the interface) reorient differently when the surface charge density changes mainly due to charge dipole interaction with the surfactant head groups. This has been depicted pictorially in Figure 5.6. C 2V axis for the water molecules above the head group reorients away (Figure 5.6c) from the positive charge which leads to a de- crease in the tilt angle of the C 2V axis with respect to the surface normal with increasing positive charge. However, water molecules in the sub-surface layer turn away (Figure 5.6d) from the interface toward the bulk water as the interface charge density increases. Change in tilt angles for the up/down oriented water molecules with CTAB concentra- tion are presented in Table 5.2. Similar studies in SDS/water interface does not provide any change in spectral lineshape in PPP polarization spectra. This can be rationalized by considering the interaction of the negative headgroup and the water molecules in the surface layer (above head group) and subsurface layer (below headgroup). Due to repul- sion between the oxygen of the water and the negative head group of the surfactant the 90 (a) (b) (c) (d) Figure 5.6: Calculated SFG amplitude (2 eff as a function of the tilt angle of the wa- ter C 2V axis with respect to the surface normal in SSP, PPP and SPS polarization combinations for SFG-Visible-infrared laser beams. Change in tilt angle for the free-OH species (a) and Hydrogen bonded species (b) as a function of the surface charge density and the corresponding molecular picture at the interface ((c) and (d) respectively). Vertical lines are color coded according to bulk CTAB concen- tration, purple (0.01 mMol), blue (0.03 mMol) and red (0.07 mMol) respectively. Reproduced from 25 . C 2V axis of both type of molecules (free-OH/ H-bonded water) are oriented towards the negative charge and the orientation of these water species does not change much with the change of surface charge density. 91 Table 5.2: Change in tilt angle of the C 2V axis for the free-OH (up oriented) and H-bonded (down oriented) water molecules as a function of CTAB concentration. Concentration (mMol) Free -OH (up oriented) H-bonded (down oriented) 0.01 52 131 0.03 47 134 0.07 38 155 A comparative pictorial description of water orientation at SDS solution interface and in CTAB solution interface is presented in Figure 5.7. The lineshape of the water bend spectra at SDS solution interface does not change in any of the polarization com- bination spectra. The average tilt angle of the free-OH or 1630 cm 1 frequency peak is 85 and 100 for the hydrogen bonded molecules or the 1660 cm 1 peak. (a) (b) Figure 5.7: Comparative picture of molecular orientation at the SDS/water (a) and CTAB /water interface (b). Reproduced from 25 . 5.4 Conclusion In conclusion, we have presented the water bend SFG spectra of a positive surfactant (CTAB) solution in water in all three polarization combinations. Spectral lineshapes in 92 different polarizations are different and they change with the change in surface charge density of the solution in PPP polarization. Quantitative spectral lineshape analysis and orientational analysis of the water bend spectra in CTAB solution interface offer a com- prehensive molecular picture of orientation for different hydrogen bond classes of water molecules. We have also estimated the positive charge required to completely alter or re- orient the surface structure of water at these interfaces. We expect that these results will encourage new experimental and theoretical studies to gain mechanistic details of the role of differently oriented hydrogen bond structure at the charged interfaces, especially at aqueous electrochemical interfaces. 5.5 Appendix 5.5.1 Experimental details and spectral fitting A detailed description of our VSFG setup has been described in previos chapter. The laser powers at the sample surface were2 and10 microjoules (J) per pulse for the IR and visible pulses, respectively. The SFG spectra are corrected for the scattering of the 800 nm light (recorded by blocking IR light only and subtracted from the raw SFG data). We have used 15 M water from our Millipore system and distilled it through a sealed distillation apparatus cleaned with piranha solution to prepare ion-free and or- ganic contamination free water. All solutions were freshly made from that pure water before every measurement. Spectral fitting of the experimenal SFG spectra were performed using resonant Lorentzians as explained in Chapter 2. All fitting parameters are presented in Table 5.3. 93 Figure 5.8: SFG spectra of water bending mode at air/water interface, nega- tively charged SDS (1mMol) /water interface and positively charged CTAB (0.4 mMol)/water interface. 5.5.2 Comparison of SFG signals from pure water, 0.4 mM CTAB and 1 mM SDS solution Figure 5.8 compares the SFG spectra of water bend from a pure air/water interface, a negatively charged interface and a positively charged interface. Spectral line shape of the water bend spectra for air/water interface and negative charged interface is similar for both SSP and PPP spectra, but the SFG spectra from positive interface in PPP polar- ization is different. Despite various studies the surface charge of pure air/water interface is still an ambiguous topic. 31, 32 Even though the determination of surface charge at the water interface is beyond the scope of this article, the similarity in line shapes between air/water interface and negative charge interface indicates structural similarity between those two interfaces 94 Table 5.3: Fitting parameters of the vibrational SFG spectra of the water bend measured for SSP, SPS and PPP polar- izations at the and CTAB/water interface in 0.01, 0.03 and 0.07 mMol concentrations. 0.01 mMol 0.03 mMol 0.07 mMol SSP SPS PPP SSP SPS PPP SSP SPS PPP ! 1 , cm 1 1630.5 1630.5 1630.5 1630.5 1630.5 1630.5 1630.5 1630.5 1630.5 ! 2 , cm 1 1662.5 1662.5 1662.5 1662.5 1662.5 1662.5 1662.5 1662.5 1662.5 B 1 -2.73 0.8 -0.67 0.4 1.80 0.5 -5.35 1.0 -0.230.20 2.760.5 -7.531.5 2.050.5 2.41.5 B 2 6.29 0.8 0.99 0.4 -3.13 0.5 12.73 1.0 0.390.20 -5.911.0 14.631.0 -10.451.0 -3.431.0 1 54 15 40 20 70 20 64 15 3825 5215 8920 4925 7730 2 56 15 40 20 74 20 66 15 3825 7315 5920 13330 4120 A NR 0.189 0.168 0.122 0.257 0.216 0.185 0.373 0.373 0.174 , rad -0.788 -1.86 -1.17 -0.767 -1.20 -1.20 -0.672 -1.675 -0.563 95 Chapter 5 References [1] Yeganeh, MS; Dougal, SM; Pink, HS Vibrational spectroscopy of water at liq- uid/solid interfaces: Crossing the isoelectric point of a solid surface Phys. Rev. Lett. 1999, 83, 1179. [2] Singh, Prashant Chandra; Nihonyanagi, Satoshi; Yamaguchi, Shoichi; Tahara, Tahei Interfacial water in the vicinity of a positively charged interface studied by steady-state and time-resolved heterodyne-detected vibrational sum frequency generation J. Chem. Phys. 2014, 141, 18C527. [3] Singh, Prashant Chandra; Nihonyanagi, Satoshi; ; Yamaguchi, Shoichi; ; Tahara, Tahei; ł Ultrafast vibrational dynamics of water at a charged interface revealed by two-dimensional heterodyne-detected vibrational sum frequency generation J. Chem. Phys. 2012, 137, 094706. [4] Nihonyanagi, Satoshi; Yamaguchi, Shoichi; Tahara, Tahei Direct evidence for orientational flip-flop of water molecules at charged interfaces: A heterodyne- detected vibrational sum frequency generation study J. Chem. Phys. 2009, 130, 204704. [5] Toney, Michael F; Howard, Jason N; Richer, Jocelyn; Borges, Gary L; Gordon, Joseph G; Melroy, Owen R; Wiesler, David G; Yee, Dennis; Sorensen, Larry B V oltage-dependent ordering of water molecules at an electrodeelectrolyte interface Nature 1994, 368, 444–446. [6] Ong, Shaowei; Zhao, Xiaolin; Eisenthal, Kenneth B Polarization of water molecules at a charged interface: second harmonic studies of the silica/water in- terface Chem. Phys. Lett. 1992, 191, 327–335. [7] Gragson, DE; McCarty, BM; Richmond, GL Ordering of interfacial water molecules at the charged air/water interface observed by vibrational sum frequency generation J. Am. Chem. Soc. 1997, 119, 6144–6152. [8] Stiopkin, Igor V .; Weeraman, Champika; Pieniazek, Piotr A.; Shalhout, Fadel Y .; Skinner, James L.; Benderskii, Alexander V . Hydrogen bonding at the water sur- face revealed by isotopic dilution spectroscopy Nature 2011, 474, 192–195. [9] Jungwirth, Pavel Physical chemistry: Water’s wafer-thin surface Nature 2011, 474, 168–169. 96 [10] Pieniazek, P. A.; Tainter, C. J.; Skinner, J. L. Interpretation of the water surface vibrational sum-frequency spectrum J. Chem. Phys. 2011, 135, 044701. [11] Pezzotti, Simone; Galimberti, Daria Ruth; Gaigeot, Marie-Pierre 2d h-bond net- work as the topmost skin to the air–water interface J. Phys. Chem. Lett. 2017, 8, 3133–3141. [12] Auer, BM; Skinner, JL Water: Hydrogen bonding and vibrational spectroscopy, in the bulk liquid and at the liquid/vapor interface Chem. Phys. Lett. 2009, 470, 13–20. [13] Auer, BM; Skinner, JL Vibrational sum-frequency spectroscopy of the liquid/vapor interface for dilute hod in d2o J. Chem. Phys. 2008, 129, 214705. [14] Ni, Yicun; Skinner, JL Ir and sfg vibrational spectroscopy of the water bend in the bulk liquid and at the liquid-vapor interface, respectively J. Chem. Phys. 2015, 143, 014502. [15] Fecko, C. J.; Eaves, J. D.; Loparo, J. J.; Tokmakoff, A.; Geissler, P. L. Ultrafast hydrogen-bond dynamics in the infrared spectroscopy of water Science 2003, 301, 1698–1702. [16] Bakker, HJ; Skinner, JL Vibrational spectroscopy as a probe of structure and dy- namics in liquid water Chem. Rev. 2009, 110, 1498–1517. [17] Asbury, J. B.; Steinel, T.; Stromberg, C.; Corcelli, S. A.; Lawrence, C. P.; Skinner, J. L.; Fayer, M. D. Water dynamics: Vibrational echo correlation spectroscopy and comparison to molecular dynamics simulations J. Phys. Chem. A 2004, 108, 1107–1119. [18] Du, Q.; Superfine, R.; Freysz, E.; Shen, Y . R. Vibrational spectroscopy of water at the vapor water interface Phys. Rev. Lett. 1993, 70, 2313–2316. [19] Raymond, E. A.; Tarbuck, T. L.; Brown, M. G.; Richmond, G. L. Hydrogen- bonding interactions at the vapor/water interface investigated by vibrational sum- frequency spectroscopy of hod/h2o/d2o mixtures and molecular dynamics simula- tions J Phys Chem B 2003, 107, 546–556. [20] Morita, Akihiro; Hynes, James T. A theoretical analysis of the sum frequency generation spectrum of the water surface Chem. Phys. 2000, 258, 371–390. [21] Vinaykin, Mikhail; Benderskii, Alexander V . Vibrational sum-frequency spectrum of the water bend at the air/water interface J. Phys. Chem. Lett. 2012, 3, 3348– 3352. 97 [22] Dutta, Chayan; Benderskii, Alexander V On the assignment of the vibrational spectrum of the water bend at the air/water interface J. Phys. Chem. Lett. 2017, 8, 801–804. [23] Dutta, C; Moberg, D. R.; Mammetkuliyev, M.; Straight, S.C.; Paesani, F.; Bender- skii, A.V . A comprehensive molecular picture of the air/water interface Submitted. [24] Nagata, Yuki; Hsieh, Cho-Shuen; Hasegawa, Taisuke; V oll, Judith; Backus, Ellen H. G.; Bonn, Mischa Water bending mode at the watervapor interface probed by sum-frequency generation spectroscopy: A combined molecular dynamics simu- lation and experimental study J. Phys. Chem. Lett. 2013, 4, 1872–1877. [25] Dutta, C; Benderskii, A.V . Refined picture of molecular orientation at charged aqueous interfaces. Submitted. [26] Kolarov, T; Yankov, R; Esipova, NE; Exerowa, D; Zorin, ZM Charge reversal at the air/water interface as inferred from the thickness of foam films Colloid & Polymer Science 1993, 271, 519–520. [27] H¨ anni-Ciunel, Katarzyna; Schelero, Natascha; vonKlitzing, Regine Negative charges at the air/water interface and their consequences for aqueous wetting films containing surfactants Faraday Diss. 2009, 141, 41–53. [28] Kristen, Nora; Simulescu, Vasile; Vullings, Andrea; Laschewsky, Andr´ e; Miller, Reinhard; vonKlitzing, Regine No charge reversal at foam film surfaces after ad- dition of oppositely charged polyelectrolytes? J. Phys. Chem. B 2009, 113, 7986– 7990. [29] Hirose, Chiaki; Akamatsu, Naotoshi; Domen, Kazunari Formulas for the analysis of the surface sfg spectrum and transformation coefficients of cartesian sfg tensor components Appl. Spectrosc. 1992, 46, 1051–1072. [30] Rao, Yi; Tao, Yi-song; Wang, Hong-fei Quantitative analysis of orientational order in the molecular monolayer by surface second harmonic generation J. Chem. Phys. 2003, 119, 5226–5236. [31] Gray-Weale, Angus; Beattie, James K An explanation for the charge on waters surface Phys. Chem. Chem. Phys. 2009, 11, 10994–11005. [32] Saykally, Richard J Air/water interface: Two sides of the acid-base story Nature Chem. 2013, 5, 82–84. 98 Chapter 6: Surface Propensity of Cations and Anions In this chapter, surface propensity of several ions have been investigated by surface sensitive VSFG spectroscopy. The first part of this chapter presents the VSFG spectra of acid and base solutions in the water bending region and a plausible model for H- bond structure at the aqueous halide acid solution has been proposed. In the second part VSFG spectra of alkali halide salt solutions have been presented. Vadim Trepalin, a summer undergraduate researcher, did the FTIR measurements of salt solutions and he also helped in the SFG experiments of the salt solutions. 6.1 Solvated H + and OH at the air/water Interface 6.1.1 Introduction In presence of excess protons or hydroxides water stretch region is markedly different from the pure air/water interface. 1 In general, the broad hydrogen bonded region (3000- 3600 cm 1 ) increases in intensity as the proton concentration increases and it decreases as the hydroxide concentration decreases, while the free-OH peak (3700 cm 1 ) remain almost unchanged. Solvated protons are surface active species and hydroxides are re- pelled from the surface as suggested by surface tension measurements. The solvation 99 properties of protons and hydroxides are different due to the difference in their sizes and hydrogen bonding envirnments. 6.1.2 Low Concentration Spectra Unlike the water stretch mode, water bend SFG spectra at the interface in presence of low concentrations of H + and OH does not show any appreciable changes from that of pure water. Figure 6.1 shows the bend spectra at HCl solution (pH =1.7) and NaOH (pH = 13.0) solution interface in SSP and PPP polarization combinations. Concentration of Figure 6.1: SFG spectra of HCl solution (pH 1.7) and NaOH solution (pH 13.0) at the water bending region in SSP and PPP polarization combinations for the SFG, visible and the IR beams respectively. Spectra were offset for clarity. solvated ionic species at the interface might be very low at low bulk concentrations, due to which the electric field at the interface is low and the elctrostatic interaction between the water molecules and the protons are weak. This could be the reason that the bend 100 SFG spectra are invariant to the presence of solvated protons or hydroxide ions at the surface or subsurface layers at low concentrations. 6.1.3 High Concentration Spectra SFG spectra of concentrated solutions (2M) of halogen acids (HCl, HBr and HI) and base (NaOH) were measured in the water bending region and presented in figure 6.2a and 6.2b. Spectral lineshape in acid and base solution are similar to those of pure water with a low ferquency negative amplitude peak and a high frequency positive amplitude peak albeit with different relative amplitudes in both SSP and PPP polarization com- binations. This indiactes that the average orientation of up and down oriented water moelcules change in presence of hydrated protons and hyroxide ions. SSP spectra of the water bend in 2M solutions of acid and base (Figure 6.2a) suggests an increase in amplitude for the hydrogen bonded or down oriented water molecules in acid solutions and a decrease in amplitude for the same peak in base solution. This is similar to the changes in water stretch spectra in acid and base solution, where the free-OH peak re- mains almost unaffected but the hydrogen bonded peaks increase and decrease in acid and base solutions, respectively, in SSP polarization. The water bend PPP spectra does not differ considerably from that of pure water. Orientation analysis on the up/down oriented water molecules was performed based on the spectral fitting of these spectra as explained in the next section. Spectra of the sodium chloride (NaCl) salt solution was also measured and com- pared with the pure water bend spectra. Concentrated NaCl solutions does not show any substantial change in the water stretching region which is indicative of the fact that neither the Na + nor Cl ions are surface active and presence of those ions do not alter the surface hydrogen bond structures. Change in SFG spectra across the entire water 101 (a) SSP (b) PPP (c) Difference spectra in SSP (d) Difference spectra in PPP Figure 6.2: SFG spectra of 2M solutions of HI, HBr, HCl, NaOH and NaCl at the water bending region in SSP (top left panel) and PPP (top right panel) polar- ization combinations for the SFG, visible and the IR beams respectively. Spectra were offset for clarity. Difference SFG intensity for 2M solutions of HI, HBr, HCl, NaOH and NaCl solutions in SSP (bottom left panel) and PPP (bottom right panel) polarizations. bend region in acid and base solutions can be better recognized in the difference SFG spectra, where, the pure water bend spectra has been subtracted from the acid and base solution spectra and plotted in Figure 6.2c and 6.2d. NaCl solution difference SFG spec- trum has been used as a reference in both SSP and PPP polarizations. As NaCl does not change the water bend spectrum, the difference between the 2M NaCl and pure water 102 SFG spectra are nearly a straight line in both polarization combinations. Difference SFG spectra of acid and base solutions in SSP polarization (Figure 6.2c) clearly shows an in- crease/decrease in intensity in the H-bonded water region for acids/base respectively. However, the PPP difference spectra are almost flat in all solutions indicating a clear change in orientation for different H-bonded species. Probable reason for the changes in SFG signal in acid and base solution It is of no doubt that the SFG signals at the acid and base interfaces change considerably from that of pure water spectra. For acid solutions the increase in intensity for differenly hydrogen bonded water species can be due to the following reasons. 1. As solvated protons are surface active species, there exists a higher concentration of protons at the interface as Zundel (H 5 O + 2 ) or Eigen (H 9 O + 4 ) cations. 2–6 The terminal water molecules of these species could be SFG active and contribute to the overall SFG spectra in the water bend region. 2. Due to the presence of charged ions at the interface, the electric field at the inter- face is enhanced. This enhanced electric field at the interface contributes to the overall SFG signal through a third order (3) term which is proportional to the surface potential () at the interface. 7–10 3. Strong electrostatic interaction between water and charged species can alter the surface water structure and the effect of this interaction can extend beyond the interface layer into the subsurface region. This hydrogen bond rearrangement or increased ordering at the interface will also contribute to the overall SFG spectra which produce a higher signal than that of pure water interface. 103 Theoretical calculations and experimental studies 2–5, 11 on isolated protonated species such as Zundel, Eigen cations or protonated water trimers suggest that the bending mode vibrational frequency of these species containting the central oxygen of the hydronium ion or the terminal water molecules are 100-150 wavenumber (cm 1 ) blue shifted from the bending frequency of the single water molecule. Also, the 2 frequency of wa- ter is blue shifted in liquid ( 1650 cm 1 ) from the gas phase value of 1596 cm 1 . Thus it can be expected that the bending frequency of the protonated water species would be at least 100-150 cm 1 blue shifted from that of gas phase values. This will place the bend- ing vibrational frequency of the terminal water bend modes of the protonated species near or above 1800 cm 1 , which is beyond the spectral coverage of the pure water bend spectra. Hence, we can rule out the possible contribution of the protonated species to- wards the overall SFG spectra in the water bending region, however, spectral fitting of all the acid solution spectra suggests small frequency changes for the free-OH ( 1635 cm 1 ) and the H-bonded ( 1680 cm 1 ) water species. Hence, the increase in SFG intensity at the acid solution interface in the water bending region can be attributed to an enhanced electric field at the charged interface and reorientation of water molecules at the surface or subsurface layer due to strong electrostatic interactions. Hydroxide ions (OH ) are not surface active as suggested by the surface tension studies. 12 But from thermodynamic point of view, presence of OH ions in the bulk water disrupts the extented H-bond network of bulk water which forces the OH ions to move towards the surface layer. Solvation shell of water molecules around a OH ion is very much different than that around a H + ion. Hence, it is plausible that the OH ions are present at the subsurface layer and randomization of H-bond network at the interface layer results in a decrease in SFG intersity at the interface. 104 6.1.4 Orientational Changes in Acid and Base Solutions Spectral fitting was performed to extract the amplitudes of the two main resonant peaks in the water bend region. The overall SFG spectra consists of spectral contributions from all different H-bonded water species present at the interface. However, as we have shown before in Chapter 3 and Chapter 4, the water bend spectra can be decomposed into two main resonant peaks. The low frequency negative amplitude peak belongs to water molecules which are oriented with their hydrogens upward towards the air phase and primarily consists of water molecules having free-OH bonds. On the other hand, the high frequency positive amplitude peak is due to water molecules having a higher degree of H-bonding and the average orientation of the C 2V axis is towards the bulk phase for these water molecules. Figure 6.3 shows the amplitudes of the free-OH (B 1 / 1 ) and the H-bonded (B 2 / 2 ) water bend peak in 2M solutions of HI, HBr, HCl and NaOH in SSP and PPP polariza- Figure 6.3: SFG amplitudes of the free-OH (B 1 / 1 ) and the H-bonded (B 2 / 2 ) water bend peak in 2M solutions of HI, HBr, HCl and NaOH. 105 tion combinations. In SSP, both amplitudes are higher for all acids and lower for the NaOH solution than that of pure water. PPP amplitudes are similar for all solutions and no definite trend is observed. These paramaters can be used to determine the average orientation of these up/down oriented H-bonded water species following the orienta- tional analysis as explained in Chapter 2.2. Fitting parameters are presented in table 6.3 in the appendix section. Ratios of amplitudes of the up/down oriented water molecules in SSP and PPP polar- izations were used to determine the average orientation and are presented in Table 6.1. At the pure air/water interface the average orientation of the two main water species are 75 and110 respectively. 13, 14 In presence of hydrated protons the average tilt angle for the free-OH or the up oriented water molecules change to around 55 and for the H-bonded or the down oriented water molecules it increases to around 127 . For the NaOH solution the average tilt angles are around 63 and 110 for the same two species respectively, however, the error in determining the tilt angles might be higher due to lower signal to noise ratio. Table 6.1: Average Orientation of the C 2V axis of the free -OH and the H- bonded water species at the acid and base solution interface Free - OH H - Bonded Water 75 5 110 5 2M HI 55 5 126 5 2M HBr 56 5 129 5 2M HCl 56 5 127 5 2M NaOH 63 5 100 5 Ishiyama and Morita 15 performed Molecular Dynamics (MD) simulation on aque- ous hydrogen halide solution interface and calculated the orientations of the permanent dipole vector of water with respect to the surface normal. These calculations showed 106 (Figure 2 in Ref 15 ) that the molecules in the subsurface layer or below the Gibbs divid- ing surface for the hydrogen halide solutions reorients strongly towards the bulk com- pared to the pure water interface. Free -OH molecules at the dividing surface are also oriented more towards the bulk phase which was explained by the presence of stronger electric field interactions at the double layer. Our orientational analysis based on the ex- perimental water bend spectra suggests similar orientational changes for the H-bonded water species but differs for the free -OH molecules which clearly shows a decreased tilt angle of the C 2V axis of those water molecules. Figure 6.4: Illustrations picture of the plausible Hydrogen bond structure at the aqueous halide solution interface. Based on our experiemntal results and existing knowledge of the surface water struc- ture, we propose a plausible surface water H-bond structure at the aqueous halide so- lution interface which is presented in Figure 6.4. Due to the amphiphilic nature of the hydronium ion, the hydrogens of the hydronium ions are directed towards the liquid phase 15–17 which in turn affects the subsurface layer water molecules to reorient them- selves towards the bulk. However, water molecules at the interface layer reorient away from the bulk waters. The halide counter ions (X , X= I, Br, Cl) can also affect the 107 surface structures at these interfaces through changes of thickness of the double layer and strength of the surface potential. This effect of the halide counter ions can be seen from the differences in the H-bonded water bend region of the SFG spectra (Figure 6.2). 6.2 SFG spectra of salt solutions 6.2.1 Introduction Surface enrichment of inorganic ions has been proven to have important implications in atmospheric chemistry. Ocean surfaces and seawater aerosols, 18–20 effect of sea spray on the arctic snow, 21 bubble coalescence, 22–24 effect of electrolytes on corrosion processes at various metal electrode interfaces 25 are few examples where presence of electrolytes at the interface play a significant role. In general, ions or electrolytes when dissolved in water get strongly solvated and the extent of solvation depends on the charge, size and nature of the ions. 16, 26, 27 Solvation around an ion changes the water H-bond struc- ture due to charge dipole interactions. When the ions (specifically anions) appear at the interface the extended H-bond structure at the interface is modified which determines many critical properties of the salt solution interface. Many experimental and theoreti- cal attempts have been made to understand these electrolyte effects on the surface water structure. However, the more we study these systems the more complicated it becomes because of the sheer size of the associated problems and lack of a unified approach. It is nearly impossible to apply a particular type of experimental or theoretical method to provide a universal picture as each system and associated problems are unique and require unique treatment. Here we aim to understand the interfacial structure and orien- tation of H-bond at alkali halide salt solution interfaces by using VSFG spectroscopy in the water bending region. 108 Surface tension measurements of inorganic aqueous salt solution interface show an increase in surface tension with increasing bulk salt concentration. This is the basis of the traditional view of these surfaces as being devoid of ions as according to equation 6.1 = 1 RT lna T (6.1) where is defined as the excess number of moles of a paricular ion per unit surface area of the Gibbs dividing surface. ‘ ’ is the surface tension and ‘a’ is the bulk con- centration for the solute. If lna is positive, the surface excess () is negative. Onsager and Samaras 28 semi-quantitatively rationalized this for the first time using Gibbs adsorp- tion equation. According to their model, ions at the water surface experience a strong repulsion due to electrostatic image forces. For many decades this has been a widely accepted notion for an ion-free water interface until recently a series of surface sensi- tive experimental studies and theoretical simulations suggest that ions can be surface active depending on their charge, size and polarizability. Theoretical simulations sug- gests the presence of soft halide ions at the aqueous interface and the surface affinity of soft polarizable ions increase with increasing size of the ions. Following the remark- able theoretical developements by Jungwirth and co-workers, few other theoretical and experimental groups explored the aqueous salt solution interfaces with surface specific techniques. These studies resulted in similar conclusions which are as follows. 1. Soft polarizable anions have higher affinity to be present at the interface. 2. An increase in surface tension with increasing salt concentration does not neces- sarily means a decrese in surface concentration of inorganic ions. This has been nicely explained by Tobias and Jungwirth. 29, 30 109 3. In general, for soft anions (such as I ) there is surface excess near the Gibbs dividing surface and there is a decrease in ion concentration in the subsurface layer. Non-polarizable anions (such as Cl ) do not show up at the interface. 4. Cations (such as alkali cations) are in general small sized hard spheres and non- polarizable. They do not prefer to stay at the interface layer but they accumulate in the subsurface layers to form a double layer of ions at the inorganic aqueous salt solutions. Here we present the VSFG spectra of halide salt solutions in SSP and PPP polarizations. A quantitative comparison of the SFG spectra from different alkali halide solution in the water bending region is presented and also the effect of cation size on the interfacial water is discussed. 6.2.2 Effect of anions in the VSFG spectra of sodium salts VSFG spectra of concentrated sodium salt solutions have been studied in the water stretching region and suggests that NaI and NaBr affect the interfacial water structure while NaCl and NaF do not have any substantial effect. 31, 32 Liu et. al 26 showed an in- crease in interfacial depth for bromide and iodide soultion interface showing more dis- turbed water structure near the interface. For NaI ( 2.1M) there is a significant increase in intensity of SFG signal around 3450 cm 1 which corresponds to H-bonded water region. The free-OH peak is also enhanced in presence of NaI but the extent of that increase is much smaller. NaBr shows similar effects 26, 31 but they are less pronounced. Water bending spectra at 2M solutions of NaI, NaBr, NaCl and NaOH were mea- sured in SSP and PPP polarization and presented in figure 6.5a and 6.5b respectively. Spectral lineshape of the bend spectra for all the solutions are similar to those of water spectra but the amplitude and frequency of the two main resonant peak change. NaCl 110 (a) SSP (b) PPP Figure 6.5: SFG spectra of 2M solutions of sodium salts (NaI, NaBr, NaCl, NaOH) at the water bending region in SSP (left panel) and PPP (right panel) polarization combinations for the SFG, visible and the IR beams respectively. Spectra were offset for clarity. and NaBr spectra are similar to water spectra in both SSP and PPP polarizations but for the NaI there is a huge increase at1680 cm 1 which is the H-bonded region. The free-OH region in the NaI solution also shows increase in amplitude from that of the water spectra and the frequency is red-shifted. Spectra for NaOH have been plotted in the same graph for comparison. Qualitative comparison of these spectra reveal that the average orientation and structure at these solution interfaces were similar to that of wa- ter interface except for the NaI solution. Theoretical calculations 16, 29, 30 show that the large anions (such as I ) have higher affinity to show up at the interface. The increase of amplitude in the NaI solution as compared to that of the other solutions is an effect of the iodide ions residing at the interface. Due to the presence of these I ions there exists an interfacial electric field and the interfacial depth is increased and the number of water molecule contributing to the SFG spectra increases. However, all the other anions are less surface active and the interfacial structure is similar to that of water. 111 6.2.3 Effect of cations in the VSFG spectra Different ions have difefrent hydration structures that depend on the charge, size and the nature of the ions. In case of anions, due to their larger size and higher polarizability (a) SSP (b) PPP Figure 6.6: SFG spectra of 2M solutions iodides (HI, LiI, NaI and KI) at the water bending region in SSP (left panel) and PPP (right panel) polarization combinations for the SFG, visible and the IR beams respectively. Spectra are offset for clarity. they appear at the surface and the counter ions reside at the subsurface layer. In general, the cations are hard spheres and not easily polarizable. Due to this, cations are highly hydrated and the smaller the size of the cation the larger the hydrated radius of the cation. Presence of these different sized cations in the subsurface layer results in a change of surface potantial which is proportional to the size of the surface layer. Comparison of the iodide salts Figure 6.6a and figure 6.6b present the VSFG spectra of the 2M solutions of LiI, NaI, KI and HI in SSP and PPP respectively, which are considerably different than that of water. Iodide being the soft polarizable ions will be at the interface for all the solutions, so any change in the SFG spectra will be due to the effect of the cations. Hydrated radius of the 112 cations 33, 34 are in order: Li + (3.8 ˚ A)> Na + (3.55 ˚ A)> K + (3.3 ˚ A). According to Stern model 35, 36 the surface ptential () is given by = 4d (6.2) Where, is the surface charge, d is the thickness of the stern layer and is the dielectric Figure 6.7: SFG amplitudes of the free-OH (B 1 / 1 ) and the H-bonded (B 2 / 2 ) water bend peak in 2M solutions of LiI, NaI, KI and HI. constant of the medium. Thickness of this Stern layer will depend directly on the radius of the anions and the cations at the surface. For the iodide salts this thickness will depend on the hydrated radius of the cations. The effect of this on the overall SFG spectra will be two-fold. Due to increased thickness there will be more number of water molecules that are probed in the SFG experiment. Also, as the surface potential increases the SFG signal might increase by an electric field induced third order process. Figure 6.7 shows the SFG amplitudes for the two main resonant peaks for all the solutions in SSP and PPP polarizations. Amplitudes for the LiI, NaI and KI solutions are decreasing in the 113 same order as their hydrated radius. Amplitude for the HI solution can not be compared as both H + and I are surface active and their combined effect on the SFG spectra could be very different than the other iodide salts. (a) SSP (b) PPP Figure 6.8: SFG spectra of 2M solutions bromides (HBr, LiBr, NaBr) at the water bending region in SSP (left panel) and PPP (right panel) polarization combinations for the SFG, visible and the IR beams respectively. Spectra were offset for clarity. (a) SSP (b) PPP Figure 6.9: SFG spectra of 2M solutions chlorides (HCl, LiCl, NaCl) at the water bending region in SSP (left panel) and PPP (right panel) polarization combinations for the SFG, visible and the IR beams respectively. Spectra were offset for clarity. 114 We have also measured the VSFG spectra of the bromide and chloride salt solution interfaces in the water bending region and presented in figure 6.8 and 6.9 respectively. In general, the lithium salts have higher intensity than the sodium salts. NaCl and NaBr spectra are similar to that of water in different polarizations (SSP and PPP) which is in accordance with previous studies. Lithium salts have higher intensities in both SSP and PPP polarizations which is probably due to higher thickness of the stern layer at the interface than the sodium salts. The corresponding halide acids (HBr and HCl) have different behaviours due to different surface activity of the respective species. Beacuse of this the orientation or H-bonded structure at the interface will be different. Orientation of the up/down oriented water molecules, corresponding to the1620 cm 1 peak and 1670 cm 1 peak respectively, have been calculated from the spectral fitting parameters and presented in table 6.2. Average orientations for these two types Table 6.2: Average Orientation of the C 2V axis of the free -OH and the H- bonded water species at the iodide salt solution interface Free - OH H - Bonded Water 75 5 110 5 2M LiI 59 5 122 5 2M NaI 66 5 122 5 2M KI 67 5 124 5 of water molecules (up/down oriented) are similar for all iodide salts where the average tilt angle of the C 2V axis of the up and down oriented water molecules are 65 and 122 respectively. Based on these results a model molecular picture at the iodide salt solution has been presented in figure 6.10. In general, the iodide anions stay at the top surface layer and reorient the water molecules at the top layer in such a way that the hydrogen atoms are oriented towards the iodide anion (type ‘A’ water molecule in figure 6.10) with an average orientation of 122 for the C 2V axis. Small sized cations will 115 Figure 6.10: Illustrations for a cartoon picture of the plausible Hydrogen bond structure at the salt solution interface. stay in the sub-surface layer with many water moelcules surrounding the cations. Water molecules which will solvate these cations from the top side (type ‘B’ in figure 6.10) can be oriented in such a way the C 2V axis will be oriented upwards with an average angle of 65 . In conclusion, our experimental results on the aqueous salt solution interfaces pro- vide molecular level understanding of orientation and H-bond structure. Previous stud- ies show that soft polarizable anions are most surface active and play a significant role at the aqueous interfaces. In contrast, the hard non-polarizable cations are not sur- face active and their role in shaping the molecular structure at the interfaces have been speculated but not been quatitatively studied. We have showed that these hard cations also contribute to the surface structure and orientation of water molecules through ex- tended H-bonding interactions. We hope that, this will instigate more theoretical and experimetal groups to inspect these effects in details. 116 6.3 Appendix 6.3.1 FTIR measurements Water bend frequencies in the salt solutions have been measured using FTIR spec- troscopy. The bending vibration in all the salts (NaCl, NaBr, NaI, KI and KCl) shifts to the red side of the spectrum with respect to the water frequency. The magnitude of this shift is small ( 2-8 cm 1 ) and it is in the following order: KI< NaI< NaBr KCl NaCl< H 2 O. Figure 6.11: FTIR spectra of 2M solution of salts. 117 Table 6.3: Fitting parameters of the vibrational SFG spectra of the water bend measured for SSP and PPP polarizations at the 2M acid and base solution interface. 2M HI 2M HBr 2M HCl 2M NaOH H 2 O SSP PPP SSP PPP SSP PPP SSP PPP SSP PPP ! 1 , cm 1 1635 1635 1635 1635 1635 1635 1630 1630 1630 1630 ! 2 , cm 1 1682 1682 1682 1683 1683 1680 1662 1662 1662 1662 ! 3 , cm 1 1512 1547 1440 1544 1433 1511 1479 1454 1457 1605 ! 4 , cm 1 1738 1788 1727 1865 1775 1791 1694 1786 1705 1757 B 1 -3.8 -2.5 -4.05 -1.19 -4.8 -1.63 -1.0 -1.72 -2 -2.9 B 2 11.2 2.25 10.9 1.9 7.7 1.8 2.8 1.8 1.5 1.86 B 3 4.9 4.4 8.8 8.07 26 17.9 4.4 6.3 4.7 19.6 B 4 10.9 2.4 7.6 9.14 6 5.8 1.4 0.97 5.21 3.13 1 49 70 50 35 58 54 35 80 50 50 2 102 62 75 93 67 53 98 42 50 50 3 78 78 144 129 246 163 66 94 80 168 4 98 75 145 139 100 97 51 40 80 78 A NR 0.14 0.2 0.17 0.19 0.18 0.18 0.16 0.12 0.16 0.07 -1.7 -2.47 -1.6 -2.85 -1.75 -2.36 -0.81 -1.2 -1.1 -1.6 118 Table 6.4: Fitting parameters of the vibrational SFG spectra of the water bend measured for SSP and PPP polarizations at the Iodide salt solution interface. 2M LiI 2M NaI 2M KI SSP PPP SSP PPP SSP PPP ! 1 , cm 1 1618 1621 1621 1621 1620 1620 ! 2 , cm 1 1673 1673 1678 1678 1694 1693 ! 3 , cm 1 1392 1445 1345 1498 1556 1375 ! 4 , cm 1 2241 1797 1808 1804 1804 B 1 -5.1 -3.1 -4.04 -3.6 -2.1 -2.07 B 2 9.3 7.3 8.35 6.5 7.9 4.1 B 3 11.5 12.5 26.4 2.3 30 22 B 4 6.8 -0.18 4.3 7.8 2 1 50 35 43 40 35 35 2 79 125 78 122 74 90 3 106 120 59 83 219 173 4 21 55 133 84 84 A NR 0.14 0.35 0.33 0.31 0.09 0.26 -1.7 -2.78 -2.01 -2.78 -2.5 -2.94 119 Chapter 6 References [1] Tarbuck, Teresa L; Ota, Stephanie T; Richmond, Geraldine L Spectroscopic studies of solvated hydrogen and hydroxide ions at aqueous surfaces J. Am. Chem. Soc. 2006, 128, 14519–14527. [2] Chaban, Galina M; Jung, Joon O; Gerber, R Benny Anharmonic vibrational spec- troscopy of hydrogen-bonded systems directly computed from ab initio potential surfaces:(h2o) n, n= 2, 3; cl-(h2o) n, n= 1, 2; h+ (h2o) n, n= 1, 2; h2o- ch3oh J. Phys. Chem. A 2000, 104, 2772–2779. [3] Falk, Michael; Gigu` ere, Paul A Infrared spectrum of the h3o+ ion in aqueous solutions Canadian J. Chem. 1957, 35, 1195–1204. [4] Asmis, Knut R; Pivonka, Nicholas L; Santambrogio, Gabriele; Br¨ ummer, Math- ias; Kaposta, Cristina; Neumark, Daniel M; W¨ oste, Ludger Gas-phase infrared spectrum of the protonated water dimer Science 2003, 299, 1375–1377. [5] Duong, Chinh H; Gorlova, Olga; Yang, Nan; Kelleher, Patrick J; Yu, Qi; McCoy, Anne B; Bowman, Joel M; Johnson, Mark A Disentangling the complex vibra- tional spectrum of the protonated water trimer, h+ (h2o) 3, with two-color ir-ir photodissociation of the bare ion and anharmonic vscf/vci theory J. Phys. Chem. Lett. 2017. [6] Levering, Lori M; Sierra-Hernandez, M Roxana; Allen, Heather C Observation of hydronium ions at the air- aqueous acid interface: Vibrational spectroscopic studies of aqueous hcl, hbr, and hi J. Phys. Chem. C 2007, 111, 8814–8826. [7] Zhao, Xiaolin; Subrahmanyan, Suchitra; Eisenthal, Kenneth B Determination of pka at the air/water interface by second harmonic generation Chem. Phys. Lett. 1990, 171, 558–562. [8] Zhao, Xiaolin; Ong, Shaowei; Eisenthal, Kenneth B Polarization of water molecules at a charged interface. second harmonic studies of charged monolay- ers at the air/water interface Chem. Phys. Lett. 1993, 202, 513–520. [9] Zhao, Xiaolin; Ong, Shaowei; Wang, Hongfei; Eisenthal, Kenneth B New method for determination of surface pka using second harmonic generation Chem. Phys. Lett. 1993, 214, 203–207. [10] Eisenthal, KB Liquid interfaces probed by second-harmonic and sum-frequency spectroscopy Chem. Rev. 1996, 96, 1343–1360. 120 [11] Kim, Jeongho; Schmitt, Udo W; Gruetzmacher, Julie A; V oth, Gregory A; Scherer, Norbert E The vibrational spectrum of the hydrated proton: Comparison of exper- iment, simulation, and normal mode analysis J. Chem. Phys. 2002, 116, 737–746. [12] Washburn, E. W.; West, C. J. International critical tables of numerical data, physics, chemistry and technology, V ol. 4; National Academies, 1928. [13] Dutta, Chayan; Benderskii, Alexander V On the assignment of the vibrational spectrum of the water bend at the air/water interface J. Phys. Chem. Lett. 2017, 8, 801–804. [14] Dutta, C; Moberg, D. R.; Mammetkuliyev, M.; Straight, S.C.; Paesani, F.; Bender- skii, A.V . A comprehensive molecular picture of the air/water interface Submitted. [15] Ishiyama, Tatsuya; Morita, Akihiro Molecular dynamics analysis of interfacial structures and sum frequency generation spectra of aqueous hydrogen halide solu- tions J. Phys. Chem. A 2007, 111, 9277–9285. [16] Mucha, Martin; Frigato, Tomaso; Levering, Lori M; Allen, Heather C; Tobias, Douglas J; Dang, Liem X; Jungwirth, Pavel Unified molecular picture of the sur- faces of aqueous acid, base, and salt solutions J. Phys. Chem. B 2005, 109, 7617– 7623. [17] Petersen, Matt K; Iyengar, Srinivasan S; Day, Tyler JF; V oth, Gregory A The hydrated proton at the water liquid/vapor interface J. Phys. Chem. B 2004, 108, 14804–14806. [18] Finlayson-Pitts, BJ The tropospheric chemistry of sea salt: A molecular-level view of the chemistry of nacl and nabr Chem. Rev. 2003, 103, 4801–4822. [19] Laskin, Alexander; Gaspar, Daniel J; Wang, Weihong; Hunt, Sherri W; Cowin, James P; Colson, Steven D; Finlayson-Pitts, Barbara J Reactions at interfaces as a source of sulfate formation in sea-salt particles Science 2003, 301, 340–344. [20] Knipping, EM; Lakin, MJ; Foster, KL; Jungwirth, P; Tobias, DJ; Gerber, RB; Dabdub, D; Finlayson-Pitts, BJ Experiments and simulations of ion-enhanced in- terfacial chemistry on aqueous nacl aerosols Science 2000, 288, 301–306. [21] Spicer, Chester W; Plastridge, Robert A; Foster, Krishna L; Finlayson-Pitts, Bar- bara J; Bottenheim, Jan W; Grannas, Amanda M; Shepson, Paul B Molecular halo- gens before and during ozone depletion events in the arctic at polar sunrise: con- centrations and sources Atmos. Environ. 2002, 36, 2721–2731. [22] Marˇ celja, Stjepan Short-range forces in surface and bubble interaction Curr. Opin. Colloid Interface Sci. 2004, 9, 165–167. 121 [23] Craig, Vincent SJ Bubble coalescence and specific-ion effects Curr. Opin. Colloid Interface Sci. 2004, 9, 178–184. [24] Henry, Christine L; Dalton, Casuarina N; Scruton, Lehoa; Craig, Vincent SJ Ion- specific coalescence of bubbles in mixed electrolyte solutions J. Phys. Chem. C 2007, 111, 1015–1023. [25] Tucceri, R A review about the surface resistance technique in electrochemistry Surf. Sci. Rep. 2004, 56, 85–157. [26] Liu, Dingfang; Ma, Gang; Levering, Lori M; Allen, Heather C Vibrational spec- troscopy of aqueous sodium halide solutions and air- liquid interfaces: Observation of increased interfacial depth J. Phys. Chem. B 2004, 108, 2252–2260. [27] Jungwirth, Pavel; Curtis, Joseph E; Tobias, Douglas J Polarizability and aqueous solvation of the sulfate dianion Chem. Phys. Lett. 2003, 367, 704–710. [28] Onsager, Lars; Samaras, Nicholas NT The surface tension of debye-h¨ uckel elec- trolytes J. Chem. Phys. 1934, 2, 528–536. [29] Vrbka, Lubos; Mucha, Martin; Minofar, Babak; Jungwirth, Pavel; Brown, Eric C; Tobias, Douglas J Propensity of soft ions for the air/water interface Curr. Opin. Colloid Interface Sci. 2004, 9, 67–73. [30] Jungwirth, Pavel; Tobias, Douglas J Specific ion effects at the air/water interface Chem. Rev. 2006, 106, 1259–1281. [31] Raymond, Elizabeth A; Richmond, Geraldine L Probing the molecular structure and bonding of the surface of aqueous salt solutions J. Phys. Chem. B 2004, 108, 5051–5059. [32] Gurau, Marc C; Lim, Soon-Mi; Castellana, Edward T; Albertorio, Fernando; Kataoka, Sho; Cremer, Paul S On the mechanism of the hofmeister effect J. Am. Chem. Soc. 2004, 126, 10522–10523. [33] Conway, Brian Evans Ionic hydration in chemistry and biophysics, V ol. 12; Else- vier Science Ltd, 1981. [34] Railsback, L Bruce Some fundamentals of mineralogy and geochemistry On-line book, quoted from: www. gly. uga. edu/railsback 2006. [35] Bard, Allen J; Faulkner, Larry R; Leddy, Johna; Zoski, Cynthia G Electrochemical methods: fundamentals and applications, V ol. 2; wiley New York, 1980. [36] Aveyard, Robert; Haydon, Denis Arthur An introduction to the principles of sur- face chemistry; CUP Archive, 1973. 122 Chapter 7: SFG Spectroscopy at Graphene Electrode Interface In this chapter, water H-bond structure, orientation of interfacial water and voltage de- pendence of the interfacial water spectra at graphene electrode interface were studied using VSFG spectroscopy. All samples were prepared and characterized by Dr. Stephen Cronin’s group at USC. Angelo Montenegro and Muhammet Mammetkuliyev helped in SFG experiments and they are continuing future experiments in this project. Muham- met played a significant role in designing the sample cell and he also wrote the voltage dependent data acquisition program. 7.1 Introduction Single layer graphene has many unique properties, which have drawn the attention of many researchers from diverse fields including electrochemistry and catalysis. 1–4 Graphene consists of a two-dimensional sheet of sp 2 -bonded carbon atoms arranged in a hexagonal honeycomb lattice with high mechanical strength and novel electronic properties. 5, 6 Neutral single layer graphene is a semimetal without a band gap between the valence and conduction bands. Due to its relatively small density of states, the elec- tronic properties of graphene are easily tunable by chemical or electrostatic doping. 7–10 Graphene based materials also have high electrical conductivity 11 and low chemical 123 reactivity under electrochemical conditions in the aqueous phase. Aqueous phase elec- trochemistry is largely affected by the interfacial water structure and orientation near the electrode surface. Hence, it is of the utmost importance to understand the hydrogen bond network and molecular interaction of water molecules at the electrode interfaces. Exper- imental and theoretical studies 12–17 suggest that water molecules are oriented differently at negatively and positively charged electrode interfaces. Near negatively charged elec- trodes, water molecules are generally oriented with their hydrogens pointing towards the electrode and, near positively charged electrodes, they are oriented with their hydrogens pointing away from the electrode. Graphene is hydrophobic in character, which can al- ter the dynamic hydrogen bond network of water at the interface. 18 The presence of a weakly interacting free or dangling -OH species at nanoconfined carbon nanotube and graphene interfaces has been predicted by several theoretical simulations, 19–22 however, only a handful of experimental attempts have been made to spectroscopically probe the graphene electrode surface to understand the intricate nature of molecular interactions at this interface. Politano 23 et al. presented evidence for the existence of non-H (non- D)-bonded water molecules at the water/graphene interface using high-resolution elec- tron energy loss spectroscopy (HREELS). However, these experiments were performed under ultra-high vacuum and low temperature conditions, which might not be directly applicable to electrochemistry in aqueous solutions. Here, we seek to understand the intricate structural details of water at graphene electrode interfaces (Figure:7.1) under electrochemical conditions by using a surface sensitive vibrational spectroscopy tech- nique. Vibrational spectroscopy in the water stretching region can be used as a probe to elu- cidate the structural differences at various interfaces. Vibrational sum frequency gener- ation (VSFG) is a second order non-linear optical process in which a narrow bandwidth 124 Figure 7.1: Schematic of Single layer graphene/water interface. visible beam and a broadband infrared beam are overlapped in space and time at the interface. Due to the inherent non-centrosymmetry, molecules at the interface interact with the two incoming fields and generate a new field that oscillates at the sum of the incident frequencies. SFG is generated due to the electric field-induced second order po- larization (P (2) ), which is proportional to the nonlinear susceptibility ( (2) ). At charged interfaces, a third order polarization component (P (3) ) contributes to the overall SFG signal and is dependent on the surface potential induced by the surface charges and the ionic strength of the bulk solution. Due to the ion-induced surface potential, the De- bye screening length changes at the surface, which in turn changes the probing depth of the SFG experiments. We have used surface sensitive VSFG spectroscopy in the - OD stretching region to study the interfacial water at solid/ liquid interfaces (i.e., CaF 2 interface, Al 2 O 3 interface and graphene electrode interface). The key questions we are trying to answer here are: 1. Is the graphene interface hydrophobic in nature and if yes, what are the spectro- scopic signatures of it? 125 2. What are the nature of interactions between graphene and water molecules in close contact with the graphene surface and how the water molecules are oriented at this interface under electrochemical conditions? 3. What is the effect of surface charge (charged graphene) on the molecular orienta- tion of interfacial water molecules? 4. How does the presence of electrolytes affect the surface water structure at the graphene interface? 7.2 Result and Discussions VSFG spectra at the air/water interface 24, 25 has two main features, a sharp peak on the blue side of the spectrum (3700 cm 1 for H 2 O and2740 cm 1 for D 2 O 26 ) and a broad peak on the red side of the spectrum (3200-3500 cm 1 for H 2 O and2300-2600 cm 1 for D 2 O). The sharp peak is assigned to the water molecules having a free-OH (- OD) or dangling OH (-OD) bond at the interface, and the broad region of the spectrum is formed by contributions from various sub-classes of water molecules with different de- grees of hydrogen bonding. In general, as the strength of the hydrogen bond increases, the overall stretch spectra shift towards lower frequencies. That is, the stronger the hydrogen bonding strength, the lower the frequency of the stretching vibrations. The sharp peak at 3700 cm 1 (2740 cm 1 ) is a characteristic peak observed at hydropho- bic interfaces 27, 28 (e.g., air/water interface, water/octadecyltrichlorosilane (OTS) inter- face, water/hexane interface). However, at hydrophilic interfaces (e.g., water/silica), this dangling-OH (-OD) species do not exist due to hydrogen bonding to the solid interface. 126 7.2.1 CaF 2 /D 2 O Interface Figure 7.2 shows the SFG spectra measured at the CaF 2 /D 2 O interface in the -OD stretch region in the presence/absence of salt (1M NaCl). The main feature in the pure D 2 O Figure 7.2: VSFG spectra of D 2 O in the -OD stretch region in the absence and presence of 1M NaCl solution at the CaF 2 /D 2 O interface. spectra is the broad hydrogen bonded peaks in the 2200-2600 cm 1 region. The overall broad feature is evident in the presence of the salt solution, however, the relative ampli- tudes of different sub-classes of hydrogen bonded molecules change in the salt solution. This change of intensity of the 2500 cm 1 peak with respect to the 2350 cm 1 peak is an evidence for a lower degree of coordination of the interfacial waters in the salt solution. This has been observed before by Covert et al. 29 where they measured the VSFG spectra of -OH stretches at various solid/H 2 O interfaces and salt concentrations. At higher ionic strengths, the surface potential increases and the surface ordering of the interfacial water molecules increases, which provides a stronger SFG signal. 127 We measured these SFG spectra at the CaF 2 surfaces in order to compare it with our graphene interface results, which are discussed later in this report. This will also help us disentangle the effects of the monolayer graphene on the interfacial water structure. 7.2.2 Al 2 O 3 /D 2 O Interface Our samples were prepared by transferring a single layer of graphene on to a substrate of 10 nm Al 2 O 3 deposited on a CaF 2 window. This thin layer of Al 2 O 3 might have Figure 7.3: VSFG spectra of D 2 O in the -OD stretch region in the absence and presence of 1M NaCl solution at the Al 2 O 3 /D 2 O interface. The black traces are fits to the raw spectra. an effect on the interfacial ordering of molecules when in contact with water and is re- ported for comparison with graphene/water spectra. The VSFG spectra 30–32 of water at the Al 2 O 3 interface is sensitive to protonation/deprotonation of the surface groups at low/high pH values. The presence of terminal hydroxyl groups at oxide mineral in- terfaces has been suggested previously, 33 which can be protonated or deprotonated de- pending on the pH value. However, for our experimental conditions (pH7), the extent 128 of protonation/deprotonation of these terminal hydroxyl groups will be minimal. The presence of a graphene layer on top of the Al 2 O 3 will prevent any interaction between the terminal hydroxyl groups and the D 2 O molecules. VSFG spectra at the Al 2 O 3 /CaF2 interface has been measured in the absence and presence of salt solution, as presented in Figure 7.3. The small peak at 2710 cm 1 is most likely due to the protonation of the terminal hydroxyl group present at the mineral interface, which does not change in amplitude when in contact with the 1M NaCl solution. On the contrary, the red shifted peaks at 2370 cm 1 and 2500 cm 1 are diminished in the presence of the ionic solution mainly due to the screening of the surface potential. Curve fitting of these two SFG spectra suggests that the amplitude of the peak at 2710 cm 1 remains almost unaffected with a slight variation of the peak frequency. 7.2.3 Graphene/D 2 O Interface Presence of Dangling -OD at Graphene Interface We measured the OD stretch spectra at the graphene/D 2 O interface and compared the spectra with those of the Al 2 O 3 /D 2 O interface as shown in Figure 7.4a. The two main spectral signatures of the Al 2 O 3 /D 2 O interface observed at 2370 cm 1 and 2500 cm 1 are also present at the graphene/D 2 O interface with an additional sharp peak at 2680 cm 1 . As discussed earlier, the presence of a sharp peak on the blue side of the stretch- ing spectra indicates the presence of dangling -OD molecules at the interfacial layer. The hydrophilicity of the interface shifts the frequency of these water molecules straddling the interface towards lower frequencies, which also suggests the presence of weak inter- actions between water molecules and the hydrophobic interface. We have also measured the VSFG signal from the dry sample cell (experiments performed without any D 2 O) and in isotopically diluted solution for the spectroscopic assignment of this 2680 cm 1 129 (a) (b) Figure 7.4: (Left Panel) Comparison of the VSFG spectrum of D 2 O at graphene and Al 2 O 3 interfaces. Spectral response from the dry graphene cell and with iso- topically diluted water at the graphene interface are also plotted in the same scale for comparison. (Right Panel) VSFG spectra of D 2 O in the -OD stretch region in the absence and presence of 1M NaCl solution at the graphene/D 2 O interface. The black traces are fits to the raw spectra. peak. Single layer graphene exhibits a characteristic Raman peak (2D) at 2670 cm 1 , which is infrared inactive. We do not see any evidence of the graphene 2D peak in our VSFG spectra (purple trace in Figure 7.4) from the dry graphene cell. This is in accor- dance with the spectroscopic selection rules, i.e., for any vibration to be SFG active, it must be both infrared and Raman active. We measured the VSFG spectra (blue trace in Figure 7.4) from the isotopically diluted D 2 O (D 2 O: H 2 O = 1:4)/graphene interface and show that the sharp peak at 2680 cm 1 vanishes, which is expected as the number of free -OD oscillators decreases with dilution. This further indicates that this peak is not the 2D peak of single layer graphene. The interfacial -OD peak (2707 cm 1 ,) from the Al 2 O 3 interface is absent in both the dry graphene spectrum and the isotopically diluted 130 D 2 O/graphene interface. This proves that the terminal hydroxyl peaks of the alumina do not interfere with the OD peaks at the graphene interface. Effect of NaCl on the interfacial water structure at graphene interface Figure 7.4b compares the VSFG spectra from graphene in D 2 O and graphene in contact with the 1M NaCl solution in D 2 O. At high salt concentrations, the Debye screening length decreases, which in turn decreases the third order polarization component of the SFG signal. Since this third order contribution to the overall SFG signal is the dominant term at higher ionic strengths, the signal intensity at the graphene interface in the salt solution decreases in the OD spectral region. However, the overall change in the distribution of different hydrogen bonded species in the salt/graphene surface is rather surprising (Figure 7.4b). Strongly hydrogen bonded species at lower frequencies are diminished to a greater extent than the sharp peak on the blue side. Spectral fitting of both spectra was performed in order to extract the frequencies of different hydrogen bonded species and are summarized in table 7.1. The sharp peak at 2678 cm 1 at the graphene/D 2 O interface is shifted to 2707 cm 1 with a slight increase in amplitude in the salt solution without any external applied potential. The organization of water molecules or the local hydrogen bonding structure near the hydrophobic graphene surface will be disrupted by the ions present in the Debye layer, which might be the reason for the blue shift of the sharp dangling -OD peak in the salt solution with respect to pure D 2 O. The complex interactions present at this kind interface involves a combination of van der Waals, electrostatic forces and hydrogen bonding interactions, which require detailed theoretical attention to fully unravel the nature of these interactions. 131 Effect of Surface Charge on interfacial water One of the main motivations behind the graphene project was to understand the inter- actions between the surface charge and the water molecules at the interface. Graphene being an excellent conductor of electricity and having high mechanical strength and low chemical reactivity is an ideal material to study these interactions. It is possbile to change to potential on the graphene by simply connecting the graphene to an external voltage source. Under these conditions, if the graphene surface is in contact with clean ion-free water, it is possible to probe the physical interactions between the surface charge and the water molecules under electrochemical conditions. We have measured the VSFG spectra of D 2 O in the -OD stretch region at the graphene/D 2 O interface at various sur- face charges in presence and absence of 1M NaCl and presented in figure 7.5a and figure 7.5b respectively. All these spectra were fitted using resonant Lorenzians as explained in Chapter 2, and the fitting parameters are presented in table 7.2 and table 7.3. VSFG (a) (b) Figure 7.5: VSFG spectra of D 2 O in the -OD stretch region at the graphene/D 2 O interface at various surface charges in presence (a) and absence (b) of 1M NaCl. The black traces are fits to the raw spectra. spectra at the graphene/D 2 O interface contains four Lorentzians (as presented earlier in 132 the 0 V olt spectra), however, the relative amplitudes of those peaks change with surface potential. VSFG spectra at the graphene/D 2 O interface in presence of 1M NaCl solution also change with surface potential, where mostly the free-OD bond is affected. In pres- ence of high concentration of the electrolyte the strongly H-bonded species are already screened as the Debye length becomes shorter, so the free-OD species present at the surface layer in close contact with the graphene were affected the most as the surface potential on the graphene changes. Presence of free-OH/OD species at the interface is an indication of the surface hy- drophobicity and the extent of the frequency shift of the free-OH/OD species with re- spect to the neutral surface would be a measure of this hydrophobicity. Graphene un- dergoes a reversible transition in the presence of an external electric field. Theoretical calculations 34, 35 showed that graphene can change its character from hydrophobic to hydrophilic in the presence of an electric field. According to Jiang’s calculations, 34 when the electric field is negative, the graphene surface becomes hydrophilic and vice versa for the dissociative adsorption of H 2 O on graphene. It should also be noted that the presence of electric field at the graphene surface also affects the interfacial struc- ture and surface wetting properties of water droplets. 36 Presence of free-OD species in VSFG spectra at the graphene D 2 O interface suggests that it is hydrophobic in nature. This free-OD frequecy is red shifted from the free-OD frequency of the pure D 2 O in- terface. Also, this frequency changes in the presence of external electric field. Figure 7.6 presents the change in the free-OD frequency at the graphene/D 2 O interface which shows a clear red shift of the free-OD peak when the external elctric field changes from positive to negative. This probably due to the reversible transition of the graphene sur- face, where a more negative potential makes the surface more hydrophilic. This induced hydrophilicilty will in turn affect the free-OD frequency in contact with the graphene 133 Figure 7.6: Frequency shift of the free-OD with surface charge surface. The change in free-OD frequency is also reversible with external electric field. Figure 7.6 also shows the electron doping concentration at the surface at those exter- nal surafce potentials. Doping concentration of electrons were calculated from Raman measurements and discussed in the appendix section. 7.3 Conclusion In conclusion, our collaborators (Prof. S. B. Cronin Group) at USC prepared the sin- gle layer graphene electrodes for spectro-chemical applications and we have performed surface selective spectroscopy on the electrode interface to directly probe the hydro- gen bonding structure of water at the interface. Our VSFG studies proved the pres- ence of a weakly hydrogen bonded or dangling -OD species for the first time at the graphene electrode interface under electrochemical conditions. VSFG studies of the salt solutions indicate a different hydrogen bond structure at the electrode interface. MD 134 simulations and water transport measurements 4, 37–39 indicate different desalination per- formance for functionalized nanoporous graphene membranes. Our experimental obser- vation of different hydrogen bonded water structure at graphene electrode interfaces in the absence/presence of salt solution will stimulate further experimental and theoretical studies. 7.4 Appendix 7.4.1 Experimental Details Sample Preparation A 10 nm Al 2 O 3 film is deposited on 1” CaF 2 windows by atomic layer deposition (ALD) at 200 C using trimethylaluminum (TMA) as the Al source and water vapor as the O source. Electron beam metal evaporation is then performed to deposit 5/200 nm thick Ti/Au electrodes on the CaF 2 windows, as shown in Figure7.7b. CaF 2 windows are (a) (b) Figure 7.7: Schematic of the sample cell used in our SFG experiments (Left). Single layer graphene on CaF 2 window (right). used as substrates. Monolayer graphene is grown on a copper foil by the chemical 135 vapor deposition (CVD) method with CH 4 at 1000 C. 40 This graphene/copper foil is then spin-coated with (poly methyl methacrylate) (PMMA) at 2000 rpm for 60 second, and baked on a hotplate at 170 C for 5 minute. The PMMA/graphene/copper foil is later cut into1cm x1 cm squares and soaked in a ferric chloride solution. The copper foil is dissolved in the solution and the PMMA/graphene film is then transferred on to the CaF 2 /Au substrate. The PMMA film is subsequently removed by acetone. Sample characterization has been done by standard Raman spectroscopic techniques. Sample Characterization The Cronin group performed Raman measurements on the graphene samples in order to make sure the samples were of good quality before and after the spectroscopic mea- surements. In general, single layer graphene has characteristic Raman peaks 1600 cm 1 (G-peak) and 2700 cm 1 (2D-peak). We made sure to have clean graphene samples before every measurements and also measured its Raman peaks after the SFG experiments to make sure that the graphene layer on the CaF 2 is still intact. (a) Raman Spectra before SFG experiments (b) Raman Spectra after SFG experiments Figure 7.8: Sample characterization before (left panel) and after SFG experiments (right panel). 136 G-band shift measurements Raman spectra (532nm excitation) of the graphene electrode under applied electrochem- ical potentials are taken in liquid solutions consisting of pure DI water using a water immersion lens. In order to protect the lens from the solution, a 13m thick Teflon sheet (American Durafilm Inc.) was used to cover the lens. Three terminal potentio- stat (Gamry, Inc.) is used to add bias between our graphene working electrode and the reference electrode. A glassy carbon (SPI, Inc.) and a Ag/AgCl reference electrode were used as the counter electrode and reference electrode, respectively. Three-terminal potentiostat setup eliminates errors associated with voltage drops across the low con- ductivity of DI water. Figure 7.9: Frequency shift of the G-band with external potential using Raman spectroscopy. Carrier concentrations at the corresponding voltages were also cal- culated and ploted in the right axis. Figure 7.9 shows the ‘G’ band Raman shift measured as a function of the applied voltage measured in a three-terminal electrochemical cell. Here, we observe substan- tial upshifts in the ‘G’ band Raman frequency for both positive and negative applied 137 potentials. Using the relation between Fermi energy (EF) and doping concentration (n) obtained by Das Sarma et al,, 41 and the linear relation between EF and ! G (with slightly different slopes observed for electron and hole doping). 42 We also plot the charge density in the graphene as a function of the applied voltage on the right axis of Figure 7.9. 43 VSFG Experiments VSFG experiments were performed in the reflection geometry, with both the visible and infrared beams passing through the back of the electrode, as depicted in Figure 7.7a. Our graphene sample, deposited on the CaF 2 window is transparent in the -OD stretch- ing frequency (2200-2800 cm 1 ), which allows us to perform SFG in this flipped ge- ometry. We performed VSFG experiments at the graphene/D 2 O interface in the absence and presence of 1M NaCl salt solution. The VSFG setup used in these measurements is the same as described previously. 44, 45 Briefly, we use a broad-band infrared pulse centered at 2500 cm 1 , generated by an optical parametric amplifier (OPA) followed by difference frequency generation (NDFG). The sum-frequency signal is upconverted by a narrow-band visible pulse as generated by a home built 4-f stretcher with a frequency resolution of 20 cm 1 . Both the visible and the infrared pulses are focused on the sample surface to a spot size of200m. The laser power at the sample surface was 1J and 8 J per pulse for the infrared and the visible beams, respectively. The angle of incidence from the surface normal was 67 for the visible and 62 for the infrared beams. All of the spectra were collected in the SSP polarization combination (SFG, visible, infrared) and recorded with a 500 nm monochromator (Princeton Instruments, 1200 l/mm grat- ing) and a liquid nitrogen cooled CCD detector (Roper Scientific). Non-resonant SFG 138 spectra collected in the flipped geometry from a 100 nm thick gold (Au) film deposited on the same CaF 2 window was used for the normalization of all the D 2 O spectra. Table 7.1: Fitting parameters for the VSFG spectra of the graphene/D 2 O and Al 2 O 3 /D 2 O interfaces in the absence and presence of 1M NaCl. Graphene/D 2 O Graphene/ 1M NaCl Al 2 O 3 /D 2 O Al 2 O 3 /1M NaCl ! 1 , cm 1 2367 2341 2370 2373 ! 2 , cm 1 2495 2518 2513 2508 ! 3 , cm 1 2613 2641 2613 2628 ! 4 , cm 1 2678 2707 2712 2694 B 1 24.2 0.73 30.8 31.6 B 2 31.5 19.5 29.7 11.5 B 3 6.8 6.2 -6.9 2.9 B 4 5.4 9.4 6.5 6.5 1 , cm 1 73 155 74 108 2 , cm 1 98 156 101 75 3 , cm 1 64 83 121 54 4 , cm 1 37 52 54 52 A NR 0.08 0.067 0.25 0.15 , rad -0.81 -12.56 -2.3 -2.09 139 Table 7.2: Fitting parameters for the VSFG spectra of the graphene/D 2 O with neg- ative and positive surface potential on the graphene. 0 V olt -0.5 V olt -1.0 V olt +0.5 V olt ! 1 , cm 1 2367 2372 2369 2375 ! 2 , cm 1 2495 2499 2488 2505 ! 3 , cm 1 2613 2614 2593 2614 ! 4 , cm 1 2678 2678.5 2673.5 2684 B 1 24.2 23.0 21.8 32.1 B 2 31.5 30.0 20.0 31.2 B 3 6.8 6.4 12.2 10.7 B 4 5.4 6.6 6.0 6.9 1 , cm 1 73 74 73 81 2 , cm 1 98 96 80 91 3 , cm 1 64 66 94 75 4 , cm 1 37 40 39 43 A NR 0.08 0.070 0.09 0.056 , rad -0.81 -0.83 -0.56 -1.6 140 Table 7.3: Fitting parameters for the VSFG spectra of the graphene/D 2 O in pres- ence of 1M NaCl with negative and positive surface potential on the graphene. 0 V olt -0.5 V olt +0.5 V olt ! 1 , cm 1 2341 2276 2314 ! 2 , cm 1 2518 2523 2528 ! 3 , cm 1 2641 2656 2654 ! 4 , cm 1 2707 2704 2710 B 1 0.73 5.1 -3.8 B 2 19.5 20.1 17.3 B 3 6.2 10.9 4.3 B 4 9.4 3.4 5.3 1 , cm 1 155 162 -209 2 , cm 1 156 151 194 3 , cm 1 84 83 77 4 , cm 1 52 34 41 A NR -0.07 -0.05 0.08 , rad -12.5 -12.1 -21.3 141 Chapter 7 References [1] Chen, Chun-Chung; Aykol, Mehmet; Chang, Chia-Chi; Levi, AFJ; Cronin, Stephen B Graphene-silicon schottky diodes Nano Lett. 2011, 11, 1863–1867. [2] Ma, Jie; Michaelides, Angelos; Alfe, Dario; Schimka, Laurids; Kresse, Georg; Wang, Enge Adsorption and diffusion of water on graphene from first principles Phys. Rev. B 2011, 84, 033402. [3] Chen, Jin-Long; Yan, Xiu-Ping A dehydration and stabilizer-free approach to pro- duction of stable water dispersions of graphene nanosheets J. Mater. Chem. 2010, 20, 4328–4332. [4] Suk, Myung E; Aluru, NR Water transport through ultrathin graphene J. Phys. Chem. Lett. 2010, 1, 1590–1594. [5] Neto, AH Castro; Guinea, F; Peres, Nuno MR; Novoselov, Kostya S; Geim, An- dre K The electronic properties of graphene Rev. Mod. Phys. 2009, 81, 109. [6] Frank, IW; Tanenbaum, David M; van derZande, Arend M; McEuen, Paul L Me- chanical properties of suspended graphene sheets J. Vac. Sci. Technol. B: Micro- electron. Nanometer Struct. Process. Meas. Phenom. 2007, 25, 2558–2561. [7] Meric, Inanc; Han, Melinda Y .; Young, Andrea F.; Ozyilmaz, Barbaros; Kim, Philip; Shepard, Kenneth L. Current saturation in zero-bandgap, top-gated graphene field-effect transistors Nat. Nano. 2008, 3, 654–659. [8] Lin, Yu-Ming; Chiu, Hsin-Ying; Jenkins, Keith A; Farmer, Damon B; Avouris, Phaedon; Valdes-Garcia, Alberto Dual-gate graphene fets withf T of 50 ghz IEEE Electron Device Lett. 2010, 31, 68–70. [9] Bao, Wenzhong; Myhro, Kevin; Zhao, Zeng; Chen, Zhen; Jang, Wanyoung; Jing, Lei; Miao, Feng; Zhang, Hang; Dames, Chris; Lau, Chun Ning In situ observation of electrostatic and thermal manipulation of suspended graphene membranes Nano Lett. 2012, 12, 5470–5474. [10] Ostrowski, Joseph H. J.; Eaves, Joel D. The tunable hydrophobic effect on electri- cally doped graphene J. Phys. Chem. B 2014, 118, 530–536. [11] Balandin, Alexander A; Ghosh, Suchismita; Bao, Wenzhong; Calizo, Irene; Teweldebrhan, Desalegne; Miao, Feng; Lau, Chun Ning Superior thermal con- ductivity of single-layer graphene Nano Lett. 2008, 8, 902–907. 142 [12] Toney, Michael F; Howard, Jason N; Richer, Jocelyn; Borges, Gary L; Gordon, Joseph G; Melroy, Owen R; Wiesler, David G; Yee, Dennis; Sorensen, Larry B V oltage-dependent ordering of water molecules at an electrodeelectrolyte interface Nature 1994, 368, 444–446. [13] Ataka, Ken-ichi; Yotsuyanagi, Takao; Osawa, Masatoshi Potential-dependent re- orientation of water molecules at an electrode/electrolyte interface studied by surface-enhanced infrared absorption spectroscopy J. Phys. Chem. 1996, 100, 10664–10672. [14] Davis, Joel G; Rankin, Blake M; Gierszal, Kamil P; Ben-Amotz, Dor On the co- operative formation of non-hydrogen-bonded water at molecular hydrophobic in- terfaces Nature Chem. 2013, 5, 796–802. [15] Tong, Yujin; Lapointe, Franois; Thmer, Martin; Wolf, Martin; Campen, R Kramer Hydrophobic water probed experimentally at the gold electrode/aqueous interface Angew. Chem. 2017. [16] Schnur, Sebastian; Gro, Axel Properties of metalwater interfaces studied from first principles New J. Phys. 2009, 11, 125003. [17] Gro, Axel; Gossenberger, Florian; Lin, Xiaohang; Naderian, Maryam; Sakong, Sung; Roman, Tanglaw Water structures at metal electrodes studied by ab initio molecular dynamics simulations J. Electrochem. Soc. 2014, 161, E3015–E3020. [18] Leenaerts, O; Partoens, B; Peeters, FM Water on graphene: Hydrophobicity and dipole moment using density functional theory Phys. Rev. B 2009, 79, 235440. [19] Gordillo, MC; Marti, J Water on graphene surfaces Journal of Physics: Condensed Matter 2010, 22, 284111. [20] Mart, J; Gordillo, MC Effects of confinement on the vibrational spectra of liquid water adsorbed in carbon nanotubes Phys. Rev. B 2001, 63, 165430. [21] Kolesnikov, Alexander I; Zanotti, Jean-Marc; Loong, Chun-Keung; Thiyagara- jan, Pappannan; Moravsky, Alexander P; Loutfy, Raouf O; Burnham, Christian J Anomalously soft dynamics of water in a nanotube: a revelation of nanoscale con- finement Phys. Rev. Lett. 2004, 93, 035503. [22] Ho, Tuan A; Striolo, Alberto Polarizability effects in molecular dynamics simula- tions of the graphene-water interface J. Chem. Phys. 2013, 138, 054117. [23] Politano, Antonio; Marino, Antonio Raimondo; Formoso, Vincenzo; Chiarello, Gennaro Hydrogen bonding at the water/quasi-freestanding graphene interface Carbon 2011, 49, 5180–5184. 143 [24] Du, Quan; Freysz, Eric; Shen, Y Ron Surface vibrational spectroscopic studies of hydrogen bonding and hydrophobicity Science 1994, 264, 826–827. [25] Shen, Yuen Ron; Ostroverkhov, Victor Sum-frequency vibrational spectroscopy on water interfaces: polar orientation of water molecules at interfaces Chem. Rev. 2006, 106, 1140–1154. [26] Stiopkin, Igor V; Jayathilake, Himali D; Bordenyuk, Andrey N; Benderskii, Alexander V Heterodyne-detected vibrational sum frequency generation spec- troscopy J. Am. Chem. Soc. 2008, 130, 2271–2275. [27] Gragson, DE; Richmond, GL Potential dependent alignment and hydrogen bond- ing of water molecules at charged air/water and ccl4/water interfaces J. Am. Chem. Soc. 1998, 120, 366–375. [28] Scatena, LF; Brown, MG; Richmond, GL Water at hydrophobic surfaces: weak hydrogen bonding and strong orientation effects Science 2001, 292, 908–912. [29] Covert, Paul A; Jena, Kailash C; Hore, Dennis K Throwing salt into the mix: Altering interfacial water structure by electrolyte addition J. Phys. Chem. Lett. 2013, 5, 143–148. [30] Zhang, Luning; Tian, Chuanshan; Waychunas, Glenn A; Shen, Y Ron Structures and charging of -alumina (0001)/water interfaces studied by sum-frequency vibra- tional spectroscopy J. Am. Chem. Soc. 2008, 130, 7686–7694. [31] Yeganeh, MS; Dougal, SM; Pink, HS Vibrational spectroscopy of water at liq- uid/solid interfaces: Crossing the isoelectric point of a solid surface Phys. Rev. Lett. 1999, 83, 1179. [32] Braunschweig, Bjrn; Eissner, Stephan; Daum, Winfried Molecular structure of a mineral/water interface: effects of surface nanoroughness of -al2o3 (0001) J. Phys. Chem. C 2008, 112, 1751–1754. [33] Hass, Kenneth C; Schneider, William F; Curioni, Alessandro; Andreoni, Wanda The chemistry of water on alumina surfaces: Reaction dynamics from first princi- ples Science 1998, 282, 265–268. [34] Jiang, QG; Ao, ZM; Chu, DW; Jiang, Q Reversible transition of graphene from hydrophobic to hydrophilic in the presence of an electric field J. Phys. Chem. C 2012, 116, 19321–19326. [35] Xu, Zhemi; Ao, Zhimin; Chu, Dewei; Younis, Adnan; Li, Chang Ming; Li, Sean Reversible hydrophobic to hydrophilic transition in graphene via water splitting induced by uv irradiation Scientific Reports 2014, 4. 144 [36] Ren, Hongru; Zhang, Leining; Li, Xiongying; Li, Yifan; Wu, Weikang; Li, Hui Interfacial structure and wetting properties of water droplets on graphene under a static electric field Phys. Chem. Chem. Phys. 2015, 17, 23460–23467. [37] Cohen-Tanugi, David; Grossman, Jeffrey C Water desalination across nanoporous graphene Nano Lett. 2012, 12, 3602–3608. [38] Surwade, Sumedh P; Smirnov, Sergei N; Vlassiouk, Ivan V; Unocic, Raymond R; Veith, Gabriel M; Dai, Sheng; Mahurin, Shannon M Water desalination using nanoporous single-layer graphene Nat. Nano. 2015, 10, 459–464. [39] Konatham, Deepthi; Yu, Jing; Ho, Tuan A; Striolo, Alberto Simulation insights for graphene-based water desalination membranes Langmuir 2013, 29, 11884–11897. [40] Chen, Chun-Chung; Li, Zhen; Shi, Li; Cronin, Stephen B Thermoelectric trans- port across graphene/hexagonal boron nitride/graphene heterostructures Nano Res. 2015, 8, 666–672. [41] Sarma, S Das; Adam, Shaffique; Hwang, EH; Rossi, Enrico Electronic transport in two-dimensional graphene Rev. Mod. Phys. 2011, 83, 407. [42] Froehlicher, Guillaume; Berciaud, St´ ephane Raman spectroscopy of electro- chemically gated graphene transistors: Geometrical capacitance, electron-phonon, electron-electron, and electron-defect scattering Phys. Rev. B 2015, 91, 205413. [43] Yan, Jun; Zhang, Yuanbo; Kim, Philip; Pinczuk, Aron Electric field effect tuning of electron-phonon coupling in graphene Phys. Rev. Lett. 2007, 98, 166802. [44] Vinaykin, Mikhail; Benderskii, Alexander V . Vibrational sum-frequency spectrum of the water bend at the air/water interface J. Phys. Chem. Lett. 2012, 3, 3348– 3352. [45] Dutta, Chayan; Svirida, Anton; Mammetkuliyev, Muhammet; Rukhadze, Marina; Benderskii, Alexander V Insight into water structure at the surfactant surfaces and in microemulsion confinement J. Phys. Chem. B 2017, 121, 7447–7454. 145 Chapter 8: Water Structure in micro-emulsion confinement In this chapter, water H-bond structure and orientation of interfacial water in micro- emulsion confinement and at the surfactant interface was studied using linear (FTIR) and non-linear (SFG) spectroscopic techniques. Anton Svirida, a summer intern in our group and Dr. Marina Rukhadze, a visiting scientist from Tbilisi State University contributed to this work. Anton helped me with the SFG experiments, and Marina performed the partical size measurements and the linear FTIR measurements. 8.1 Introduction Water being the ubiquitous medium for many chemical and most of the biological pro- cesses holds the key for fundamental understanding of those systems. Many of these interactions occur at the interface of these systems and water’s ability of form ex- tented H-bonding network with many different types of H-bonded species plays a key role in those processes. Biological membranes and biomolecules, 1–6 micelles 7 and re- verse micelles, 8–11 protein folding 12, 13 and catalysis 14, 15 are a few examples where con- fined water interacts strongly on molecular scale with the interfacial molecules or ions. Biomembranes are complex structures formed by lipid bilayers with an array of aggre- gated lipids arranged in two dimensions, where polar head groups remain exposed to 146 the aqueous phase and the lipid chains face towards each other. Confined water in re- verse micelles provides a good model system for complex biological systems, such as biomembranes. 2, 6 Reverse micelles have one lipid/water interface and can be thought of as a simplified biomembrane model with similar type of H-bonding interactions between the surfactant head group and the water at the interface. Similar interactions occur at the planar surfactant monolayer interfaces, but without the geometric nano-confinement ef- fects (Figure 8.1). Figure 8.1: Interfacial and bulk water in reverse micelles (top left) and surfactant monolayer/water interfaces (top right) used in this study. Reproduced from Ref. 16. Reverse micelles are spherical molecular aggregates formed by ionic (e.g., sodium 1,4 bis 2 ethylhexylsulfosuccinate: AOT) or nonionic (e.g., Polyoxyethylene 4-lauryl ether: Brij L-4) surfactants in non-polar organic solvents. These reverse micelles form a nanometer sized water pool confining a number of water molecules inside the cavity, where the charged hydrophilic headgroups are in close contact with the enclosed water molecules and the organic tail groups are oriented towards the organic phase. Physical 147 and chemical properties of these reverse micelles are highly sensitive to the w 0 number, where w 0 number is defined as the ratio of molarities of water and surfactants. W 0 = [H 2 O] [Surfactant] (8.1) Vibrational spectroscopy in the OH-stretch region is a convenient and powerful tool for studying water structure and dynamics in bulk environment and at interfaces. Bulk water has been studied using a diverse range of linear and nonlinear techniques both theoretically and experimentally. 17–20 The OH-stretch vibration of a water molecule is very sensitive to its hydrogen bonding environment in the bulk and can be delocalized over 10 or more surrounding water molecules. 19, 21 Intermolecular and intramolecular couplings between the OH-stretches play a central role in determining the vibrational line shapes. OH-stretch spectra also show a strong non-Condon effect, i.e., the transi- tion dipole strongly depends on the transition frequency. 17, 22 The inherent properties of water molecules at the interface are expected to be different than that of bulk water, primarily due to the asymmetry of the interface. Surface sensitive nonlinear vibrational sum frequency generation (VSFG) spectroscopy has been successfully implemented to study structure and dynamics at the air/water interface. 3, 4, 23–28 VSFG is a non-linear spectroscopic technique with monolayer surface specificity. 29 The overall SFG spectrum which is polarization dependent is formed by contributions from all the different types of water hydrogen bonding classes possible at the interface as calculated by Skinner et.al. 18, 20 Reverse micelle in an organic phase can be directly studied by sum frequency scattering 30, 31 to obtain structural and orientation information of water molecules inside the reverse micelle, however this is highly dependent on the size of the particles and for small radius particles the scattering process will be highly inefficient. 148 Analysis of the OH-stretching band of water in confined systems like reverse mi- celles offers information about molecular interactions of water at the interface of nanoassemblies. According to the popular core/shell model 32 of nanoconfined water in reverse micelles there are two distinct regions inside the reverse micelle; the core water is of bulk character which absorbs at the red side of the spectrum, and the in- terfacial water absorbs at the blue side due to hydrogen bonding interaction with the surfactant head group. Sechler et.al. 33 determined the spectral contributions of these different water molecules from the core and interface of these reverse micelles to the overall absorption using volumetric analysis with infrared absorption spectroscopy. The structure of interfacial water in the presence of soluble ionic and non-ionic sur- factants has also been studied extensively using VSFG spectroscopy. 34–36 Due to the presence of charged surfactants at the interface, the surface waters become highly or- dered which in turn produces stronger SFG signal from the surfactant/water interface as compared to pure air/water interface. Most of these studies are focused on the SSP po- larization spectra which probes water molecules which only have a component transition dipole moment perpendicular to the surface plane. However, there are other polariza- tion combinations (e.g., PPP) that can access water molecules with components both parallel and perpendicular to the surface plane. Hence a polarization dependent VSFG study of surfactant/water interface can provide additional information on the molecular orientation. Water insoluble ionic and non-ionic surfactants, when placed on the water surface, are expected to orient with the polar head groups submerged in the top most layer of water and the non-polar organic groups pointing outwards at the air phase. The water structure at these interfaces should mimic the environments of the water pools inside the 149 reverse micelles. In general, as the number of hydrogen bonds around a water molecule increases the stretching vibration of that water shifts towards lower frequency. Here, we present a comparative study of infrared absorption and VSFG of interfa- cial water in reverse micelles and in Langmuir monolayer, respectively. Our polariza- tion dependent VSFG experiments indicate the presence of a particular type of water molecules with one donor hydrogen and a free or loosely bound -OH bond (central fre- quency around 3560 cm 1 ) inside the water pool of ionic and non-ionic reverse micelles. We have performed orientational analysis on this type of water molecules to extract the average tilt angle of this particular type of waters for a detailed understanding of the interfacial water structure. 8.2 Experimental Details 8.2.1 Preparation of the reverse microemulsion and determination of the particle size AOT is an ionic surfactant with a negatively charged head group whereas Brij L-4 is a non-ionic surfactant. Structures of both the molecules are shown in Figure 8.2. Reverse microemulsion solutions were prepared with AOT and Brij L-4 surfactants in hexane (HPLC grade) with varying water content. The molar ratio of water to surfactant was calculated by W 0 =[H 2 0]/[surfactant]. Transparent one-phase solutions of the surfac- tants in hexane were achieved through gentle sonication. All spectroscopy experiments were performed on stable, one-phase systems at room temperature. The reverse micelle diameter was measured using dynamic light scattering technique and the water pool radius for different solutions was calculated. 150 Figure 8.2: (a)Structure of sodium 1,4-bis-2-ethylhexylsulfosuccinate (AOT), bot- tom left and Polyoxyethylene(4)lauryl ether (Brij L-4), bottom right. (b) Reverse micelle structure showing the total radius (Rh), radius (R) of the water pool inside the reverse micelle and the length (Ltail) of the surfactant tail group. Reproduced from Ref. 16. Particle-size measurements were performed with a commercial goniometer (CGS-2, ALV-GmbH, D-Langen) equipped with a vertical-polarized 22-mW HeNe laser (wave- length =632.8 nm), a fiber optical detection unit with an avalanche photodiode, and an ALV-5000/E multiple correlator. Cylindrical light-scattering cells of 10 mm outer di- ameter were used. All measurements were performed at 90.028 detection angle and at 25 C. All solutions were left to equilibrate at least 2 h before measurements. For a reverse micelle with radius R h we can write R h =R+L tail (see Figure8.1); where R is the radius of the water pool inside the reverse micelle and L tail is the length of the surfactant tail 37 groups. Measured radius for the reverse micelles and the corresponding water pool radius values for multiple reverse micelles with different W 0 values are summerized in Table 8.1 151 Table 8.1: Dynamic light scattering results and the calculated radius of water pool inside the reverse micelles AOT Brij L-4 W 0 R h (nm) R(nm) W 0 R h (nm) R(nm) 4 2.4 1.20 1 2.39 1.19 8 3.33 2.13 2 2.74 1.54 10 3.47 2.27 4 3.33 2.13 12 3.7 2.50 6 4.97 3.77 15 4.07 2.97 8 6.37 5.17 17 4.53 3.54 10 6.8 5.6 20 5.21 4.28 8.2.2 Volumetric Analysis Micro-emulsions have two separate regions; the core water pool and the interfacial water region, and the overall absorption spectrum is the sum of absorptions from these two regions. The net absorption can be written as A(!) =A i (!) +A c (!) (8.2) A(r;!) =zN[P i (R)h i (!) +P c (R)h c (!)] (8.3) P c (R) = ( 1d R ) 3 (8.4) P i (R) = 1 ( 1d R ) 3 (8.5) where, A i and A c are the absorptions from the interfacial and the core water region respectively. z is the sample thickness, P i and P c are the fractions water molecules in different water regions, h i and h c are the spectral responses from water molecules in the interface and in the core water pool, respectively. The water fraction at different 152 regions were calculated from the experimentally measured water radius values at various water concentrations, i.e., at different W 0 (= [H 2 O]/ [surfactant]) values. The spectral responses were determined from the fits of A/zN vs R for different W 0 values. R is the radius of the water pool inside the reverse micelle and d is the thickness of the interfacial water layer. IR absorption spectra were recorded in the water -OH stretching band (3000-3800 cm 1 ) in Bruker Vertex80 IR spectrometer with vacuum capability equipped with a 0.05 cm path length calcium fluoride CaF 2 window. 8.2.3 SFG Experiment We used broad-band SFG spectroscopy to probe a wide range of vibrational frequencies. A schematic detail of our experimental setup was described previously in details in Chapter 2.1. The laser power at the sample surface was 1.0 J and 12.0 J per pulse for the infrared and the visible, respectively. To obtain clean water for the samples, we used 15 M water from our Millipore system and distilled it through a sealed distillation apparatus cleaned with piranha solution before any measurement. In order to record the whole spectrum in the water stretching region (3100-3800 cm 1 ), we have to tune the central frequency of the infrared pulse in two different re- gions (3300 cm 1 and 3600 cm 1 center frequency) to overcome the bandwidth limita- tion in our experiments. Two spectra recorded at two different central frequencies were joined using Matlab programming. Two different spectra (measured consecutively in the same day), in which the IR pulses were tuned to two different center frequencies, were obtained. The two spectra overlap; the region of overlap is approximately 200 cm 1 . The spectra were joined together using a matlab script. Figure 8.3 shows the 153 Figure 8.3: SFG spectra of OH-stretching band at the air/water interface for ssp and ppp polarization combinations of the SFG, visible and the IR pulses measured at different center frequency for the IR beam and stitched together afterwards dur- ing the data analysis. The right panel shows the full spectra at the water stretching region where the sharp peak around 2750 cm-1 is due the dangling Oh molecules at the interface and the red shifted broad peaks are due to hydrogen bonded water molecules. All other SFG spectra presented in the paper have been analyzed by this method. Reproduced from Ref. 16. SFG spectra taken at the air/water interface in SSP and PPP polarization combination at two different regions and joined together by the above mentioned procedure. Curve fitting of the actual measured spectra with resonant and non-resonant part is necessary in order to get the actual frequencies of different transitions. Second order non-linear susceptibility (2) is expressed as a sum of Lorentzians, and the non-resonant part is fitted as a constant term with a phase difference in the real part as shown in equation 8.6. (2) /A NR e i NR + X j B j ! IR ! j +i j (8.6) where, A NR is the non-resonant contribution to the overall spectra with a phase differ- ence NR from the resonant parts. B j is the resonant amplitude with center frequency ! j and linewidth j . 154 8.3 Results and Discussions 8.3.1 FTIR results We prepared reverse micelle micro-emulsion of AOT and Brij L-4 in hexane with vary- ing water content and performed volumetric analysis to separate the bulk and interfacial water spectral response in the OH-stretching region following the procedure developed by Sechler et.al. 33 Reverse micelle is made of spherical pool of water surrounded by surfactant head group, with two distinct water phases. For large values of pool radius, water molecules at the center of the pool are more homogeneous and surrounded by other molecules as in bulk water. Water molecules at the edges where they interact strongly with the surfactant are of truly interfacial character and have distinctly sepa- rate spectral signature. Figure 8.4 shows the infrared spectra of AOT (top left panel) reverse micelle with different W 0 values. As the water content increases the size of the water pool inside the micelle increases and the stronger hydrogen bonding interactions from the bulk water become dominant which shifts the overall spectra towards lower frequencies. This effect of increasing bulk hydrogen bonding interactions can be better understood from the normalized absorption spectra as depicted in the top right panel in Figure 8.4, where the bulk frequencies dominate the overall spectra as W 0 increases. The infrared absorptions spectra for multiple microemulsion solutions of Brij L-4 with varying W 0 and the normalized absorption spectra are shown in the bottom left and right panel of Figure 8.4 respectively. Brij L-4 reverse micelles behave similarly as those of AOT and overall water spectra is interface water like for smaller size of the micelles and bulk water like for larger micelles. Figure 8.5 shows the spectral responses of the bulk and the interface regions deter- mined from the fits of A/zN vs R, where z is the sample thickness and N is the number 155 Figure 8.4: SFG spectra of OH-stretching band at the air/water interface for ssp and ppp polarization combinations of the SFG, visible and the IR pulses measured at different center frequency for the IR beam and stitched together afterwards dur- ing the data analysis. The right panel shows the full spectra at the water stretching region where the sharp peak around 2750 cm-1 is due the dangling Oh molecules at the interface and the red shifted broad peaks are due to hydrogen bonded water molecules. All other SFG spectra presented in the paper have been analyzed by this method. Reproduced from Ref. 16. density of the surfactant and R is the radius of the water droplet inside the reverse mi- celle. The bulk water response for AOT and Brij L-4 (blue and red dotted lines) are very similar to the pure bulk water IR absorption spectrum (solid black line). Even though the interfacial counterpart is blue-shifted from the bulk due to different degree of hy- drogen bonding interactions as described previously, the interface spectra for the ionic (AOT: blue solid line) and non-ionic (Brij: red solid line) micelle are different. Due to 156 Figure 8.5: Separated spectral responses of bulk (dotted) and interfacial (solid) water regions for AOT (blue) and Brij (red) reverse micelle determined from the fits of A/zN versus r, where r is the radius of different micro-emulsion solutions. Solid black line shows the pure bulk water IR absorption spectra. Reproduced from Ref. 16. stronger charge-dipole interactions with the ionic surfactant head group, AOT/water in- terface contains large ensemble of molecules with various degree of hydrogen bonding. On the other hand, Brij/water interface has weaker dipole-dipole interactions and most of the hydrogen bonding interactions resemble that of the bulk water interactions which make it red shifted compared to the ionic interface. Infrared studies show that there are two distinctly different regions of water with var- ious degrees of hydrogen bonding inside the reverse micelle. The core water molecules behave more like bulk waters and absorb more on the red side of the spectrum, whereas the water molecules that are in close proximity to the surfactant head groups absorbs more on the blue side. 157 8.3.2 SFG results Detailed structural and orientational information of these interfacial waters in close proximity to the surfactant head groups can be obtained by surface selective SFG spec- troscopy on surfactant monolayers. Usually pure air/water spectra give a double hump structure in the water stretch (3200-3600 cm 1 ) band in SFG. Several other groups have performed SFG experiments (homodyne 27, 38–41 and phase sensitive 26, 28, 36, 42 ) on the air/water interface and fitted the spectra with four Lorentzians. The sharp spec- trally separated peak on the blue side has been attributed to free water molecules having one -OH dangling at the interface. The other three frequencies are well separated from the free-OH bond due to stronger H-bonding interactions. It is difficult to assign these broad regions to a particular kind of molecules. Skinner and coworkers have calculated the spectral density of different types of H-bonding molecules for pure air/H 2 O 24 and air/D 2 O 23, 25 interfaces. They also defined different H-bonded classes of water molecules according to the total number of donor and acceptor hydrogen bonds. In general, as the H-bonding interactions between adjacent water molecules become stronger, OH-stretch frequencies decrease. The spectral intensity around 3700 cm 1 is attributed to free-OH molecules, specifically to 2 S and 1 N type of molecules according to Skinners defini- tion, where 2 S represents water molecules having 2 hydrogen bonds with single donor hydrogen, and 1 N type of waters having only one hydrogen bond without any donor hydrogen. The other OH-stretching frequency in the same free water molecule at the air/water interface has been assigned 26, 38, 43 to be around 3550 cm 1 . Figure 8.6 shows the SFG spectra of surfactant/water interface in SSP (red) and PPP (blue) polarizations and the separated infrared response (green) of the interfacial water obtained from volumetric analysis is plotted in the same graph. Both AOT/water inter- face (fig. 8.6: top left panel) and Brij/water interface (fig. 8.6:top right panel) show four 158 Figure 8.6: (a) SFG spectra of OH-stretching band in AOT/water interface (left panel) and Brij/water interface (right panel) for ssp and ppp polarization combi- nations of the SFG, visible and the IR pulses. Interface spectral responses sepa- rated from FTIR spectra of the AOT and Brij reverse micelle is plotted in the same graphs for comparison. (b)Comparison of the SFG spectra of air/water, AOT/water and Brij/water interfaces for SSP (left) and PPP (right) polarization combinations of the SFG, visible and the IR pulses. Reproduced from Ref. 16. resonant contributions in the hydrogen bonded region in both SSP and PPP polariza- tion, but spectral fitting in Brij/water interface is somewhat complicated due to strong non-resonant background signal. However, we have identified the main types of water molecules at both surfaces that resemble the interfacial component in infrared spectra of reverse micelles. At the surfactant interfaces, the freeOH peak is suppressed due to in- teractions with the hydrophilic surfactant head group, so we do not observe a sharp peak 159 in the far blue side of the spectrum ( 3700 cm 1 for pure air/water interface). The other part of the spectra of AOT/water interface have been fitted with four main resonant fea- tures at 3230cm 1 , 3430 cm 1 , 3560 cm 1 , and 3640 cm 1 for both polarizations. For Brij/water interface SSP spectra shows two peaks around 3200 cm 1 and 3400 cm 1 , but the 3560 cm 1 peak is buried in strong background signal. On the other hand, PPP spectra in Brij interface show most prominently the 3560 cm 1 peak. Figure 8.6 also compares the spectra for SSP (bottom left panel) and PPP (bottom right panel) polar- izations for all three interfaces: AOT, Brij, and air/water. The PPP polarization spectra from the ionic and non-ionic surfactant interfaces show a prominent peak at 3560 cm 1 , in clear resemblance of the interfacial component in the infrared spectra obtained from the volumetric FTIR analysis of reverse micelles. This indicates that similar hydrogen-bonding structures may be present at both surfactant interfaces. A detailed study of water/phospholipid interface in homodyne VSFG 44, 45 and heterodyne-detected VSFG 46, 47 experiments in SSP polarization demon- strated the presence of two distinct positive bands: a strong band around 3300 cm 1 and a weak band around 3580 cm 1 . Recently Ishiyama et.al. 48 calculated the Im[ 2 ] spec- tra at water/POPC [3-palmitoyl-2-oleoyl-D-glycero-1-phosphatidylcholin] interface and provided number density of water molecules at different sites at the zwitter-ionic lipid interface. They assigned the low frequency band to water molecules bonded to po- lar phosphate and choline group of the POPC molecule while the other high frequency band was said to be associated with water molecules near the carbonyl oxygens. AOT or Brij do not contain any highly polar nitrogen or phosphate centers, but we still observe strong intensity peaks in low frequency side of the spectrum in SSP polarization, while the high frequency peak is more pronounced in the PPP polarization spectra. The posi- tion of the 3560 cm 1 peak is constant in ionic or non-ionic surfactant surface, whereas 160 the other two peak positions change. This suggests that this peak originates from a wa- ter molecule having one donor hydrogen weakly bound to the surfactant head group, as recently suggested by Nojima et al. 47 8.3.3 Isotopic dilution studies This assignment of the 3560 cm 1 peak is further investigated in isotopic dilution study using HOD/H 2 O mixture. We measured the SFG spectra from a HOD/H 2 O surfactant interface, where the OD stretching frequency will be decoupled from the rest of the system. SFG spectra from HOD/AOT interface in SSP and PPP polarization are plotted in Figure 8.7. Three Lorentzian fitting scheme has been applied to both the spectra with three frequencies at 2540 cm 1 , 2616 cm 1 and 2670 cm 1 (also shown in Figure 8.7). The first two peaks appear in the same region as of pure air/water interface (peak at Figure 8.7: Isotopic dilution study at the HOD/AOT interface. Homodyne detected SFG spectra of OD stretch at HOD/AOT interface. D 2 O and H 2 O were mixed at 1:3 ratio so that D 2 O/HOD/H 2 O is 1:6:9. SSP and PPP polarization spectra are plotted in the left and right panel respectively. Spectra were fit with three resonant lorentzians with center frequencies at 2540 cm 1 , 2616 cm 1 and 2670 cm 1 . Reproduced from Ref. 16. 3430 cm 1 and 3560 cm 1 ) when scaled (divided by the isotopic factor 1.36) 42 for an 161 OD-stretch instead of an OH-stretch. A very small intensity peak at 2670 cm 1 has been observed in both polarization as previously reported by Stiopkin et.al. 26 This peak had been tentatively assigned to D 2 O molecules with two donor hydrogen and one acceptor hydrogen bond, however its contribution to the overall spectra is negligible and it will not be discussed any further for our present study. The main contribution in the PPP spectra is from the 2616 cm 1 peak which is from a molecule with one donor D and one donor hydrogen bond probably to the surfactant head group. Orientational analysis for this OH/-OD stretch has been performed for both the water/AOT and HOD/AOT interface as discussed in the next section. 8.4 Orientational analysis Orientation of a particular molecular group on the surface can be determined from the amplitude ratio of SFG spectra in PPP and SSP polarizations of the same molecular tran- sitions. Gan et.al. 38, 39 studied orientation of water molecules at the air/water interface in details and analyzed the symmetry properties of four main peaks of OH-stretching region. Their polarization analysis reveals that the sharp peak at 3700 cm 1 and the broad peak at 3550 cm 1 have C 1V symmetry whereas the other two broad hydrogen bonded peak in the red side belong to C 2V symmetry group. The peak at 3550 cm 1 was assigned to the hydrogen bonded -OH of the interfacial water molecule with a free -OH bond. This peak also appears in the infrared spectra of water dimer 49 where the frequency for the donor OH bond is around 3550 cm 1 , which in turn supports the peak assignment. In a surfactant/water interface there is little chance of existence for a free water molecule in the surface or the concentration of those molecules will be so lit- tle that it could not be detected by SFG. However, dangling-OD was detected in lipid monolayer interfaces by Ma et.al 50 assigned to resign in the hydrophobic region in the 162 long hydrocarbon tails of the lipid monolayer. Other VSFG studies provide evidence of dangling OH bond of water at the water/hydrophobic surfaces. 51, 52 Based on this, we suggest that the peak at 3560 cm 1 in the present study arises due to water molecules having a donor hydrogen weakly bound to the surfactant head group. This allows us to perform orientational analysis on the 3560 cm 1 peak using a C 1V symmetry and use the hyperpolarizability tensor values similar to the free -OH group. 53, 54 We have followed the general equations as provided by Gan et. al. 38, 39 for our orientational analysis. Table 3 in the Supporting Information shows the values and the corresponding C and D parameters for the vibrational mode under consideration in SSP and PPP polarizations We simulated the orientational angle dependence of the OH group for our experimental geometry assuming a distribution for the tilt angle for the 3560 cm 1 peak. Figure 8.8 shows the change of SFG intensity with the tilt angle for both SSP and PPP polarization. The gray line shows the experimentally observed SSP/PPP ratio. Ori- entational analysis for both H 2 O/AOT and HDO/AOT interfaces (3560 cm 1 peak for the -OH stretch and 2616 cm 1 peak for the -OD stretch) yields the same tilt angle, 155 5 from the surface normal, indicating that the hydrogen of this group is on average pointing down into the water phase. We have also performed orientational analysis as- suming a Gaussian distribution for the distribution function, and find that the average tilt angle is nearly insensitive to the assumed distribution. This may be due to the fact that at the surfactant interface interaction between the polar head group and one of the OH group could restrict the orientational motion; also there is evidence of reduced rotational mobility of water at the interface 55 The assignment is in general agreement with the re- cently proposed assignment of Nojima et al. of the weakly H-bonded peak in observed in phospholipid monolayers. 47 163 Figure 8.8: Simulated SFG intensity of the 3560 cm 1 /2616 cm 1 bond at different tilt angle with distribution for SSP and PPP polarization at the air/H 2 O (Left) and at the air/HOD (Right) interface respectively. The gray line represents the experimental intensity ratio in PPP and SSP polarization (I PPP /I SSP ) values which are 2.46 and 1.82. Reproduced from Ref. 16. In conclusion, our study found similarities between the interfacial water structure in- side reverse micelles and at surfactant/water interfaces, as evidenced by the same spec- tral features observed in infrared spectra of the reverse micelles and SFG spectra at the surfactant interfaces. Lower frequency bands 3200-3400 cm 1 are dominant in the SSP SFG spectra at both ionic and non-ionic surfactant monolayers. The 3560 cm 1 peak is more pronounced in PPP polarization SFG and it is observed for both surfactants. The same peak is also observed in the infrared spectra of interfacial water in reverse micelles. Orientational analysis on the 3560 cm 1 peak provides the average tilt angle of the surface water molecules close to the surfactant surface to be around 155 with re- spect to the surface normal. Our present study suggests that the tentative assignment of this spectral feature to water molecules associated with the surfactant head group, with 164 the OH bond pointing down into the water phase. A more rigorous theoretical analysis with more precisely calculated values of molecular hyperpolarizability would be desir- able for future studies refining the spectroscopic assignment of the hydrogen bonding species present at various water interfaces. 8.5 Appendix Spectral fitting of the SFG data and decomposition of the spectra The fitting parameters for H 2 O/ AOT and HOD/ AOT solution spectra in SSP and PPP polarizations have been listed in Table 8.2 Table 8.2: Fitting parameters for VSFG spectra in the water stretching region at AOT/ water and AOT/ HOD interfaces measured in SSP and PPP polarizations. AOT/H 2 O AOT/ HOD SSP PPP SSP PPP ! 1 , cm 1 3230.0 5.0 3250.0 25.0 ! 2 , cm 1 3430.0 10.0 3466.0 5.0 2540.0 3.0 2540.0 10.0 ! 3 , cm 1 3560.0 10.0 3560.0 5.0 2616.0 5.0 2616.0 5.0 ! 4 , cm 1 3646.0 20.0 3630.0 10.0 2670.0 15.0 2670.0 25.0 B 1 106.0 5.0 1.0 0.5 B 2 71.0 5.0 20.0 1.0 29.5 1.0 4.0 0.5 B 3 15.0 3.0 37.0 0.5 16.0 1.0 29.0 0.5 B 3 -8.75 3.0 4.0 0.5 1.8 0.5 0.02 0.1 1 , cm 1 131.0 5.0 45.0 30.0 2 , cm 1 143.0 7.0 123.0 10.0 94.0 5.0 77.0 10.0 3 , cm 1 91.0 10.0 110.0 2.0 91.0 5.0 122.0 2.0 4 , cm 1 61.0 15.0 48.0 10.0 41.0 20.0 14.0 30.0 A NR 0.1020.01 0.160.05 0.0970.01 0.052 0.05 , rad 2.6 0.10 4.3 0.1 3.33 0.1 1.49 0.1 165 Three main Lorentzians for the air/AOT interface are at 3230 cm 1 , 3430 1 and 3560 1 . Contributions of each of these peaks towards the overall spectra are shown in Figure 8.9. (a) (b) Figure 8.9: 4 lorentzian fitting for the SFG spectra in SSP polarization (a) and in PPP polarization (b) at the H 2 O/AOT interface and the contribution of each peak towards the overall spectra. Vertical colored lines are placed as a guide to the center frequencies for each of the fitted lorentzians. (b). Reproduced from 16 . Parameters used in the SFG orientational analysis Orientational calculations for the 3560 cm 1 peak at the air/water interface have been done using values for the free-OH stretching peak 53, 54 and for the 2616 cm 1 peak at the air/HOD interface values of the OD peak6 has been used. The corresponding ’C’ and ’D’ parameters are listed in Table 8.3. 166 Table 8.3: ’C’ and ’D’ parameters used for orientational analysis. aac = bbc =0.32, ccc =1 for 3700 cm 1 and 3560 cm 1 peak and aac = bbc =0.40, ccc =1 for 2616 cm-1 peak. 3700 cm 1 peak (H 2 O) 3560 cm 1 peak (H 2 O) 2616 cm 1 peak (D 2 O C D C D C D SSP 0.52 0.153 0.52 0.16 0.4286 0.1416 PPP 10.09 -0.0206 10.14 -0.0215 25.85 -0.0062 167 Chapter 8 References [1] Levinger, Nancy E.; Costard, Rene; Nibbering, Erik T. J.; Elsaesser, Thomas Ultra- fast energy migration pathways in self-assembled phospholipids interacting with confined water J. Phys. Chem. A 2011, 115, 11952–11959. [2] Fayer, Michael D Dynamics of water interacting with interfaces, molecules, and ions Acc. Chem. Res. 2011, 45, 3–14. [3] Ni, Yicun; Gruenbaum, Scott M.; Skinner, James L. Slow hydrogen-bond switch- ing dynamics at the water surface revealed by theoretical two-dimensional sum- frequency spectroscopy Proc. Natl. Acad. Sci. 2013, 110, 1992–1998. [4] Roy, S; Gruenbaum, SM; Skinner, JL Theoretical vibrational sum-frequency gen- eration spectroscopy of water near lipid and surfactant monolayer interfaces J. Chem. Phys. 2014, 141, 18C502. [5] Pal, Samir Kumar; Peon, Jorge; Bagchi, Biman; Zewail, Ahmed H Biological wa- ter: femtosecond dynamics of macromolecular hydration J. Phys. Chem. B 2002, 106, 12376–12395. [6] Walde, Peter; Giuliani, Anna Maria; Boicelli, C Andrea; Luisi, Pier Luigi Phospholipid-based reverse micelles Chem. Phys. Lipids 1990, 53, 265–288. [7] Balasubramanian, Sundaram; Pal, Subrata; Bagchi, Biman Hydrogen-bond dy- namics near a micellar surface: origin of the universal slow relaxation at complex aqueous interfaces Phys. Rev. Lett. 2002, 89, 115505. [8] Harpham, Michael R; Ladanyi, Branka M; Levinger, Nancy E; Herwig, Kenneth W Water motion in reverse micelles studied by quasielastic neutron scattering and molecular dynamics simulations J. Chem. Phys. 2004, 121, 7855–7868. [9] Willard, Dale M; Riter, Ruth E; Levinger, Nancy E Dynamics of polar solvation in lecithin/water/cyclohexane reverse micelles J. Am. Chem. Soc. 1998, 120, 4151– 4160. [10] Tan, Howe-Siang; Piletic, Ivan R; Riter, Ruth E; Levinger, Nancy E; Fayer, MD Dynamics of water confined on a nanometer length scale in reverse micelles: ul- trafast infrared vibrational echo spectroscopy Phys. Rev. Lett. 2005, 94, 057405. [11] Skinner, JL; Pieniazek, PA; Gruenbaum, SM Vibrational spectroscopy of water at interfaces Acc. Chem. Res. 2011, 45, 93–100. 168 [12] Rhee, Young Min; Sorin, Eric J; Jayachandran, Guha; Lindahl, Erik; Pande, Vi- jay S Simulations of the role of water in the protein-folding mechanism Proc. Natl. Acad. Sci. 2004, 101, 6456–6461. [13] Levy, Yaakov; Onuchic, Jos N Water mediation in protein folding and molecular recognition Annu. Rev. Biophys. Biomol. Struct. 2006, 35, 389–415. [14] England, Jeremy L; Lucent, Del; Pande, Vijay S A role for confined water in chaperonin function J. Am. Chem. Soc. 2008, 130, 11838–11839. [15] Argyris, Dimitrios; Cole, David R; Striolo, Alberto Ion-specific effects under con- finement: the role of interfacial water ACS nano 2010, 4, 2035–2042. [16] Dutta, Chayan; Svirida, Anton; Mammetkuliyev, Muhammet; Rukhadze, Marina; Benderskii, Alexander V Insight into water structure at the surfactant surfaces and in microemulsion confinement J. Phys. Chem. B 2017, 121, 7447–7454. [17] Corcelli, SA; Skinner, JL Infrared and raman line shapes of dilute hod in liquid h2o and d2o from 10 to 90 c J. Phys. Chem. A 2005, 109, 6154–6165. [18] Auer, B; Kumar, R; Schmidt, JR; Skinner, JL Hydrogen bonding and raman, ir, and 2d-ir spectroscopy of dilute hod in liquid d2o Proc. Natl. Acad. Sci. 2007, 104, 14215–14220. [19] Bakker, HJ; Skinner, JL Vibrational spectroscopy as a probe of structure and dy- namics in liquid water Chem. Rev. 2009, 110, 1498–1517. [20] Tainter, CJ; Ni, Y; Shi, L a; Skinner, JL Hydrogen bonding and oh-stretch spec- troscopy in water: Hexamer (cage), liquid surface, liquid, and ice J. Phys. Chem. Lett. 2012, 4, 12–17. [21] Buch, V; Tarbuck, T; Richmond, GL; Groenzin, H; Li, I; Shultz, MJ Sum fre- quency generation surface spectra of ice, water, and acid solution investigated by an exciton model J. Chem. Phys. 2007, 127, 204710. [22] Loparo, Joseph J; Roberts, Sean T; Nicodemus, Rebecca A; Tokmakoff, Andrei Variation of the transition dipole moment across the oh stretching band of water Chem. Phys. 2007, 341, 218–229. [23] Auer, BM; Skinner, JL Vibrational sum-frequency spectroscopy of the liquid/vapor interface for dilute hod in d2o J. Chem. Phys. 2008, 129, 214705. [24] Auer, BM; Skinner, JL Water: Hydrogen bonding and vibrational spectroscopy, in the bulk liquid and at the liquid/vapor interface Chem. Phys. Lett. 2009, 470, 13–20. 169 [25] Pieniazek, P. A.; Tainter, C. J.; Skinner, J. L. Interpretation of the water surface vibrational sum-frequency spectrum J. Chem. Phys. 2011, 135, 044701. [26] Stiopkin, Igor V .; Weeraman, Champika; Pieniazek, Piotr A.; Shalhout, Fadel Y .; Skinner, James L.; Benderskii, Alexander V . Hydrogen bonding at the water sur- face revealed by isotopic dilution spectroscopy Nature 2011, 474, 192–195. [27] Shen, Yuen Ron; Ostroverkhov, Victor Sum-frequency vibrational spectroscopy on water interfaces: Polar orientation of water molecules at interfaces Chem. Rev. 2006, 106, 1140–1154. [28] Ji, N; Ostroverkhov, V; Tian, CS; Shen, YR Characterization of vibrational reso- nances of water-vapor interfaces by phase-sensitive sum-frequency spectroscopy Phys. Rev. Lett. 2008, 100, 096102. [29] Stiopkin, Igor V; Jayathilake, Himali D; Bordenyuk, Andrey N; Benderskii, Alexander V Heterodyne-detected vibrational sum frequency generation spec- troscopy J. Am. Chem. Soc. 2008, 130, 2271–2275. [30] Roke, Sylvie; Bonn, Mischa; Petukhov, Andrei V Nonlinear optical scattering: The concept of effective susceptibility Phys. Rev. B 2004, 70, 115106. [31] deAguiar, Hilton B; Scheu, Rdiger; Jena, Kailash C; deBeer, Alex GF; Roke, Sylvie Comparison of scattering and reflection sfg: a question of phase-matching Phys. Chem. Chem. Phys. 2012, 14, 6826–6832. [32] Piletic, Ivan R; Moilanen, David E; Spry, DB; Levinger, Nancy E; Fayer, MD Testing the core/shell model of nanoconfined water in reverse micelles using linear and nonlinear ir spectroscopy J. Phys. Chem. A 2006, 110, 4985–4999. [33] Sechler, Timothy D; DelSole, Edward M; Dek, John C Measuring properties of interfacial and bulk water regions in a reverse micelle with ir spectroscopy: a vol- umetric analysis of the inhomogeneously broadened oh band J. Colloid. Interface. Sci. 2010, 346, 391–397. [34] Gragson, DE; McCarty, BM; Richmond, GL Ordering of interfacial water molecules at the charged air/water interface observed by vibrational sum frequency generation J. Am. Chem. Soc. 1997, 119, 6144–6152. [35] Gragson, DE; Richmond, GL Probing the structure of water molecules at an oil/water interface in the presence of a charged soluble surfactant through isotopic dilution studies J. Phys. Chem. B 1998, 102, 569–576. 170 [36] Nihonyanagi, Satoshi; Yamaguchi, Shoichi; Tahara, Tahei Direct evidence for orientational flip-flop of water molecules at charged interfaces: A heterodyne- detected vibrational sum frequency generation study J. Chem. Phys. 2009, 130, 204704. [37] Maitra, Amarnath Determination of size parameters of water-aerosol ot-oil reverse micelles from their nuclear magnetic resonance data J. Phys. Chem. 1984, 88, 5122–5125. [38] Gan, Wei; Wu, Dan; Zhang, Zhen; Feng, Ran-ran; Wang, Hong-fei Polarization and experimental configuration analyses of sum frequency generation vibrational spectra, structure, and orientational motion of the air/water interface J. Chem. Phys. 2006, 124, 114705. [39] Gan, Wei; Wu, Dan; Zhang, Zhen; Guo, Yuan; Wang, Hong-fei Orientation and motion of water molecules at air/water interface Chinese J. Chem. Phys. 2006, 19, 20–24. [40] Raymond, EA; Tarbuck, TL; Richmond, GL Isotopic dilution studies of the va- por/water interface as investigated by vibrational sum-frequency spectroscopy J. Phys. Chem. B 2002, 106, 2817–2820. [41] Richmond, GL Molecular bonding and interactions at aqueous surfaces as probed by vibrational sum frequency spectroscopy Chem. Rev. 2002, 102, 2693–2724. [42] Tian, Chuan-Shan; Shen, Y Ron Isotopic dilution study of the water/vapor inter- face by phase-sensitive sum-frequency vibrational spectroscopy J. Am. Chem. Soc. 2009, 131, 2790–2791. [43] Feng, Ran-ran; Guo, Yuan; Lu, Rong; Velarde, Luis; Wang, Hong-fei Consis- tency in the sum frequency generation intensity and phase vibrational spectra of the air/neat water interface J. Phys. Chem. A 2011, 115, 6015–6027. [44] Watry, Mark R; Tarbuck, Teresa L; Richmond, Geraldine L Vibrational sum- frequency studies of a series of phospholipid monolayers and the associated water structure at the vapor/water interface J. Phys. Chem. B 2003, 107, 512–518. [45] Tyrode, Eric; Johnson, C Magnus; Kumpulainen, Atte; Rutland, Mark W; Claes- son, Per M Hydration state of nonionic surfactant monolayers at the liquid/vapor interface: Structure determination by vibrational sum frequency spectroscopy J. Am. Chem. Soc. 2005, 127, 16848–16859. [46] Mondal, Jahur A; Nihonyanagi, Satoshi; Yamaguchi, Shoichi; Tahara, Tahei Three distinct water structures at a zwitterionic lipid/water interface revealed by heterodyne-detected vibrational sum frequency generation J. Am. Chem. Soc. 2012, 134, 7842–7850. 171 [47] Nojima, Yuki; Suzuki, Yudai; Yamaguchi, Shoichi Weakly hydrogen-bonded water inside charged lipid monolayer observed with heterodyne-detected vibrational sum frequency generation spectroscopy J. Phys. Chem. C 2017. [48] Ishiyama, Tatsuya; Terada, Daichi; Morita, Akihiro Hydrogen-bonding structure at zwitterionic lipid/water interface J. Phys. Chem. Lett. 2015, 7, 216–220. [49] Page, Ralph H; Frey, Jeremy G; Shen, Y-R; Lee, Yuan-Tseh Infrared predissoci- ation spectra of water dimer in a supersonic molecular beam Chem. Phys. Lett. 1984, 106, 373–376. [50] Ma, Gang; Chen, Xiangke; Allen, Heather C Dangling od confined in a langmuir monolayer J. Am. Chem. Soc. 2007, 129, 14053–14057. [51] Scatena, LF; Brown, MG; Richmond, GL Water at hydrophobic surfaces: weak hydrogen bonding and strong orientation effects Science 2001, 292, 908–912. [52] Du, Quan; Freysz, Eric; Shen, Y Ron Surface vibrational spectroscopic studies of hydrogen bonding and hydrophobicity Science 1994, 264, 826–827. [53] Murphy, William F The rovibrational raman spectrum of water vapour v 1 and v 3 Mol. Phys. 1978, 36, 727–732. [54] Wei, Xing; Shen, YR Motional effect in surface sum-frequency vibrational spec- troscopy Phys. Rev. Lett. 2001, 86, 4799. [55] Bakulin, Artem A; Pshenichnikov, Maxim S Reduced coupling of water molecules near the surface of reverse micelles Phys Chem Chem Phys 2011, 13, 19355– 19361. 172

Abstract (if available)

Abstract The unique ability of water to form extended hydrogen bond network in the bulk environment is responsible for many unusual physical properties of liquid water. This extended hydrogen bond network is terminated at the interface producing an inherent inhomogeneity in structure, orientation and distribution of hydrogen bonds at the interface. Because of these intrinsic properties interfacial water plays a signiﬁcant role in a variety of systems, e.g., biological membranes, environmental interfaces, electrochemical interfaces, heterogeneous catalysis etc. Understanding the basic interactions of water at the interface is crucial. We have used surface sensitive spectroscopic techniques to investigate into the nature of hydrogen bonds and molecular orientations at the interface. This thesis is mainly focused on assigning the water bending mode at the air/water interface, establishing the bend mode as a complementary probe of hydrogen bonding and investigating molecular orientation of different hydrogen bonded species at neutral interfaces, chemically and electrochemically charged surfaces and in micro-emulsion conﬁnement.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Molecular orientation from sum frequency generation spectroscopy: case study of water and methyl vibrations

PDF

Asymmetric response and the reorientation of interfacial water with respect to applied electric fields

PDF

Vibrational sum frequency spectroscopy of molecules on metal, semiconductor, and aqueous surfaces

PDF

Vibrational sum frequency generation studies of electrode interfaces

PDF

Water partitioning between the bulk and an electrode surface

PDF

Probing charge transfer and electric fields at the bulk and interface using vibrational spectroscopy

PDF

Understanding organic and inorganic semiconductor surfaces by coherent nonlinear spectroscopy

PDF

Nonlinear spectroscopy of chemical phenomena at interfaces: carrier trapping and local electric fields

PDF

Dissociation energy and dynamics of HCl and aromatic-water clusters

PDF

Interfacial polarization and photobasicity: spectroscopic studies of the electrode environment and photochemical effects

PDF

Vibrational spectroscopy and molecular orientation at interfaces: correlating the structure of a material's surface to its properties

PDF

Interfacial vibrational and electronic excited states of photovoltaic materials probed using ultrafast nonlinear spectroscopy

PDF

Dissociation energy and dynamics of water clusters

PDF

Monitoring local electric fields, charge densities at electrode surfaces using in-situ graphene/surface enhanced Raman spectroscopy (GERS/SERS) based Stark-shifts

PDF

Potential-induced formation and dissociation of adducts at interfaces and adsorption of phenols onto metal electrodes

PDF

Ultrafast dynamics of excited state intra- and intermolecular proton transfer in nitrogen containing photobases

PDF

Photodissociation dynamics and energetics of HCl-(H₂O)₃

PDF

Elucidating local electric fields and ion adsorption using benzonitrile: vibrational spectroscopic studies at interfaces and in bulk

PDF

Following redox chemistry and excited state dynamics in solution using liquid jet photoelectron spectroscopy

PDF

Carbon-hydrogen bond activation: radical methane functionalization; unactivated alkene coupling; saccharide degradation; and carbon dioxide hydrogenation

Asset Metadata

Creator Dutta, Chayan (author)

Core Title Molecular orientation and structure of hydrogen bonds at water interfaces

School College of Letters, Arts and Sciences

Degree Doctor of Philosophy

Degree Program Chemistry

Publication Date 11/10/2018

Defense Date 10/23/2017

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

Tag air/water interface,hydrogen bonds,molecular orientation,non-linear spectroscopy,OAI-PMH Harvest,SFG,vibrational spectroscopy,water surface

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Benderskii, Alexander V. (committee chair), Dawlaty, Jahan M. (committee member), Nakano, Aiichiro (committee member)

Creator Email cdutta@usc.edu,som06july@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c40-453709

Unique identifier UC11264126

Identifier etd-DuttaChaya-5888.pdf (filename),usctheses-c40-453709 (legacy record id)

Legacy Identifier etd-DuttaChaya-5888.pdf

Dmrecord 453709

Document Type Dissertation

Rights Dutta, Chayan

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