University of Southern California Dissertations and Theses

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

(USC Thesis Other)

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

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content Following redox chemistry and excited state dynamics in solution using liquid jet photoelectron spectroscopy by Anirban Roy 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 2015 Copyright 2015 Anirban Roy ii Dedicated to my loving parents, Chhabi Roy and Jayanta Roy Thank you for your unconditional love and support iii ACKNOWLEDGEMENTS Living in Los Angeles and pursuing my PhD at USC for last 6 years has been a wonderful experience. This journey to the wonderland of physical chemistry would not have been possible without all the help and encouragement from my friends, colleagues and faculties. I would like to thank the entire Bradforth Research group to make journey so enjoyable. Thanks to Dr. Steve Bradforth, “THE BIG BOSS,” for gambling on having me in his research group, I am sure he is as much relieved as I am for not destroying everything in the laser lab. Thanks for all the knowledge and insight you passed on to me, but I am sure I will never succeed on getting “The TOUCH” on aligning the ReGa in SSC 714. I have to admit that at times it was rather annoying when I failed to troubleshoot the problem sometimes for a full day if not more and then have you spent a minute or two to think about it and everything gets back to normal following your ideas, on the flipside, however, it was still rather reassuring to know I had that option. Steve, thanks for making me learn more than just chemistry. Being a graduate student for years, I believe USC has undoubtedly one of the best physical chemistry division in the whole country. It is not only the state of the art research facilities, or the world class course curriculum, not even the two Nobel laureates but the collaborative atmosphere among the peers and the mentors made it one of the premiere institutes,. I am really grateful to USC for providing such an amiable environment to nurture young scientists. During my budding years in the graduate school, I was lucky enough to have Diana Suffern as my mentor in the Bradforth lab. Although she was a final year graduate student, she was kind and patient enough to teach me the basics of optics as well as explaining jokes she often cracked while we are looking for signals. Years later I figured how painful it is to explain my own jokes (Thanks to Gaurav!!). As the years passed, I was fortunate to work with Tom iv Zhang, the only representative from the far east in our truly international research group (please see below for details). Tom always made me feel like working with a fellow colleague as he accepted my stupidity quite gracefully while working with the optics. He was kind enough to teach my how to run gravity jet and regular laser operation in SSC 716. I hope I won't forget the steps even if I try. Then there was the Russian, the French and the German invasion in the British research group in the America. As I always said our group meeting won't be as colorful without Konstantin as an audience or as the performer (yes, not a speaker, but strictly a performer). I really appreciate his effort to show up in full formal dress-code on his group meeting days for last four years. Unfortunately, he could not convince any of us to return the favor when he is in the audience. Elsa, our French post-doc always charmed me with her cute accent and homemade pies and obviously when she dug deep into the photophysics of hybrid solar cell materials. And I was truly fortunate to partner with Dr. Robert Seidel on the liquid microjet project. It was really a remarkable journey together starting from putting the laser table together to writing up the manuscript (I still cannot believe we did it!!). Robert left really a deep impression on me or rather on the whole Bradforth group by his mean jokes, birthday cards, chocolates from overseas and obviously his meticulous and super-perfect machine designing. I am really happy that he is still continuing his journey to explore and exploit the intricacies of the liquid jet photoemission technique. Over the years I have been in an extended family of graduate students and post-docs who made sure that my life was never boring. I can never thank enough Debashree Di, Suman Da, Amit Da, Parichita Di, Atanu, Piyush, Subodh, Chayan, Subhashis, Dhritiman and (fnu) Gaurav Kumar. Not to mention how much happiness REMPI brought to our lives (Thanks to Amit Da v and Parichita Di). Purnim and Saptaparna for being my best friends for almost eleven years now, it would have been difficult without you guys around. I literally could not have made it through without the help of our Machine shop staffs, Don and Ramon. Without them I would have left the program midway since buying whole spectrometer would have cost twice a grad student yearly scholarship. Thanks to them, we made it cheap and twice faster and definitely better!! So, I owe them my PhD. This acknowledgement would be incomplete if I don't thank the person who runs the Chemistry department, Michele Dea. Although she is the busiest person but she still solves all our problems instantly. Without her I would have been stuck with those administrative issues for forever. I would also like to thank Katie McKissick and Magnolia Benitez for their advice as graduate advisers. I am grateful to National Science Foundation (NSF) for financially supporting the liquid microjet photoelectron spectroscopy project over the years (Grant: CHE-1301465). Most importantly, I would like to thank my parents who always supported my decision and provided me the courage to pursue education abroad. Both of them always encouraged me to become independent and I hope that I will fulfill her dreams. Finally, I would like to acknowledge my best friend and lab partner Saptaparna Das for being always there for me. We joined the Bradforth group in the same year. And because of her I was able to solve many difficult experimental and scientific problems. She was kind enough to listen to my n number of practice talks for screening and qualifying examination. She is the person who always heard my issues and cheer me up. I could fill up pages with gratitude towards her and I hope that this companionship with her stays for forever. vi TABLE OF CONTENTS List of Figures viii List of Tables xv Abstract xvi Chapter 1. Introduction to Liquid Jet Photoelectron Spectroscopy 1 1.1. Principles of Photoelectron Spectroscopy 1 1.2. Time Resolved Photoelectron Spectroscopy 3 1.3. Photoelectron Spectroscopy in liquids 4 1.4. Application of liquid jet PES: Investigating oxidative DNA damage and repair 7 1.5. Thesis Objectives 8 Chapter 1 References 11 Chapter 2. Building the liquid microjet photoelectron spectrometer 17 2.1. Introduction 17 2.2. Experimental Design 17 2.2.1. Liquid Microjet: Performance in Vacuum and Compatibility with Photoelectron Spectroscopy 17 2.2.2. Building liquid-jet Photoelectron Spectrometer 19 2.2.3. Designing Time of Flight spectrometer (TOF) 24 2.2.3.1. Principle of time of flight detection 24 2.2.3.2. Energy resolution 25 2.2.3.3. Collection Efficiency: Field free TOF vs. Magnetic bottle TOF 26 2.2.3.4. Magnetic bottle TOF: How it works 28 2.2.4. Laser system 29 2.2.5. Detection strategy 31 2.2.6. Experimental strategy for transient photoelectron measurement 31 2.2.7. Spectrometer Calibration 32 2.8. Project Goals Chapter 2 References 35 36 Chapter 3. Measuring the Vertical and Adiabatic Ionization Energies of Aromatic Amino Acids in Water 38 3.1. Introduction 38 3.2. Experimental 41 3.3. Results and Discussion 44 3.4. Conclusion 55 Chapter 3 References 58 vii Chapter 4. Chapter 4. Relaxation dynamics of hydrated electron: comparison of CTTS vs. CTTS* excitation of aqueous Iodide solution 63 4.1. Introduction 63 4.2. Experimental 65 4.3. Results and Discussion 67 4.3.1. Excitation to the first CTTS band 67 4.3.2. Excitation to the second CTTS (CTTS*) band 72 4.4. Conclusion 79 Chapter 4 References 82 Chapter 5. Exploring the long time excited state dynamics of phenol in aqueous solution 86 5.1. Introduction 86 5.2. Experiment 89 5.3. Results and Discussion 91 5.4. Conclusion 97 Chapter 5 References 98 viii List of Figures Figure 1.1: Schematic illustration of photoelectron spectroscopy. The energy level diagram shows ionization from the higher lying molecular orbitals, HOMO (Highest Occupied Molecular Orbitals) and HOMO-1. The incident photon energy (h ), electron binding energy (eBE) and electron kinetic energy (eKE) are related via energy conservation: eBE = h  - eKE. According to Koopmans' theorem the binding energy of the outgoing electron is a measure of the parent orbital energy, i.e., eBE = - E MO 2 Figure 1.2: Schematic illustration of time resolved photoelectron spectroscopy (TRPES). After photoexcitation to the excited state with the pump pulse, the probe ejects electron and the molecule ends up in cationic state. The kinetic energy distribution of the outgoing electrons is shown the right panel. By varying the delay ( t) between pump and probe pulses, excited state dynamics can be probed by monitoring the evolution of the kinetic energy spectra. For example, if the molecule undergoes vibrational relaxation or nonadiabatic transition to another electronic state ( ), observed change in electron kinetic energy will reflect the dynamics. 4 Figure 1.3: Variation of effective electron attenuation length (EAL) of with electron kinetic energy for liquid water from two recent studies using x-ray photoelectron spectroscopy with liquid microjet technique. For comparison 'universal inelastic mean free path (IMFP) curve' is also shown. 6 Figure 2.1: Schematic 3D model of the home built photoelectron spectrometer on laser table. The change of water-vapor density as a function of radial distance from the microjet is illustrated in the inset 18 Figure 2.2: Inside the source chamber. Interaction region showing liquid jet, skimmer and the permanent magnet configuration (a). Formation of ice-needle and subsequent jet- blockage (b). An in vacuo ice-cutter is implemented to break the ice-needles before they reach the nozzle (c,d). The 'ice-cutter' is driven by a DC power supply continuously during the experiment (d). 20 ix Figure 2.3: Schematic of the liquid delivery system. See text for detailed description. 22 Figure 2.4: Mounted MCP detector assembly (a). Enlarged view of the multichannel plate (b). To enhance the detection efficiency of slow photoelectrons a grounded copper mesh is incorporated in front of the MCP (c). The schematic of MCP operation before and after installing 'copper mesh' is also shown (d). 23 Figure 2.5: The circuit diagram of the home-built voltage divider box. The electrical components and circuit configuration are shown for the configuration with grounded copper mesh (Figure 2.4 c,d). The resistors and capacitors have average accuracy of ±5% of the specified value. 24 Figure 2.6: Collection efficiency of a field free time of flight spectrometer. The parameters considered here are based on the instrument built at USC 26 Figure 2.7: Magnetic field time of flight spectrometer. The variation of magnetic field is plotted against the distance from the interaction region. The gradient of the magnetic field is high in the 'source chamber' region, however, the magnetic field is homogenous in the flight tube. 27 Figure 2.8: Illustrating electron trajectory in homogenous (a) and in inhomogenous (b) magnetic field. In case of magnetic bottle spectrometer the inhomogeniety effectively parallelize electron trajectories along the magnetic field axis (flight axis). For example, if θ i =   , B i = 1T and B f = 1mT then θ f = 1.8 0 , i.e., almost parallel to the flight axis 28 Figure 2.9: Experimental laser set up for one beam and two beam (transient) photoemission experiment. M: Mirror; DC: Dichroic Mirror. Generation of the 200 nm pulses and the translation stage is not shown 30 Figure 2.10: Transient photoemission detection scheme using two choppers. The dark shades of the chopped signals denote blocked pulses. See text for details 32 x Figure 2.11: Calibration of the photoelectron spectrometer with Nitric Oxide (NO) in presence of the water jet. a) Energy level diagram of the NO states involved in the resonant photoionization using 200 nm ultrashort pulses. The observed vibrational structure (b and d) represents vibronic transitions to the cationic surface (D 0 ). The time of flight values for each peak is calibrated against the literature kinetic energy values and fitted with eqn. 2.6. The calibrated kinetic energy spectra is shown in d). The peak highlighted with the asterisk originates from multiphoton ionization of the water vapor in the vicinity. 34 Figure 2.12: Estimation of streaming potential of water jet containing ~30 mM NaCl. Resonant photoemission of Nitric Oxide (NO) is performed with 200 nm in presence of the water jet. The kinetic energy of the NO(v=0)→NO + (v'=2) band is plotted against the position of the jet (x) with respect to the laser focus. The data is fitted with the following function: eKE(x)=eKE( )-(L/L+x)φ, where L, φ are the distance of the laser focus from skimmer entrance and the streaming potential respectively. From the above fitting the streaming potential is found to be 112±30 mV. 35 Figure 3.1: Schematic illustration of X-ray and resonant two photon ionization photoemission process. High energy X-ray radiation ionizes from all the valence orbitals and produces photoelectrons at high kinetic energies (yellow arrows). On the contrary, in resonant photoemission, ionization takes place only from those orbitals which are involved in the electronic excitation (purple arrows). Since the total energy deposited in the latter process is much smaller, it yields lower kinetic energy electrons. 40 Figure 3.2: X-ray photoemission (a-c) and resonant two photon ionization (d-f) spectra of the aromatic amino acids in water (solid circles). Individual Gaussian fits and the total fit are shown in blue dashed lines and solid red lines respectively. following 500 nm (green) and 700 nm (red) excitation. 44 Figure 3.3: UV-VIS absorption spectra of aromatic amino acids in water. For comparison absorption spectra of Benzene, Indole and Phenol (in water) absorption spectra is also shown. Solid violet band shows the excitation beam spectrum centered at 267 nm (FWHM~3 nm). 46 xi Figure 3.4: Estimation of Adiabatic Ionization Energies (AIE) of aromatic amino acids. Intensity axis is plotted in semilog scale to emphasize the noise level at low signal intensity. The red line indicates the rising edge of the signal and the blue region denotes the spread in noise levels (see text). Due to reference subtraction, the dynamic range in the signal intensity is low (<400:1) in XPS compared to R2PI measurements (>2000:1) which makes the assignmenet of noise level more ambiguious in XPS data 48 Figure 3.5: R2PI-PES spectra of Cytidine (Cyt), Adenosine (Ado) and deoxyguanosine monophosphate (dGMP) in water. AIE has been estimated as described in the text. We found that our measurement overestimates the standard oxidation potential by ~ 0.6 V(± 0.2 V) when compared to the values reported in ref 13 (cf Table 2). See text for explanation. 51 Figure 3.6: Schematic representation of the resonant two photon ionization process. With delta pulse excitation (Case 1), the initial wavepacket launched in the S 1 surface (FC region) undergoes ionization prior to any nuclear and electronic relaxation and reaches final state D 0 (radical cation). The excess energy is manifested as the electron kinetic energy. Maximum available kinetic energy corresponding to S 0 (v=0)→D 0 (v=0) transition is shown as KE 0 max . The adiabatic ionization energy can be defined as AIE(0)=2*h - KE 0 max . In case of finite pulse laser (Case 2), relaxation takes place within the pulse width ( t). Observed adiabatic ionization energy AIE( t)=2*h -KE  t max . Since KE 0 max ≥ KE  t max , observed ionization threshold for Case 2 will always be equal to or higher than the actual value (no relaxation, Case 1). 53 Figure 4.1: Evolution of integrated photoelectron signal at 251 nm excitation (blue solid circles) with pump-probe delay. Red solid line shows best fit with sum of three exponential functions convoluted with instrument response (~160 fs). The green dashed line shows simulation based on modified Staib-Borgis model (see text for details). 67 xii Figure 4.2: a) Transient photoemission spectra of aqueous iodide with 251 nm pump at different delays (solid lines). Red dashed lines are Gaussian fitting functions from the global fit (see text for details). The broad photoemission band at earlier time (<0.5 ps) undergoes spectral blue shifting and narrowing. Evolution of the band center (binding energy, eBE) and the full width half maximum (FWHM) of the Gaussian fitting function are shown in b) and c), respectively. Red dash-dot lines are exponential fits and yield characteristic relaxation time scales and magnitude (see below). 70 Figure 4.3: a) Full transient photoelectron spectrum of ~100mM sodium iodide in water. Global fitting of the data and the residual are shown in b) and c), respectively. d) Squared value of the residual from global fitting algorithm at each delay point. At delay>0.2 ps, the average residual 2 value is 0.06 71 Figure 4.4: Kinetic scheme of electron dynamics subsequent to CTTS excitation (at 251 nm) of aqueous iodide solution. Parameters are described in the text. 72 Figure 4.5: Time resolved photoelectron spectra of 100mM NaI solution at different pump probe delay times. Solid red and blue lines show photoemission bands at 251 nm and 200 nm excitation energy respectively. A shoulder between 1-2.5 eV binding energy is observed at early times (t < 2 ps) with 200 nm pump. Prompt disappearance of this band indicates the presence of a short lived transient not observed for lowest energy CTTS excitation (251 nm). 73 Figure 4.6: Photoelectron spectrum at 200 fs delay with 200 nm pump (solid line with solid circles). The spectrum is fitted with sum of two Gaussians (solid red) representing p and s-state of hydrated electron (see text for details). The ratio of the area under each Gaussian represents the relative population of the two states upon photoexcitation at 200 fs (assuming identical photoionization cross section). 74 xiii Figure 4.7: a) Transient photoemission spectra of aqueous iodide with 200 nm pump at different delays (solid lines). Red dash-dot lines are sum of two Gaussian functions from the global fit (see text for details). The band centered at 2.1 eV binding energy decays within 500 fs (cf. Figure 4.8); no spectral evolution is observed. However, the higher binding energy peak undergoes spectral blue shifting and narrowing as shown in b) and c), respectively. Red dotted lines are exponential fits to the data. 75 Figure 4.8: Integrated photoemission signal of the two Gaussian peaks at 200 nm pump as a function of pump probe delay. a) Variation of the lower binding energy peak population (green solid circle) fitted with a single exponential function with time constant of 350 fs (red broken line). b) Population decay of the higher binding energy peak (solid square). The experimental data is simulated following modified Staib-Borgis model (orange broken line) as discussed in the main text (eqn. 2) 77 Figure 4.9: Kinetic scheme of electron dynamics subsequent to CTTS* excitation (at 200 nm). 79 Figure 5.1: Experimental set up for transient absorption measurements. The upper panel shows optical layout for the broadband probe (270-700 nm) generation. M1, M2, M3: Variable density filter; L: Plano convex lens The far UV probe (270-380 nm) is generated by focusing 400 nm onto a rotating CaF2 disc. Visible probing range is produced by bypassing the 400 nm BBO crystal (flipping FM1 and FM2). The lower panel illustrates pump-probe measurement scheme 90 Figure 5.2: Transient absorption spectra of 20 mM aqueous phenol solution. Time slices at short (  10 ns) and long delays (˃ 10 ns) are shown in a) and b) respectively. Quenching experiment with HCl solution confirms the presence of hydrated electron as shown in c). Kinetic traces (and exponential fits) at 600 nm probe wavelength shown in d). The delay axis in d) is broken between 50-250 ns to emphasize the two different delay scales. 92 xiv Figure 5.3: Comparing the transient absorption signals of 20 mM phenol solution containing 250 mM Cesium Chloride (CsCl) with the control solution containing equal amount of Sodium Chloride (NaCl) at 10 and 100 ns delay (a). After adding CsCl we observed an initial decrease in the transient absorption signal underlying the radical band (<5 ns), followed by an increase at times 10 ns<t<100 ns and finally a decay at longer time (b). The shaded areas in a) illustrate the increase and subsequent decay of the TA signal for delays>10 ns 93 Figure 5.4: a) Transient absorption spectra of aqueous phenol solution with far UV probe at selective pump-probe delay times. b) Extracting T-T absorption spectrum from nanosecond TA spectra (see text for details). For comparison reported triplet absorption spectrum 7 is included. An arbitrary scaling factor and offset has been incorporated to the literature spectrum. 95 Figure 5.5: Kinetic scheme of the photo-physical/chemical processes of aqueous phenol solution. k rad indicates radiative decay rate, k 1 and k 2 indicate rates of parallel reactions forming the triplet (T 1 ) and radical (R) (along with the solvated electron but not included for analysis, see text for details) respectively. k3 denotes pseudo-first order rate constant of triplet quenching by oxygen. 96 Figure 5.6: a) Experimental time slices (solid lines) and time slices from the global fitting (red dashed lines) at selective pump-probe delay times. b) Kinetics of each transient species from the global fitting model discussed above. 97 xv List of Tables Table 3.1: Vertical Ionization Energy of the aromatic amino acids in aqueous solutions. The standard error of VIE is ±0.1 eV. Average FWHM of the photoelectron bands in solutions is 0.9±0.1 eV. For comparison gas phase values are also included from ref 33-35 57 Table 3.2. Adiabatic Ionization Energies (in eV) and standard one electron oxidation potentials (vs SHE, in V) of important biomolecules in aqueous solution. Error associated with AIE is ±0.1 V 57 Table 4.1: Simulation parameters to model population dynamics using modified Staib-Borgis model (eqn. 4.1-4.3). *For 200 nm excitation, we only simulate the population for the higher binding energy peak 81 Table 4.2: Fitting parameters for relaxation dynamics. *For 200 nm excitation, we only fit the higher binding energy peak 81 xvi Abstract Photoelectron spectroscopy, developed as a spectroscopic technique based on photoelectric effect, provides useful insights into chemical composition, molecular electronic structure and reaction dynamics. Although traditionally used for solid substrates and ultracold molecular beams, recent development of liquid microjet technique expands the scope of electron detection technique to the realm of pure liquids and liquid solutions. Although liquid microjet technique unfolds a new direction of spectroscopic endeavor, the application and widespread adoption has been limited primarily due to the highly demanding infrastructure and maintenance cost associated with high energy synchrotron radiation as the light source. This dissertation describes an alternative design and realization of a liquid jet photoelectron spectrometer coupled with ultrafast lasers as light sources to study liquid solution with greater selectivity and sensitivity compared to traditional x-ray photoemission measurements. Using this novel technique along with complementary transient absorption measurements, the following chapters delve into the details of redox chemistry of biomolecules as well as explore the excited state dynamics of solvated electron and aqueous phenol solutions. 1 Chapter 1. Introduction to Liquid Jet Photoelectron Spectroscopy 1.1 Principles of Photoelectron Spectroscopy The concept of photoelectron spectroscopy dates back to the discovery of photoelectric effect in the early 20th century. 1 Over the years, the realization and application of the photoelectron spectroscopy has been extended from chemical analysis 2-3 to exploring short lived transient species in chemical reactions. 4-7 The interaction of light with a molecule to form a molecular ion and an electron (photoionization) is the most common form of photoelectron spectroscopy. The physical process of photoionization can be expressed formally as: A(n,v,l) + h → A + (n',v',l') + e - (1.1) Where A, A + and e - denote neutral, cationic molecule and the free electron respectively. Each molecule is associated with the corresponding electronic, vibrational and rotational degrees of freedom and is specified by respective quantum numbers (n,v,l) and (n',v',l') for the initial and the final molecular state. Following conservation of energy we can write E{A; n,v,l} h  = E{A + ; n',v',l'}+ E(e - ) h eBE + eKE (1.2) where eBE = E{A + ; n', v', l'}- E{A; n, v, l} and eKE = E(e - ) indicating electron binding and electron kinetic energy. If we consider the distribution of electrons in a molecule in terms of the molecular orbital model as shown in Figure 1.1, then eqn. 1.1 implies the removal of an electron from a single molecular orbital without affecting electron distribution in the rest of the orbitals (Koopmans' theorem). This "one-electron" picture is often used in photoelectron spectroscopy and characterizes the only electronic selection rule: any transition between ion and neutral is 2 allowed that occurs by removal of an electron from a single orbital without rearrangement of the electron occupation of the other orbitals. If several electronic states of the cation are accessible energetically from the neutral with the photon energy employed, only those that are related to the neutral electronic configuration by a one-electron process will be seen in the photoelectron spectrum. As a result a number of cationic states (or in general, final molecular states) which are inaccessible ('dark') via electronic transition due to selection rules, can be populated using photoelectron spectroscopy and vice versa. Thus the complementarity between electronic and photoelectron selection rules in principle provide a deeper insight into molecular spectroscopy and dynamics. Figure 1.1. Schematic illustration of photoelectron spectroscopy. The energy level diagram shows ionization from the higher lying molecular orbitals, HOMO (Highest Occupied Molecular Orbitals) and HOMO-1. The incident photon energy (h ), electron binding energy (eBE) and electron kinetic energy (eKE) are related via energy conservation: eBE = h  - eKE. According to Koopmans' theorem the binding energy of the outgoing electron is a measure of the parent orbital energy, i.e., eBE = - E MO 3 The intensity of the photoelectron bands will depend on the overlapping integral of the neutral wave function (Ψ v ) with the cation vibrational wave functions (Ψ' v' ), formally called Franck-Condon factors (FCF) FCF = |< Ψ' v' | Ψ v ˃| 2 (1.3) If the initial and the final molecular state have very different equilibrium geometries, there will be a long vibrational progression in the photoelectron spectrum yielding broad photoelectron band envelope. The spacing between the progression peaks provides information on the nature of the normal modes and anharmonicity of the final state. Therefore, photoelectron spectra combined with ab initio quantum chemistry calculations yield a complete picture of the molecular electronic structure in the neutral and cationic state (for photoionization). 1.2 Time Resolved Photoelectron Spectroscopy Since the binding energy of the different molecular states varies depending on the electronic nature of the state, monitoring binding energy during a chemical or physical process effectively illustrates the evolution of the system. Following this idea, time resolved photoelectron spectroscopy (TRPES) has been exploited to probe physical/chemical transformations, transient species and molecular reaction dynamics. 4, 8-11 The experimental strategy follows the traditional pump-probe scheme where the pump excites a molecule to the excited state potential energy surface and the probe ionizes from the excited state. By varying the delay between the pump and the probe pulse, it is possible to map the excited state landscape as the molecule undergoes relaxation as well as to probe the photophysics and photochemistry 4 subsequent to photoexcitation. Figure 1.2 illustrates the physical concept and observables in a typical TRPES measurements. 1.3 Photoelectron Spectroscopy in liquids Historically, photoelectron spectroscopy has been used to study solid substrates 12-16 and molecules in the gas phase, nowadays introduced into vacuum using supersonic jet expansion technique. 17-19 High energy radiation is usually focused to the substrate or on the ultra-cold Figure 1.2. Schematic illustration of time resolved photoelectron spectroscopy (TRPES). After photoexcitation to the excited state with the pump pulse, the probe ejects electron and the molecule ends up in cationic state. The kinetic energy distribution of the outgoing electrons is shown the right panel. By varying the delay ( t) between pump and probe pulses, excited state dynamics can be probed by monitoring the evolution of the kinetic energy spectra. For example, if the molecule undergoes vibrational relaxation or nonadiabatic transition to another electronic state ( ), observed change in electron kinetic energy will reflect the dynamics. 5 molecular beam and the ejected electrons are collected and analyzed using a time of flight spectrometer, a hemispherical analyzer or a position sensitive detector. The sensitivity and efficiency of electron collection falls rapidly with increase in the number of collisions with surrounding molecules. As a result (photo)electron detection techniques are always performed under high vacuum environment to reduce losses due to unwanted electron scattering. Implementing photoelectron spectroscopy with liquids, therefore, impose challenges to create ideal environment for electron detection. Siegbahn and coworkers were the first to demonstrate how to perform photoelectron spectroscopy of liquids. 2 Although revolutionary, their technique was limited to liquids with low vapor pressure (such as formamide) while most of the important and chemically relevant reactions take place in highly volatile liquids, such as, water and alcohols which are incompatible with Siegbahn's measurement strategy. Besides, their technique suffered from substantial degradation of signal over time due to absence of replenishment of the fresh sample surface. Almost a couple of decades later, Faubel and coworkers resolved the issue of using highly volatile liquids along with continuously replenishing liquid surface during measurement by introducing liquid microjet technique. 20 Their seminal work laid the groundwork for an astonishing revival of the field. Faubel owed his success to the implementation of a fast flowing liquid microjet into vacuum, a technique he had developed earlier 21 demonstrating collision-free evaporation for sufficiently small water sample. The details of the liquid microjet operation will be discussed in the subsequent chapter, however, it is worth mentioning that the key idea of detecting photoelectrons with high enough sensitivity and efficiency relies on reducing the travel distance of the electrons from source (higher pressure, 10 - 5 -10 -6 mbar) to the detection chamber (lower pressure, 10 -8 -10 -9 mbar) and by keeping pressure low closer to the liquid source by using liquid nitrogen cryo-traps. After successful 6 implementation of the liquid microjet technique, much attention has been bestowed upon investigating different volatile liquids, most importantly water in great details. Interested readers are referred to a number of review articles which cover the application of liquid microjet to explore the electronic structure of liquid water. 22-25 One important aspect which became a topic of immense interest is the variation of the probing depth with outgoing electron kinetic energy. Initially liquid water was believed to follow the universal curve compiled over decades 26 based on experiments done on different solid substrates and by Monte- Carlo simulations. 27-30 However, a number of recent studies 31-33 using liquid microjet photoelectron spectroscopy illustrated that the effective probing depth of water is distinct from the universal trend, most significantly in the low electron energy region (<20 eV) as shown in Figure 1.3. Hence, one can probe molecules at different layer depths in the liquid jet (typical diameter 20 ± 5 m) by varying the incident photon energy enabling the technique to investigate both interface and bulk solutions Figure 1.3. Variation of effective electron attenuation length (EAL) of with electron kinetic energy for liquid water from two recent studies using x-ray photoelectron spectroscopy with liquid microjet technique. 32-33 For comparison 'universal inelastic mean free path (IMFP) curve' is also shown 26 7 1.4 Application of liquid jet PES: Investigating oxidative DNA damage and repair In recent years a lot of scientific and technological efforts have been invested into understanding the DNA damage and repair mechanism. 34-36 One of the major sources of direct DNA damage is the oxidative damage induced by ionizing radiation. Oxidative damage has significant deleterious effects on genomic DNA, such as strand breaks and nucleobase damage, 37 which can lead to mutations and cancer. 38 Double strand breaks are by far the most damaging lesion as their repair is particularly difficult and the mutation probability is high. 39 For a double strand break to occur, at least two oxidative lesions must lie within close range on the double helix and the probability of such an occurrence increases if the hole (the locus of the oxidized base) can migrate efficiently by charge transport. 40-44 One of the proposed ways the eukaryotic cells repair the oxidative damage is by rapid electron transfer from nearby amino acid residues, such as tryptophan and tyrosine, to the local oxidation center (nucleobase). 34-35 Rates of electron (hole) transport between two nucleobases or between a nucleobase and aromatic amino acids can be approximated by Marcus theory. 45 To predict such rates, accurate knowledge of two key energetic parameters is required: the energy change for the electron to move between biomolecules and the total reorganization energy connected with this process. Equivalently, this information can be synthesized from knowledge of the energetic parameters for the redox half- reactions of the two individual redox centers, namely the vertical ionization energies (VIEs) of each and the individual reorganization energies on loss (gain) of an electron. 46 Although there is a vast literature of traditional electrochemical measurements reporting the half cell redox potentials of biomolecules (and their simpler molecular analogs) in water, often the large spread of the reported values leads to ambiguous if not erroneous conclusion. 47-49 Most importantly the notion of the pH dependent redox potentials (beyond a simple Nernstian relationship 50 ) indicates 8 the sensitivity of the thermodynamic quantities to the chemical environment 51 and the need of a new experimental technique to extract these ionization parameters cannot be overemphasized. Recently Bradforth and coworkers combined steady state x-ray photoelectron spectroscopy in liquid microjet with ab initio calculations to extract vertical ionization energies, reorganization energy and standard redox potentials of nucleic acid components in buffered aqueous environment. 52-53 The extracted standard redox potential values do not suffer from follow-up reactions, such as, deprotonation and subsequent radical reactions which render electrochemical measurements irreversible. Thus, liquid jet photoelectron spectroscopy not only establishes a new spectroscopic technique to explore molecular electronic structure in liquids, but also provides a new tool to obtain equilibrium thermodynamic parameters which are inaccessible via conventional analytical methods. 1.5 Thesis Objectives Although very effective, x-ray photoelectron spectroscopy suffers from its own limitations, such as, demanding infrastructure for high energy and high fluence x-ray light source, limited chemical selectivity and comparatively low sensitivity to dilute solutions. To circumvent these limitations we implemented a table top tunable deep UV laser pulses as light source and coupled it with home built liquid microjet photoelectron spectrometer. Chapter 2 describes the building and instrumentation of the home built time resolved photoelectron spectrometer in detail. In chapter 3, a comparison of x-ray photoelectron and resonant two photon ionization spectroscopy using ultrafast deep UV lasers has been demonstrated with the example of aromatic amino acids in aqueous solutions. Our experimental results validate that 9 resonant photoemission process provides high chemical selectivity as well as superior sensitivity and could potentially be used as a reliable analytical tool. We have also demonstrated how to extract thermodynamic parameters such as the standard one electron redox potential of a half cell and corresponding reorganization energy in the electron transfer reaction purely from our photoemission measurements. The spectroscopic determination of standard ionization parameters turns out to be not only consistent with relevant theoretical estimation, it also provides a benchmark for further improvement in the theoretical methods. By coupling ultrafast lasers at deep UV or even shorter wavelength radiation with a liquid microjet, several different groups attempted to explore excited state dynamics in aqueous and nonaqueous solutions. 54-63 These time resolved photoemission studies complemented traditional transient absorption and fluorescence measurements and unfold a new direction to study photoinduced dynamics in the condensed phase. The relaxation dynamics of solvated electron, the simple most quantum system in liquids, especially has attracted a lot of attention from experimentalists as well as from theorists due to its complex and often controversial nature. Recent TRPES measurements of aqueous iodide solution provide decisive conclusion on the relaxation dynamics of the solvated electron. 64 However, the questions remain relevant on the nature of the relaxation pathways when photoexcitation leads to significant population in the higher lying anion excited states. Also, the influence of the counter ion in the first solvation shell is yet to be explored. Chapter 4 delves into all these pertinent questions by analyzing ultrafast relaxation dynamics of hydrated electron at different excitation wavelengths using transient photoemission spectroscopy of aqueous iodide solutions. Another essential scientific quest is to explore the mechanism of photoinduced reactions in photoactive biomolecules. One of the most versatile photoactive agents in biological systems 10 is tyrosine. 65 To investigate the photophysics and photochemistry of tyrosine we choose to follow through the excited state dynamics of its model chromophore, phenol in aqueous solution. Previous studies in our group illustrated that aqueous phenol, when photoexcited to its first excited state undergoes slow (time scale >1 ns) autoionization to produce phenoxyl radical and the solvated electron. 66-67 This observation is in sharp contrast to excitation to the higher lying state at 200 nm. 67 In chapter 5, we followed through the previous ultrafast broadband transient absorption experiment to much longer time scale (~1 s) by implementing a new sub-ns laser source. Our extended probe time scale coupled with far UV to visible spectral probing (250-620 nm) enable us to portray a complete picture of ensuing excited dynamics and photochemistry of aqueous phenol solution. 11 References 1. Einstein, A., Über einen die Erzeugung und Verwandlung des Lichtes betreffenden heuristischen Gesichtspunkt. Annalen der Physik 1905, 322, 132-148. 2. Siegbahn, H.; Siegbahn, K., ESCA applied to liquids. Journal of Electron Spectroscopy and Related Phenomena 1973, 2, 319-325. 3. Siegbahn, K., Electron spectroscopy for atoms, molecules, and condensed matter. Rev Mod Phys 1982, 54, 709-728. 4. Jefferys Greenblatt, B.; Zanni, M. T.; Neumark, D. M., Photodissociation dynamics of the I2− anion using femtosecond photoelectron spectroscopy. Chem Phys Lett 1996, 258, 523- 529. 5. Zanni, M. T.; Batista, V. S.; Greenblatt, B. J.; Miller, W. H.; Neumark, D. M., Femtosecond photoelectron spectroscopy of the I2− anion: Characterizat ion of the Ã′ 2Πg,1/2 excited state. The Journal of Chemical Physics 1999, 110, 3748-3755. 6. Fischer, I.; Villeneuve, D. M.; Vrakking, M. J. J.; Stolow, A., Femtosecond wave ‐packet dynamics studied by time ‐resolved zero ‐kinetic energy photoelectron spectroscopy. The Journal of Chemical Physics 1995, 102, 5566-5569. 7. Baumert, T.; Thalweiser, R.; Gerber, G., Femtosecond two-photon ionization spectroscopy of the B state of Na3 clusters. Chem Phys Lett 1993, 209, 29-34. 8. Bradforth, S. E.; Weaver, A.; Arnold, D. W.; Metz, R. B.; Neumark, D. M., Examination of the Br+HI, Cl+HI, and F+HI Hydrogen Abstraction Reactions by Photoelectron-Spectroscopy of BrHI - , ClHI - , and FHI - . J Chem Phys 1990, 92, 7205-7222. 9. Neumark, D. M., Transition State Spectroscopy of Bimolecular Chemical Reactions. Annual Review of Physical Chemistry 1992, 43, 153-176. 10. Bragg, A. E.; Verlet, J. R. R.; Kammrath, A.; Cheshnovsky, O.; Neumark, D. M., Hydrated electron dynamics: From clusters to bulk. Science 2004, 306, 669-671. 11. Yandell, M. A.; King, S. B.; Neumark, D. M., Time-Resolved Radiation Chemistry: Photoelectron Imaging of Transient Negative Ions of Nucleobases. Journal of the American Chemical Society 2013, 135, 2128-2131. 12. Ge, N.-H.; Wong, C. M.; Lingle, R. L.; McNeill, J. D.; Gaffney, K. J.; Harris, C. B., Femtosecond Dynamics of Electron Localization at Interfaces. Science 1998, 279, 202-205. 13. Hertel, T.; Knoesel, E.; Wolf, M.; Ertl, G., Ultrafast Electron Dynamics at Cu(111): Response of an Electron Gas to Optical Excitation. Phys Rev Lett 1996, 76, 535-538. 12 14. Haight, R.; Bokor, J.; Stark, J.; Storz, R. H.; Freeman, R. R.; Bucksbaum, P. H., Picosecond Time-Resolved Photoemission Study of the InP(110) Surface. Phys Rev Lett 1985, 54, 1302-1305. 15. Williams, R. H.; Srivastava, G. P.; McGovern, I. T., Photoelectron spectroscopy of solids and their surfaces. Reports on Progress in Physics 1980, 43, 1357. 16. Chadi, D. J.; Bauer, R. S.; Williams, R. H.; Hansson, G. V.; Bachrach, R. Z.; Mikkelsen, J. C.; Houzay, F.; Guichar, G. M.; Pinchaux, R.; Pétroff, Y., Atomic and Electronic Structure of the 7x7 Reconstructed Si (111) Surface. Phys Rev Lett 1980, 44, 799-802. 17. Potts, A. W.; Price, W. C., Photoelectron Studies of Ionic Materials using Molecular Beam Techniques. Phys Scripta 1977, 16, 191. 18. Wang, L.-S.; Reutt, J. E.; Lee, Y. T.; Shirley, D. A., High resolution UV photoelectron spectroscopy of CO+2, COS+ and CS+2 using supersonic molecular beams. Journal of Electron Spectroscopy and Related Phenomena 1988, 47, 167-186. 19. Reutt, J. E.; Wang, L. S.; Lee, Y. T.; Shirley, D. A., Molecular beam photoelectron spectroscopy and femtosecond intramolecular dynamics of H2O+ and D2O+. The Journal of Chemical Physics 1986, 85, 6928-6939. 20. Faubel, M.; Steiner, B.; Toennies, J. P., Photoelectron spectroscopy of liquid water, some alcohols, and pure nonane in free micro jets. J. Chem. Phys. 1997, 106, 9013. 21. Faubel, M.; Schlemmer, S.; Toennies, J. P., A molecular beam study of the evaporation of water from a liquid jet. Z. Phys. D: At., Mol. Clusters 1988, 10, 269. 22. Winter, B.; Faubel, M., Photoemission from liquid aqueous solutions. Chem Rev 2006, 106, 1176-1211. 23. Winter, B., Liquid microjet for photoelectron spectroscopy. Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment 2009, 601, 139-150. 24. Faubel, M.; Siefermann, K. R.; Liu, Y.; Abel, B., Ultrafast Soft X-ray Photoelectron Spectroscopy at Liquid Water Microjets. Accounts Chem Res 2012, 45, 120-130. 25. Hayashi, H.; Watanabe, N.; Udagawa, Y.; Kao, C.-C., The complete optical spectrum of liquid water measured by inelastic x-ray scattering. Proceedings of the National Academy of Sciences 2000, 97, 6264-6266. 26. Seah, M. P.; Dench, W. A., Quantitative electron spectroscopy of surfaces: A standard data base for electron inelastic mean free paths in solids. Surface and Interface Analysis 1979, 1, 2-11. 27. Michaud, M.; Wen, A.; Sanche, L., Cross Sections for Low-Energy (1–100 eV) Electron Elastic and Inelastic Scattering in Amorphous Ice. Radiation Research 2003, 159, 3-22. 13 28. Emfietzoglou, D.; Kyriakou, I.; Garcia-Molina, R.; Abril, I.; Nikjoo, H., Inelastic Cross Sections for Low-Energy Electrons in Liquid Water: Exchange and Correlation Effects. Radiation Research 2013, 180, 499-513. 29. Tomita, H.; Kai, M.; Kusama, T.; Ito, A., Monte Carlo simulation of physicochemical processes of liquid water radiolysis. Radiat Environ Biophys 1997, 36, 105-116. 30. Pimblott, S. M.; LaVerne, J. A.; Mozumder, A., Monte Carlo Simulation of Range and Energy Deposition by Electrons in Gaseous and Liquid Water. The Journal of Physical Chemistry 1996, 100, 8595-8606. 31. Ottosson, N.; Faubel, M.; Bradforth, S. E.; Jungwirth, P.; Winter, B., Photoelectron spectroscopy of liquid water and aqueous solution: Electron effective attenuation lengths and emission-angle anisotropy Journal of Electron Spectroscopy and Related Phenomena 2010, 177, 60-70. 32. Thürmer, S.; Seidel, R.; Faubel, M.; Eberhardt, W.; Hemminger, J. C.; Bradforth, S. E.; Winter, B., Photoelectron Angular Distributions from Liquid Water: Effects of Electron Scattering. Phys Rev Lett 2013, 111, 173005. 33. Suzuki, Y.-I.; Nishizawa, K.; Kurahashi, N.; Suzuki, T., Effective attenuation length of an electron in liquid water between 10 and 600 eV. Physical Review E 2014, 90, 010302. 34. Morozova, O. B.; Kiryutin, A. S.; Sagdeev, R. Z.; Yurkovskaya, A. V., Electron Transfer between Guanosine Radical and Amino Acids in Aqueous Solution. 1. Reduction of Guanosine Radical by Tyrosine. J. Phys. Chem. B 2007, 111, 7439-7448. 35. Morozova, O. B.; Kiryutin, A. S.; Yurkovskaya, A. V., Electron Transfer between Guanosine Radicals and Amino Acids in Aqueous Solution. II. Reduction of Guanosine Radicals by Tryptophan. J. Phys. Chem. B 2008, 112, 2747-2754. 36. Kao, Y.-T.; Saxena, C.; Wang, L.; Sancar, A.; Zhong, D., Femtochemistry in enzyme catalysis: DNA photolyase. Cell Biochem Biophys 2007, 48, 32-44. 37. Lehnert, S., Biomolecular Action of Ionizing Radiation (Series in Medical Physics and Biomedical Engineering). 2007. 38. Morgan, W. F., Non-Targeted and Delayed Effects of Exposure to Ionizing Radiation: II. Radiation-Induced Genomic Instability and Bystander Effects In Vivo, Clastogenic Factors and Transgenerational Effects. Radiation Research 2003, 159, 581-596. 39. Sutherland, B. M.; Bennett, P. V.; Sidorkina, O.; Laval, J., Clustered Damages and Total Lesions Induced in DNA by Ionizing Radiation: Oxidized Bases and Strand Breaks†. Biochemistry 2000, 39, 8026-8031. 40. E. Lomax, M.; K. Gulston, M.; O'Neill , P., Chemical Aspects of Clustered DNA Damage Induction by Ionising Radiation. Radiation Protection Dosimetry 2002, 99, 63-68. 14 41. O’Neill, M.; Barton, J., DNA -Mediated Charge Transport Chemistry and Biology. In Long-Range Charge Transfer in DNA I, Schuster, G. B., Ed. Springer Berlin Heidelberg: 2004; Vol. 236, pp 67-115. 42. Kawai, K.; Majima, T., Hole Transfer Kinetics of DNA. Accounts Chem Res 2013, 46, 2616-2625. 43. Thazhathveetil, A. K.; Trifonov, A.; Wasielewski, M. R.; Lewis, F. D., Increasing the Speed Limit for Hole Transport in DNA. Journal of the American Chemical Society 2011, 133, 11485-11487. 44. Grozema, F. C.; Tonzani, S.; Berlin, Y. A.; Schatz, G. C.; Siebbeles, L. D. A.; Ratner, M. A., Effect of GC Base Pairs on Charge Transfer through DNA Hairpins: The Importance of Electrostatic Interactions. Journal of the American Chemical Society 2009, 131, 14204-14205. 45. Marcus, R. A.; Sutin, N., Electron transfers in chemistry and biology. Biochimica et Biophysica Acta (BBA) - Reviews on Bioenergetics 1985, 811, 265-322. 46. Wang, X.-B.; Wang, L.-S., Probing the electronic structure of redox species and direct determination of intrinsic reorganization energies of electron transfer reactions. The Journal of Chemical Physics 2000, 112, 6959-6962. 47. Seidel, C. A.; Schulz, A.; Sauer, M. H., Nucleobase-Specific Quenching of Fluorescent Dyes. 1. Nucleobase One-Electron Redox Potentials and Their Correlation with Static and Dynamic Quenching Efficiencies. J. Phys. Chem. 1996, 100, 5541. 48. Crespo-Hernández, C. E.; Close, D. M.; Gorb, L.; Leszczynski, J., Determination of Redox Potentials for the Watson−Crick Base Pairs, DNA Nucleosides, and Relevant Nucleoside Analogues. The Journal of Physical Chemistry B 2007, 111, 5386-5395. 49. Harriman, A., Further comments on the redox potentials of tryptophan and tyrosine. The Journal of Physical Chemistry 1987, 91, 6102-6104. 50. Wardman, P., Reduction potentials of one-electron couples involving free radicals in aqueous solution. J. Phys. Chem. Ref. Data 1989, 18, 1637-1755. 51. Steenken, S., Purine bases, nucleosides, and nucleotides: aqueous solution redox chemistry and transformation reactions of their radical cations and e- and OH adducts. Chem Rev 1989, 89, 503-520. 52. Schroeder, C. A.; Pluhařová, E.; Seidel, R.; Schroeder, W. P.; Faubel, M.; Slavíček, P.; Winter, B.; Jungwirth, P.; Bradforth, S. E., Oxidation Half-Reaction of Aqueous Nucleosides and Nucleotides via Photoelectron Spectroscopy Augmented by ab Initio Calculations. Journal of the American Chemical Society 2015, 137, 201-209. 53. Pluhařová, E.; Schroeder, C.; Seidel, R.; Bradforth, S. E.; Winter, B.; Faubel, M.; Slavíček, P.; Jungwirth, P., Unexpectedly Small Effect of the DNA Environment on Vertical Ionization Energies of Aqueous Nucleobases. J. Phys. Chem. Lett. 2013, 4, 3766. 15 54. Siefermann, K. R.; Liu, Y. X.; Lugovoy, E.; Link, O.; Faubel, M.; Buck, U.; Winter, B.; Abel, B., Binding energies, lifetimes and implications of bulk and interface solvated electrons in water. Nature Chemistry 2010, 2, 274-279. 55. Tang, Y.; Suzuki, Y. I.; Shen, H.; Sekiguchi, K.; Kurahashi, N.; Nishizawa, K.; Zuo, P.; Suzuki, T., Time-resolved photoelectron spectroscopy of bulk liquids at ultra-low kinetic energy. Chem Phys Lett 2010, 494, 111-116. 56. Tang, Y.; Shen, H.; Sekiguchi, K.; Kurahashi, N.; Mizuno, T.; Suzuki, Y. I.; Suzuki, T., Direct measurement of vertical binding energy of a hydrated electron. Physical Chemistry Chemical Physics 2010, 12, 3653-3655. 57. Shen, H. A.; Kurahashi, N.; Horio, T.; Sekiguchi, K.; Suzuki, T., Direct Measurement of Vertical Electron Binding Energies of Solvated Electrons in Methanol and Ethanol. Chemistry Letters 2010, 39, 668-670. 58. Yamamoto, Y.-i.; Suzuki, Y.-I.; Tomasello, G.; Horio, T.; Karashima, S.; Mitríc, R.; Suzuki, T., Time- and Angle-Resolved Photoemission Spectroscopy of Hydrated Electrons Near a Liquid Water Surface. Phys Rev Lett 2014, 112, 187603. 59. Buchner, F.; L bcke, A.; Heine, N.; Schultz, T., Time-resolved photoelectron spectroscopy of liquids. Rev Sci Instrum 2010, 81, 113107. 60. Lübcke, A.; Buchner, F.; Heine, N.; Hertel, I. V.; Schultz, T., Time-resolved photoelectron spectroscopy of solvated electrons in aqueous NaI solution. Physical Chemistry Chemical Physics : PCCP 2010, 12, 14629. 61. Buchner, F.; Ritze, H.-H.; Lahl, J.; Lubcke, A., Time-resolved photoelectron spectroscopy of adenine and adenosine in aqueous solution. Physical Chemistry Chemical Physics 2013, 15, 11402-11408. 62. Buchner, F.; Nakayama, A.; Yamazaki, S.; Ritze, H.-H.; Lübcke, A., Excited-State Relaxation of Hydrated Thymine and Thymidine Measured by Liquid-Jet Photoelectron Spectroscopy: Experiment and Simulation. Journal of the American Chemical Society 2015, 137, 2931-2938. 63. Shreve, A. T.; Elkins, M. H.; Neumark, D. M., Photoelectron spectroscopy of solvated electrons in alcohol and acetonitrile microjets. Chem Sci 2013, 4, 1633-1639. 64. Elkins, M. H.; Williams, H. L.; Shreve, A. T.; Neumark, D. M., Relaxation Mechanism of the Hydrated Electron. Science (New York, N.Y.) 2013, 342, 1496-1499. 65. Tolbert, L. M.; Solntsev, K. M., Excited-State Proton Transfer: From Constrained Systems to “Super” Photoacids to Superfast Proton Transfer†. Accounts Chem Res 2002, 35, 19- 27. 66. Zhang, Y. PhD Thesis. University of Southern California. 16 67. Zhang, Y.; Oliver, T. A. A.; Ashfold, M. N. R.; Bradforth, S. E., Exploring the autoionization and photo-induced proton-coupled electron transfer pathways of phenol in aqueous solution. In preparation. 17 Chapter 2. Building the liquid microjet photoelectron spectrometer 2.1 Introduction All photoemission experiments reported here have been performed using liquid microjet technique invented by Faubel and coworkers. 1 A detailed description of the technique and its diverse application have been reported in a number of articles, 2-6 however, a brief account will be presented in the following section. We have used both high energy x-ray and ultrafast deep UV lasers as light sources in our experiments. All X-ray photoemission measurements have been performed at high intensity U41-PGM beamline at BESSY II electron storage ring facility, a third generation synchrotron radiation source. A full description of the x-ray light source can be found elsewhere. 7 All laser experiments have been performed at USC with a home built liquid jet photoelectron spectrometer. In the following sections a comprehensive description of the photoelectron spectrometer will be presented. 2.2 Experimental Design 2.2.1 Liquid Microjet: Performance in Vacuum and Compatibility with Photoelectron Spectroscopy The experimental breakthrough in ESCA of liquids at vapor Pressure >1 mbar, came in 1997 when Manfred Faubel et al. 1 reported the first He I (21.22 eV) photoelectron spectrum of neat water, which laid the groundwork for an astonishing revival of the field. The main challenge among all experiments is to maintain liquid flow under high vacuum condition and to guide electrons to the energy detector despite the solution’s high vapor pressure. This is 18 accomplished by reducing the distance, over which electrons travel, without inelastic scattering by reducing the effective pressure by several orders of magnitude over a very short distance. With the liquid microjet, the latter is a natural consequence of the rapidly decreasing vapor density as a function of the radial distance from the jet. The emerging reduction of electron mean free path is illustrated in Figure 2.1. For typical experimental conditions the effective electron mean free path is approximately 1-2 mm, the distance at which the spectrometer entrance is located. A more detailed consideration of the effective electron transfer length is given in literature. 8 The vacuum liquid-jet with a velocity of typically 80 m/s is produced by forcing the solution at approximately 5–20 bar backing-pressure through a circular glass capillary of 20 µm inner diameter. Size may be varied by ±5 depending on the solution’s vapor pressure and Figure 2.1. Schematic 3D model of the home built photoelectron spectrometer on laser table. The change of water-vapor density as a function of radial distance from the microjet is illustrated in the inset 19 the velocity may be changed depending on viscosity. Under these conditions the vacuum jet consists of a smooth continuous region of liquid water, which decays into droplets at a distance of approximately 5-10 mm downstream from the nozzle orifice. 1-2 Solute diffusion within the liquid jet is fast enough to establish equilibrium concentration between bulk and solution interface within the laminar region. 9 The use of a fast flowing jet as a free liquid surface in vacuum counteracts the efficient evaporative cooling of high-vapor pressure liquids by the continuous rapid replacement of liquid. Importantly, liquid–liquid molecule collisions occur on the 1 ps timescale, and local thermodynamic equilibrium is established within tens of picoseconds matching gas-phase impact collision rates four orders of magnitude lower. 1 2.2.2 Building liquid-jet Photoelectron Spectrometer The photoelectron spectrometer can be divided into two sub-sections, a) Source Chamber, where liquid solutions are injected and laser-matter interaction takes place and b) Detection Chamber, where ejected electrons are collected and guided to the detector. a) Source Chamber: The source chamber consists of liquid-jet assembly, pump assembly and the entrance window for the ultrafast laser pulses. A 3D-model is shown in Figure 2.1. Liquid jet and the laser pulses interact in the center of a cubic vacuum chamber (approximately 23 2323 cm 3 ), and the entrance orifice of the electron analyzer is at ~3 mm away from the interaction point. Notice that the small distance between liquid-jet and spectrometer entrance (~3 mm) is a necessary condition to warrant undisturbed electron travel in a region of high vapor pressure (10 -5 /10 -6 mbar base pressure) under operation conditions. Such a relatively high vacuum – within a chamber 20 containing a streaming liquid-water jet – is achieved by combined mechanical (turbomolecular pumps) and cryo-pumping. It is perhaps interesting to mention that mechanical pumping alone – and we use a big turbo pump capable of 1500 l/s (Hi Pace Turbo 1500 from Pfeiffer Vacuum)– would not be possible; the major load of water vapor must be carried away by trapping water molecules on cryo surfaces in order to reach the optimum base pressure (10 -3 -10 -4 mbar) for the turbo pump operation. Two liquid-nitrogen cryo-traps attached to the extended vacuum chamber warrant Figure 2.2. Inside the source chamber. Interaction region showing liquid jet, skimmer and the permanent magnet configuration (a). Formation of ice-needle and subsequent jet-blockage (b). An in vacuo ice-cutter is implemented to break the ice-needles before they reach the nozzle (c,d). The 'ice-cutter' is driven by a DC power supply continuously during the experiment (d). 21 good performance. One of the two cryo-traps is a 56 cm long (6 cm diameter) metal tube immersed in liquid nitrogen (LN2) dewar connected to the main cube, in the direction of the flow of the liquid jet to catch the liquid (hereafter, called 'catcher-tube') and condense it to ice. Another cryo-trap (outer surface area ~1450 cm 2 ) is mounted in the extended part of the source chamber filled with LN2. Under typical jet conditions, 80 ms -1 , 6 o C, 20 µm jet diameter, the liquid-jet freezes well before reaching the bottom of the catcher tube. Often, it is the ‘ice load’ that determines when an experimental run has to be started over, which involves venting, removing the ice and cleaning and drying the inner walls of the vacuum chamber as well as drying the cold traps. In the worst case scenario, the jet might not hit the bottom of the catcher tube which leads to formation of ice-needles reaching the jet nozzle and stops operation. By introducing a motorized ice-cutter (custom-made, shown in Figure 2.2) at the junction of the cubic interaction chamber and the catcher tube, we extended the operation time to 8-10 hours with regular refilling of the cryo-traps. The sketch of Figure 2.3 shows the main components needed to produce the liquid jet. Since slight position changes of the jet during collection of PE spectra would lead to unacceptable fluctuations of measured signal, jet stability is a most crucial concern. Stable flows with smooth laminar surface are achieved here using a standard high-pressure High-Performance Liquid Chromatography (HPLC; Shimazdu) pump used in constant flow-rate mode. Flow rate is usually varied between 0.4 – 0.6 ml/min depending on the solvent viscosity and temperature to maintain laminarity. Solutions are usually degassed using sonicator prior to injection in order to remove air bubbles. 2 µm pore-sized filters are used in-line to collect particles from getting into the glass capillary, thus preventing the latter from clogging. Jet positions and normal operation is 22 monitored visually using a Digital Microscope camera (Plugable, 50x Optical Magnification) mounted close to a customize transparent acrylic glass flange. b) Detection Chamber: The detection chamber comprises three modular subsections - a skimmer assembly, a stainless steel flight tube (50 cm) and a multichannel plate (MCP) detector. The skimmer assembly contains a stainless steel cone-shaped skimmer with 400 m orifice at the apex and a mechanical valve underneath to isolate the detection chamber from the high pressure source chamber during venting. The flight tube is made of a hollow stainless steel tube (50 cm  10 cm) mounted on a metal flange and attached to the interaction chamber. A commercially available Figure 2.3. Schematic of the liquid delivery system. See text for detailed description. 23 five-way cross (P104690, Ideal Vacuum LLC., Albuquerque, NM) is mounted on the other side of the flight tube as a housing for the MCP detector, pressure gauge (PKR 251, Active, Pirani, Pfeiffer Vacuum) and two 300 L/s turbomolecular pumps (HiPace 300, Pfeiffer Vacuum). The detector contains a pair of 40 mm diameter multichannel plates (MCP) in Chevron configuration (BOS-40-6, Beam Imaging Solutions, Longmont, CO). Initially the front plate was grounded and the back plate is biased at +1500V using a high voltage DC power supply (Matsusada, 3kV) and a home-made voltage divider (Figure 2.5). The amplified electron signal (gain ~10 7 ) is capacitively coupled out of a biased (+1700V) metal anode and passed through additional preamplifier (100x, Phillips Scientific) for further processing. To increase the collection efficiency of the slow photoelectrons a grounded copper mesh has been mounted ~5 mm away from the MCP and an additional +200V bias has been applied to all the MCP components (Figure 2.4). Figure 2.4. Mounted MCP detector assembly (a). Enlarged view of the multichannel plate (b). To enhance the detection efficiency of slow photoelectrons a grounded copper mesh is incorporated in front of the MCP (c). The schematic of MCP operation before and after installing 'copper mesh' is also shown (d). 24 2.2.3 Designing Time of Flight spectrometer (TOF) 2.2.3.1 Principle of time of flight detection As mentioned before we have implemented time of flight (TOF) detection scheme to analyze electron kinetic energy (eKE). The working principle of time of flight detection has been first reported by Wiley and McLaren 10 for high resolution mass spectrometry. Although the fundamental physical principle for charged particle detection is identical, the instrumental Figure 2.5. The circuit diagram of the home-built voltage divider box. The electrical components and circuit configuration are shown for the configuration with grounded copper mesh (Figure 2.4 c,d). The resistors and capacitors have average accuracy of ±5% of the specified value. 25 adoption for electron detection 11-13 was much simpler without electrostatic lenses. For photoelectron detection the mass of the particles (electrons) are the same, therefore the flight time distribution directly reflects the distribution of electron kinetic energy following the expression below: (2.1) Where E: electron kinetic energy; m=mass of an electron; v=velocity of the electron; l=flight length and t: flight time. A precise knowledge of flight length and flight time are therefore instrumental to extract kinetic energy. Time of flight also provides dispersed energy detection since the detector detects all the electrons with different kinetic energy to provide full spectrum in a single acquisition as opposed to other conventional detection technique, such as, hemispherical analyzer which discriminates electrons with a desired kinetic energy (pass energy) and scans the pass energy to acquire the full spectrum. Therefore, the former detection scheme is ideal for pulsed laser experiment, hence, adopted in our experimental set up. 2.2.3.2 Energy resolution The relative energy resolution of the spectrometer can be expressed as (2.2) In our experiment length of the tube (l) is a quasi fixed parameter (value obtained from calibration and used uniformly for all measurements), any error in the latter ( l) will effectively be manifested as a further timing error ( t). Therefore, the working energy resolution expression will be 26 (2.3) We note that since t  E -1/2 , E/E  E 1/2 , i.e., relative energy resolution will be higher for low kinetic energy electrons. 2.2.3.3 Collection Efficiency: Field free TOF vs. Magnetic bottle TOF From equation 2, we can conclude that the energy resolution increases with longer tube (or effectively with longer flight time), however, increase in resolution compromises collection efficiency. In field free time of flight detection, the collection efficiency is solely determined by the experimental geometry as shown in Figure 2.6. Assuming the isotropic electron ejection (i.e., the electrons are ejected in all 4  direction with equal probability) from the interaction point, the collection efficiency will be expressed as a ratio of the solid angle subtended by the MCP detector to 4 . Based on the geometric parameters of our flight tube and the detector diameter, the collection efficiency has been calculated to be 0.04% (cf. Figure 2.6). Figure 2.6. Collection efficiency of a field free time of flight spectrometer. The parameters considered here are based on the instrument built at USC 27 To improve the collection efficiency we adopted magnetic bottle time of flight detection strategy following the seminal work by Kruit and Read. 14 In this scheme the ejected photoelectrons are guided to the detector by the imposed magnetic field without changing their kinetic energy, therefore, increases collection efficiency upto 50%, however, with a sacrifice in energy resolution. The magnetic bottle is formed between a strong SmCo permanent magnet 25 mm diameter, 60 mm height with a field strength of 0.7 T in the interaction chamber and a milli- tesla field created by a solenoid outside the 50 cm long spectrometer flight tube. The magnetic field (B) induced by the solenoid coil can be expressed as (2.4) Where  0 = vacuum permeability = 4  10 -7 Wb  A -1  m -1 ,  r = relative permeability of air = 1.02, N = number of turns in the solenoid coil, i = applied current and l = tube length. Putting N = 81 and i = 3 A and l = 0.5 m, we found the magnetic field B = 0.6 mT which is pretty close to the measured value of 1 mT (Figure 2.7). Figure 2.7. Magnetic field time of flight spectrometer. The variation of magnetic field is plotted against the distance from the interaction region. The gradient of the magnetic field is high in the 'source chamber' region, however, the magnetic field is homogenous in the flight tube. Collection efficiency ~50% 28 To increase the gradient of the magnetic flux d ensity in the interaction region, a 6 mm high conical soft iron tip is placed on top of the cylindrical magnet. A -metal shielding of the flight tube inhibits distortion of the magnetic field by external fields, e.g., the earth magnetic field. In operation, the soft iron tip is located about 2 mm behind the liquid jet and 5 mm behind the micro skimmer. Three kinematic micrometers mounted in a xyz stage adjust the position of the permanent magnet and thus allow an alignment of the magnetic bottle. 2.2.3.4 Magnetic bottle TOF: How it works In order to understand the mechanism of magnetic bottle, we need to consider the trajectory of electron in an inhomogeneous magnetic field created by the combination of the permanent magnet and the electromagnet. In an axial uniform magnetic field an electron follows a spiral trajectory following the equation of motion under Lorentz force as shown in Figure 2.8. Figure 2.8. Illustrating electron trajectory in homogenous (a) and in inhomogenous (b) magnetic field. In case of magnetic bottle spectrometer the inhomogeniety effectively parallelize electron trajectories along the magnetic field axis (flight axis). For example, if θ i =   , B i = 1T and B f = 1mT then θ f = 1.8 0 , i.e., almost parallel to the flight axis 29 The radius of curvature and the pitch of the helical motion stays constant as long as the magnetic field remains the same. However, the radius of curvature and the pitch change abruptly when the magnetic field varies along the axis. In Figure 2.7 we have shown how the magnetic field decreases along the time of flight axis. In their paper Kruit and Read illustrated that the radius of curvature increases with decrease in magnetic field (eqn. 2.5), therefore the inhomogeniety of the magnetic field effectively parallelizes the electron trajectory irrespective of their initial ejection angle with respect to the axial magnetic field 14 (cf. Figure 2.8). (2.5a) (2.5b) where {θ i , r i } and {θ f , r f } represent initial and final electron ejection angle with respect to the flight axis and radius of curvature of the electron respectively. Technically, all the electrons ejected in 2  hemisphere could be collected yielding 50% collection efficiency. On the other hand, the parallelization process contributes to the timing offset as well as timing uncertainty which effectively degrades the overall energy resolution. 2.2.4 Laser system For all the experiments in this dissertation ultrafast deep UV (240-350 nm) laser pulses have been used. For two beam (pump-probe) experiments, a tunable pump (probe) beam (240- 300 nm) is generated by doubling the visible output (480-700 nm) of an Optical parametric Amplifier (Spectra-Physics 800C, pumped by 800 nm from a Ti: Sapphire regenerative amplifier, Coherent Legend-Elite, 3.2W, 35 fs, 1 kHz). The residual visible beam after second 30 harmonic generation is typically rejected by using couple of broadband DUV mirrors (220-255 nm, Layertec Gmbh) or dielectric mirrors (290-300 nm). The second beam fixed at 267 nm is generated using sum frequency set up as shown in Figure 2.9. A mechanical translation stage (Newport Corp) is incorporated in the tunable beam path for controlling the delay between the pump and the probe pulses and precisely controlled by a motion controller (ESP 300, Newport Corp.). The time resolution of the experimental set up was estimated from the cross-correlation width of the pump and probe beams on a 1 mm quartz film 15-16 and varies from 160-200 fs. For calibration purpose (discussed in section 2.2.7) we used 200 nm pulses by mixing the 267 nm with the residual fundamental (800 nm) in a Type I BBO crystal (not shown in Figure 2.9). The polarization of all the DUV pulses are rotatable using /2 waveplate (except 200 nm due to lack of proper l/2 waveplate), however, kept vertical with respect to our lab frame for all the experiments. Figure 2.9. Experimental laser set up for one beam and two beam (transient) photoemission experiment. M: Mirror; DC: Dichroic Mirror. Generation of the 200 nm pulses and the translation stage is not shown 31 2.2.5 Detection strategy As discussed before the current pulse generated at the MCP anode is coupled out via a capacitor from a home built electronics box. After pre-amplification the analog signal has been fed to a high speed ADC card (EON121G20, 1GHz, DynamicSignals LLC) and analyzed using LabVIEW programming interface. We operated our detector in counting mode, i.e., we limit the number of photoelectrons hitting the detector to 10 or less per laser pulse. The histogram of the electron arrival times (for a typical 1000-5000 laser pulses) with respect to the laser trigger represents the time of flight spectra. A typical threshold of 200 mV (-ve polarity) has been set in the LabVIEW program to distinguish the signal (<500 mV, -ve polarity) from the background noise. The time resolution of digitizer card (1 ns), sets the electronic limit of the overall energy resolution. A faster sampling (>1 GHz) will be able to overcome this limit. The LabVIEW data acquisition program saves the acquired time of flight spectra as '.txt' file for further processing. 2.2.6 Experimental strategy for transient photoelectron measurement In time resolved mode, typically both the pump and the probe beam yield photoelectrons due to resonant one color two photon ionization. In most cases we try to choose probe wavelength outside the molecular absorption band to reduce two photon contribution, however, detuning the probe wavelength is limited by the available photon energy needed to ionize from the excited state surface and the photo products. In order to separate the one color two photon ionization signals (from both pump and the probe beams) from the desired transient photoemission signal, one needs to subtract the pump only and probe only signals from the total signal. To perform the subtraction for each pump-probe delay point, we implemented double chopper configuration. In this scheme, both the pump and the probe beam passes through two 32 identical mechanical choppers (Thorlabs) each running at 250 Hz (quarter of the laser repetition rate, 1 kHz) with a constant relative phase difference of /2 between them. Using this configuration (cf. Figure 2.10) we can acquire a complete transient photoemission spectra for 4 laser shots without worrying about long time laser intensity degradation and 1/f noise. 2.2.7 Spectrometer Calibration To calibrate our spectrometer and evaluate the programming performance we performed two photon ionization of Nitric Oxide (NO) gas. The photoemission spectra of NO is well known 17 and has been used previously to characterize magnetic bottle spectrometer. 18 Following the earlier report we performed photoionization using ultrashort pulses at 200 nm wavelength Figure 2.10. Transient photoemission detection scheme using two choppers. The dark shades of the chopped signals denote blocked pulses. See text for details 33 and the time of flight spectra of NO has been recorded for typically 100,000 laser shots. The electron counts were kept at ~10/pulse by controlling the laser light intensity. The vibrationally structured NO time of flight spectra is then calibrated against the literature reported kinetic energy and fit with the function expressed below. (2.6) where t 0 and E 0 represent timing and energy offset respectively. The typical values for the fitting parameters are as follows: l = 50 ± 2 cm, t 0 = 100 ± 20 ns and E 0 = 200 ± 20 meV. A typical calibration spectra is shown in Figure 2.11. The vibrational progression observed in the kinetic energy spectrum helps us to estimate energy resolution of our spectrometer, e.g., E = 0.17 eV at E=2.55 eV yielding E/E ~ 7%. However, the relative energy resolution ( E/E) is better for lower kinetic energy electrons as discussed in section 2.2.3.2. The calibration experiment with NO is done in presence of water jet flowing inside the source chamber, however, the jet is moved away from the laser focus (towards the permanent magnet). The calibration parameters are very sensitive to the permanent magnet position and the jet position. Usually the E 0 value changes significantly by moving the nozzle in and out of the laser focus due to jet's electrokinetic potential (streaming potential). 19-22 Usually, we can reduce the effect by adding small amount of electrolytes (e.g., 20-30 mM Sodium Chloride). 2 Recently, Suzuki and coworkers 19 and Lübcke and coworkers 22 established standard protocol to measure and correct for streaming potential (φ). Although we did not exclusively measure this extra offset potential in every measurement, the experimental value ranges between 100-200 mV for water jet and is in agreement with the reported values in the literature (Figure 2.12). 34 2.8 Project Goals After successful calibration of the magnetic bottle spectrometer with gas phase Nitric Oxide, we aim to implement it to study photoionization of aromatic amino acids in aqueous solutions (Chapter 3) and ultrafast relaxation dynamics of hydrated electron (Chapter 4). Figure 2.11. Calibration of the photoelectron spectrometer with Nitric Oxide (NO) in presence of the water jet. a) Energy level diagram of the NO states involved in the resonant photoionization using 200 nm ultrashort pulses. The observed vibrational structure (b and d) represents vibronic transitions to the cationic surface (D 0 ). The time of flight values for each peak is calibrated against the literature kinetic energy values 17 and fitted with eqn. 2.6. The calibrated kinetic energy spectra is shown in d). The peak highlighted with the asterisk originates from multiphoton ionization of the water vapor in the vicinity. 35 Figure 2.12. Estimation of streaming potential of water jet containing ~30 mM NaCl. Resonant photoemission of Nitric Oxide (NO) is performed with 200 nm in presence of the water jet. The kinetic energy of the NO(v=0)→NO + (v'=2) band is plotted against the position of the jet (x) with respect to the laser focus. The data is fitted with the following function: eKE(x)=eKE( )- (L/L+x)φ, where L, φ are the distance of the laser focus from skimmer entrance and the streaming potential respectively. From the above fitting the streaming potential is found to be 112±30 mV. 36 References 1. Faubel, M.; Steiner, B.; Toennies, J. P., Photoelectron spectroscopy of liquid water, some alcohols, and pure nonane in free micro jets. J. Chem. Phys. 1997, 106, 9013. 2. Winter, B.; Faubel, M., Photoemission from liquid aqueous solutions. Chem Rev 2006, 106, 1176-1211. 3. Winter, B., Liquid microjet for photoelectron spectroscopy. Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment 2009, 601, 139-150. 4. Winter, B.; Aziz, E. F.; Hergenhahn, U.; Faubel, M.; Hertel, I. V., Hydrogen bonds in liquid water studied by photoelectron spectroscopy. J Chem Phys 2007, 126, 124504 5. Link, O.; Vohringer-Martinez, E.; Lugovoj, E.; Liu, Y. X.; Siefermann, K.; Faubel, M.; Grubmuller, H.; Gerber, R. B.; Miller, Y.; Abel, B., Ultrafast phase transitions in metastable water near liquid interfaces. Faraday Discuss 2009, 141, 67-79. 6. Faubel, M.; Siefermann, K. R.; Liu, Y.; Abel, B., Ultrafast Soft X-ray Photoelectron Spectroscopy at Liquid Water Microjets. Accounts Chem Res 2012, 45, 120-130. 7. Seidel, R. PhD Thesis. Technical University Berlin, Berlin. 8. Kurtz, R. L.; Usuki, N.; Stockbauer, R.; Madey, T. E., Measurements of electron attenuation lengths in condensed molecular solids. Journal of Electron Spectroscopy and Related Phenomena 1986, 40, 35-58. 9. Winter, B.; Weber, R.; Schmidt, P. M.; Hertel, I. V.; Faubel, M.; Vrbka, L.; Jungwirth, P., Molecular structure of surface-active salt solutions: Photoelectron spectroscopy and molecular dynamics simulations of aqueous tetrabutylammonium iodide. J. Phys. Chem. B 2004, 108, 14558. 10. Wiley, W. C.; McLaren, I. H., Time ‐of ‐Flight Mass Spectrometer with Improved Resolution. Rev Sci Instrum 1955, 26, 1150-1157. 11. White, M. G.; Rosenberg, R. A.; Gabor, G.; Poliakoff, E. D.; Thornton, G.; Southworth, S. H.; Shirley, D. A., Time ‐of ‐flight photoelectron spectroscopy of gases using synchrotron radiation. Rev Sci Instrum 1979, 50, 1268-1273. 12. Bachrach, R. Z.; Brown, F. C.; Hagström, S. B. M., Photoelectron spectroscopy by time−of−flight technique using synchrotron radiation. Journal of Vacuum Science & Technology 1975, 12, 309-312. 13. Kennerly, R. E., High ‐resolution pulsed electron beam time ‐of ‐flight spectrometer. Rev Sci Instrum 1977, 48, 1682-1688. 14. Kruit, P.; Read, F. H., Magnetic field paralleliser for 2π electron -spectrometer and electron-image magnifier. Journal of Physics E: Scientific Instruments 1983, 16, 313. 37 15. Dragomir, A.; McInerney, J. G.; Nikogosyan, D. N., Femtosecond measurements of two- photon absorption coefficients at lambda=264 nm in glasses, crystals, and liquids. Applied Optics 2002, 41, 4365-4376. 16. Mohammadhassan, V.; Davide, D. A.; Felice, G.; Raffaele, V.; Carlo, A., Temporal and spectral characterization of femtosecond deep-UV chirped pulses. Laser Physics Letters 2015, 12, 025302. 17. Jarvis, G. K.; Evans, M.; Ng, C. Y.; Mitsuke, K., Rotational-resolved pulsed field ionization photoelectron study of NO+(X 1Σ+,v+=0 –32) in the energy range of 9.24–16.80 eV. The Journal of Chemical Physics 1999, 111, 3058-3069. 18. Buchner, F.; L bcke, A.; Heine, N.; Schultz, T., Time-resolved photoelectron spectroscopy of liquids. Rev Sci Instrum 2010, 81, 113107. 19. Tang, Y.; Suzuki, Y. I.; Shen, H.; Sekiguchi, K.; Kurahashi, N.; Nishizawa, K.; Zuo, P.; Suzuki, T., Time-resolved photoelectron spectroscopy of bulk liquids at ultra-low kinetic energy. Chem Phys Lett 2010, 494, 111-116. 20. Shen, H. A.; Kurahashi, N.; Horio, T.; Sekiguchi, K.; Suzuki, T., Direct Measurement of Vertical Electron Binding Energies of Solvated Electrons in Methanol and Ethanol. Chemistry Letters 2010, 39, 668-670. 21. Shreve, A. T.; Elkins, M. H.; Neumark, D. M., Photoelectron spectroscopy of solvated electrons in alcohol and acetonitrile microjets. Chem Sci 2013, 4, 1633-1639. 22. Preissler, N.; Buchner, F.; Schultz, T.; Lübcke, A., Electrokinetic Charging and Evidence for Charge Evaporation in Liquid Microjets of Aqueous Salt Solution. The Journal of Physical Chemistry B 2013, 117, 2422-2428. 38 Chapter 3. Measuring the Vertical and Adiabatic Ionization Energies of Aromatic Amino Acids in Water 3.1 Introduction The redox properties of aromatic amino acids, specifically Tryptophan and Tyrosine are of broad scientific interest as the side groups of these amino acids provide the most versatile and ubiquitous redox activity in the functional proteins of living systems 1-3 . For example, both Tryptophan and Tyrosine are found to play a significant role in DNA damage repair mechanism by transferring an electron (or both an electron and a proton) to the nucleobase radical cation (or neutral radical) that is the primary product of the oxidative DNA damage 4-5 . Knowledge about equilibrium and nonequilibrium ionization parameters, such as, adiabatic and vertical ionization energies of the constituent amino acids, is also important in the study of radiation damage to proteins. Similarly to oxidative damage in DNA, the initial photo- oxidized site in protein is found to be non-local due to hole migration along the peptide backbone 6 , and this can lead to harmful cross-links between amino acid residues far away from the primary oxidation site. The rates of hole transfer processes are often approximated by Marcus theory where the free energy of the electron transfer reaction, and therefore the redox potential of the electron donor/acceptor couple, plays a prominent role 7-8 . The redox properties of Tryptophan and Tyrosine have been studied extensively, however, there is not yet consensus in the reported standard oxidation potential 9-11 . One of the major reasons for the experimental disagreement can be attributed to the complex protonation/deprotonation equilibria that change the measured electrode potentials as a function of pH. Often the standard redox potential (E 0 at pH=0) and measured redox potential at neutral 39 solutions (E 7 at pH=7) do not follow a linear Nernstian relationship due to existence of several acid-base equilibria, i.e., several pK a contributing over the measured pH range. Recently, Bradforth and coworkers have shown that an alternative measurement technique, liquid microjet photoelectron spectroscopy, can be used as a reliable tool not only for measuring vertical ionization energies but also to estimate standard redox potentials for phenol and DNA nucleotides and nucleosides in aqueous solutions. 12-13 Being inherently an ultrafast nonequilibrium technique, photoemission measurements circumvent any contribution to E 0 value coming from fast deprotonation reactions and the irreversibility of the electrochemical process unavoidable on the timescale of standard measurement techniques such as cyclic voltammetry. This is a well known problem for oxidation of many organic systems. 14 In this report we have employed two different photoionization techniques: Liquid- microjet X-ray photoelectron spectroscopy (XPS) and resonant two photon ionization photoelectron spectroscopy (R2PI-PES) to measure the ionization energies of the aromatic amino acids in solution. XPS and R2PI-PES differ with respect to several physical parameters such as, chemical selectivity and sensitivity, relative ionization cross section and probing depth. In resonant two photon ionization the first photon excites the molecules to the excited state (HOMO→LUMO transition) and the second photon subsequently ionizes from the resonant intermediate state under the same pulse envelope. Therefore, we can predict a priori the chemical site or the molecular orbital from which ionization will take place based on the linear absorption spectra and choice of the resonant excitation/ionization wavelength. 15 In Figure 3.1 we explain the chemical selectivity in case of aromatic amino acids: when the excitation energy is on resonance with the → * transition of the aromatic moiety, the ionization takes place only from the electron promoted from the  orbital producing a final state with  -1 configuration; little or no 40 ionization occurs from the amino or carboxylic acid backbone which absorb at higher energy. On the other hand, in XPS the high energy radiation indiscriminately ionizes from all the valence (and even core orbitals if the photon energy is high enough) with propensities determined entirely by ionization cross section, and therefore, imparts no chemical selectivity to the ionization. The relative ionization cross-sections are also very different in R2PI-PES measurements due to involvement of the intermediate resonant state. The total ionization signal in resonant measurement depends on both absorption and subsequent ionization cross-section, whereas, for XPS the ionization is solely dependent on one-photon ionization cross section. In general, high absorption cross-sections in the R2PI-PES process, and higher available photon flux, provides superior sensitivity over single photon ionization for dilute solutions. In this report we have illustrated the enhanced sensitivity of the resonant two photon ionization technique in case of sparingly soluble L-Tyrosine solution. Figure 3.1. Schematic illustration of X-ray and resonant two photon ionization photoemission process. High energy X-ray radiation ionizes from all the valence orbitals and produces photoelectrons at high kinetic energies (yellow arrows). On the contrary, in resonant photoemission, ionization takes place only from those orbitals which are involved in the electronic excitation (purple arrows). Since the total energy deposited in the latter process is much smaller, it yields lower kinetic energy electrons. 41 Recent soft X-Ray photoemission measurements probe the surface sensitivity of the photoemission technique as a function of outgoing electron kinetic energy 16-18 . It has been found that the probing depth is at a minimum at lower kinetic energies than previously expected with reference to so called “universal” probing depth curves or with respect to theoreti cal estimates of the energy dependent inelastic mean free path. Although the issue of probing depth at ultralow kinetic energy is still under debate 19 and a topic of active research, current results indicated that the effective electron attenuation length (EAL) should be different in XPS and in R2PI-PES measurements. 3.2 Experimental All X-ray photoelectron spectroscopy measurements were performed at the U41-PGM beam line at the BESSY synchrotron radiation facility in Berlin using 175 eV photon energy. Experimental details of the photoelectron spectrometer and the liquid microjet have been reported elsewhere. 20 Briefly, a liquid jet of 25 µm diameter was injected into vacuum from a fused-silica nozzle; jet velocity and temperature were approximately 40 ms -1 and 25°C. Photoelectrons are detected parallel to the synchrotron light polarization vector and perpendicular to the flow of the liquid jet. Emitted photoelectrons pass from the main interaction chamber (operating at 10 -4 mbar) through a 500 µm diameter orifice to the differentially pumped detector chamber (operating at 2 10 -6 mbar) which houses a hemispherical electron energy analyzer equipped with a multichannel detector. The small distance of 0.5 mm between the liquid jet and the orifice assures that detected electrons have not suffered from inelastic scattering with water gas-phase molecules near the jet surface 20-21 . The energy resolution of the U41-PGM beam 42 line was better than 65 meV at 175 eV photon energy used for the valence PE measurements and the energy resolution of the hemispherical analyzer, ~100 meV at 10 eV pass energy, was constant with kinetic energy. Resonant two photon ionization experiments were done at USC using ultrafast deep ultraviolet (DUV) pulsed excitation at 267 nm. The DUV pulse was generated by sum frequency mixing of the fundamental (800 nm, 30 fs) from a Ti: Sapphire amplifier system (Coherent Legend Elite) and its second harmonic (400 nm) in a 100 m thick type II BBO crystal. The pulse width was estimated to be ~200 fs from two photon absorption in a 1 mm thick quartz film 22-23 . The polarization of 267 nm used in the experiment was vertical with respect to the laboratory frame and perpendicular to the time of flight axis (orthogonal polarization geometry with respect to the synchrotron experiment). The spot size of the beam at the focus was estimated to be not more than 80 m. In our home built photoelectron spectrometer we have implemented a magnetic bottle time of flight electron detection strategy which enables energy-dispersed detection with significantly higher collection efficiency (~50%) than field free time of flight detection 24-26 . Photoelectrons were detected at the end of a 50 cm flight tube using a pair of 40 mm diameter multichannel plate (MCP) detectors in Chevron configuration (Beam Imaging Solutions). The signal from the anode of the MCP stack was capacitively coupled out and amplified with a 100  gain preamplifier (Phillips Scientific, model 6954B-100) and digitized with a high speed digitizer card (1GHz, DynamicSignals LLC) so that multiple electrons per shot can be resolved and recorded. In all the experiments the count rate was maintained to ~10 electrons/s by varying the laser pulse intensity and the DUV pulse energy was always kept well below 80 nJ to minimize higher order effects. The detection chamber was pumped down to 8  10 -7 mbar using two turbomolecular pumps (500 L/s, Pfeiffer Vacuum) during operation. The 43 source chamber was maintained to 2 10 -6 mbar pressure by using liquid nitrogen cryo traps and a turbomolecular pump (1500 L/s, Pfeiffer Vacuum). The spectrometer was calibrated using vibrationally-resolved two photon photoemission spectra of NO gas with 200 nm pulses 26 . The relative energy resolution of the spectrometer ( E/E) is ~7% at 2.55 eV kinetic energy determined from NO photoemission spectra (cf. Figure 2.10). We note that the spectrometer energy resolution is a convolution of kinetic energy spread due to flight time distribution, spectral bandwidth of the excitation/ionization pulse and time response of the detection electronics. The last two contributions are constant while the energy resolution due to flight time distribution varies with square root of the electron kinetic energy 24 , i.e., higher energy resolution at lower kinetic energy. Aqueous solutions of L-Phenylalanine (100 mM), L-Tryptophan (40 mM) and L-Tyrosine (~2 mM) (from Sigma-Aldrich) were prepared without further purification. The pH of the solutions were maintained at pH=7.4-7.6 using Tris/HCl buffer in R2PI-PES experiments. In our XPS experiments, however, the solution pH was not controlled by the buffer solution: measured pHs are 6.1 for Phenylalanine and 6.5 for Tryptophan solutions prepared. In both experiments sodium chloride/fluoride salts are added (~20 mM) to minimize streaming potential. 20-21, 27-31 Prepared solutions were injected into the vacuum chamber using a HPLC solvent delivery pump (Shimazdu) with a constant flow rate of 0.5 mL/min and a backing pressure of 4-5 bar. The jet thickness was ~20 m based on the liquid- jet nozzle diameter. 44 3.3 Results and Discussion X-ray photoemission (top row) and resonant two photon ionization (bottom row) spectra of the aromatic amino acids are shown in Figure 3.2. The XPS spectra reported here are derived by carefully subtracting the solvent reference spectrum under identical experimental conditions (i.e., same salt concentration). Raw spectra were energy calibrated against the water 1b 1 binding energy (11.31 eV) 31 and intensity normalized against the water 1b 2 peak and high binding energy background of water. R2PI-PES measurements, on the other hand, are background free as the excitation energy (4.64 eV) is only resonant with the → * transition and is several eV lower than the water absorption band edge. 32 Figure 3.2. X-ray photoemission (a-c) and resonant two photon ionization (d-f) spectra of the aromatic amino acids in water (solid circles). Individual Gaussian fits and the total fit are shown in blue dashed lines and solid red lines respectively. 45 Photoemission bands are fitted with a Gaussian or sum of Gaussian functions. The band centers of the fitting functions have been assigned as the vertical ionization energies and are listed in Table 1. Tryptophan has the lowest (7.3 eV) and Phenylalanine the highest vertical ionization energy, some 1.4 eV higher, for the aromatic amino acids family. We find the vertical ionization energies are almost identical for XPS and R2PI-PES measurements, which primarily suggests the absence of any ultrafast dynamics in the intermediate resonance state accessed in the latter experiment (see below). Since in these two measurements the EAL are expected to be quite different 18 as discussed before, we can also conclude that the orbital energies do not alter dramatically with variation in the local environment. In XPS measurements several photoemission bands are apparent, but the available photon energy in the resonant photoionization experiment (2  4.64 = 9.28 eV) limits the range of detection often to the lowest one or two ionization channels. The sharp falling edge in the higher binding energy of the R2PI-PES spectra reflects the detector transmission function at low electron kinetic energy and causes greater uncertainty of higher binding energy peak positions (± 0.1 eV) when fitted with a Gaussian function. Due to lower solubility of L-Tyrosine in water (~3 mM), it was practically impossible to obtain an X-ray photoelectron spectra with an acceptable signal to noise ratio (S/N) even after longer acquisition time (Figure 2, c). However, the resonance enhancement in the R2PI-PES measurement yields PE spectra with excellent S/N and comparable to the other amino acids (Figure 2, f). This clearly illustrates the superior sensitivity of the R2PI-PES measurements over single photon ionization techniques for samples at low concentration. There is an extensive literature on the gas phase photoemission of isolated amino acids. Prior works by Ham 33 , Campbell 34-35 , Inokuchi 36 and Prince 37 reported vertical ionization 46 energies of isolated aromatic amino acids. These experimental results along with theoretical calculations 38 facilitated the spectral assignments of the gas phase photoemission bands. The peaks between 8-11 eV were assigned to ionization from the -orbitals centered on the aromatic moiety (using the notation of ref 35,  3 : HOMO,  2 :HOMO-1) and the nonbonding orbital of amine nitrogen (n N ) in the amino acid residue. 33-36 Comparing our solution X-ray photoemission spectra against the gas phase results, we can assign the XPS bands as indicated in Table 1. The energetics of the ionization process is expected to be affected by the introduction of a highly polar environment when transitioning from gas phase to aqueous solution. However, the extent of the solvent influence on the VIE will depend strongly on the relative stabilization of the initial (neutral) and destabilization of the final (radical cation) state in water. 39-40 For example, we see the solvation shift in the vertical ionization energy is greater for the polar chromophores Tryptophan and Tyrosine as compared to non-polar chromophore in Phenylalanine. Figure 3.3. UV-VIS absorption spectra of aromatic amino acids in water. For comparison absorption spectra of Benzene, Indole and Phenol (in water) absorption spectra is also shown. Solid violet band shows the excitation beam spectrum centered at 267 nm (FWHM~3 nm). 47 Assigning the R2PI-PES band, on the other hand is conceptually easier, since in the one electron picture, the resonant photoionization must take place solely from the orbital promoted from in the optical excitation. In Figure 3.3 we present the UV/Vis absorption spectra of the aromatic amino acids and their aromatic side chain precursors. At our excitation wavelength (267 nm), we only excite the aromatic moieties in the amino acids ( → * transition). Hence, the resonant photoemission spectra is attributed to ionization from occupied  orbitals and confirms our XPS assignments based on gas phase literature discussed above. We note that for Tryptophan and Tyrosine we observed two resonant photoemission bands which we assign to ionization from HOMO (  3 ) and HOMO-1(  2 ) orbital. This might seem quite counter-intuitive since electronic excitation primarily involves HOMO→LUMO transition, therefore, subsequent ionization should only produce final state with  3 -1 configuration. In case of Tryptophan, two overlapping transitions S 0 → 1 L a and S 0 → 1 L b (based on Platt-Murrell's nomenclature for aromatic molecules) 41-42 constitute the electronic absorption band between 240-300 nm 43 (cf Figure 3). Roos and coworkers have shown that the major contribution to the above two transitions correspond to HOMO→LUMO (54%) and HOMO - 1→LUMO (44%) transitions respectively. 44 Therefore, we can argue that the admixture of those transitions at our resonant photoionization wavelength leads to both  3 -1 and  2 -1 electronic configuration final states. However, this argument does not apply to Tyrosine since the upper * state is too high in energy to be excited with 267 nm pump. 45 In resonant multiphoton ionization of isolated phenol, Weber and coworkers observed photoelectron bands due to configuration interaction. 46 They estimated 12% contribution to the total photoelectron signal from configuration interaction when excited to the S 1 state intermediate. We have also performed R2PI-PES measurements on aqueous phenol and a comparison with the reported X-ray 48 photoemission spectra 12 reveals that the excited state configuration interaction also contributes in the condensed phase. The presence of  2 -1 configuration in the final states for both Tryptophan and Tyrosine hence indicates significant configuration interaction in the intermediate excited state. To the best of our knowledge this is the first time the experimental evidence of configuration interactions has been presented in the condensed phase using photoemission spectroscopy. We can also extract the adiabatic ionization energy (AIE) from the onset of the photoelectron spectra as listed in Table 2. The onset has been defined as the intersection of the noise level and the rising edge of the PE spectra when plotted in the semilog scale in the intensity axis. 47 The rising edge has been defined by the slope line drawn at 1/e 2 value of the lowest binding energy peak intensity (red line in Figure 3.4). The noise level, on the other hand, is an ill-defined quantity and depends on the instrument sensitivity at the low signal intensity. We have noticed that the resonant photoemission measurements offers a greater dynamic range in Figure 3.4. Estimation of Adiabatic Ionization Energies (AIE) of aromatic amino acids. Intensity axis is plotted in semilog scale to emphasize the noise level at low signal intensity. The red line indicates the rising edge of the signal and the blue region denotes the spread in noise levels (see text). Due to reference subtraction, the dynamic range in the signal intensity is low (<400:1) in XPS compared to R2PI measurements (>2000:1) which makes the assignmenet of noise level more ambiguious in XPS data 49 electron counts (>2000:1) than X-Ray photoemission experiments (<400:1) since we are not subtracting comparable solvent background signal and the associated noise in the former technique; this affects the noise level especially in the low signal intensity region (cf Figure 4). This particular advantage allows us to better assign the noise floor while extracting the ionization onset. Therefore, we use our R2PI-PES results to estimate the adiabatic ionization energy. We note that a range of estimates for the baseline noise can be assigned in each of the R2PI-PES spectra, so we choose to define this uncertainty (±0.1 eV) in the estimated ionization threshold (cf. Figure 3.4). The adiabatic ionization energy is the energy difference between the neutral (initial) and the cationic (final) states in their respective equilibrium geometries, [AIE=E(cation, v"=0)- E(neutral, v'=0)]. If we assume the general reaction coordinate of the potential energy surfaces to be the collective motion of the solvent molecules, the energy surfaces of the neutral and cationic states can be expressed as Marcus parabolas in one electron transfer reaction 7 . Now, the energy difference between two parabolas at their respective equilibrium solvent configuration represents the change in free energy due to oxidation ( G 0 ) and therefore equivalent to the standard one electron oxidation potential (E 0 ; G 0 =-nFE 0 ). Hence, we can compare our estimated AIE values with the reported E 0 values from the electrochemistry literature. Harriman reported standard oxidation potentials of Tyrosine and Tryptophan in water using cyclic voltametry and illustrated the variation of E 0 values with pH. 9 The equilibrium redox reaction in water under cyclic voltametry condition involves both electron and proton transfer as shown below Trp Trp(-H) ● + H + + e - (3.1a) Tyr Tyr(-H) ● + H + + e - (3.1b) 50 The quantity of interest, G 0 for the one electron oxidation reaction without follow up reactions is best achieved in photoionization process, which yields the instantaneous distribution of oxidized species, before the slower proton transfer step, as shown below. Trp Trp +● + e - (3.2a) Tyr Tyr +● + e - (3.2b) Evidently the electrochemical and photoionization estimations of standard redox potential involve different half-cell reactions (eqn. 3.1 and 3.2) which explains the difference between the E 0 values listed in Table 2. To the best of our knowledge we could not find any literature reporting E 0 (A +● /A, A≡ Trp or Tyr) values in aqueous solution using fast equilibrium techniques or cyclic voltammetry in aprotic solvents to validate our spectroscopic estimates. However, the reorganization energy which is defined by the difference between the vertical and adiabatic ionization energy, is found to be ~1.2-1.5 eV which is within the acceptable range of values for biomolecules with similar aromatic moieties 12-13 , therefore validating that the spectroscopic E 0 (A +● /A) are not unreasonable. To extend our spectroscopic measurement technique for E 0 estimation, we performed R2PI-PES experiments on the following nucleosides/nucleotides: Cytidine (Cyt), Adenosine (Ado) and deoxyguanosine monophosphate (dGMP) in water where the E 0 (A +● /A, A≡ Nucleobase) have been subject to more detailed study, in particular by electrochemical measurements in aprotic solvents. In a recent report, Schroeder et al. reported the XPS measurements of these DNA components and derived the standard one-electron oxidation potentials in aqueous solutions. 13 The authors addressed the subtlety in calculating standard state E 0 values by carefully considering all possible acid–base equilibria of the reduced and oxidized 51 form of the redox couple and illustrated the deviation from the E 0 values reported earlier in the literature [Table 4 in Ref 13]. Most importantly they established that E 0 values arrived at from VIEs from photoelectron spectroscopy and computed reorganization energies within a polarized continuum model for the solvent were in good agreement with electrochemical E 0 measured in the aprotic solvent acetonitrile. We compare E 0 values derived from R2PI-PES of the nucleosides/nucleotide (Figure 3.5) when referenced to the standard hydrogen electrode (E 0 = AIE - E 0 (H + /½ H 2 | Pt) where E 0 (H + /½ H 2 | Pt) = 4.28 eV 12, 48 ) with those reported in Ref 13 in Table 3.2. We notice that the values extracted from the R2PI-PES measurements in all cases overestimate both the E 0 reported in Schroeder et al., as well as the reported E 0 in acetonitrile, by almost 0.4-0.8 eV. Comparing the R2PI spectra with the XPS spectra in each case also suggests poorer alignment of the thresholds than we see in Figure 3.4 here for amino acids. In order to explain this disparity in the E 0 values, we need a closer look into the physical picture of the R2PI-PES process. In resonant photoemission, the initial excitation to the resonant state and the subsequent ionization takes place within the same optical pulse envelope. Our Figure 3.5. R2PI-PES spectra of Cytidine (Cyt), Adenosine (Ado) and deoxyguanosine monophosphate (dGMP) in water. AIE has been estimated as described in the text. We found that our measurement overestimates the standard oxidation potential by ~ 0.6 V(± 0.2 V) when compared to the values reported in ref 13 (cf Table 2). See text for explanation. 52 initial picture might be to assume the excitation pulse to be close to a delta pulse in time ( t ~ 0) 49 , therefore, subsequent ionization happens before the commencement of any wavepacket motion away from the Frank Condon region of the intermediate electronically excited state surface. However, with finite laser pulse width ( t > 0), nuclear reorganization and/or electronic relaxation can occur within the pulse envelope. This ensuing ultrafast nuclear and electronic dynamics will alter the onset of ionization as illustrated in Figure 3.6. With the characteristic relaxation time  relax and the relaxation energy as E, we can formulate an empirical relationship between observed AIE( t) and the ideal AIE(0) (when there is no relaxation) as follows (3.3) Only for a delta pulse excitation or infinite relaxation time, t/  relax →0, the observed and actual ionization onset would be are identical, i.e., AIE( t) = AIE(0). We can now examine the validity of eqn. 3.3 for a test case of Adenosine. From ultrafast transient absorption studies in aqueous solution 50 we have found the initially photoexcited singlet state ( 1 L a ) of Adenosine relaxes to the close lying 1 L b surface with a rate of (55 fs) -1 . Using the AIE( t) estimated here from our R2PI-PES experiment (Table 2), AIE(0) derived from E 0 values in Ref 13 and t=200 fs, we found the relaxation energy ( E) to be 0.3±0.1 eV. This value is in excellent agreement to the calculated energy difference of 0.36 eV between 1 L a and 1 L b in water. 51 This observation corroborates our physical picture of alteration of the AIE due to variation of the laser pulse width, excited state relaxation time and the relaxation energy, hence explaining the difference in the E 0 values emerging from R2PI-PES measurements and those derived in Ref 13. A similar explanation applies to dGMP and cytidine where ultrafast electronic 53 relaxation has been invoked to describe the excited state dynamics of the nucleobase chromophore. We now consider how this picture translates to the case of amino acids. The initial excited state dynamics of the aromatic amino acids are expected to resemble that of their model aromatic precursors. The excited state photophysics of aqueous Tryptophan and its model chromophore indole has been explored extensively at 266-290 nm excitation using various time Figure 3.6. Schematic representation of the resonant two photon ionization process. With delta pulse excitation (Case 1), the initial wavepacket launched in the S 1 surface (FC region) undergoes ionization prior to any nuclear and electronic relaxation and reaches final state D 0 (radical cation). The excess energy is manifested as the electron kinetic energy. Maximum available kinetic energy corresponding to S 0 (v=0)→D 0 (v=0) transition is shown as KE 0 max . The adiabatic ionization energy can be defined as AIE(0)=2*h -KE 0 max . In case of finite pulse laser (Case 2), relaxation takes place within the pulse width ( t). Observed adiabatic ionization energy AIE( t)=2*h - KE t max . Since KE 0 max ≥ KE t max , observed ionization threshold for Case 2 will always be equal to or higher than the actual value (no relaxation, Case 1). 54 resolved techniques. 52-55 In a fluorescence up-conversion study, Bräm and coworkers found an ultrafast relaxation component with 160±40 fs time scale with a 720±110 cm -1 shift for Tryptophan in water which is attributed to the fast inertial response of the solvent 52 . Using these parameters we estimate the change in adiabatic ionization energy within our pulse width (200 fs), compared to the strictly instantaneous limit, to be 0.09±0.01 eV for Tryptophan, well within the error of our AIE estimation (±0.1 eV). Recent pump-probe studies by Barry and coworkers on aqueous Tyrosine (at pH=9) at 280 nm pump show excited state dynamics with the fastest time scale >10 ps attributed to sequential or coupled electron proton transfer. 56 The authors did not observe any sub-picosecond dynamics within their time resolution of 360 fs. Similarly, our recent broadband transient absorption studies on aqueous phenol solution at 267 nm pump (resolution <50 fs) do not show any indication of subpicosecond solvent or vibrational relaxation 57-58 . The observed long time (>1 ps) relaxation dynamics observed in Tyrosine (and its model chromophore Phenol) essentially has no effect on the R2PI measurements as it happens at times much longer than the pulse width (200 fs), i.e., t/  relax →0. In case of benzene, the aromatic chromophore of Phenylalanine, the first excited state has a lifetime in the order of 2 ns in water when excited close to the S 1 origin ca. 260 nm. 59 The experimental fluorescence lifetime did not change when the pump wavelength was reduced to 249 nm (1773 cm -1 higher than the 0- 0 transition), therefore indicates that the onset of the 'channel 3' relaxation pathway 60-61 lies at even lower wavelength (higher energy). From the absorption spectra of Phenylalanine (cf. Figure 3), we can say at 267 nm we are exciting close to the 0-0 transition and expect no vibrational relaxation ( E~0). The reported electronic relaxation time scale is also orders of magnitude higher than our pulse width 60 . Therefore, the two fold effect of minimal excited state relaxation energy ( E~0) and slow relaxation time ( t/  relax ~0) result AIE( t) to be almost identical to 55 AIE(0) according to eqn. 3. Hence, we conclude that in cases where there is only minor intermediate state relaxation dynamics, R2PI-PES experiments are able to provide useful standard oxidation potentials; the precision of these values, which is not as high as those determined by typical electrochemical methods, are determined by the experimental error in establishing the threshold, ~  0.1 eV. 3.4 Conclusion We have measured the vertical ionization energies of the aromatic amino acids in aqueous solutions using two different photoemission techniques. We find that Tryptophan has the lowest ionization energy followed by Tyrosine and Phenylalanine. This result is consistent with the order of expected ease of oxidation of the amino acids in aqueous solution. We do not observe any significant variation of the orbital energies as we go from near threshold ionization in R2PI-PES to X-ray ionization which involves photon energies in large excess of the ionization thresholds and leading to ejected electrons with high kinetic energy (>150 eV). We have also demonstrated the superior sensitivity of the resonant two photon ionization scheme for solutions with low solute concentration in contrast to the single-photon XPS measurements used hitherto and often requiring solute concentrations of 0.1 – 1 M. This higher sensitivity along with chemical selectivity of the R2I-PES technique can be exploited both for spectroscopic assignment and as an analytical tool. The near-instantaneous nature of the photoionization process has been exploited leading to estimation of intrinsic standard redox potentials for systems in aqueous solutions that otherwise rapidly deprotonate, on a timescale faster than the establishment of the reversible equilibria 56 required for standard electrochemical techniques like cyclic voltammetry. Ionization thresholds can be established with greater certainty using R2PI-PES, and for situations where there are no significant intermediate state relaxation dynamics taking place on the timescale of the pulse width, such as the aromatic amino acids, reliable standard oxidation potentials E 0 (A +● /A) can be derived. On the other hand, for systems that do exhibit rapid electronic or vibrational dynamics that lead to > 0.1 eV energetic relaxation, we have explored the dynamical nature of the adiabatic ionization energy revealed in the R2PI-PES and empirically formulated the correlation between estimated ionization onset with electronic and nuclear relaxation on the excited state surface. In principle, reliable oxidation energies could be established for such systems from pulse width dependent photoelectron spectra. This assumes that intermediate state relaxation does not take place so rapidly that the energetic uncertainty in the pulse energies exceeds the other instrumental factors limiting the electron kinetic energy resolution. 57 Table 3.1. Vertical Ionization Energy of the aromatic amino acids in aqueous solutions. The standard error of VIE is ±0.1 eV. Average FWHM of the photoelectron bands in solutions is 0.9±0.1 eV. For comparison gas phase values are also included from ref 33-35 Amino Acids VIE (eV)/XPS VIE (eV)/R2PI-PES VIE (eV)(gas phase) L-Phenylalanine 8.7(  3 ), 9.4(n N), 9.8(  2 ) 8.8(  3 ) 8.9(  3 ), 9.3(n N), 9.7(  2 ) L-Tryptophan 7.3(  3 ), 8.0(  2 ), 8.9(n N), 9.5(n N) 7.3(  3 ), 8.2(  2 ) 7.9(  3 ), 8.3(  2 ), 9.4(n N), 9.8(n N) L-Tyrosine - 8.0(  3 ),8.6(  2 ) 8.5(  3 ), 9.4(  2 ), 9.6(n N) Table 3.2. Adiabatic Ionization Energies (in eV) and standard one electron oxidation potentials (vs SHE, in V) of important biomolecules in aqueous solution. Error associated with AIE is ±0.1 V AIE a E R2PI-PES 0 (A +● /A) b vs SHE E 0 (A +● /A) vs SHE E 7 e vs SHE L-Phenylalanine 7.5 3.22 - - L-Tryptophan 5.9 1.62 - 1.015 f L-Tyrosine 6.5 2.22 - 0.93 f Cytidine 7.3 3.02 2.4 c , 2.14 d ~1.6 g Adenosine 6.8 2.52 2.1 c , 1.96 d 1.44 g Deoxyguanosine Monophosphate 6.6 2.32 1.5 c , 1.49 d 1.31 g a AIE extracted from R2PI-PES experiment in solution (see text for details) b E R2PI-PES 0 =AIE-E 0 SHE where E 0 SHE =4.28 V 12, 48 . We note that the uncertainty of E 0 SHE values in the literature is ±0.2 V. 12, 48, 62 However, this uncertainty is common for all systems and will not change the relative values of the standard redox potential. c E 0 (A +● /A) values from X-ray photoemission measurements (cf Table 4 in ref 13). d E 0 (A +● /A) values in acetonitrile 13 , 63 e E 7 , one electron oxidation potential measured at pH=7 f from ref 9, values are measured vs NHE g from ref 13 58 References 1. Westerlund, K.; Berry, B. W.; Privett, H. K.; Tommos, C. Exploring amino-acid radical chemistry: protein engineering and de novo design. Biochim. Biophys. Acta. 2005, 1707, 103- 116. 2. Kovacic, P. Protein electron transfer (mechanism and reproductive toxicity): Iminium, hydrogen bonding, homoconjugation, amino acid side chains (redox and charged), and cell signaling. Birth Defects Res. C: Embryo Today. 2007, 81, 51-64. 3. Bogdanov, A. M.; Mishin, A. S.; Yampolsky, I. V.; Belousov, V. V.; Chudakov, D. M.; Subach, F. V.; Verkhusha, V. V.; Lukyanov, S.; Lukyanov, K. A. Green fluorescent proteins are light-induced electron donors. Nat. Chem. Biol. 2009, 5, 459-461. 4. Morozova, O. B.; Kiryutin, A. S.; Sagdeev, R. Z.; Yurkovskaya, A. V. Electron Transfer between Guanosine Radical and Amino Acids in Aqueous Solution. 1. Reduction of Guanosine Radical by Tyrosine. J. Phys. Chem. B 2007, 111, 7439-7448. 5. Morozova, O. B.; Kiryutin, A. S.; Yurkovskaya, A. V. Electron Transfer between Guanosine Radicals and Amino Acids in Aqueous Solution. II. Reduction of Guanosine Radicals by Tryptophan. J. Phys. Chem. B 2008, 112, 2747-2754. 6. Schlag, E. W.; Yang, D. Y.; Sheu, S. Y.; Selzle, H. L.; Lin, S. H.; Rentzepis, P. M. Dynamical principles in biological processes: A model of charge migration in proteins and DNA. Proc. Natl. Acad. Sci. 2000, 97, 9849-9854. 7. Marcus, R. A. On the Theory of Oxidation ‐Reduction Reactions Involving Electron Transfer. I. J. Chem. Phys. 1956, 24, 966-978. 8. Marcus, R. A. On the Theory of Oxidation ‐Reduction Reactions Involving Electron Transfer. III. Applications to Data on the Rates of Organic Redox Reactions. J. Chem. Phys. 1957, 26, 872-877. 9. Harriman, A. Further comments on the redox potentials of tryptophan and tyrosine. J. Phys. Chem. 1987, 91, 6102-6104. 10. Jovanovic, S. V.; Harriman, A.; Simic, M. G. Electron-transfer reactions of tryptophan and tyrosine derivatives. J. Phys. Chem. 1986, 90, 1935-1939. 11. Butler, J.; Land, E. J.; Swallow, A. J.; Prutz, W. A. Comment on "Electron-transfer reactions of tryptophan and tyrosione derivatives". J. Phys. Chem. 1987, 91, 3113-3114. 12. Ghosh, D.; Roy, A.; Seidel, R.; Winter, B.; Bradforth, S. E.; Krylov, A. I. First-Principle Protocol for Calculating Ionization Energies and Redox Potentials of Solvated Molecules and Ions: Theory and Application to Aqueous Phenol and Phenolate. J. Phys. Chem. B 2012, 116, 7269-7280. 59 13. Schroeder, C. A.; Pluhařová, E.; Seidel, R.; Schroeder, W. P.; Faubel, M.; Slavíček, P.; Winter, B.; Jungwirth, P.; Bradforth, S. E. Oxidation Half-Reaction of Aqueous Nucleosides and Nucleotides via Photoelectron Spectroscopy Augmented by ab Initio Calculations. J. Am. Chem. Soc. 2015, 137, 201-209. 14. Wardman, P., Reduction potentials of one-electron couples involving free radicals in aqueous solution. J. Phys. Chem. Ref. Data. 1989, 18, 1637-1755. 15. Chatterley, A. S.; Johns, A. S.; Stavros, V. G.; Verlet, J. R. R. Base-Specific Ionization of Deprotonated Nucleotides by Resonance Enhanced Two-Photon Detachment. J. Phys. Chem. A 2013, 117, 5299-5305. 16. Ottosson, N.; Faubel, M.; Bradforth, S. E.; Jungwirth, P.; Winter, B. Photoelectron spectroscopy of liquid water and aqueous solution: Electron effective attenuation lengths and emission-angle anisotropy J. Electron Spectrosc. Relat. Phenom. 2010, 177, 60-70. 17. Thürmer, S.; Seidel, R.; Faubel, M.; Eberhardt, W.; Hemminger, J. C.; Bradforth, S. E.; Winter, B. Photoelectron Angular Distributions from Liquid Water: Effects of Electron Scattering. Phys. Rev. Lett. 2013, 111, 173005. 18. Suzuki, Y.-I.; Nishizawa, K.; Kurahashi, N.; Suzuki, T. Effective attenuation length of an electron in liquid water between 10 and 600 eV. Phys. Rev. E 2014, 90, 010302. 19. Buchner, F.; Schultz, T.; Lubcke, A. Solvated electrons at the water-air interface: surface versus bulk signal in low kinetic energy photoelectron spectroscopy. Phys. Chem. Chem. Phys. 2012, 14, 5837-5842. 20. Winter, B.; Faubel, M. Photoemission from liquid aqueous solutions. Chem. Rev. 2006, 106, 1176-1211. 21. Faubel, M.; Schlemmer, S.; Toennies, J. P. A molecular beam study of the evaporation of water from a liquid jet. Z. Phys. D: At., Mol. Clusters 1988, 10, 269. 22. Dragomir, A.; McInerney, J. G.; Nikogosyan, D. N. Femtosecond measurements of two- photon absorption coefficients at lambda=264 nm in glasses, crystals, and liquids. Appl. Opt. 2002, 41, 4365-4376. 23. Mohammadhassan, V.; Davide, D. A.; Felice, G.; Raffaele, V.; Carlo, A. Temporal and spectral characterization of femtosecond deep-UV chirped pulses. Laser Phys. Lett. 2015, 12, 025302. 24. Kruit, P.; Read, F. H. Magnetic field paralleliser for 2π electron -spectrometer and electron-image magnifier. J. Phys. E: Sci. Instrum. 1983, 16, 313. 25. Rijs, A. M.; Backus, E. H. G.; de Lange, C. A.; Westwood, N. P. C.; Janssen, M. H. M. ‘Magnetic bottle’ spectrometer as a versatile tool for laser photoelectron spectroscopy. J. Electron Spectrosc. Relat. Phenom. 2000, 112, 151-162. 26. Buchner, F.; L bcke, A.; Heine, N.; Schultz, T. Time-resolved photoelectron spectroscopy of liquids. Rev. Sci. Instrum. 2010, 81, 113107. 60 27. Tang, Y.; Shen, H.; Sekiguchi, K.; Kurahashi, N.; Mizuno, T.; Suzuki, Y. I.; Suzuki, T. Direct measurement of vertical binding energy of a hydrated electron. Phys. Chem. Chem. Phys. 2010, 12, 3653-3655. 28. Shen, H. A.; Kurahashi, N.; Horio, T.; Sekiguchi, K.; Suzuki, T. Direct Measurement of Vertical Electron Binding Energies of Solvated Electrons in Methanol and Ethanol. Chem. Lett. 2010, 39, 668-670. 29. Preissler, N.; Buchner, F.; Schultz, T.; Lübcke, A. Electrokinetic Charging and Evidence for Charge Evaporation in Liquid Microjets of Aqueous Salt Solution. J. Phys. Chem. B 2013, 117, 2422-2428. 30. Shreve, A. T.; Elkins, M. H.; Neumark, D. M. Photoelectron spectroscopy of solvated electrons in alcohol and acetonitrile microjets. Chem. Sci. 2013, 4, 1633-1639. 31. Kurahashi, N.; Karashima, S.; Tang, Y.; Horio, T.; Abulimiti, B.; Suzuki, Y.-I.; Ogi, Y.; Oura, M.; Suzuki, T. Photoelectron spectroscopy of aqueous solutions: Streaming potentials of NaX (X = Cl, Br, and I) solutions and electron binding energies of liquid water and X−. J. Chem. Phys. 2014, 140, 174506. 32. Elles, C. G.; Rivera, C.; Zhang, X.; Pieniazek, P. A.; Bradforth, S. E. Electronic structure of liquid water from polarization-dependent two-photon absorption spectroscopy. J. Chem. Phys. 2009, 130, 084501 33. Cannington, P. H.; Ham, N. S. He(I) and He(II) photoelectron spectra of glycine and related molecules. J. Electron Spectrosc. Relat. Phenom. 1983, 32, 139-151. 34. Campbell, S.; Beauchamp, J. L.; Rempe, M.; Lichtenberger, D. L. Correlations of lone pair ionization energies with proton affinities of amino acids and related compounds. Site specificity of protonation. Int. J. Mass Spectrom. Ion Proc. 1992, 117, 83-99. 35. Campbell, S.; Marzluff, E. M.; Rodgers, M. T.; Beauchamp, J. L.; Rempe, M. E.; Schwinck, K. F.; Lichtenberger, D. L. Proton Affinities and Photoelectron Spectra of Phenylalanine and N-Methyl- and N,N-Dimethylphenylalanine. Correlation of Lone Pair Ionization Energies with Proton Affinities and Implications for N-Methylation as a Method to Effect Site Specific Protonation of Peptides. J. Am. Chem. Soc. 1994, 116, 5257-5264. 36. Seki, K.; Inokuchi, H. Photoelectron spectrum of l-tryptophan in the gas phase. Chem. Phys. Lett. 1979, 65, 158-160. 37. Plekan, O.; Feyer, V.; Richter, R.; Coreno, M.; Prince, K. C. Valence photoionization and photofragmentation of aromatic amino acids. Mol. Phys. 2008, 106, 1143-1153. 38. Dehareng, D.; Dive, G. Vertical Ionization Energies of α -L-Amino Acids as a Function of Their Conformation: an Ab Initio Study. Int. J. Mol. Sci. 2004, 5, 301-332. 39. Winter, B.; Weber, R.; Hertel, I. V.; Faubel, M.; Jungwirth, P.; Brown, E. C.; Bradforth, S. E. Electron binding energies of aqueous alkali and halide ions: EUV photoelectron spectroscopy of liquid solutions and combined ab initio and molecular dynamics calculations. J. Am. Chem. Soc. 2005, 127, 7203-7214. 61 40. Jagoda-Cwiklik, B.; Slaviek, P.; Cwiklik, L.; Nolting, D.; Winter, B.; Jungwirth, P. Ionization of imidazole in the gas phase, microhydrated environments, and in aqueous solution. J. Phys. Chem. A 2008, 112, 3499-3505. 41. Platt, J. R. Classification of Spectra of Cata-Condensed Hydrocarbons. J. Chem. Phys. 1949, 17, 489. 42. Murrell, J. N. The Theory of the Electronic Spectra of Organic Molecules. 1971. 43. Weber, G. Fluorescence-Polarization Spectrum and Electronic-Energy Transfer in Tyrosine, Tryptophan and related Compounds. Biochem. J. 1960, 75, 335. 44. Serrano-Andrés, L.; Roos, B. O. Theoretical Study of the Absorption and Emission Spectra of Indole in the Gas Phase and in a Solvent. J. Am. Chem. Soc. 1996, 118, 185-195. 45. Dixon, R. N.; Oliver, T. A. A.; Ashfold, M. N. R. Tunnelling under a conical intersection: Application to the product vibrational state distributions in the UV photodissociation of phenols. J. Chem. Phys. 2011, 134, 194303. 46. Schick, C. P.; Carpenter, S. D.; Weber, P. M. Femtosecond multiphoton ionization photoelectron spectroscopy of the S-2 state of phenol. J. Phys. Chem. A 1999, 103, 10470-10476. 47. Buchner, F.; Ritze, H.-H.; Lahl, J.; Lubcke, A. Time-resolved photoelectron spectroscopy of adenine and adenosine in aqueous solution. Phys. Chem. Chem. Phys. 2013, 15, 11402-11408. 48. Isse, A. A.; Gennaro, A. Absolute Potential of the Standard Hydrogen Electrode and the Problem of Interconversion of Potentials in Different Solvents. J. Phys. Chem. B 2010, 114, 7894. 49. This assumption however leads to infinite uncertainty in the kinetic energy spectrum. 50. Jailaubekov, A. PhD Thesis. University of Southern California. 51. Yamazaki, S.; Kato, S. Solvent Effect on Conical Intersections in Excited-State 9H- Adenine: Radiationless Decay Mechanism in Polar Solvent. J. Am. Chem. Soc. 2007, 129, 2901- 2909. 52. Bräm, O.; Oskouei, A. A.; Tortschanoff, A.; van Mourik, F.; Madrid, M.; Echave, J.; Cannizzo, A.; Chergui, M. Relaxation Dynamics of Tryptophan in Water: A UV Fluorescence Up-Conversion and Molecular Dynamics Study. J. Phys. Chem. A 2010, 114, 9034-9042. 53. Leonard, J.; Sharma, D.; Szafarowicz, B.; Torgasin, K.; Haacke, S. Formation dynamics and nature of tryptophan's primary photoproduct in aqueous solution. Phys. Chem. Chem. Phys. 2010, 12, 15744-15750. 54. Sharma, D.; Léonard, J.; Haacke, S. Ultrafast excited-state dynamics of tryptophan in water observed by transient absorption spectroscopy. Chem. Phys. Lett. 2010, 489, 99-102. 55. Peon, J.; Hess, G. C.; Pecourt, J.-M. L.; Yuzawa, T.; Kohler, B. Ultrafast Photoionization Dynamics of Indole in Water. J. Phys. Chem. A 1999, 103, 2460-2466. 62 56. Pagba, C. V.; Chi, S.-H.; Perry, J.; Barry, B. A. Proton-Coupled Electron Transfer in Tyrosine and a β -Hairpin Maquette: Reaction Dynamics on the Picosecond Time Scale. J. Phys. Chem. B 2015, 119, 2726-2736. 57. Zhang, Y. PhD Thesis. University of Southern California. 58. Zhang, Y.; Oliver, T. A. A.; Ashfold, M. N. R.; Bradforth, S. E. Exploring the autoionization and photo-induced proton-coupled electron transfer pathways of phenol in aqueous solution. In preparation. 59. Luria, M.; Ofran, M.; Stein, G. Natural and experimental fluorescence lifetimes of benzene in various solvents. J. Phys. Chem. 1974, 78, 1904-1909. 60. Lee, S.-H.; Tang, K.-C.; Chen, I. C.; Schmitt, M.; Shaffer, J. P.; Schultz, T.; Underwood, J. G.; Zgierski, M. Z.; Stolow, A. Substituent Effects in Molecular Electronic Relaxation Dynamics via Time-Resolved Photoelectron Spectroscopy: ππ* States in Benzenes. J. Phys. Chem. A 2002, 106, 8979-8991. 61. Parker, D. S. N.; Minns, R. S.; Penfold, T. J.; Worth, G. A.; Fielding, H. H. Ultrafast dynamics of the S1 excited state of benzene. Chem. Phys. Lett. 2009, 469, 43-47. 62. Lewis, A.; Bumpus, J. A.; Truhlar, D. G.; Cramer, C. J. Molecular Modeling of Environmentally Important Processes: Reduction Potentials. J. Chem. Educ. 2004, 81, 596. 63. Seidel, C. A.; Schulz, A.; Sauer, M. H. Nucleobase-Specific Quenching of Fluorescent Dyes. 1. Nucleobase One-Electron Redox Potentials and Their Correlation with Static and Dynamic Quenching Efficiencies. J. Phys. Chem. 1996, 100, 5541. 63 Chapter 4. Relaxation dynamics of hydrated electron: comparison of CTTS vs. CTTS* excitation of aqueous Iodide solution 4.1 Introduction The relaxation dynamics of solvated electron, the simple most quantum system in liquids, has attracted a lot of attention from experimentalists as well as from theorists due to its complex and often controversial nature. 1-5 The ultrafast relaxation dynamics of hydrated electron has been explored using various time resolved spectroscopies. 6-12 In most of the experimental works, the hydrated electron has been generated by either direct solvent ionization using two or more photons 13-14 or exciting to the first charge transfer to the solvent (CTTS) state of common halide 8, 15-17 or complex ions. 9, 18 Bradforth and coworkers found that the electron ejection mechanism and the subsequent recombination dynamics of the nascent photoelectrons strongly depend on the process of their generation and the surrounding medium. 19 Quite surprisingly, the electron dynamics subsequent to excitation to the higher lying CTTS (here after will be referred as CTTS*) state did not attract much attention 20 as the lowest energy excitation. In case of aqueous Iodide solution, the charge transfer bands are separated by almost the spin-orbit splitting of ~0.9 eV of the Iodine atom ( 2 P 3/2 vs 2 P 1/2 ) in gas phase. 20-22 Therefore, it is a widespread assumption that in those two charge transfer states the electron is diffused over Iodine cores with different spin-orbit states. In a broadband transient absorption experiment Moskun et al investigated the transients after exciting aqueous Iodide solution to the CTTS* state at 200 nm. 20 Although the authors were primarily focused on the exploring the presence (absence) of I* ( 2 P 1/2 ) radical following the photodetachment, they indeed observed a electron dynamics distinct from the lowest CTTS excitation. 64 In transient absorption measurements the observed spectroscopy and dynamics of the transients are solely dependent on the probing window. Therefore, the physical model adopted based on those measurements could be incomplete or even misleading if one or more transients are undetectable in the probing range. Transient photoemission measurements on the other hand provides broader probing window provided the probe photon energy is higher than the binding energy. Recent development of liquid microjet technique by Faubel and coworkers expanded the photoemission measurements to liquids and solutions. 23-25 In particular time resolved photoemission studies in aqueous and nonaqueous solutions have been carried out by several groups to explore excited state dynamics of solvated electron 11-12, 26-27 and nucleotides/nucleosides. 28-29 In a recent three pulse photoemission experiment Neumark and coworkers found that the relaxation of the electronically excited electron follows nonadiabatic pathway as opposed to adiabatic pathway which resolved a more than decades long controversy. 30 However, the question remains whether the relaxation follows the identical nonadiabatic route when the geminate partner is in contact with the electron within the solvent shell. Although the ensuing dynamics of iodide CTTS state in water is explored in details, a number of different physical models have been proposed to explain the observed dynamics. 6, 12, 31-32 Suzuki and coworkers proposed several intermediate states in the process of CTTS state evolution to form a hydrated electron. 6 Each early time intermediate reflects different solvent configuration and captures dynamics in a chemically intuitive kinetic scheme. However, treating the instantaneous solvent configuration as discreet intermediates can be misleading as the solvation is a continuous evolution as opposed to discreet state-wise first order kinetics. Lübcke et al 12 on the other hand treated the solvation as a continuous process and incorporated the effect of thermalization in the dynamical nature of the photoemission bandwidth and peak position. 65 Although qualitatively the latter model seems closest to the actual dynamical process subsequent to photoexcitation, there are still open questions regarding the spectral assignments and early time recombination dynamics. In this paper, we are contrasting the electron dynamics of aqueous iodide solution when pumped to the CTTS and CTTS* states using time resolved photoemission spectroscopy. We found that the initial dynamics (up to 2 ps) in these two excitation processes are different owing to distinct photophysical processes. However, due to rapid ultrafast relaxation the long time dynamics (>10 ps) and the survival probability manifest in a very similar fashion. Our work is focused to reevaluate the existing physical models and deliver a concise and intuitive picture of photophysical processes subsequent to optical excitation to iodide solutions in water. 4.2 Experimental All experiments were performed using a home built time resolved photoelectron spectrometer equipped with liquid microjet invented by Manfred Faubel. 23 The design details of the photoelectron spectrometer has been discussed in previous chapters (chapter 2 and 3). Briefly, we implemented magnetic bottle time of flight detection scheme. Photoejected electrons are guided into the flight tube through a 380 m skimmer and detected using a pair of 40 mm diameter multichannel plate (MCP) detectors in Chevron configuration (Beam Imaging Solutions). In all the experiments the count rate was maintained to ~10 electron/s by varying the laser pulse intensity. For experiments at CTTS excitation, the pump beam at 251 nm (4.94 eV) is generated by doubling the visible output (502 nm) of an OPA (Spectra-Physics 800C, pumped by 800 nm from a Ti: Sapphire regenerative amplifier, Coherent Legend-Elite, 3.2W, 35 fs, 1 kHz). 66 For higher energy excitation at 200 nm (6.2 eV), the fourth harmonic is generated in a sum frequency set up using Type –I and Type-II BBO crystals. For all experiments the probe wavelength was fixed at 267 nm (4.644 eV) and generated using the same sum frequency set up. The time resolution of the experiments was estimated from the cross-correlation width (~160 fs) of the pump and probe beams on 1 mm quartz film. 33-34 Both the pump and probe beams are focused on the liquid jet at a small angle (~4-5º). The spot sizes are estimated to be not more than 80 m. In all experiments the pump and probe polarizations are held perpendicular to each other. For time resolved measurements each beam is chopped at 250 Hz with two identical mechanical choppers synced at /2 constant phase delay to acquire transient spectra for every four laser shots. Pump-probe time of flight spectra is acquired for every 4000 shots at each delay point and averaged over 10 scans. The acquired transient time of flight (TOF) spectra is then converted to electron kinetic energy (eKE) spectra with appropriate jacobian transformation owing to the fact that the area under the TOF and eKE spectra (i.e., number of electrons) would be the same. The transformed spectra are then smoothed before further processing. 100 mM aqueous salt solution is prepared by dissolving sodium iodide ( 99.5%, Sigma- Aldrich) in ultrapure water (18.2 MΩ·cm, Millipore) without any purification. Prepared solutions were injected into the vacuum chamber using a HPLC solvent delivery pump (Shimazdu) with a constant flow rate of 0.5 mL/min and a backing pressure of 4-5 bar. The jet thickness was ~20 m based on the liquid-jet nozzle diameter. 67 Figure 4.1. Evolution of integrated photoelectron signal at 251 nm excitation (blue solid circles) with pump-probe delay. Red solid line shows best fit with sum of three exponential functions convoluted with instrument response (~160 fs). The green dashed line shows simulation based on modified Staib- Borgis model (see text for details). 4.3 Results and Discussion 4.3.1 Excitation to the first CTTS band At 251 nm pump wavelength, we have observed a photo-electron band centered around 3.2-3.5 eV binding energy. The integrated peak intensity between 1-4 eV has been plotted against the pump probe delay in Figure 4.1. The decay profile can be fitted with a sum of three exponentials convoluted with instrument response function. The characteristic time constants are found to be 0.55 ps (45%), 5.1 ps (22%) and 77 ps (33%). We note that the last two time constants are in reasonable agreement with the values reported in transient absorption measurements. 8 However, the subpicosecond decay component (0.55 ps) observed here was not previously reported in the time resolved absorption measurements. 68 Previously the sub picoseconds dynamics in TRPES experiments has been attributed to the decay of the initially populated CTTS state to contact pair I-e -6, 35 and/or a rapid nonadiabatic geminate recombination of the initially ejected 'hot' photoelectrons to the ground state iodide. 12 Although the integrated intensity decay profile of the photoemission band illustrates both the evolution of population and/or change in the photoionization cross section with time, in both reports the authors assumed that the change in photoionization cross-section during the evolution of the CTTS state is minimal and the observed decrease in photoemission signal reflects only the depopulation of the decaying state. Although this assumption is quite intuitive, a careful determination of the early time relaxation of the photodetached electron from aqueous iodide by monitoring the solvated electron absorption band at different wavelengths did not show any evidence of the sub-picosecond geminate recombination. 36 On the other hand, Messina et al have shown that the initially populated CTTS state of iodide dissociates to iodine and electron contact pair with an average life time of 200 fs. 10 The authors stressed that the inhomogeniety of the local solvation structure around the CTTS state plays an important role in the decay rate of the CTTS state into I:e - pair as predicted by Bradforth and Jungwirth. 37 The evolution of CTTS state to I:e - geminate pairs is manifested by the rapid change in spatial extent of the initial wave function and it is expected to be directly correlated with the change in photoionization cross section. Recent time resolved photoemission studies of iodide in methanol 38 reported an ultrafast decay component subsequent to photoexcitation and attributed it to the decrease in the photodetachment cross section during decay of the relaxed CTTS state to a solvated electron configuration. Here, we attempt to empirically include the photoionization cross section as a dynamical parameter to explain the observed relaxation dynamics. 69 Previously, the solvated electron population dynamics has been analyzed using the competitive kinetic model proposed by Staib and Borgis. 32 According to this model the survival probability [ of the solvated electron can be expressed as (4.1) where k p , k n and k d refers to the rate of contact pair formation, geminate recombination and mutual diffusion respectively and can successfully model the experimental data up to 100 ps. 36, 39 However, the survival probability does not account for the possible change in the ionization cross-section at early time. Assuming an exponential nature of the ionization cross section evolution due to solvent reorientation, the time dependent photoemission signal, S(t), can be expressed as (4.2) where (4.3) (t) is the time dependent photoionization cross section and (0) and (∞) represent ionization cross sections at initial time (t=0) and after dissociation of the CTTS state (t=∞), respectively. We should note that in our model the initial and final ionization cross sections are treated as fitting parameters and the their values themselves did not represent the actual ionization cross sections. However, we believe their relative amplitudes would reflect the ratio of the original values. Using eqn. 2 and assuming the rate of change of ionization cross-section (k  ) to be (550 fs) -1 as found from the exponential fitting, we can simulate the integrated intensity decay profile reasonably well with k p =(0.2 ps) -1 , k n =(40 ps) -1 and k d =(80 ps) -1 (cf. Figure 4.1, Table 4.1). These 70 Figure 4.2. a) Transient photoemission spectra of aqueous iodide with 251 nm pump at different delays (solid lines). Red dashed lines are Gaussian fitting functions from the global fit (see text for details). The broad photoemission band at earlier time (<0.5 ps) undergoes spectral blue shifting and narrowing. Evolution of the band center (binding energy, eBE) and the full width half maximum (FWHM) of the Gaussian fitting function are shown in b) and c), respectively. Red dash-dot lines are exponential fits and yield characteristic relaxation time scales and magnitude (see below). values match remarkably well with the values reported in the literature 39 and justify our physical model. A closer look in to the early time photoelectron spectra shows evidence of the spectral blue shifting and band narrowing (0-5 ps, Figure 4.2) and have been assigned to the signature of electronic relaxation in the ground state surface of the hydrated electron in earlier reports. 12, 30 In transient absorption studies, this relaxation has been manifested by the 0.36 eV blue shift of the absorption maximum with a time scale of 850 fs. 36 To disentangle the solvent relaxation dynamics from the population decay, we adopted a global fitting algorithm to fit the full 2D transient photoemission dataset using a Gaussian 71 Figure 4.3. a) Full transient photoelectron spectrum of ~100mM sodium iodide in water. Global fitting of the data and the residual are shown in b) and c), respectively. d) Squared value of the residual from global fitting algorithm at each delay point. At delay>0.2 ps, the average residual 2 value is 0.06. function with time dependent center position, width and intensity (Figure 4.3). We perform nonlinear least square fitting of the photoemission band at each experimental time point and extracted the variation of the above parameters with time. This modeling strategy is similar in spirit to the model used by Lübcke et al 12 and Elkins et al 38 , however, we emphasize that we don't impose any exponential decay kinetics on the floating parameters as used in those reports. This fitting strategy is comparatively robust and widely applicable since it does not restrict the time evolution to a single exponential or a sum of exponential functions. However, the output of the global fitting algorithm, i.e., the time dependent fitting parameters: band center, full width half maximum (FWHM) and the total population (area under the Gaussian fit) can be independently fitted/simulated using any physical or numerical model, such as modified Staib- Borgis model (eqn. 1-3) for population decay as discussed earlier. 72 Figure 4.4. Kinetic scheme of electron dynamics subsequent to CTTS excitation (at 251 nm) of aqueous iodide solution. Parameters are described in the text. The evolution of the FWHM can be fitted with a single exponential function with a rate of (1±0.2 ps) -1 (k relax ). The band center follows the similar time scale (0.6±0.1 ps) with a blue shift ( ) of 0.3 eV (cf. Figure 4.2 b,c, Table 4.2) which is in reasonable agreement with the solvent relaxation dynamics reported in transient absorption measurements. 36 The population dynamics can be reasonably modeled with the modified Staib-Borgis equation with identical rate constants discussed before. We can consolidate all the physical processes subsequent to optical excitation to the CTTS state in a kinetic scheme as shown in Figure 4.4. 4.3.2 Excitation to the second CTTS (CTTS*) band When excited to the higher lying CTTS state at 200 nm (6.2 eV), we observed a photoelectron band similar to that observed at 251 nm excitation. However, a careful comparison between the time slices at different excitation energy (251 nm vs 200 nm) reveals that the PE 73 Figure 4.5. Time resolved photoelectron spectra of 100mM NaI solution at different pump probe delay times. Solid red and blue lines show photoemission bands at 251 nm and 200 nm excitation energy respectively. A shoulder between 1-2.5 eV binding energy is observed at early times (t < 2 ps) with 200 nm pump. Prompt disappearance of this band indicates the presence of a short lived transient not observed for lowest energy CTTS excitation (251 nm). band at 200 nm pump has a shoulder in the lower binding energy, between 1-2.5 eV, which disappears by 2 ps (Figure 4.5). This early time transient feature is absent when excited to the lower CTTS state. To model the ensuing relaxation dynamics, we expanded our basis set to a combination of two Gaussians both having time dependent intensity, width and center position. With this new model, we can globally fit the transient 2D data set reasonably well and extract the kinetics of the fitting parameters. The basis spectra were shown for the 200 fs time slice in Figure 4.6. Our spectrally decomposed global fitting algorithm shows the following dynamics: i) The higher binding energy (>3 eV) band is produced instantaneously within our time resolution and shows both spectral relaxation and population decay. We observe a blue shifting 74 Figure 4.6. Photoelectron spectrum at 200 fs delay with 200 nm pump (solid line with solid circles). The spectrum is fitted with sum of two Gaussians (solid red) representing p and s-state of hydrated electron (see text for details). The ratio of the area under each Gaussian represents the relative population of the two states upon photoexcitation at 200 fs (assuming identical photoionization cross section) of the band center by ~0.3 eV accompanied by spectral narrowing at an average rate of (~2 ps) - 1 (cf. Figure 4.7 b,c) which is similar to what we observed for 251 nm excitation (Table 4.2). At delay 10 ps the spectral shape of the photoemission band is also identical to that found in the lowest CTTS excitation experiment as shown in Figure 4.5. Figure 4.8 (b) illustrates the population decay of this band along with Staib-Borgis simulation following eqn. 2-3. We found that the intensity decay can be modeled reasonably well with time constants identical to that of 251 nm excitation (Table 4.1). All the similarity in dynamics and spectroscopy of the high binding energy band at CTTS* excitation with that of the CTTS excitation indicate that the photoemission originates from an identical molecular state. In analogy to our photoemission results and earlier transient absorption experiments we can attribute this feature to I:e - contact pair showing characteristic rate constants of formation, thermal relaxation, geminate recombination and mutual diffusion. 75 Figure 4.7. a) Transient photoemission spectra of aqueous iodide with 200 nm pump at different delays (solid lines). Red dash-dot lines are sum of two Gaussian functions from the global fit (see text for details). The band centered at 2.1 eV binding energy decays within 500 fs (cf. Figure 4.8); no spectral evolution is observed. However, the higher binding energy peak undergoes spectral blue shifting and narrowing as shown in b) and c), respectively. Red dotted lines are exponential fits to the data. ii) The low binding energy band ~2.1 eV also shows instrument limited appearance time, however, it undergoes an ultrafast population decay with a life time of 350 fs (Figure 4.8 a). We don't observe any significant spectral evolution of this band during this time. In an identical experiment Lübcke et al observed an ultrafast decay of a low binding energy photoemission band. 12 As discussed in the previous section they assigned it to 'hot' photoelectrons which undergoes thermalization to the higher binding energy band ('cold' photoelectrons) as well as rapid geminate recombination at a rate of (900 fs) -1 . Although the thermalization timescale in their experiment matches with earlier reports, the magnitude of thermal relaxation (~1 eV) is significantly higher than what previously found (~0.36 eV) 36 . We also note that although the authors used 200 nm pump to populate the charge transfer to solvent band of iodide, they 76 apparently consider the relaxation dynamics happening in the lower lying CTTS state. Fox and Hayon reported the electronic absorption spectrum of aqueous iodide at 318K which shows two CTTS bands at 5.4 and 6.4 eV (229 and 195 nm). 40 Following their observation, it is assumed that upon photoexcitation to the first and second CTTS bands, the core fragments produced are I( 2 P 3/2 ) and I*( 2 P 1/2 ) respectively based on their spin-orbit splitting value of 0.9 eV in the gas phase. 20-22 Therefore, at 200 nm excitation we expect to populate primarily the CTTS* state which consists of higher energy spin orbit state of the iodine ( 2 P 1/2 ) core and electron. Subsequent to photoexcitation we expect that the CTTS* state would form I*:e - contact pair due to instantaneous solvent fluctuation, followed by competing geminate recombination and mutual diffusion processes as originally proposed by Staib and Borgis. 32 Following this physical picture the nonadiabatic geminate recombination would involve electron transfer to I* forming ground state iodide. As considered previously, the halide atom and electron recombination is expected to lie in the Marcus inverted regime 20, 41 , therefore, the rate of electron transfer would be slower as the free energy difference between reactants (geminate pair) and the product (halide ion in the ground state) increases. Based on this idea, we expect that the recombination rate of the I*:e - geminate pair would rather be slow compared to I:e - . Previous transient absorption results indeed show that the recombination rate for CTTS* excitation is comparatively slower than that observed for the lowest CTTS excitation. 20 This observation does not corroborate with the subpicosecond recombination of 'hot' photoelectrons as proposed by Lübcke and coworkers. We adopt an alternative physical model to spectrally and temporally characterize the ensuing CTTS* state dynamics after photoexcitation. 77 Figure 4.8. Integrated photoemission signal of the two Gaussian peaks at 200 nm pump as a function of pump probe delay. a) Variation of the lower binding energy peak population (green solid circle) fitted with a single exponential function with time constant of 350 fs (red broken line). b) Population decay of the higher binding energy peak (solid square). The experimental data is simulated following modified Staib-Borgis model (orange broken line) as discussed in the main text (eqn. 2) In a recent three pulse transient photoemission experiment of aqueous iodide solution, Elkins et al probed the evolution of so called 'p' state of hydrated electron and reported the vertical binding energy of the relaxed 'p' state to be 2.2 ± 0.2 eV. 30 We note that this vertical binding energy is almost identical to that of the weakly bound feature (~2.1 eV) observed at our 200 nm excitation experiment within the experimental error. Therefore, we assign it to the weakly bound p- state of the hydrated electron. The formation and fate of the p-state electron will be discussed below. In time resolved absorption experiment with aqueous iodide solution upon 200 nm excitation, Moskun et al did not find any evidence of predicted I* transient absorption spectrum upon with 300 fs time resolution. 20 The authors proposed that the absence of I* spectral feature could be either due to ultrafast internal conversion from I* to I via efficient electronic to vibrational energy transfer (E→V) to surrounding wate r molecules involving quadruple-dipole 78 coupling 20, 42-44 or due to auto-detachment to I( 2 P 3/2 ) surface or both. The E→V energy transfer mechanism is efficient when the I* and I energy splitting (~0.9 eV) is near resonance with a solvent acceptor mode, e.g., two quanta of water stretching vibration. Incidentally, the energy difference between the s and p state of hydrated electron when formed (3.2-2.1=1.1 eV) matches perfectly well 30 with the iodine spin orbit splitting energy and can allow the s → p excitation in the hydrated electron as a possible energy acceptor channel. In light of this proposition, we can summarize the excited state dynamics of iodide as follows (Figure 4.9): Following excitation at 200 nm, ~30% of CTTS* state decays to a contact pair of I*( 2 P 1/2 ):e - s which subsequently exchanges energy to form I( 2 P 3/2 ):e - p within our instrument response (~160 fs). Electronically excited hydrated electron (p state) then undergoes internal conversion to the ground state (s state) at a rate of (450 fs) -1 . Remaining CTTS* population (~70%) undergo relaxation back to the lower lying CTTS state with instrument limited time scale and form I( 2 P 3/2 ):e - s contact pair. Ground state electron population then follows subsequent dynamics as in case of 251 nm excitation. Finally, according to this picture the internal conversion (IC) time scale (450 fs) of e - p → e - s is almost 6 times slower than what Elkins et al reported in ref 30 and suggests 'adiabatic relaxation' picture in contrast to their original conclusion of 'nonadiabatic relaxation' pathway. This apparent disagreement between internal conversion timescale (and therefore, the nature of the electronic relaxation pathway) can be attributed to the different solvent environment around p-state. In the three pulse experiment, the pump pulse to create p-state hydrated electron was applied 200 ps after the synthesis pulse in order to capture fully equilibrated electron. At this time most of the captured electrons diffuse away from its geminate partner. When excited to p- state, the solvent configuration around hydrated electron without the geminate partner is 79 Figure 4.9. Kinetic scheme of electron dynamics subsequent to CTTS* excitation (at 200 nm). presumably distinct from that around the I( 2 P 3/2 ):e - p geminate pair and would certainly influence the internal conversion rate. This argument is in agreement with prior work by Bradforth and coworkers where they found that the relaxation dynamics of ejected photoelectrons from pure water is significantly different from that of the aqueous iodide solution due to the presence of an iodine atom in the first solvent shell. 36 4.4 Conclusion We contrasted the short and long time dynamics of aqueous iodide solution after photoexcitation to two different charge transfer to solvent states. We implemented a global fitting strategy based on nonlinear least square fitting algorithm (Levenberg–Marquardt) to deconvolute population and thermalization dynamics. We found that excitation to the lower lying CTTS band produces only one molecular state which undergoes thermalization and follows competitive kinetic model proposed by Staib and Borgis with time scales quite similar to those 80 reported earlier. 36, 45 However, we explicitly incorporated the change in the photoionization cross-section during the CTTS state evolution as a dynamical parameter in the Staib-Borgis model to account for the initial sub-picosecond decay. The relative magnitudes of the initial and final ionization cross-section provides useful insight into the relative spatial extent of the electronic wave function during evolution of the charge transfer state and could be addressed by theory. The observed spectral signature and dynamics at initial time with higher excitation energy is quite distinct from that of the lower energy excitation. Ultrafast depopulation of a low binding energy band at ~2.1 eV indicates the involvement of the p-state electron in the relaxation process (not observed in excitation to the lower CTTS band). We speculate that the efficient coupling of the spin-orbit splitting energy of the iodine core with the electronic excitation of the hydrated electron is most likely responsible for the generation of p-electron. We found that in presence of the geminate partner the relaxation timescale of the p-electron to ground state supports adiabatic pathway in contrast to nonadiabatic mechanism for a free hydrated electron reported recently. 81 Excitation wavelength 1/k p (ps) 1/k n (ps) 1/k d (ps) 1/k  (ps) (0) (a.u) ( ) (a.u) 251 nm 0.2 40 80 0.55 4.65 0.65 200 nm* 0.2 40 80 0.55 3.15 0.65 Excitation wavelength (nm) FWHM Peak Center w(0) (eV) w( ) (eV) 1/k relax (ps) BE(0) (eV) BE( ) (eV) 1/k relax (ps) 251 1.5 ± 0.3 1.0 ± 0.2 1.0 ± 0.5 3.10 ± 0.1 3.45 ± 0.1 0.6 ± 0.2 200* 1.3 ± 0.1 1.0 ± 0.1 1.3 ± 0.5 3.17 ± 0.1 3.43 ± 0.1 2.3 ± 0.5 Table 4.1. Simulation parameters to model population dynamics using modified Staib-Borgis model (eqn. 4.1-4.3). *For 200 nm excitation, we only simulate the population for the higher binding energy peak Table 4.2. Fitting parameters for relaxation dynamics. *For 200 nm excitation, we only fit the higher binding energy peak 82 References 1. Chen, X.; Bradforth, S. E., The Ultrafast Dynamics of Photodetachment. Annual Review of Physical Chemistry 2008, 59, 203-231. 2. Young, R. M.; Neumark, D. M., Dynamics of Solvated Electrons in Clusters. Chem Rev 2012, 112, 5553-5577. 3. Abel, B., Hydrated Interfacial Ions and Electrons. Annual Review of Physical Chemistry 2013, 64, 533-552. 4. Rossky, P. J.; Schnitker, J., The hydrated electron: quantum simulation of structure, spectroscopy, and dynamics. The Journal of Physical Chemistry 1988, 92, 4277-4285. 5. Turi, L.; Rossky, P. J., Theoretical Studies of Spectroscopy and Dynamics of Hydrated Electrons. Chem Rev 2012, 112, 5641-5674. 6. Tang, Y.; Suzuki, Y. I.; Shen, H.; Sekiguchi, K.; Kurahashi, N.; Nishizawa, K.; Zuo, P.; Suzuki, T., Time-resolved photoelectron spectroscopy of bulk liquids at ultra-low kinetic energy. Chem Phys Lett 2010, 494, 111-116. 7. Kimura, Y.; Alfano, J. C.; Walhout, P. K.; Barbara, P. F., Ultrafast Transient Absorption Spectroscopy of the Solvated Electron in Water. The Journal of Physical Chemistry 1994, 98, 3450-3458. 8. Kloepfer, J. A.; Vilchiz, V. H.; Lenchenkov, V. A.; Bradforth, S. E., Femtosecond dynamics of photodetachment of the iodide anion in solution: resonant excitation into the charge- transfer-to-solvent state. Chem Phys Lett 1998, 298, 120-128. 9. Lenchenkov, V.; Kloepfer, J.; Vilchiz, V.; Bradforth, S. E., Electron photodetachment from [Fe(CN) 6 ] 4- : photoelectron relaxation and geminate recombination. Chem Phys Lett 2001, 342, 277-286. 10. Messina, F.; Bräm, O.; Cannizzo, A.; Chergui, M., Real-time observation of the charge transfer to solvent dynamics. Nat Commun 2013, 4. 11. Siefermann, K. R.; Liu, Y. X.; Lugovoy, E.; Link, O.; Faubel, M.; Buck, U.; Winter, B.; Abel, B., Binding energies, lifetimes and implications of bulk and interface solvated electrons in water. Nature Chemistry 2010, 2, 274-279. 12. Lübcke, A.; Buchner, F.; Heine, N.; Hertel, I. V.; Schultz, T., Time-resolved photoelectron spectroscopy of solvated electrons in aqueous NaI solution. Physical Chemistry Chemical Physics : PCCP 2010, 12, 14629. 13. Reuther, A.; Laubereau, A.; Nikogosyan, D. N., Primary Photochemical Processes in Water. The Journal of Physical Chemistry 1996, 100, 16794-16800. 14. Crowell, R. A.; Bartels, D. M., Multiphoton Ionization of Liquid Water with 3.0−5.0 eV Photons. The Journal of Physical Chemistry 1996, 100, 17940-17949. 83 15. Kloepfer, J. A.; Vilchiz, V. H.; Lenchenkov, V. A.; Bradforth, S. E., Molecular Mechanisms for Photodetachment in Water. In Ultrafast Phenomena XIII, Miller, D.; Murnane, M. M.; Scherer, N. F.; Weiner, A. M., Eds. Springer Verlag: Berlin, 2003; pp 471-473. 16. Fischer, M. K.; Laubereau, A.; Iglev, H., Femtosecond electron detachment of aqueous bromide studied by two and three pulse spectroscopy. Physical Chemistry Chemical Physics 2009, 11, 10939. 17. Iglev, H.; Fischer, M. K.; Laubereau, A., Electron detachment from anions in aqueous solutions studied by two- and three-pulse femtosecond spectroscopy. Pure and Applied Chemistry 2010, 82, 1919-1926. 18. Lenchenkov, V. A.; Vilchiz, V. H.; Kloepfer, J. A.; Bradforth, S. E., Femtosecond Study of Electron Photodetachment from Complex Anions: Fe(CN) 6 4 - and CuBr 2 - in H 2 O. In Ultrafast Phenomena XII, Elsaesser, T.; Mukamel, S.; Murnane, M. M.; Scherer, N. F., Eds. Springer- Verlag: Berlin, 2000; pp 476-478. 19. Vilchiz, V. H.; Chen, X. Y.; Kloepfer, J. A.; Bradforth, S. E., Solvent effects on geminate recombination dynamics after photodetachment. Radiation Physics and Chemistry 2005, 72, 159-167. 20. Moskun, A. C.; Bradforth, S. E.; Thogersen, J.; Keiding, S., Absence of a Signature of Aqueous I( 2 P 1/2 ) after 200-nm Photodetachment of I - (aq). J Phys Chem A 2006, 110, 10947- 10955. 21. Jortner, J.; Raz, B.; Stein, G., The far u.-v. absorption spectrum of the iodide ion in aqueous solution. Transactions of the Faraday Society 1960, 56, 1273-1275. 22. Blandamer, M. J.; Fox, M. F., Theory and applications of charge-transfer-to-solvent spectra. Chem Rev 1970, 70, 59-93. 23. Faubel, M.; Steiner, B.; Toennies, J. P., Photoelectron spectroscopy of liquid water, some alcohols, and pure nonane in free micro jets. J. Chem. Phys. 1997, 106, 9013. 24. Winter, B.; Faubel, M., Photoemission from liquid aqueous solutions. Chem Rev 2006, 106, 1176-1211. 25. Winter, B., Liquid microjet for photoelectron spectroscopy. Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment 2009, 601, 139-150. 26. Shen, H. A.; Kurahashi, N.; Horio, T.; Sekiguchi, K.; Suzuki, T., Direct Measurement of Vertical Electron Binding Energies of Solvated Electrons in Methanol and Ethanol. Chemistry Letters 2010, 39, 668-670. 27. Shreve, A. T.; Yen, T. A.; Neumark, D. M., Photoelectron spectroscopy of hydrated electrons. Chem Phys Lett 2010, 493, 216-219. 84 28. Buchner, F.; Ritze, H.-H.; Lahl, J.; Lubcke, A., Time-resolved photoelectron spectroscopy of adenine and adenosine in aqueous solution. Physical Chemistry Chemical Physics 2013, 15, 11402-11408. 29. Buchner, F.; Nakayama, A.; Yamazaki, S.; Ritze, H.-H.; Lübcke, A., Excited-State Relaxation of Hydrated Thymine and Thymidine Measured by Liquid-Jet Photoelectron Spectroscopy: Experiment and Simulation. Journal of the American Chemical Society 2015, 137, 2931-2938. 30. Elkins, M. H.; Williams, H. L.; Shreve, A. T.; Neumark, D. M., Relaxation Mechanism of the Hydrated Electron. Science (New York, N.Y.) 2013, 342, 1496-1499. 31. Iglev, H.; Trifonov, A.; Thaller, A.; Buchvarov, I.; Fiebig, T.; Laubereau, A., Photoionization dynamics of an aqueous iodide solution: the temperature dependence. Chem Phys Lett 2005, 403, 198-204. 32. Staib, A.; Borgis, D., Reaction pathways in the photodetachment of an electron from aqueous chloride: A quantum molecular dynamics study. The Journal of Chemical Physics 1996, 104, 9027-9039. 33. Dragomir, A.; McInerney, J. G.; Nikogosyan, D. N., Femtosecond measurements of two- photon absorption coefficients at lambda=264 nm in glasses, crystals, and liquids. Applied Optics 2002, 41, 4365-4376. 34. Mohammadhassan, V.; Davide, D. A.; Felice, G.; Raffaele, V.; Carlo, A., Temporal and spectral characterization of femtosecond deep-UV chirped pulses. Laser Physics Letters 2015, 12, 025302. 35. Suzuki, Y.-I.; Shen, H.; Tang, Y.; Kurahashi, N.; Sekiguchi, K.; Mizuno, T.; Suzuki, T., Isotope effect on ultrafast charge-transfer-to-solvent reaction from I- to water in aqueous NaI solution. Chem Sci 2011, 2, 1094-1102. 36. Vilchiz, V. H.; Kloepfer, J. A.; Germaine, A. C.; Lenchenkov, V. A.; Bradforth, S. E., Map for the relaxation dynamics of hot photoelectrons injected into liquid water via anion threshold photodetachment and above threshold solvent ionization. J Phys Chem A 2001, 105, 1711-1723. 37. Bradforth, S. E.; Jungwirth, P., Excited states of iodide anions in water: A comparison of the electronic structure in clusters and in bulk solution. J Phys Chem A 2002, 106, 1286-1298. 38. Elkins, M. H.; Williams, H. L.; Neumark, D. M., Dynamics of electron solvation in methanol: Excited state relaxation and generation by charge-transfer-to-solvent. The Journal of Chemical Physics 2015, 142, 234501. 39. Kloepfer, J. A.; Vilchiz, V. H.; Lenchenkov, V. A.; Germaine, A. C.; Bradforth, S. E., The ejection distribution of solvated electrons gernerated by the one photon photodetachment of aqueous I - and two photon ionization of solvent. J Chem Phys 2000, submitted (May, 2000). 85 40. Fox, M. F.; Hayon, E., Far ultraviolet solution spectroscopy of the iodide ion. Journal of the Chemical Society, Faraday Transactions 1: Physical Chemistry in Condensed Phases 1977, 73, 1003-1016. 41. Chen, X. Ultrafast transient spectroscopy and electron photodetachment of inorganic and organic anions in aqueous solutions. University of Southern California, Los Angeles, CA, 2006. 42. Chiappero, M. S.; Badini, R. G.; Argüello, G. A., Quenching of I(2P½) by R-OH compounds. Influence of the alkyl group on the quenching efficiency. International Journal of Chemical Kinetics 1997, 29, 155-160. 43. Chiappero, M. S.; Argüello, G. A., Deactivation of I(2P1/2) by CH3Cl, CH2Cl2, CHCl3, CCl3F, and CCl4. International Journal of Chemical Kinetics 1998, 30, 799-803. 44. Donovan, R. J.; Husain, D., Electronically excited iodine atoms I(52P1/2). Part 3. Transactions of the Faraday Society 1966, 62, 2023-2029. 45. Kloepfer, J. A.; Vilchiz, V. H.; Lenchenkov, V. A.; Germaine, A. C.; Bradforth, S. E., The ejection distribution of solvated electrons generated by the one-photon photodetachment of aqueous I - and two-photon ionization of the solvent. J Chem Phys 2000, 113, 6288-6307. 86 Chapter 5. Exploring the long time excited state dynamics of phenol in aqueous solution 5.1 Introduction Tyrosine is one of the most versatile and ubiquitous redox centers among the aromatic amino acids essential for its role in biochemical reactions, such as, DNA damage repair in eukaryotic cells 1 and oxygen evolution in photosynthesis 2 (Kok Cycle) in the reaction center of Photosystem II. A number of these key redox reactions involving tyrosine are found to be photo- driven, therefore the photophysical/photochemical properties of its light absorbing chromophore, phenol, are of immense interest. 3-6 Nanosecond flash photolysis studies of aqueous tyrosine, phenol and its derivatives by Bent and Hayon and by Getoff and coworkers reveal the spectroscopic characteristics and decay kinetics of the long time transients – triplet state, phenoxyl radical and solvated electron, 7-9 however, the precise knowledge of the formation mechanism and the early time dynamics of these photoproducts was limited due to the poor time resolution (>10ns) in the early experiments. The generation of phenoxyl radical and solvated electron subsequent to photoexcitation has been initially proposed to go via biphotonic ionization from the triplet state. 10 However, intensity dependent studies later showed that electron/radical generation varies linearly with pump photon intensity ruling out two photon ionization process. 8 Recently the possibilities of excited state proton transfer (ESPT), autoionization and ground or excited-state proton-coupled electron transfer (PCET) pathways have also been evaluated as possible deactivation pathways, 11-12 however, a precise understanding of all the possible photochemical/physical processes comprising a kinetic scheme is still underway. 87 In gas phase, photodissociation of a phenol molecule with excitation wavelengths 285> λ phot >240 nm (S 1 ←S 0 origin is at 275.113 nm) leads to the generation of a phenoxyl radical and an H-atom. 13 This is an important non-radiative deactivation pathway for phenol and several other hetero-aromatic molecules having X-H (X=N,O) bond. 14 Ashfold and coworkers illustrated that the tunneling under the conical intersection (CI) between bound 1 * (S 1 ) and dissociative 1 *(S 2 ) surface leads to slow photodissociation of O-H bond when excited to optically bright 1 * state. 13, 15 From quantum mechanical calculations the spin orbit coupling between S 1 and T 1 state for isolated phenol is found to be very low (~0.1 cm -1 ), therefore, suggests a small role for the triplet state in the nonradiative decay pathways. The direct photoionization channel was also ignored as the ionization threshold for isolated phenol (8.74 eV) 16-17 is above the pump photon energy (4.35-5.17 eV). In non polar medium, such as, in cyclohexane solution, the excited state dynamics of phenol is quite similar to that of in gas phase. 3, 5 However, in the polar medium the dynamics and therefore the quantum yield of the transients change significantly, e.g., the triplet yield and fluorescence quantum yield changes from 0.27 (Q T ) and 0.083(Q f ) to 0.67(Q T ) and 0.19(Q f ) respectively when the surrounding medium is changed from cyclohexane to ethanol. 3 In polar medium the energetics of the potential energy surfaces change dramatically opening new pathways for deactivation from the S 1 surface. 18-20 Moreover, the effect of strong coupling between phenol and protic solvent (such as ethanol and water) on excited state photophysics due to H-bonding interaction is still elusive. For example, in aqueous solution the ionization threshold lies fairly close to the S 1 excitation origin 20 opening up a possible pathway for autoionization. In recent femtosecond pump-probe experiments at 267 nm pump, Zhang et al found spectral signatures of phenoxyl radical and solvated electron only after a nanosecond 88 indicating slow generation of the photoproducts. 12, 20 The authors did not observe any kinetic isotope effect for normal and deuterated phenol (PhOH vs. PhOD) which confirms that the photoionization does not occur by the homolytic fission of the O-H bond as observed in cyclohexane, nor by excited state proton transfer followed by electron ejection from the excited phenolate anion (PhO ─ *). The observed slow generation (~ns) of photoelectrons along with phenoxyl radicals and the absence of isotope effect led to the conclusion of near threshold autoionization mechanism followed by ultrafast (~100 fs) deprotonation although an excited state proton coupled electron transfer pathway could not be ruled out. No spectral signature of triplet state was reported in their transient absorption experiment, however, the authors noticed apparent change in dynamics in the blue edge of the probing window (  350 nm) when cesium chloride (CsCl) salt was added. Cesium chloride is known to enhance intersystem crossing (ISC) rate, therefore, it will populate (and depopulate) triplet state faster than usual. Earlier nanosecond flash photolysis studies on aqueous phenol solutions by Bent and Hayon 21 reported spectral signatures of hydrated electron, phenoxyl radical and triplet state. The authors observed the transients within their experimental instrument response (15 ns), therefore, they were unable to decipher the early time dynamics (<15 ns). However, the broad spectral window into the far UV extending up to 250 nm enabled them to assign absorption spectra of each transient species at longer time. Here we combine sub-nanosecond time resolution (~700 ps) with ultra broad band probing window from deep UV to visible (270-620 nm) to capture the longtime excited state dynamics and compare our results with previous fs-ps range transient absorption measurements to provide a unified picture of the surprisingly complex photophysics/photochemistry of phenol in aqueous solution. 89 5.2 Experiment All experiments described in this chapter were performed using a home-built braodband transient absorption (TA) spectrometer. The details of the spectrometer components are described elsewhere. 20, 22 A brief description of the working principles are given below. In transient absorption spectroscopy, a pump laser pulse is tuned to the electronic absorption band of the molecule to induce optical excitation. A broadband probe pulse, then monitors the behavior of the molecules in the excited state surface by following the time varying absorption and/or stimulated emission from the excited state. Depending on the range of the probing window and the delay between the pump and the probe pulse, we can explore the photophysics and photochemistry subsequent to photoexcitation. Long time (ns- s) transient absorption measurements were performed using fourth harmonic (266 nm) of a picosecond Nd:YAG laser (Alphalas, PULSELAS-A-266-300-SP, pulsewidth ~700 ps) synchronized to our femtosecond laser systems by a digital delay generator (Stanford Research Systems, DG645) as a pump and broadband super continuum as a probe. In our experiment we use two different super continuum generation methods to achieve probing window from 270-700 nm: i) By focusing a fraction of 800 nm output (35 fs) from the regenerative amplifier onto a 2 mm thick rotating CaF 2 crystal, we generate broadband probe pulse from 350-700 nm. ii) In order to probe wavelength shorter than 350 nm, we use <200 nJ 400 nm ultrashort pulses (<70 fs estimated by material dispersion) generated by doubling the fundamental in a 1 mm type-I BBO crystal (RedOptronics) to drive super continuum generation in a 5 mm thick CaF 2 crystal (Layertec Gmbh). 23 A probing window of 270-380 nm can be achieved with stability and shot to shot noise comparable to that of 800 nm pumped continuum. The experimental set up is shown in Figure 5.1. All experiments were performed in wire guided 90 gravity jet to minimize the buildup of photochemical products which contaminate the measured TA signal. 24 The thickness of the sample jet is estimated to be 100-120 m at the point of intersection with the two laser beams. The overall time resolution of the TA experiment is limited by the pulse width at 266 nm and is estimated ~700 ps. Figure 5.1. Experimental set up for transient absorption measurements. The upper panel shows optical layout for the broadband probe (270-700 nm) generation. M1, M2, M3: Metal mirror; RR: retro reflector; FM1, FM2: Flip mirrors; WP: /2 Wave plate; VDF: Variable density filter; L: Plano convex lens The far UV probe (270-380 nm) is generated by focusing 400 nm onto a rotating CaF2 disc. Visible probing range is produced by bypassing the 400 nm BBO crystal (flipping FM1 and FM2). The lower panel illustrates pump-probe measurement scheme 91 A 20 mM solution is prepared by dissolving solid phenol (>99%, Avocado Research) in ultrapure water (18.2 MΩ·cm, Millipore) without further purification. To reduce triplet and solvated electron quenching by oxygen, 7 all solutions are purged with dry nitrogen for hours and the gravity jet reservoirs were kept under N 2 environment. For scavenging experiments HCl (33 mM) and CsCl (250 mM) are added to 20mM phenol solutions. 5.3 Results and Discussion The observed spectral features between 350-650 nm in the early time (delay <2 ns) are quite similar to what observed in the sub-nanosecond transient absorption experiment reported before. 20 The observed broad transient absorption band covering the entire probing window (350-620 nm) has been attributed to the phenol excited state absorption (S 1 →S n ) along with the excimer absorption band around 600 nm. 20 A careful observation also indicates the onset of a sharp peak around 400 nm previously assigned to the phenoxyl radical absorption band. 25-26 The vibrational progression of the phenoxyl radical at 380 and 400 nm and a broad absorption band of hydrated electron (peak at ~700 nm) are prominent by 10 ns. These spectral features evolve with timescales in few hundreds of nanosecond. Presence of hydrated electron has been confirmed by quenching experiments with HCl as shown in Figure 5.2 c). A small shoulder at 600 nm still exists for the acid quenching experiments (cf. Figure 5.2 c) which could be the left over population (10% of the initial mOD) of the excimer band. As per this speculation the lifetime of the excimer is less than 10 ns. Indeed, the kinetics at 600 nm band at normal pH reveal that the transient signal decays in a biexponential fashion with time constants 4.5±1 ns and 400±50 ns (Figure 5.2 d). The long time component vanishes in the acidic solution (with 33 mM 92 HCl), therefore, it is assigned to the population decay of the equilibrated hydrated electron. These observations led us to the conclusion that the excimer lifetime (4.5±1 ns) is almost identical to the singlet monomer (S 1 ) lifetime of phenol in water (3.3 ns). 20 We also noticed a broad shoulder between 350-500 nm underlying the vibrationally resolved radical absorption spectrum in Figure 5.2 c). Therefore, from our initial observation we found three transients consisting long time TA spectra: hydrated electron, phenoxyl radical and an unknown species underneath the radical spectra. Figure 5.2. Transient absorption spectra of 20 mM aqueous phenol solution. Time slices at short (  10 ns) and long delays (˃ 10 ns) are shown in a) and b) respectively. Quenching experiment with HCl solution confirms the presence of hydrated electron as shown in c). Kinetic traces (and exponential fits) at 600 nm probe wavelength shown in d). The delay axis in d) is broken between 50- 250 ns to emphasize the two different delay scales. 93 Bent and Hayon reported the excited state absorption spectra of phenol and related compounds in water. 7 They reported that the triplet-triplet (T-T) absorption spectrum of phenol peaks at 250 nm and extends to at least 400 nm in the visible region. To investigate whether the unknown spectral feature discussed above is due to T-T absorption, we added Cesium Chloride (CsCl) to enhance intersystem crossing rate (ISC). At delay <100 ns the observed dynamics is quite distinct with the added cesium salt as shown in Figure 5.3 b). However, at longer delay it is rather similar possibly due to spectral overlapping of the T-T absorption band with the relatively long lived phenoxyl radical absorption band. The initial decrease in transient absorption signal upon addition of CsCl can be attributed to enhanced ISC rate which depopulates the S 1 state at a rate ~4 times faster than the normal as found in the time resolved fluorescence experiment (  f = 3.3 ns vs. 0.80 ns after adding CsCl). 20 Within our instrument resolution (~700 ps) a significant amount of S 1 population decays to T 1 . For phenol the oscillator strength of the T-T absorption band is reported to be low (<1000 M-1 cm-1 ~ 400 nm), 7 assuming the S 1 →S n absorption band has higher oscillator strength, we would expect overall reduction of the signal is due to enhanced S 1 →T 1 transfer rate. The dynamics beyond 5 ns is quite intriguing: it shows a rise and a sub- Figure 5.3. Comparing the transient absorption signals of 20 mM phenol solution containing 250 mM Cesium Chloride (CsCl) with the control solution containing equal amount of Sodium Chloride (NaCl) at 10 and 100 ns delay (a). After adding CsCl we observed an initial decrease in the transient absorption signal underlying the radical band (<5 ns), followed by an increase at times 10 ns<t<100 ns and finally a decay at longer time (b). The shaded areas in a) illustrate the increase and subsequent decay of the TA signal for delays>10 ns 94 -sequent decay for both the cases (with and without Cs + ). Evidently, the rise does not follow the same kinetics as the S 1 population decay, therefore, it does not imply the rise of the T-T absorption band, however, upon addition of Cs + the total signal intensity has been increased. All these observations indicate that the excited state photophysics/photochemistry of phenol is much more intricate than initially anticipated due to the overlapping transient absorption bands of the photoproducts and their complex kinetics. To retrieve the triplet absorption spectrum in the far UV, we performed transient absorption measurements with 400 nm pumped continuum (250-380 nm). Selected time slices are shown in Figure 5.4 a). In order to extract T-T absorption spectrum we subtracted the longest time trace (960 ns) from the time slice at 20 ns under the assumptions that at long time the only contributing transient is the phenoxyl radical and at delay>10 ns the contribution from the singlet excited state absorption will be minimal based on the fluorescence lifetime of phenol in water 20 (  f = 3.3 ns). Therefore, the difference spectra essentially represents the T-T absorption spectra. Figure 5.4 b) shows the time slices and the extracted T-T absorption spectra and compared with previously reported triplet absorption spectra derived in a similar fashion. 7 Evidently, the spectral shape of our triplet absorption spectrum matches quite well with the literature spectrum. We noticed that the triplet band extends further red beyond the probing window with 400 nm pumped continuum, therefore, validates our earlier assumption of triplet contribution to the transient absorption signal at wavelengths >350 nm. It is worth mentioning that in this far UV probe experiment we neglected the contribution from the hydrated electron since its extinction falls rapidly at wavelengths smaller than 400 nm. 95 Since we found all the transients spectra overlapping in the far UV region, it is quite difficult to disentangle the dynamics of each species from single wavelength analysis without a physical kinetic model. In order to deconvolute the spectral contribution of each transients and their corresponding dynamics, we performed global fitting analysis (PDP global fitting software) with a simple kinetic model as described below: Subsequent to photoexcitation, the excited singlet state (S 1 ) forms triplet (T 1 ), phenoxyl radical (R) and hydrated electron (e). Photoexcited molecules on S 1 surface also undergo radiative decay to the ground state. Both the triplet and the solvated electron are known to be quenched by oxygen with characteristics bimolecular quenching rates reported earlier. 7 Interestingly, the triplet phenol is quenched to form the phenoxyl radical. 7 Considering all this photo-physical/chemical processes and assuming no significant contribution from the hydrated electron signal in the probing range, we can express the kinetic scheme as shown in Figure 5.5. Figure 5.4. a) Transient absorption spectra of aqueous phenol solution with far UV probe at selective pump-probe delay times. b) Extracting T-T absorption spectrum from nanosecond TA spectra (see text for details). For comparison reported triplet absorption spectrum 7 is included. An arbitrary scaling factor and offset has been incorporated to the literature spectrum. 96 s The results of the this global fitting analysis is shown in Figure 5.6. Based on our model, the radiative decay rate (k rad ) is found to be (25.5 ± 2 ns) -1 . This value is consistent with the reported fluorescence quantum yield (  f = 0.12-0.14) and fluorescence lifetime (  f = 3.3 ns) in water (  f =  f *k rad ). 20 The rates of singlet to triplet conversion (k 1 ) and radical formation (k 2 ) are (4.2 ± 0.2 ns) -1 and (33 ± 2 ns) -1 respectively. Using these rate constant values, we found the fluorescence lifetime [  f = 1/(k rad +k 1 +k 2 )] to be 3.11 ns which is again in excellent agreement with the experimental value. The triplet state is found to undergo pseudo first order quenching by oxygen forming phenoxyl radical at a rate (k 3 ) of (415 ± 20 ns) -1 . At saturated oxygen concentration in fresh water at 20ºC and 1 bar pressure, the pseudo first order quenching rate (k Q *[Q]) is estimated to be (285 ns) -1 . 7 Therefore, our global fitting results indicate that the prepared solution contains dissolved oxygen approximately 60% of the maximum possible concentration. Although the stock solution and the gravity jet reservoir is purged with dry Figure 5.5. Kinetic scheme of the photo-physical/chemical processes of aqueous phenol solution. k rad indicates radiative decay rate, k 1 and k 2 indicate rates of parallel reactions forming the triplet (T 1 ) and radical (R) (along with the solvated electron but not included for analysis, see text for details) respectively. k 3 denotes pseudo-first order rate constant of triplet quenching by oxygen. 97 nitrogen during the experiment, the jet surface is exposed to atmospheric oxygen which diffusively dissolves into the solution and induce bimolecular quenching of the photogenerated triplet state. From the above mentioned rate constants we estimate the quantum yields of each transients as follows:  f = 13 ± 2 %,  T = 78 ± 2% and  R = 10 ± 2%. 5.4 Conclusion Using long time transient absorption measurements with ultra broadband probing window (250-380 nm and 350-620 nm), we have successfully deconvoluted transient species produced after photoexcitation to the first excited singlet surface (S 1 ) of aqueous phenol solution. We have illustrated that extending the probe to the far UV provides much more spectral information than the conventional visible window. However, to disentangle the dynamics in the enriched spectral region, we need to employ a global fitting strategy with appropriate physical model. Following this strategy we are able to extract population dynamics of each spectral components and their corresponding quantum yields. Figure 5.6. a) Experimental time slices (solid lines) and time slices from the global fitting (red dashed lines) at selective pump-probe delay times. b) Kinetics of each transient species from the global fitting model discussed above. 98 References 1. Morozova, O. B.; Kiryutin, A. S.; Sagdeev, R. Z.; Yurkovskaya, A. V., Electron Transfer between Guanosine Radical and Amino Acids in Aqueous Solution. 1. Reduction of Guanosine Radical by Tyrosine. J. Phys. Chem. B 2007, 111, 7439-7448. 2. Kok, B.; Forbush, B.; McGloin, M., COOPERATION OF CHARGES IN PHOTOSYNTHETIC O2 EVOLUTION–I. A LINEAR FOUR STEP MECHANISM. Photochem Photobiol 1970, 11, 457-475. 3. Hermann, R.; Mahalaxmi, G. R.; Jochum, T.; Naumov, S.; Brede, O., Balance of the Deactivation Channels of the First Excited Singlet State of Phenols: Effect of Alkyl Substitution, Sterical Hindrance, and Solvent Polarity. J. Phys. Chem. A 2002, 106, 2379. 4. Brede, O.; Leichtner, T.; Kapoor, S.; Naumov, S.; Hermann, R., Antithetical product situation in the femtosecond and nanosecond photoionization of sterically hindered phenols in non-protic solvents. Chem Phys Lett 2002, 366, 377-382. 5. Zhang, Y.; Oliver, T. A. A.; Ashfold, M. N. R.; Bradforth, S. E., Contrasting the excited state reaction pathways of phenol and para-methylthiophenol in the gas and liquid phases. Faraday Discuss 2012, 157, 141-163. 6. Harris, S. J.; Murdock, D.; Zhang, Y.; Oliver, T. A. A.; Grubb, M. P.; Orr-Ewing, A. J.; Greetham, G. M.; Clark, I. P.; Towrie, M.; Bradforth, S. E.; Ashfold, M. N. R., Comparing molecular photofragmentation dynamics in the gas and liquid phases. Physical Chemistry Chemical Physics 2013, 15, 6567-6582. 7. Bent, D. V.; Hayon, E., Excited state chemistry of aromatic amino acids and related peptides. I. Tyrosine. Journal of the American Chemical Society 1975, 97, 2599-2606. 8. Grabner, G.; Köhler, G.; Zechner, J.; Getoff, N., PATHWAYS FOR FORMATION OF HYDRATED ELECTRONS FROM EXCITED PHENOL AND RELATED COMPOUNDS. Photochem Photobiol 1977, 26, 449-458. 9. Grabner, G.; Koehler, G.; Marconi, G.; Monti, S.; Venuti, E., Photophysical properties of methylated phenols in nonpolar solvents. The Journal of Physical Chemistry 1990, 94, 3609- 3613. 10. Feitelson, J.; Hayon, E.; Treinin, A., Photoionization of phenols in water. Effects of light intensity, oxygen, pH, and temperature. Journal of the American Chemical Society 1973, 95, 1025-1029. 11. Pagba, C. V.; Chi, S.-H.; Perry, J.; Barry, B. A., Proton-Coupled Electron Transfer in Tyrosine and a β -Hairpin Maquette: Reaction Dynamics on the Picosecond Time Scale. The Journal of Physical Chemistry B 2015, 119, 2726-2736. 12. Zhang, Y.; Oliver, T. A. A.; Ashfold, M. N. R.; Bradforth, S. E., Exploring autoionization and photo-induced proton coupled electron transfer pathways of phenol in aqueous solution J. Am. Chem. Soc. 2015, in preparation. 99 13. Nix, M. G. D.; Devine, A. L.; Cronin, B.; Dixon, R. N.; Ashfold, M. N. R., High resolution photofragment translational spectroscopy studies of the near ultraviolet photolysis of phenol. J Chem Phys 2006, 125, 133318. 14. Ashfold, M. N. R.; Cronin, B.; Devine, A. L.; Dixon, R. N.; Nix, M. G. D., The Role of π * Excited States in the Photodissociation of Heteroaromatic Molecules. Science 2006, 312, 1637-1640. 15. Ashfold, M. N. R.; Devine, A. L.; Dixon, R. N.; King, G. A.; Nix, M. G. D.; Oliver, T. A. A., Exploring nuclear motion through conical intersections in the UV photodissociation of phenols and thiophenol. Proceedings of the National Academy of Sciences of the United States of America 2008, 105, 12701-12706. 16. Eland, J. H. D., Photoelectron spectra of conjugated hydrocarbons and heteromolecules. International Journal of Mass Spectrometry and Ion Physics 1969, 2, 471-484. 17. Ghosh, D.; Roy, A.; Seidel, R.; Winter, B.; Bradforth, S. E.; Krylov, A. I., First-Principle Protocol for Calculating Ionization Energies and Redox Potentials of Solvated Molecules and Ions: Theory and Application to Aqueous Phenol and Phenolate. J Phys Chem B 2012, 116, 7269-7280. 18. Sobolewski, A. L.; Domcke, W., Photoinduced Electron and Proton Transfer in Phenol and Its Clusters with Water and Ammonia†. The Journal of Physical Chemistry A 2001, 105, 9275-9283. 19. Sobolewski, A. L.; Domcke, W.; Dedonder-Lardeux, C.; Jouvet, C., Excited-state hydrogen detachment and hydrogen transfer driven by repulsive (1)pi sigma* states: A new paradigm for nonradiative decay in aromatic biomolecules. Physical Chemistry Chemical Physics 2002, 4, 1093-1100. 20. Zhang, Y. PhD Thesis. University of Southern California. 21. Bent, D. V.; Hayon, E., Excited state chemistry of aromatic amino acids and related peptides. III. Tryptophan. Journal of the American Chemical Society 1975, 97, 2612-2619. 22. Das, S. PhD Thesis. 23. Riedle, E.; Bradler, M.; Wenninger, M.; Sailer, C. F.; Pugliesi, I., Electronic transient spectroscopy from the deep UV to the NIR: unambiguous disentanglement of complex processes. Faraday Discuss 2013, 163, 139-158. 24. Tauber, M. J.; Mathies, R. A.; Chen, X. Y.; Bradforth, S. E., Flowing liquid sample jet for resonance Raman and ultrafast optical spectroscopy. Rev Sci Instrum 2003, 74, 4958-4960. 25. Pullin, D.; Andrews, L., The absorption spectra of the phenoxyl radical in solid argon. Journal of Molecular Structure 1983, 95, 181-185. 26. Chen, X.; Larsen, D. S.; Bradforth, S. E.; van Stokkum, I. H. M., Broadband spectral probing revealing ultrafast photochemical branching after ultraviolet excitation of the aqueous phenolate anion. J Phys Chem A 2011, 115, 3807-3819. 100 27. Das, T. N., Oxidation of Phenol in Aqueous Acid: Characterization and Re actions of Radical Cations vis-à-vis the Phenoxyl Radical. The Journal of Physical Chemistry A 2005, 109, 3344-3351. 28. Köhler, G.; Kittel, G.; Getoff, N., Decay processes of singlet excited phenol in solution. J Photochem 1982, 18, 19-27. 29. Köhler, G.; Getoff, N., Energy dependence and solvent effects in the deactivation of phenol molecules in solution. Chem Phys Lett 1974, 26, 525-528. 30. Zhang, Y.; Oliver, T. A. A.; Ashfold, M. N. R.; Bradforth, S. E., Exploring the autoionization and photo-induced proton-coupled electron transfer pathways of phenol in aqueous solution. In preparation. 31. Laws, W. R.; Ross, J. B. A.; Wyssbrod, H. R.; Beechem, J. M.; Brand, L.; Sutherland, J. C., Time-resolved fluorescence and proton NMR studies of tyrosine and tyrosine analogs: correlation of NMR-determined rotamer populations and fluorescence kinetics. Biochemistry 1986, 25, 599-607. 32. Pines, E.; Fleming, G. R., Self quenching of 1-naphthol. Connection between time- resolved and steady-state measurements. Chem Phys 1994, 183, 393-402. 33. Hasselbacher, C. A.; Waxman, E.; Galati, L. T.; Contino, P. B.; Ross, J. B. A.; Laws, W. R., Investigation of hydrogen bonding and proton transfer of aromatic alcohols in nonaqueous solvents by steady-state and time-resolved fluorescence. The Journal of Physical Chemistry 1991, 95, 2995-3005.

Abstract (if available)

Abstract Photoelectron spectroscopy, developed as a spectroscopic technique based on photoelectric effect, provides useful insights into chemical composition, molecular electronic structure and reaction dynamics. Although traditionally used for solid substrates and ultracold molecular beams, recent development of liquid microjet technique expands the scope of electron detection technique to the realm of pure liquids and liquid solutions. Although liquid microjet technique unfolds a new direction of spectroscopic endeavor, the application and widespread adoption has been limited primarily due to the highly demanding infrastructure and maintenance cost associated with high energy synchrotron radiation as the light source. This dissertation describes an alternative design and realization of a liquid jet photoelectron spectrometer coupled with ultrafast lasers as light sources to study liquid solution with greater selectivity and sensitivity compared to traditional x-ray photoemission measurements. Using this novel technique along with complementary transient absorption measurements, the following chapters delve into the details of redox chemistry of biomolecules as well as explore the excited state dynamics of solvated electron and aqueous phenol solutions.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Liquid phase photoelectron spectroscopy as a tool to probe excess electrons in liquid ammonia along with Birch chemistry carbanions: from dilute electrolytic solutions to the metallic regime

PDF

Ultrafast spectroscopy of aromatic amino acids and their chromophores in the condensed phase

PDF

Spectroscopic signatures and dynamic consequences of multiple interacting states in molecular systems

PDF

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

PDF

Excited state proton transfer in quinoline photobases

PDF

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

PDF

Two-photon absorption spectroscopy and excited state photochemistry of small molecules

PDF

Ultrafast spectroscopic interrogation and simulation of excited states and reactive surfaces

PDF

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

PDF

Understanding organic and inorganic semiconductor surfaces by coherent nonlinear spectroscopy

PDF

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

PDF

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

PDF

Ultrafast electronic deactivation of DNA bases in aqueous solution

PDF

Computational spectroscopy in gas and condensed phases

PDF

Excited state dynamics of organic photovoltaics probed using ultrafast spectroscopy

PDF

Molecular orientation and structure of hydrogen bonds at water interfaces

PDF

Adiabatic and non-adiabatic molecular dynamics in nanoscale systems: theory and applications

PDF

Spectroscopy and photodissociation dynamics of hydroxyethyl radicals

PDF

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

PDF

Photoelectron and ion imaging investigations of spectroscopy, photoionization, and photodissociation dynamics of diazomethane and diazirine

Asset Metadata

Creator Roy, Anirban (author)

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

School College of Letters, Arts and Sciences

Degree Doctor of Philosophy

Degree Program Chemistry

Publication Date 09/22/2015

Defense Date 08/04/2015

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

Tag excited state dynamics,liquid micro-jet,OAI-PMH Harvest,photoelectron spectroscopy

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Bradforth, Stephen E. (committee chair), Benderskii, Alexander V. (committee member), El-Naggar, Moh (committee member)

Creator Email anirbanr@usc.edu,roy86_usc@yahoo.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c40-186983

Unique identifier UC11273301

Identifier etd-RoyAnirban-3945.pdf (filename),usctheses-c40-186983 (legacy record id)

Legacy Identifier etd-RoyAnirban-3945.pdf

Dmrecord 186983

Document Type Dissertation

Format application/pdf (imt)

Rights Roy, Anirban

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