Chinese Journal of Chemical Physics  2017, Vol. 30 Issue (6): 671-677

#### The article information

Junjun Tan, Yi Luo, Shuji Ye

A Highly Sensitive Femtosecond Time-Resolved Sum Frequency Generation Vibrational Spectroscopy System with Simultaneous Measurement of Multiple Polarization Combinations

Chinese Journal of Chemical Physics, 2017, 30(6): 671-677

http://dx.doi.org/10.1063/1674-0068/30/cjcp1706114

### Article history

Accepted on: July 12, 2017
A Highly Sensitive Femtosecond Time-Resolved Sum Frequency Generation Vibrational Spectroscopy System with Simultaneous Measurement of Multiple Polarization Combinations
Junjun Tan, Yi Luo, Shuji Ye
Dated: Received on June 2, 2017; Accepted on July 12, 2017
Hefei National Laboratory for Physical Sciences at the Microscale, and Department of Chemical Physics, University of Science and Technology of China, Hefei 230026, China
*Author to whom correspondence should be addressed. Shuji Ye, E-mail:shujiye@ustc.edu.cn, Tel./FAX: +86-551-63603462
Abstract: Characterization of real-time and ultrafast motions of the complex molecules at surface and interface is critical to understand how interfacial molecules function. It requires to develop surface-sensitive, fast-identification, and time-resolved techniques. In this study, we employ several key technical procedures and successfully develop a highly sensitive femtosecond time-resolved sum frequency generation vibrational spectroscopy (SFG-VS) system. This system is able to measure the spectra with two polarization combinations (ssp and ppp, or psp and ssp) simultaneously. It takes less than several seconds to collect one spectrum. To the best of our knowledge, it is the fastest speed of collecting SFG spectra reported by now. Using the time-resolved measurement, ultrafast vibrational dynamics of the N-H mode of α-helical peptide at water interface is determined. It is found that the membrane environment does not affect the N-H vibrational relaxation dynamics. It is expected that the time-resolved SFG system will play a vital role in the deep understanding of the dynamics and interaction of the complex molecules at surface and interface. Our method may also provide an important technical proposal for the people who plan to develop time-resolved SFG systems with simultaneous measurement of multiple polarization combinations.
Key words: Femtosecond time-resolved    Sum frequency generation vibrational spectroscopy    Ultrafast vibrational dynamics    Multiple polarization combination measurement    Chiral
Ⅰ. INTRODUCTION

Sum frequency generation vibrational spectroscopy (SFG-VS) is a second-order optical laser technique. Owing to its intrinsic sensitivity to the surface and interface, it has been demonstrated to be a powerful and label-free tool to identify the molecular structures and dynamics at the surface and interface in different chemical environments [1-5]. It has been widely applied to investigate the structure and orientation of various interfacial molecules, ranging from small molecules such as water and alcohol [6, 7], to complex systems of polymers and biomolecules [8-15]. In previous studies, the SFG spectra were generated mainly using a picosecond narrow-band frequency scanning system (NB-SFG) [16-21]. Although NB-SFG can offer a measurement with high spectral resolution, it takes more than ten minutes to acquire one spectrum. Therefore, only molecular information at static states is obtained. However, static information is not always sufficient to characterize the molecular behaviors and interactions occurred at surface and interface [11]. Generally, the interfacial chemical reactivity and biological processes involve many ultrafast motion in the timescale from hundreds femtosecond (fs) to picosecond (ps), for example, the formation of equilibrium surface structures, hydrogen-bond breaking and reorganizing, occurrence of the chemical transformations on surfaces, ultrafast electron and energy transfer at the interface, allosteric communication in proteins, and the vibrational mode coupling [11, 22-33]. To monitor such ultrafast motion at the interface, it requires to develop highly sensitive, fast-identification, and time-resolved SFG-VS technique.

Femtosecond broad-band infrared SFG system (BB-SFG) developed recently can greatly fasten the speed of collecting the spectra owing to the high repetition rate of femtosecond laser and a broad-band infrared beam used [34-38]. It exhibits great potentials in capturing the fast dynamic processes. For instance, Yan group achieved the dynamics of hIAPP misfolding at the DPPG monolayer/water interface in minute-time scale [39]. Totally, it still takes more than one minute to collect one spectrum of single polarization combination in previous BB-SFG reports. There is still large space to enhance the speed in collecting spectra. In addition, orientation angle of a function group is one of important conformational information that SFG measures. It can be deduced by measuring the spectral intensity ratio of different polarization combinations [4, 40-44]. In previous common experimental procedures, researchers often first collected the spectra with a polarization combination (for example, ssp) and then turned to another polarization combination (for example, ppp) [10, 45, 46]. Therefore, such procedures do not allow to deduce the real-time orientation information of irreversible processes. Furthermore, femtosecond time-resolved measurements by using an intense femtosecond or picosecond laser pulse pump followed by a BB-SFG probe can provide direct information about the structural change induced by the pump laser pulse, yielding a new light on the translational and vibrational dynamics of interfacial molecule, as well as interfacial electron/energy transfer rates and pathways [27, 28, 47]. Currently only several groups have constructed the femtosecond time-resolved systems [24, 48-50] and the time-resolved researches in these groups are mainly focused on the studies of small molecules such as water [51]. Few applications of the time-resolved SFG experiments on the interfacial complex molecular system were reported. Herein, we develop a highly sensitive femtosecond time-resolved SFG-VS system with simultaneous measurement of multiple polarization combinations. By adopting a near-total-internal-reflection geometry to enhance the signals, it takes only several ten milliseconds for the best case and several seconds for the normal case to collect one spectrum. It can measure the spectra with two polarization combinations (ssp and ppp, or psp and ssp) simultaneously. For the time-resolved measurement, the system has a time-resolution of ~100 fs and spectral resolution of ~5 cm$^{-1}$. We believe our time-resolved SFG system will play a vital role in the deep understanding of the dynamics and interaction of the complex molecules at surface and interface.

Ⅱ. MATERIALS AND SAMPLE PREPARATIONS

1, 2-Ditetradecanoyl-sn-glycero-3-phospho-(1$'$-rac-gly cerol) (sodium salt) (DMPG) and 1-palmitoyl-2-oleoyl-sn-glycero-3-phospho-(1$'$-rac-glycerol) (sodium salt) (POPG) lipids were purchased from Avanti Polar Lipids (Alabaster, AL). D$_2$O (99.9% D) was purchased from Sigma-Aldrich. WALP23 (sequence: GWW(LA)$_8$LWWA) and LK$_7$$\beta (sequence: LKLKLKL) with a purity of \geq98% were purchased from Shanghai Apeptide Co., Ltd. GP41(sequence: AVGIGALFLGFLGAAGSTMGARS) with a purity of \geq95.1% was ordered from MyBioSource, Inc. DI H_2O (with resistivity of 18.2 M\Omega$$\cdot$cm) was produced by a Milli-Q reference system (Millipore, Bedford, MA). WALP23 and LK_7\beta (with the concentration of 2.0 mg/mL) were dissolved in methanol (purchased from Alfa Aesar). GP41 was dissolved in hexafluoro-isopropanol (Aldich-Sigma). The DMPG and POPG solution (with the concentration of 2.0 mg/mL) were prepared in mixed solvents of chloroform and methanol (with a volume ratio of 2:1) (purchased from Sinopharm Chemical Reagent Co., Ltd.). The lipids and WALP23 solutions were kept at -20 ^{\circ}C. Right-angle CaF_2 prisms were purchased from Chengdu Ya Si Optoelectronics Co., Ltd. (Chengdu, China). All of the chemicals were used as received. The substrate prism cleaning and lipid bilayer preparation were carried out using a standard procedure given in our previous report [17]. Ⅲ. RESULTS AND DISCUSSION A. Femtosecond broad-bandwidth sum frequency generation system We first construct the femtosecond BB-SFG system using a high power regenerative amplifier (Spectra Physics, Spitfire Ace seeded by Mai-Tai SP). The average power of the amplifier output at 800 nm is 5.0 W with the bandwidth of 13 nm, pulse width of 100 fs, and a repetition rate of 1 kHz. The output is split into three fractions of 2.0, 2.0, and 1.0 W, respectively. The first fraction of 1.0 W is passed through a home-built 4-f pulse shaping system (FIG. 1(a)) to generate a picosecond SFG visible probe, which is the key for the SFG spectral resolution [52-55]. The 4-f pulse shaping system is composed of two gratings (1800 grooves/mm, PC 1800 25\times25\times6 NIR, Spectrogon, optimized at 800 nm), two plano convex cylindrical lens with anti-reflection coating (FL=200 mm, LJ1653L1-B, Thorlabs), and one silt (VA100/M, Thorlabs). By tuning the slit, a picosecond narrow band visible light with a Gaussian profile can be produced.  FIG. 1 (a) 4-f pulse shaping system (G=grating, S=slit, CL=cylindrical lens). (b) Polarization controller for the SFG signals (GLP=Glan-laser polarizer, HWP=half-wave plate). The second fraction of the pulse (2.0 W) is used for the excitation of a commercial optical parametric amplifier (Spectra Physics, TOPAS Prime) and collinear difference frequency generation system (DFG, AgGaS_2 crystal) to produce tunable broadband infrared pulses (2.5-10 μm) for SFG probe. The left 2.0 W of the output is used for the second TOPAS/DFG system to produce the tunable infrared pulses (2500 nm-10 μm) for exciting the vibrational modes in time-resolved SFG system. After we obtain the narrow band visible pulse and broad-band infrared pulses, we align these two beams in the same plane and send toward the sample with incident angles of 45^{\circ} and 60^{\circ} relative to the sample surface normal, respectively. Before the beams are sent to the sample surface, the narrow band visible beam energy is modulated by a combination of a zero-order half-wave plate (10RP02-46, Newport) and a Glan-Laser polarizer (10GL08AR.16, Newport). Its polarization is controlled by a combination of a Glan-Laser polarizer (10GL08AR.16, Newport) and a zero-order half-wave plate (10RP02-46, Newport). The IR beam is passed through an optical delay (XPS, Newport) to control the timing. The beams are focused and overlapped at the sample surface by using off-axis parabolic bare gold mirror (FL=150 mm, 84-591, Edmund Optics) for IR, and plano-convex lens (FL=750 mm, SPX032AR.16, Newport) for visible beam, respectively. The beam diameters at the focusing point are determined to be approximately 200 μm. The IR beams are protected by a home-built chamber purged with dry gas (dry gas generator, Peak Scientific) to avoid the IR energy loss due to water vapor absorption. The SFG signals in reflection direction pass through plano-convex lens (FL=150 mm, Thorlabs), a Glan-laser polarizer (10GL08AR.16, Newport), an achromatic half-wave plate (15RP52-1, Newport)(FIG. 1(b)), a notch filter (NF808-34, Thorlabs), and a plano-convex lens (FL=100 mm, Thorlabs), are then focused into a spectrograph (Shanrock 303i). The resulting spectra are captured by an ICCD (Andor istar 734). The gate pulse width of ICCD is set to be 20 ns. The time-zero is determined by monitoring the infrared-visible sum-frequency signals as a function of the delay between the IR probe and the visible pulses. After finishing construction of the BB-SFG system, we first calibrate the monochromator by using 435.83 nm line of mercury vapor lamp and then test the properties of the system. When the slit in 4-f pulse shaping system is set to 30 μm, a picosecond pulse centered at 801.0 nm with the output of ~15 μJ and bandwidth of 0.33 nm is achieved, resulting a spectral resolution of ~5 cm^{-1} (FIG. 2(a)). A nonlinear SFG signal from the surface of GaAs is used to determine the IR profile. The power and bandwidth of the IR pulses at different frequency are given in Table Ⅰ and FIG. 2(b). Table Ⅰ Typical IR powers and full width at half-maximum (FWHM) for the spectral regions shown in FIG. 2(b).  FIG. 2 (a) Spectrum of the 800 nm beam after pulse shaper. (b) IR energy profiles in different spectral regions. Finally, we use two real samples to examine the spectral resolution and speed to collect the spectra. To enhance the SFG signals, a near-total-internal-reflection geometry is adopted. The energy of visible (801 nm) and infrared (6000 nm) is 15 and 12.8 μJ respectively. The first sample is peptide LK_7\beta. The proton exchange dynamics of LK$_7$$\beta at air/water interface has been studied using BB-SFG by Yan et al. It is found that the peptide forms an antiparallel \beta-sheet at amphiphilic interfaces [56]. FIG. 3(a) shows the psp spectrum of LK_7$$\beta$ film coated on CaF$_2$ prism with an acquisition time of 18 ms. The SFG spectrum has been normalized by the IR energy profile of the IR pulses that was determined by measuring the SFG signals from the gold layer coated at the prism. The spectrum is dominated by two strong peaks at 1630 and 1660 cm$^{-1}$ and a weak peak at 1690 cm$^{-1}$, indicating that LK$_7$$\beta film adopts an antiparallel \beta-sheet and \beta-turn structure [56]. It is worth mentioning that the spectrum with a very high signal/noise ratio can be obtained with an acquisition time of 18 ms, which is much faster than previous report [56]. To further confirm that our system can collect the spectra in second-time scale, we performed another experiment by measuring the spectra of WALP23 at DMPG/H_2O bilayer interface, which is the common environment for the interactions between the lipid membranes and peptides. It has been reported that WALP forms continuous \alpha-helices in the lipid bilayers [57, 58]. FIG. 3(b) shows the ppp spectrum of amide I band with the acquisition time of 1 s. The spectrum is dominated by a single peak at 1660 cm^{-1}. The peak position is consistent with the 2D SFG results for the gold surface-bound \alpha-helical peptide [59]. Either the spectrum of LK_7$$\beta$ film or WALP demonstrates that our BB-SFG system has very fast spectral acquisition speed. To the best of our knowledge, it is the fastest speed of collecting spectra reported by now. This system will provide an important tool to look insight into the interfacial real-time processes in second-time scale.

 FIG. 3 (a) The psp spectrum of LK$_7$$\beta$ coated at CaF$_2$ prism surface with an acquisition time of 18 ms. (b) The ppp spectrum of WALP23 at the DMPG bilayer/H$_2$O interface with 1 s acquisition time.
B. BB-SFG system with simultaneous measurement of multiple polarization combinations

Knowledge of orientation angle and conformation change in irreversible processes is critical to understand the dynamic behavior, which requires to measure the SFG spectra with multiple polarization combinations simultaneously. To reach this approach, Bonn et al., adopted a beam-displacing prism to separate p-and s-polarized SFG signals, which were recorded at different heights on the CCD camera in BB-SFG system [60]. However, the separation of the polarization components of the signals is insufficient. To solve this problem, we employ two Glan-Laser polarizers to separate the polarization components (FIG. 4(a)). Glan-Laser polarizer has a extinction ratio ($T_\textrm{p}$/$T_\textrm{s}$) of $>$10$^5$:1. The SFG signals are separated by the first Glan-Laser polarizer (GLP1) where the p-polarized light traverses while the s-polarized light reflects to the mirror (M1). After separated by GLP1, the p-polarized light continues to pass through the second Glan-laser polarizer. The s-polarized light is reflected by another mirror (M2) and GLP2, which can cause a 2.6 mm vertical displacement between p-and s-polarized lights. The spectra of p-and s-polarized SFG signals are collected simultaneously by imaging at different vertical pixel array on the ICCD camera. In addition, because of the difference between ssp and ppp or psp in the propagation path and reflection efficiency of grating, it is necessary to calibrate the ssp signals. Referring to the light path in FIG. 1(b), the actual intensity of ssp spectra equals to about 2/3 times of the measured spectra in FIG. 4 (b) and (c) (steel blue line in FIG. 4(a)). The ssp and ppp spectra can be collected simultaneously when the polarization angle of the visible beam is set to be 45$^{\circ}$. Similarly, the ssp and psp spectra can be collected simultaneously when the polarization angle of the visible beam is set to be 90$^{\circ}$. FIG. 4(b) shows simultaneous ssp and ppp spectra of WALP23 at DMPG bilayer/H$_2$O interface. FIG. 4(c) presents the simultaneous ssp and psp spectra of GP41 at POPG bilayer/H$_2$O interface. This system will open broad opportunities for probing the real-time orientation change in irreversible processes and structural transition involving the formation of chiral SFG signals.

 FIG. 4 (a) A schematic of the polarization control module of sum frequency signal in BB-SFG system. (b) The ssp and ppp spectra of WALP23 at DMPG bilayer/H$_2$O interface collected simultaneously. (c) The ssp and psp spectra of gp41 at POPG bilayer/H$_2$O interface collected simultaneously.
C. Femtosecond time-resolved systems

In a time-resolved SFG-VS experiment, an intense fs or ps laser pulse in the visible or near-infrared or infrared is used to pump the sample to cause electronic, vibrational, and thermal excitation of the molecules and substrates, and SFG-VS is then used to monitor these transient changes in real time at high temporal resolution [11]. The detailed introduction and the theory for femtosecond time-resolved SFG have been given by Shen and Bonn Groups [23, 48, 51]. In our time-resolved SFG system (FIG. 5), the pump IR is focused by a concave gold mirror (R=-500 mm, 092-0130R-500, EKSMA Optical) with incident angle of 66$^{\circ}$ to the sample surface normal. The 1 kHz pump IR pulse is chopped to 500 Hz with a phase-locked optical chopper (MC2000B-EC, Thorlabs) synchronized with the laser repetition rate to generate a pump-on and pump-off SFG signals. The pump-on and pump-off SFG signals are separated by a galvanometric scanner (GVS011, Thorlabs) to different heights on the ICCD. The time-zero and instrument response are determined according to previous reports, i.e., by monitoring the third-order cross-correlation of infrared-infrared visible sum-frequency signal as a function of the delay between the IR pump and IR probe pulses [23, 48, 51]. A LABVIEW program is used to control the delay time between the IR pump and the IR probe. For every delay time, a system consisted of chopper and galvanometric scanner is used to minimize the influence of laser fluctuations on the pump-on and pump-off spectra. FIG. 6 shows a typical pump-probe IIV-SFG cross-correlation signal of WALP23 at the POPG bilayer/water interface. The instrument response function (IRF) can be fitted very well by a Gaussian function with a full width at half maximum (FWHM) of ~172 fs, resulting a time resolution of ~100 fs. FIG. 7(a) exhibits one example of the transient SFG spectra of the WALP23 at the POPG bilayer/D$_2$O interface with $\nu_{\textrm{pump}}$=3300 cm$^{-1}$ and $\nu_{\textrm{probe}}$=3300 cm$^{-1}$. Theoretically, SFG intensity is proportional to the square of the surface population density of the vibrational state [1-5]. As a result, the amide A intensity is reduced because the pump IR pulse at 3300 cm$^{-1}$ excites the N-H stretch mode from the ground ($\nu$=0) to its first vibrational state ($\nu$=1) and then causes the population of ground state decrease. Following the bleaching, the $\nu$=1 state relaxes to the ground state again and then the intensity of the amide A band gradually recovers. FIG. 7(b) shows the intensity decay of the ground state of N-H stretching, which can be well modeled by using a four-level vibrational model with the decay time (1.67$\pm$0.03) ps. The relaxation time is similar to the results of WALP at the DPPG bilayer/D$_2$O interface [61], indicating the membrane environment does not affect the N-H vibrational relaxation dynamics of $\alpha$-helical structure.

 FIG. 5 A schematic of the time-resolved SFG setup.
 FIG. 6 ypical IIV-SFG cross-correlation trace WALP23 at POPG/water bilayer interface with prism geometry.
 FIG. 7 (a) The transient SFG spectra of the WALP23 at the POPG bilayer/D$_2$O interface with $\nu_{\textrm{pump}}$=3300 cm$^{-1}$ and $\nu_{\textrm{probe}}$=3300 cm$^{-1}$. (b) The intensity decay of the ground state of N-H stretching.
Ⅳ. CONCLUSION

We have introduced a highly sensitive femtosecond time-resolved sum frequency generation vibrational spectroscopy system developed in our group in details. This system has the capability to measure the spectra with two polarization combinations (ssp and ppp, or psp and ssp) simultaneously. We adopt a near-total-internal-reflection geometry to enhance the signals. We have used the peptide film and membrane-associated peptide in the lipid bilayer/water interface to examine the system. It can be found that it takes only several ten milliseconds for the best case and several seconds for the normal case to collect one spectrum. To the best of our knowledge, it is the fastest speed of collecting SFG spectra reported by now. For the time-resolved measurement, the system has a time-resolution of ~100 fs and spectral resolution of ~5 cm$^{-1}$. Using the time resolved measurement, the ultrafast vibrational dynamics of protein N-H mode at the water interface is determined. It is found that the membrane environment does not affect the N-H vibrational relaxation dynamics of $\alpha$-helical structures. It is expected that this time-resolved SFG system will play a vital role in the deep understanding of the dynamics and interaction of the complex molecules at surface and interface. Our method will also provide important technical proposal for the researchers who plan to develop time-resolved SFG system with simultaneous measurement of multiple polarization combinations.

Ⅴ. ACKNOWLEDGMENTS

This work was supported by the National Natural Science Foundation of China (No.21473177 and No.21633007), the National Key Research and Development Program of China (2017YFA0303500), the Fundamental Research Funds for the Central Universities (WK2340000064), and the Key Research Program of the Chinese Academy of Sciences. J. J. Tan thanks Dr. Hongchun Li, Dr. Kangzhen Tian, Mr. Baixiong Zhang, Mrs. Xia Hu and Ting Yu for their kind help.

Reference
 [1] Y. R. Shen, The Principles of Nonlinear Optics. New York: Wiley, (1984). [2] A. G. Lambert, P. B. Davies, and D. J. Neivandt, Appl. Spectrosc. Rev. 40 , 103 (2005). DOI:10.1081/ASR-200038326 [3] S. Gopalakrishnan, D. Liu, H. C. Allen, M. Kuo, and M. J. Shultz, Chem. Rev. 106 , 1155 (2006). DOI:10.1021/cr040361n [4] H. F. Wang, L. Velarde, W. Gan, and L. Fu, Annu. Rev. Phys. Chem. 66 , 189 (2015). DOI:10.1146/annurev-physchem-040214-121322 [5] C. S. Tian, and Y. R. Shen, Surf. Sci. Rep. 69 , 105 (2014). DOI:10.1016/j.surfrep.2014.05.001 [6] Y. R. Shen, and V. Ostroverkhov, Chem. Rev. 106 , 1140 (2006). DOI:10.1021/cr040377d [7] R. Lu, W. Gan, B. H. Wu, Z. Zhang, Y. Guo, and H. F. Wang, J. Phys. Chem. B 109 , 14118 (2005). DOI:10.1021/jp051565q [8] X. Lu, C. Zhang, N. Ulrich, M. Xiao, Y. H. Ma, and Z. Chen, Anal. Chem. 89 , 466 (2017). DOI:10.1021/acs.analchem.6b04320 [9] B. Ding, J. Jasensky, Y. Li, and Z. Chen, Acc. Chem. Res. 49 , 1149 (2016). DOI:10.1021/acs.accounts.6b00091 [10] E. C. Y. Yan, L. Fu, Z. Wang, and W. Liu, Chem. Rev. 114 , 8471 (2014). DOI:10.1021/cr4006044 [11] S. J. Ye, and Y. Luo, Sci. China Chem. 57 , 1646 (2014). DOI:10.1007/s11426-014-5233-5 [12] S. J. Ye, F. Wei, H. Li, K. Tian, and Y. Luo, Adv. Protein Chem. Struct. Biol. 93 , 213 (2013). DOI:10.1016/B978-0-12-416596-0.00007-5 [13] T. Yu, G. Zhou, X. Hu, and S. J. Ye, Langmuir 32 , 11681 (2016). DOI:10.1021/acs.langmuir.6b02560 [14] S. Roy, P. A. Covert, W. R. FitzGerald, and D. K. Hore, Chem. Rev. 114 , 8388 (2014). DOI:10.1021/cr400418b [15] Y. Liu, J. Jasensky, and Z. Chen, Langmuir 28 , 2113 (2012). DOI:10.1021/la203823t [16] S. J. Ye, and F. Wei, Analyst 136 , 2489 (2011). DOI:10.1039/c1an15032d [17] J. Tan, S. J. Ye, and Y. Luo, J. Phys. Chem. C 119 , 28523 (2015). DOI:10.1021/acs.jpcc.5b10632 [18] Y. Liu, T. L. Ogorzalek, P. Yang, M. M. Schroeder, E. N. G. Marsh, and Z. Chen, J. Am. Chem. Soc. 135 , 12660 (2013). DOI:10.1021/ja403672s [19] J. Liu, and J. C. Conboy, J. Am. Chem. Soc. 126 , 8894 (2004). DOI:10.1021/ja031570c [20] H. I. Okur, J. Hladlkov, K. B. Rembert, Y. Cho, J. Heyda, J. Dzubiella, P. S. Cremer, and P. Jungwirth, J. Phys. Chem. B 121 , 1997 (2017). DOI:10.1021/acs.jpcb.6b10797 [21] R. R. Feng, Y. Guo, and H. F. Wang, J. Chem. Phys. 141 , 18C507 (2014). DOI:10.1063/1.4895561 [22] J. Kubota, and K. Domen, Anal. Bioanal. Chem. 388 , 17 (2007). DOI:10.1007/s00216-006-0957-5 [23] M. Smits, A. Ghosh, J. Bredenbeck, S. Yamamoto, M. Mller, and M. Bonn, New J. Phys. 9 , 390 (2007). DOI:10.1088/1367-2630/9/10/390 [24] H. Arnolds, and M. Bonn, Surf. Sci. Rep. 65 , 45 (2010). DOI:10.1016/j.surfrep.2009.12.001 [25] B. Wu, X. Wang, J. Yang, Z. Hua, K. Tian, R. Kou, J. Zhang, S. J. Ye, Y. Luo, V. S. Craig, G. Z. Zhang, and G. M. Liu, Sci. Adv. 2 , e1600579 (2016). DOI:10.1126/sciadv.1600579 [26] L. De Marco, M. Thämer, M. Reppert, and A. Tokmakoff, J. Chem. Phys. 141 , 07B6121 (2014). [27] M. Smits, A. Ghosh, M. Sterrer, M. Müller, and M. Bonn, Phy. Rev. Lett. 98 , 098302 (2007). DOI:10.1103/PhysRevLett.98.098302 [28] Z. Zhang, L. Piatkowski, H. J. Bakker, and M. Bonn, Nat. Chem. 3 , 888 (2011). DOI:10.1038/nchem.1158 [29] G. Li, D. Magana, and R. B. Dyer, Nat. Commun. 5 , 3100 (2014). [30] G. A. Miño-Galaz, J. Phys. Chem. 119 , 6179 (2015). [31] Y. Chen, H. I. Okur, N. Gomopoulos, C. MaciasRomero, P. S. Cremer, P. B. Petersen, G. Tocci, D. M. Wilkins, C. Liang, and M. Ceriotti, Sci. Adv. 2, e1501891(2016). [32] K. Bhattacharyya, Chem. Commun. 25 , 2848 (2008). [33] J. Schaefer, E. H. Backus, Y. Nagata, and M. Bonn, J. Phys. Chem. Lett. 7 , 4591 (2016). DOI:10.1021/acs.jpclett.6b02513 [34] C. Y. Tang, and H. C. Allen, J. Phys. Chem. A 113 , 7383 (2009). [35] J. P. Symonds, H. Arnolds, V. L. Zhang, K. Fukutani, and D. A. King, J. Chem. Phys. 120 , 7158 (2004). DOI:10.1063/1.1669377 [36] J. Wang, H. Dubost, A. Ghalgaoui, W. Zheng, S. Carrez, A. Ouvrard, and B. Bourguignon, Surf. Sci. 626 , 26 (2014). DOI:10.1016/j.susc.2014.03.017 [37] G. Ma, and H. C. Allen, Langmuir 22 , 5341 (2006). DOI:10.1021/la0535227 [38] M. Okuno, and T. A. Ishibashi, J. Phys. Chem. Lett. 5 , 2874 (2014). DOI:10.1021/jz501158r [39] L. Fu, G. Ma, and E. C. Yan, J. Am. Chem. Soc. 132 , 5405 (2010). DOI:10.1021/ja909546b [40] X. Chen, J. Wang, A. P. Boughton, C. B. Kristalyn, and Z. Chen, J. Am. Chem. Soc. 129 , 1420 (2007). DOI:10.1021/ja067446l [41] J. Wang, S. H. Lee, and Z. Chen, J. Phys. Chem. B 112 , 2281 (2008). DOI:10.1021/jp077556u [42] K. T. Nguyen, J. T. King, and Z. Chen, J. Phys. Chem. B 114 , 8291 (2010). DOI:10.1021/jp102343h [43] H. F. Wang, W. Gan, R. Lu, Y. Rao, and B. H. Wu, Int. Rev. Phys. Chem. 24 , 191 (2005). DOI:10.1080/01442350500225894 [44] D. Xiao, L. Fu, J. Liu, V. S. Batista, and E. C. Yan, J. Mol. Biol. 421 , 537 (2012). DOI:10.1016/j.jmb.2011.12.035 [45] S. J. Ye, H. Li, F. Wei, J. Jasensky, A. P. Boughton, P. Yang, and Z. Chen, J. Am. Chem. Soc. 134 , 6237 (2012). DOI:10.1021/ja2110784 [46] P. Yang, F. G. Wu, and Z. Chen, J. Phys. Chem. C 117 , 3358 (2013). DOI:10.1021/jp3099522 [47] P. Bodis, R. Timmer, S. Yeremenko, W. J. Buma, J. S. Hannam, D. A. Leigh, and S. Woutersen, J. Phys. Chem. C 111 , 6798 (2007). [48] A. Ghosh, M. Smits, J. Bredenbeck, N. Dijkhuizen, and M. Bonn, Rev. Sci. Instrum. 79 , 093907 (2008). DOI:10.1063/1.2982058 [49] M. A. Donovan, Y. Y. Yimer, J. Pfaendtner, E. H. Backus, M. Bonn, and T. Weidner, J. Am. Chem. Soc. 138 , 5226 (2016). DOI:10.1021/jacs.6b01878 [50] J. Kleber, K. Laβ, and G. Friedrichs, J. Phys. Chem. A 117 , 7863 (2013). DOI:10.1021/jp404087s [51] J. A. McGuire, and Y. R. Shen, Science 313 , 1945 (2006). DOI:10.1126/science.1131536 [52] A. N. Bordenyuk, and A. V. Benderskii, J. Chem. Phys. 122 , 134713 (2005). DOI:10.1063/1.1873652 [53] O. Esenturk, and R. A. Walker, J. Chem. Phys. 125 , 174701 (2006). DOI:10.1063/1.2356858 [54] Y. Rao, M. Comstock, and K. B. Eisenthal, J. Phys. Chem. B 110 , 1727 (2006). DOI:10.1021/jp055340r [55] G. Ma, J. Liu, L. Fu, and E. C. Yan, Appl. Spectrosc. 63 , 528 (2009). DOI:10.1366/000370209788347057 [56] L. Fu, D. Xiao, Z. Wang, V. S. Batista, and E. C. Yan, J. Am. Chem. Soc. 135 , 3592 (2013). DOI:10.1021/ja3119527 [57] W. Im, and C. L. Brooks, Proc. Natl. Acad. Sci. USA 102 , 6771 (2005). DOI:10.1073/pnas.0408135102 [58] L. V. Schäfer, D.H. de Jong, A. Holt, A. J. Rzepiela, A.H. de Vries, B. Poolman, J. A. Killian, and S. J. Marrink, Proc. Natl. Acad. Sci. USA 108 , 1343 (2011). DOI:10.1073/pnas.1009362108 [59] J. E. Laaser, D. R. Skoff, J. J. Ho, Y. Joo, A. L. Serrano, J. D. Steinkruger, P. Gopalan, S. H. Gellman, and M. T. Zanni, J. Am. Chem. Soc. 136 , 956 (2014). DOI:10.1021/ja408682s [60] M. Smits, M. Sovago, G. W. H. Wurpel, D. Kim, M. Mller, and M. Bonn, J. Phys. Chem. C 111 , 8878 (2007). DOI:10.1021/jp067453w [61] J. J. Tan, B. X. Zhang, Y. Luo, and S. J. Ye, Angew. Chem. Int. Ed. 56 , 12977 (2017). DOI:10.1002/anie.201706996