I. INTRODUCTION

Section:

ChooseTop of pageABSTRACTI. INTRODUCTION <<II. CLASSIFICATION OF THE...III. SETUP LAYOUT OF THE ...IV. CONSTRUCTON OF A SFG-...V. OPERATION AND DATA PRO...VI. DISCUSSION ON THE PUM...VII. CONCLUSIONSREFERENCESSum frequency generation vibrational spectroscopy (SFG-VS) or vibrational sum frequency generation (VSFG) is vibrational spectroscopic technique based on the second-order nonlinear optical process of sum frequency generation (SFG). Under the electric-dipole approximation, SFG is a forbidden process in an isotropic bulk medium yet is allowed at the surfaces and interfaces between two isotropic bulk media where inversion symmetry is always broken.1,21. Y. R. Shen, Nature 337, 519 (1989). https://doi.org/10.1038/337519a02. K. B. Eisenthal, Chem. Rev. 96, 1343 (1996). https://doi.org/10.1021/cr9502211 Therefore, SFG-VS is an intrinsically surface-selective technique and can be used to probe literally any surface and interface. In short, SFG-VS can be simply remembered as a surface vibrational spectroscopic technique; and “a surface technique,” “a vibrational spectroscopic technique,” and “a second-order nonlinear optical technique,” are the three key features pertaining to SFG-VS.When conducting a SFG-VS experiment, two pulsed laser beams with one being in the visible (VIS) (or near-infrared, near-IR) wavelength region and the other being in the infrared (IR) wavelength region are required. As shown in Fig. 1, when the two laser pulses meet at the surface, they will drive the SFG process at the surface and a new pulse with its frequency to be the sum of the frequencies of the two incoming pulses (VIS and IR) will be generated from the surface. This newly generated pulse is the SFG pulse with its frequency to be ω3 = ω1 + ω2. Moreover, when the frequency of the IR pulse is in resonant with (or matches with) the vibrational mode of a surface species, an intensity enhancement of the SFG pulse will occur. This means that by scanning the IR frequency over the spectral region of our interested vibrational mode and recording the corresponding intensity of the generated SFG pulse, we will be able to obtain a vibrational spectrum of a surface species. This is the basic working principle of SFG-VS. Mathematically, this working principle can be described using the following Eqs. (1)–(3).3–123. C. Hirose, N. Akamatsu, and K. Domen, Appl. Spectrosc. 46, 1051 (1992). https://doi.org/10.1366/00037029241243854. X. Zhuang, P. B. Miranda, D. Kim, and Y. R. Shen, Phys. Rev. B 59, 12632 (1999). https://doi.org/10.1103/PhysRevB.59.126325. D. Simonelli and M. J. Shultz, J. Chem. Phys. 112, 6804 (2000). https://doi.org/10.1063/1.4812556. J. Wang, M. L. Clarke, and Z. Chen, Anal. Chem. 76, 2159 (2004). https://doi.org/10.1021/ac049887y7. A. J. Moad and G. J. Simpson, J. Phys. Chem. B 108, 3548 (2004). https://doi.org/10.1021/jp035362i8. H.-F. Wang, W. Gan, R. Lu, Y. Rao, and B.-H. Wu, Int. Rev. Phys. Chem. 24, 191 (2005). https://doi.org/10.1080/014423505002258949. A. G. Lambert, P. B. Davies, and D. J. Neivandt, Appl. Spectrosc. Rev. 40, 103 (2005). https://doi.org/10.1081/ASR-20003832610. H.-F. Wang, L. Velarde, W. Gan, and L. Fu, Annu. Rev. Phys. Chem. 66, 189 (2015). https://doi.org/10.1146/annurev-physchem-040214-12132211. F. Tang, T. Ohto, S. Sun, J. R. Rouxel, S. Imoto, E. H. G. Backus, S. Mukamel, M. Bonn, and Y. Nagata, Chem. Rev. 120, 3633 (2020). https://doi.org/10.1021/acs.chemrev.9b0051212. J. D. Pickering, M. Bregnhoj, A. S. Chatterley, M. H. Rasmussen, K. Strunge, and T. Weidner, Biointerphases 17, 011201 (2022). https://doi.org/10.1116/6.0001401 As shown in Eq. (1), the SFG-VS intensity, ISFG−VS, is proportional to the absolute square of the macroscopic second-order nonlinear susceptibility, χ(2), as well as the intensities of the incident visible and IR beams, IVIS and IIR, ISFG−VS∝|χ(2)|2⋅IVIS⋅IIR.(1)The second-order nonlinear susceptibility, χ(2), consists of a non-resonant term, χNR(2), and a sum of resonant terms, χq(2), as shown in Eq. (2), χ(2)=χNR(2)+∑qχq(2).(2)The expression of the resonant term, χq(2), is shown in Eq. (3), χq(2)∝AqωIR−ωq+iΓq,(3)where Aq, ωq, and Γq are the amplitude, resonance frequency, and damping constant of the qth vibrational mode, respectively. As we can see here, when the frequency of the incident IR beam, ωIR, is in resonance with the qth vibrational mode of a surface species (in other words, when ωIR=ωq), the SFG-VS intensity, ISFG−VS, will experience a signal enhancement. Scanning the IR frequency over the vibrational mode will lead to a SFG spectrum.SFG-VS has the capability to provide fruitful information to help us tackle a surface phenomenon.3–133. C. Hirose, N. Akamatsu, and K. Domen, Appl. Spectrosc. 46, 1051 (1992). https://doi.org/10.1366/00037029241243854. X. Zhuang, P. B. Miranda, D. Kim, and Y. R. Shen, Phys. Rev. B 59, 12632 (1999). https://doi.org/10.1103/PhysRevB.59.126325. D. Simonelli and M. J. Shultz, J. Chem. Phys. 112, 6804 (2000). https://doi.org/10.1063/1.4812556. J. Wang, M. L. Clarke, and Z. Chen, Anal. Chem. 76, 2159 (2004). https://doi.org/10.1021/ac049887y7. A. J. Moad and G. J. Simpson, J. Phys. Chem. B 108, 3548 (2004). https://doi.org/10.1021/jp035362i8. H.-F. Wang, W. Gan, R. Lu, Y. Rao, and B.-H. Wu, Int. Rev. Phys. Chem. 24, 191 (2005). https://doi.org/10.1080/014423505002258949. A. G. Lambert, P. B. Davies, and D. J. Neivandt, Appl. Spectrosc. Rev. 40, 103 (2005). https://doi.org/10.1081/ASR-20003832610. H.-F. Wang, L. Velarde, W. Gan, and L. Fu, Annu. Rev. Phys. Chem. 66, 189 (2015). https://doi.org/10.1146/annurev-physchem-040214-12132211. F. Tang, T. Ohto, S. Sun, J. R. Rouxel, S. Imoto, E. H. G. Backus, S. Mukamel, M. Bonn, and Y. Nagata, Chem. Rev. 120, 3633 (2020). https://doi.org/10.1021/acs.chemrev.9b0051212. J. D. Pickering, M. Bregnhoj, A. S. Chatterley, M. H. Rasmussen, K. Strunge, and T. Weidner, Biointerphases 17, 011201 (2022). https://doi.org/10.1116/6.000140113. M. J. Shultz, C. Schnitzer, D. Simonelli, and S. Baldelli, Int. Rev. Phys. Chem. 19, 123 (2000). https://doi.org/10.1080/014423500229882 First, when fitting the experimentally obtained SFG spectrum according to Eqs. (1)–(3), the three important parameters of a surface vibrational mode, frequency (ωq), bandwidth (Γq), and intensity (Aq), will be obtained. These parameters are fundamental in understanding a surface phenomenon. This is actually the most common application of SFG-VS. Second, as shown in Eq. (4), χIJK,q(2)=NS∑⟨RIiRJjRKk⟩βijk,q(2),(4)the tensor element of χq(2), denoted as χIJK,q(2), is related to the surface number density of the interfacial species, NS, and the microscopic hyperpolarizability tensor element, βijk,q(2), where I, J, K and i, j, k are the laboratory and molecular coordinates, respectively. ⟨RIiRJjRKk⟩ is the Euler transformation between the molecular frame and the laboratory frame. The hyperpolarizability tensor element, βijk,q(2), is related to the IR and Raman properties of the qth vibrational mode through Eq. (5), where ∂αij(1)∂Qq and ∂μk∂Qq are the partial derivatives of the Raman polarizability tensor and the IR transition dipole moment of the qth vibrational mode; and Qq is the normal coordinate, βijk,q(2)∝∂αij(1)∂Qq∂μk∂Qq.(5)The relationship described in Eqs. (4) and (5) indicates that the measured SFG spectrum can be used to deduce the surface orientation as well as the surface number density of a surface species.3,103. C. Hirose, N. Akamatsu, and K. Domen, Appl. Spectrosc. 46, 1051 (1992). https://doi.org/10.1366/000370292412438510. H.-F. Wang, L. Velarde, W. Gan, and L. Fu, Annu. Rev. Phys. Chem. 66, 189 (2015). https://doi.org/10.1146/annurev-physchem-040214-121322 Third, when using proper polarization combinations for the visible, IR, and SFG beams, SFG-VS can be used to obtain the chiral vibrational spectrum of a surface or a surface species.14–1614. J. Wang, X. Y. Chen, M. L. Clarke, and Z. Chen, Proc. Natl. Acad. Sci. U.S.A. 102, 4978 (2005). https://doi.org/10.1073/pnas.050120610215. E. C. Y. Yan, L. Fu, Z. Wang, and W. Liu, Chem. Rev. 114, 8471 (2014). https://doi.org/10.1021/cr400604416. L. M. Haupert and G. J. Simpson, Annu. Rev. Phys. Chem. 60, 345 (2009). https://doi.org/10.1146/annurev.physchem.59.032607.093712 Fourth, when combined with pump-probe technique, SFG-VS can be used to address surface dynamics of a transient surface process.17–1917. E. H. G. Backus, J. D. Cyran, M. Grechko, Y. Nagata, and M. Bonn, J. Phys. Chem. A 122, 2401 (2018). https://doi.org/10.1021/acs.jpca.7b1230318. Y. Rao, Y. Q. Qian, G. H. Deng, A. Kinross, N. J. Turro, and K. B. Eisenthal, J. Chem. Phys. 150, 094709 (2019). https://doi.org/10.1063/1.508022819. S. Nihonyanagi, S. Yamaguchi, and T. Tahara, Chem. Rev. 117, 10665 (2017). https://doi.org/10.1021/acs.chemrev.6b00728Obtaining a SFG spectrum is a challenging task and it was not experimentally realized until the 1980s. In 1984, the Shen group performed their initial attempt to obtain a SFG spectrum from a surface monolayer; and in 1986, they formally reported the success of their work.20,2120. X. D. Zhu, H. Suhr, and Y.-R. Shen, J. Opt. Soc. Am. B 3, 252 (1986).21. X. D. Zhu, H. Suhr, and Y. R. Shen, Phys. Rev. B 35, 3047 (1987). https://doi.org/10.1103/PhysRevB.35.3047 In their pioneering work, the Shen group used the 532 nm visible pulse from a Nd:YAG laser and the tunable IR pulse from a CO2 laser to generate a SFG spectrum from the monolayer of coumarin 504 dye deposited on a fused quartz surface. This milestone work laid the foundation for the instrumentation of SFG-VS and had sparked great interests among scientists on this technique in the following decades. Nowadays, SFG-VS had found many applications in a variety of research fields including chemistry, physics, material sciences, biological sciences, and environmental sciences.15,22–3015. E. C. Y. Yan, L. Fu, Z. Wang, and W. Liu, Chem. Rev. 114, 8471 (2014). https://doi.org/10.1021/cr400604422. Y. R. Shen and V. Ostroverkhov, Chem. Rev. 106, 1140 (2006). https://doi.org/10.1021/cr040377d23. R. D. Wampler, A. J. Moad, C. W. Moad, R. Heiland, and G. J. Simpson, Acc. Chem. Res. 40, 953 (2007). https://doi.org/10.1021/ar600055t24. S. Baldelli, Acc. Chem. Res. 41, 421 (2008). https://doi.org/10.1021/ar700185h25. F. M. Geiger, Annu. Rev. Phys. Chem. 60, 61 (2009). https://doi.org/10.1146/annurev.physchem.59.032607.09365126. C. M. Johnson and S. Baldelli, Chem. Rev. 114, 8416 (2014). https://doi.org/10.1021/cr400490227. C. Z. Li, J. Yang, F. H. Su, J. J. Tan, Y. Luo, and S. J. Ye, Nat. Commun. 11, 5481 (2020). https://doi.org/10.1038/s41467-020-19330-728. S. Hosseinpour, S. J. Roeters, M. Bonn, W. Peukert, S. Woutersen, and T. Weidner, Chem. Rev. 120, 3420 (2020). https://doi.org/10.1021/acs.chemrev.9b0041029. T. Y. Lu, W. Guo, P. M. Datar, Y. Xin, E. N. G. Marsh, and Z. Chen, Chem. Sci. 13, 975 (2022). https://doi.org/10.1039/D1SC04300E30. B. Siritanaratkul, C. Eagle, and A. J. Cowan, Acc. Chem. Res. 55, 955 (2022). https://doi.org/10.1021/acs.accounts.1c00692Despite the nearly four-decade development of SFG-VS, one issue still hinders the wide use of SFG-VS. That is that a SFG-VS system has never been a piece of user-friendly instrument. In fact, SFG-VS systems (either home-built or commercial) are sophisticated instrumentations and require experienced personnel to operate. Understanding and mastering the SFG-VS instrumentation for new comers is, thus, always a very challenging task. Yet, in the literature, tutorial-style articles about SFG-VS instrumentation are limited.31–3531. G. Ma, J. Liu, L. Fu, and E. C. Y. Yan, Appl. Spectrosc. 63, 528 (2009). https://doi.org/10.1366/00037020978834705732. J. D. Pickering, M. Bregnhoj, A. S. Chatterley, M. H. Rasmussen, S. J. Roeters, K. Strunge, and T. Weidner, Biointerphases 17, 011202 (2022). https://doi.org/10.1116/6.000140333. A. P. Carpenter and J. E. Baio, Biointerphases, 17, 031201 (2022). https://doi.org/10.1116/6.000184434. B. Doughty, L. Lin, U. I. Premadasa, and Y.-Z. Ma, Biointerphases 17, 021201 (2022). https://doi.org/10.1116/6.000181735. S. Hosseinpour, Biointerphases 17, 031203 (2022). https://doi.org/10.1116/6.0001851 To help new students and new researchers to the SFG-VS field, we here would like to provide a Tutorial about SFG-VS instrumentation. This Tutorial will discuss the details about the classification, setup layout, construction, operation, and data processing about SFG-VS. We will use the steady state Ti:sapphire based broad bandwidth SFG-VS system as an example for our illustration. SFG-VS systems based on other types of pump lasers are also briefly discussed. We also hope this Tutorial is beneficial for any scientist who is interested in SFG-VS.

II. CLASSIFICATION OF THE SFG-VS SYSTEM

Section:

ChooseTop of pageABSTRACTI. INTRODUCTIONII. CLASSIFICATION OF THE... <<III. SETUP LAYOUT OF THE ...IV. CONSTRUCTON OF A SFG-...V. OPERATION AND DATA PRO...VI. DISCUSSION ON THE PUM...VII. CONCLUSIONSREFERENCESThere is no formal classification for SFG-VS in this field. In our opinion, we classify the SFG-VS systems into two types, scanning system and broad bandwidth system, based on the IR laser source.20,21,32,36–4320. X. D. Zhu, H. Suhr, and Y.-R. Shen, J. Opt. Soc. Am. B 3, 252 (1986).21. X. D. Zhu, H. Suhr, and Y. R. Shen, Phys. Rev. B 35, 3047 (1987). https://doi.org/10.1103/PhysRevB.35.304732. J. D. Pickering, M. Bregnhoj, A. S. Chatterley, M. H. Rasmussen, S. J. Roeters, K. Strunge, and T. Weidner, Biointerphases 17, 011202 (2022). https://doi.org/10.1116/6.000140336. D. K. Hore, J. L. King, F. G. Moore, D. S. Alavi, M. Y. Hamamoto, and G. L. Richmond, Appl. Spectrosc. 58, 1377 (2004). https://doi.org/10.1366/000370204264134437. S. Schrodle and G. L. Richmond, Appl. Spectrosc. 62, 389 (2008). https://doi.org/10.1366/00037020878404679538. E. W. M. Van Der Ham, Q. H. F. Vrehen, and E. R. Eliel, Opt. Lett. 21, 1448 (1996). https://doi.org/10.1364/OL.21.00144839. E. W. M. van der Ham, Q. H. F. Vrehen, and E. R. Eliel, Surf. Sci. 368, 96 (1996). https://doi.org/10.1016/S0039-6028(96)01034-540. L. J. Richter, T. P. Petralli-Mallow, and J. C. Stephenson, Opt. Lett. 23, 1594 (1998). https://doi.org/10.1364/OL.23.00159441. E. L. Hommel, G. Ma, and H. C. Allen, Anal. Sci. 17, 1325 (2001). https://doi.org/10.2116/analsci.17.132542. G. Ma and H. C. Allen, J. Phys. Chem. B 107, 6343 (2003). https://doi.org/10.1021/jp027364t43. L. Velarde, X.-Y. Zhang, Z. Lu, A. G. Joly, Z. Wang, and H.-F. Wang, J. Chem. Phys. 135, 241102 (2011). https://doi.org/10.1063/1.3675629 As we have shown in Fig. 1, a SFG-VS system required two incoming laser beams, one being visible (or near-IR) beam and the other being IR beam. The visible (or near-IR) beam needs to be a narrow spectral bandwidth pulse; and the IR beam is either a narrow spectral bandwidth pulse or a broad spectral bandwidth pulse. In Fig. 2, we show the fundamental difference between a scanning system and a broad bandwidth system. We can see, with a scanning system, the narrow bandwidth IR beam is too narrow and it can only produce one data point in the vibrational spectral region for each acquisition, while the broad bandwidth IR beam is wide enough to be able to cover the whole vibration spectral region and can produce a full SFG spectrum for each acquisition. Accordingly, a SFG-VS system with narrow bandwidth IR pulse has to scan its IR frequency over the whole spectral region in order to produce a full SFG spectrum. This type of SFG-VS system is, thus, called scanning SFG-VS. In contrast, the broad bandwidth system directly leads to the generation of the full SFG spectrum and frequency scanning is not needed.Both the scanning system and the broad bandwidth system have advantages and disadvantages. The broad bandwidth SFG-VS system is more suitable to probe transient surface species and follow surface kinetic and dynamic process. Yet, the spectral coverage of broad bandwidth IR pulse is not very broad and it is not suitable to probe a surface species such as interfacial water as this requires a very wide IR spectral coverage. In this case, a scanning SFG-VS system might be more suitable because it can scan its output frequency over a much wider spectral region. In a scanning SFG-VS system, the IR light source can be a pico-second (ps) or nano-second (ns) laser. These systems are less expensive. The first successful SFG-VS system built by the Shen group was a scanning system.20,2120. X. D. Zhu, H. Suhr, and Y.-R. Shen, J. Opt. Soc. Am. B 3, 252 (1986).21. X. D. Zhu, H. Suhr, and Y. R. Shen, Phys. Rev. B 35, 3047 (1987). https://doi.org/10.1103/PhysRevB.35.3047 In a broad bandwidth SFG-VS system, the IR source needs to be a femto-second (fs) laser. These systems are much more expensive. The broad bandwidth SFG-VS system is also costly to maintain. Yet, despite these drawbacks, nowadays broad bandwidth SFG system can be considered as the main-stream system and new SFG-VS labs prefer choosing a broad bandwidth system.

III. SETUP LAYOUT OF THE SFG-VS SYSTEM

Section:

ChooseTop of pageABSTRACTI. INTRODUCTIONII. CLASSIFICATION OF THE...III. SETUP LAYOUT OF THE ... <<IV. CONSTRUCTON OF A SFG-...V. OPERATION AND DATA PRO...VI. DISCUSSION ON THE PUM...VII. CONCLUSIONSREFERENCESFigure 3 presents the five-component block diagram of the SFG-VS system. This is a universal layout applicable to any SFG-VS system including the scanning system and the broad bandwidth system. In other words, any SFG-VS system must consist of these five parts, namely, light sources, optical control for the incoming visible (or near-IR) and IR beams, sample stage, optical control for the out-coming SFG beam, and detection system. The light sources provide the input visible (or near-IR) and IR beams; the optical control of the incoming beams are used to direct and overlap the visible and IR beam to the sample surface, spatially and temporally, with correct polarizations and with sufficient power density. This part consists of a series of optics including steering mirrors, polarizer, waveplate, and focusing lens. The sample stage is a device to accommodate the sample. It can be just a platform or it can be a sophisticated stage equipped with sample compartment for in situ and real-time monitoring. The optical control for the out-coming SFG beam is used to properly direct the SFG beam into the detection system. This part usually consists of filters, polarizer, waveplate, and focusing lens. The purpose of these optics is to block the reflected visible and IR beams and focus the SFG signal with correct polarization into the detection system. The detection system of a scanning system can be just a CCD or PMT; and the detection system of a broad bandwidth system needs to have a monochromator and a CCD. This part needs to be software-controlled and will produce the final SFG spectrum. In the following, we will present two steady state Ti:sapphire based broad bandwidth SFG-VS setups and discuss the details of each part of the setup. We will also provide a brief selection guide for the optics involved in the construction of the SFG-VS system.In Figs. 4 and 5, two broad bandwidth SFG-VS setups are presented, corresponding to the two subtypes of the broad bandwidth system. Figure 4 shows the single-amplifier system and Fig. 5 shows the double-amplifier system. The classification is based on whether the broad bandwidth SFG-VS system utilizes one regenerative amplifier or two regenerative amplifiers as their master light sources. The layouts of the two systems are based on previously reported systems that we are familiar with.31,4131. G. Ma, J. Liu, L. Fu, and E. C. Y. Yan, Appl. Spectrosc. 63, 528 (2009). https://doi.org/10.1366/00037020978834705741. E. L. Hommel, G. Ma, and H. C. Allen, Anal. Sci. 17, 1325 (2001). https://doi.org/10.2116/analsci.17.1325 It should be pointed out that the two layouts are for illustrative purposes and the presented details may not be exactly the same as the published ones.We first describe the single-amplifier system. In Fig. 4, the single-amplifier broad bandwidth SFG-VS system uses one Ti:sapphire based fs regenerative amplifier as a master pump laser which produces a kHz repetition rate (e.g., at 1 or 5 kHz) and ultrashort (e.g., at ∼100 fs) laser pulse at a near-IR wavelength (e.g., at 800 nm). The operation of the regenerative amplifier requires a seed laser and a pump laser, as shown in Fig. 4. The 800 nm fs pulse from the master light source cannot be directly used for the SFG process. Rather, this 800 nm beam is divided into two beams by a beamsplitter. One beam will pass through a device called pulse shaper. The pulse shaper consists of a grating, a planoconvex cylindrical lens, a slit, and a reflection mirror. It is built with a 4f setup and behaves like a zero-chirp “stretcher.”44–4644. A. N. Bordenyuk and A. V. Benderskii, J. Chem. Phys. 122, 134713 (2005). https://doi.org/10.1063/1.187365245. O. Esenturk and R. A. Walker, J. Chem. Phys. 125, 174701 (2006). https://doi.org/10.1063/1.235685846. Y. Rao, M. Comstock, and K. B. Eisenthal, J. Phys. Chem. B 110, 1727 (2006). https://doi.org/10.1021/jp055340r The beam after the pulse shaper will become a narrow bandwidth ps 800 nm beam and will be used for the 800 nm beam line. Alternatively, an etalon filter can be used to convert fs broad bandwidth 800 nm beam into ps narrow bandwidth 800 nm beam.47,4847. A. B. Voges, H. A. Al-Abadleh, M. J. Musorrafiti, P. A. Bertin, S. T. Nguyen, and F. M. Geiger, J. Phys. Chem. B 108, 18675 (2004). https://doi.org/10.1021/jp046564x48. A. Lagutchev, S. A. Hambir, and D. D. Dlott, J. Phys. Chem. C 111, 13645 (2007). https://doi.org/10.1021/jp075391j Another part of the 800 fs pulse will go into a device called optical parametric amplifier (OPA) to produce a broad bandwidth fs IR beam. A common operation process of an OPA is as follows. The OPA uses a small port of the 800 nm incoming beam to produce a white light continuum through a white light generation crystal (e.g., a sapphire crystal). This OPA then uses steering mirrors and focusing optics to properly combine the generated white light and the remaining 800 nm energy in a second-order nonlinear crystal (e.g., a BBO crystal) to produce two new beams with their wavelengths in the near-IR region. The two new beams are referred as signal and idler beams. The signal and idler beams are then combined into a difference frequency generation (DFG) crystal (e.g., AgGaS2) to produce fs broad bandwidth IR pulse. A Ge filter can be installed after the DFG stage to filter out the unwanted signal and idler beams. This IR beam has a broad bandwidth and will be used in the IR beam line for broad bandwidth SFG-VS.We now come to the optical control system for the two incoming beams. This part of the SFG-VS system consists of a variety of optics aiming to manage the direction, polarization, beam energy, beam size, and optical delay of 800 nm beam and IR beam. The related optical components are presented in Fig. 4. The mirrors are used to steer the beam’s direction to let the beam eventually reach the sample surface. The attenuator in the 800 nm beam line (i.e., the combination of No. 2 waveplate and No. 3 polarizer in Fig. 4) is to give us a good control of the 800 nm beam’s energy. We can use it to change the incoming beam’s energy from minimum to maximum continuously. As we can see, there are additional Glan-laser polarizer (No. 4) and half waveplate (No. 5) in the 800 nm beam line. We use this half waveplate to set the 800 nm polarization to be at s, p, or mixed polarization. Here, s polarization means that the linearly polarized light has its polarization perpendicular to the plane of incidence, while p polarization means that the linearly polarized light has its polarization parallel to the plane of incidence. Newcomers may prefer using the laser table as the reference plane to define a polarization as “horizontal polarization” or “vertical polarization.” Yet, we should always keep in mind that we use the plane of incidence as the reference plane to define polarization in SFG-VS. As for the additional Glan-laser polarizer, its purpose is to keep the polarization of the incoming beams pure. It is like a filter to filter out the unwanted polarization. The SFG-VS is polarization-dependent technique. The SFG-VS system uses linearly s-polarized or p-polarized visible (or near-IR, e.g., 800 nm in the example here) and IR beams to drive the SFG process and captures s-polarized or p-polarized SFG signal. There are actually eight polarization combinations for SFG-VS experiment, namely, ssp (s-polarized SFG, s-polarized visible, and p-polarized IR), sps, pss, ppp, sss, psp, spp, and pps. When we use one polarization combination, we do not want to detect the SFG signal coming from another polarization combination. This means that we need to ensure the “purity” of the polarizations of the visible, IR, and SFG beams. The polarizers in the beam lines are for this purpose. In the part of optical control for incoming beam, we also have a device called optical delay. This is a retro reflector set up on a translational stage. This device is used to control the optical path of the 800 nm beam. Fine tuning the optical delay can change the optical path of the 800 nm and allow a perfect temporal overlap between the 800 nm beam and the IR beam at the sample surface. In other words, the two beams will reach the sample surface exactly at the same time. We also have focusing lens. Both 800 nm and IR beams need to be focused onto the sample surface in order to have enough power density to drive the SFG process. The IR beam line also has half waveplate, polarizer, and focusing lens and mirrors. The sample stage can be just a flat platform. It can be mounted onto a vertical translation stage. This allows us to fine adjust the height of the sample to let the sample surface meet the 800 nm and IR overlapping point. The sample stage can also be used to hold a complicated sampling system for in situ and real-time investigations. After the sample stage, there are filters, polarizer, half waveplate, and focusing lens for SFG beam detection. The filter is an important component in this part. It is used to filter out the reflected 800 nm beam. We can use short pass filter or notch filter to fulfill this task. The polarizer is again to ensure the purity of the wanted polarization. The lens is to be used to properly focus the SFG into the detection system for efficient collection of the SFG beam. The half waveplate can allow us to flip the SFG polarization. In this way, we can always allow the beam with the same polarization to go into the detection system. The detection system has different efficiencies for s- and p-polarized SFG beams. Giving an example, if the detection system is efficient for p-polarized beam, we would use the half waveplate to flip the s-polarized SFG beam into p-polarized SFG beam before we detect it. The detection part of the broad bandwidth SFG-VS system consists of a monochromator and a CCD. The monochromator uses its inside grating to disperse the broad bandwidth SFG beam onto the CCD, thus generating a full SFG spectrum.Figure 5 presented the double-amplifier system. As we can see, the difference between the double-amplifier system and the single-amplifier system lies in the fact that the double-amplifier system has an extra Ti:sapphire based regenerative amplifier which can directly produce ps narrow bandwidth 800 nm beam. Therefore, this type of SFG-VS system no longer needs a band narrowing device such as a pulse shaper or an etalon filter. The double-amplifier system in general has more power in the ps 800 nm beam line than that of single-amplifier system. As for other parts of the double-amplifier system, they are very similar to that of the single-amplifier system. The most challenging issue when setting up a double-amplifier system is to synchronize the two regenerative amplifiers. We should know that inside the regenerative amplifier there are two pockel cells. Intuitively, these two pockel cells behave like two “doors.” The first one is the entrance to let the seed pulse come into the amplifier at a specific time and the second one is the exit to let the seed pule go out of the amplifier after sufficient power amplification through a chirp system. The two pockel cells are controlled by an electronic control box which sends electronic trigger signal for pockel cell activation (or, intuitively, for the opening and closing of the doors). Each regenerative amplifier is controlled by its own electronic control box. If the two electronic boxes are not synchronized, there will be no definite time difference between the two output 800 nm beams when they leave the output ports of the two regenerative amplifiers due to temporal electronic jittering. When this situation occurs, there is no way to set up the optical delay as the optical path difference between ps 800 nm and fs 800 nm varies from time to time. This situation can only be fixed by synchronizing the two electronic control boxes. The Allen lab once proposed a simple yet smart method for synchronization when they built the first double-amplifier broad bandwidth SFG-VS system.4949. E. L. Hommel and H. C. Allen, Anal. Sci. 17, 137 (2001). https://doi.org/10.2116/analsci.17.137 They found they can use one electronic control box to control the pockel cells of the two regenerative amplifiers. This strategy can successfully synchronize the two regenerative amplifiers. Nowadays, the laser company can also provide commercial solution to synchronize the two regenerative amplifiers. We here would like to provide an intuitive example to further interpret the meaning of synchronization in the double-amplifier system. If we consider the ps and fs 800 nm beams as two runners in a racing game, “synchronization” means the two runners always start from the same starting line; “without synchronization” means the two runners start from different starting lines and the distance between the two starting lines also change from time to time.Now we would like to provide some tips for selecting and purchasing of the components involved in Figs. 4 and 5. How to set up a lab space and how to choose a suitable laser table will not be covered in details in this Tutorial. In general, a common laser table installed in a basement of a laboratory building is a suitable choice to give the SFG-VS system good isolation from environmental vibration. A clean room is a suitable choice to make optics have a long lifetime. Temperature and humidity stability is necessary for a stable energy output of the laser systems.1. Light source: As for the broad bandwidth system in Fig. 4, the master light source is a Ti:sapphire based pump laser with a 1 kHz (or 5 kHz) repetition rate and 5050. H. Maekawa, S. K. K. Kumar, S. S. Mukherjee, and N.-H. Ge, J. Phys. Chem. B 125, 9507 (2021). https://doi.org/10.1021/acs.jpcb.1c05430 as for the broad bandwidth system in Fig. 5, the fs light source can have a 1 kHz (or 5 kHz) repetition rate and μJ energy per pulse) and 10 mW IR power at 1 kHz repetition rate (i.e., 10 μJ energy per pulse) at the sample surface are usually sufficient to produce high-quality SFG spectra within a reasonable acquisition time. High input powers from the 800 nm and IR beams are certainly always better than low output powers. If we are afraid of possible sample damage, we can always attenuate the input power. In addition, sufficient power outputs from the regenerative amplifier and OPA also give us the freedom to run other ultrafast spectroscopy experiments when we want to, e.g., two-dimensional IR spectroscopy, second-harmonic generation spectroscopy, and transient absorption spectroscopy.2.

Mirror: for 800 nm beam, a broadband mirror with a dielectric coating is a suitable choice. For IR beam, a gold or silver mirror is a suitable choice. As for the reflectivity, the higher the better. In addition, we usually do not need to choose a mirror with very high damage threshold. This is not because these types of mirrors are not suitable, but because they are often very expensive.

3.

Half waveplate: for visible or 800 nm, there are three different types of half waveplate we can choose, multi-order, zero-order, and achromatic waveplates. Their working wavelength ranges are different, following the order of multi-order, zero-order, and achromatic waveplates. Multi-order waveplate has a very narrow working wavelength coverage, and achromatic waveplate has a wide working wavelength coverage. In our experience, the zero-order waveplate is suitable for the 800 nm beam line and the achromatic waveplate is suitable for the broad bandwidth SFG beam line. As for the IR beam line, Berek’s compensator type of half waveplate made of MgF2 crystal can be used. It should also be pointed out that the IR transmission range of MgF2 waveplate can only cover from 4000 to ∼1250 cm−1. We also need to make sure that the clear aperture of the wavelplate needs to be bigger than the size of the beam that passes it.

4.

Glan-laser polarizer: when choosing the Glan-laser polarizer for 800 nm and SFG beams, we should pay attention on the clear aperture of the polarizer cube. We need to pay attention on the working wavelength range of the antireflection (AR) coating on the surface of the polarizer cube and always make sure the wavelength range of AR coating cover the bandwidth of the beam that passes it. We need to pay attention on the extinction ratio of the polarizer and a 100 000:1 ratio is a good choice.

5.

Wire-grid polarizer: when choosing the wire-grid polarizer for the IR beam, we should pay attention on the clear aperture of the polarizer as well as the working wavelength range of the polarizer.

6.

Filter: there are two types of filters that can be chosen to filter out the unwanted 800 nm beam on the detection part, short pass filter and notch filter. Notch filter can be more efficient than short pass filter yet it is more expensive. To filter out 800 nm beam, a short-pass filter with a cut-off wavelength at 750 nm can be chosen; a notch filter with a central wavelength at 800 nm can be chosen.

7.

Optical delay: we certainly can buy a commercial retro reflector for the optical delay. Yet, inside the commercial one, the mirrors might be fixed. This makes us lose the freedom to adjust these mirrors for optical alignment. A better option is to use two mirrors, two kinematic mirror mounts, and a translational stage (with micrometer control) to build a home-made optical delay. This home-made one gives us the freedom to individually adjust each mirror, thus, easing our optical alignment.

8.

Lens: for 800 nm and SFG beam lines, we would choose BK7 planoconvex lens with proper anti-reflection (AR) coating. The working wavelength range of the AR coating needs to be able to cover the 800 nm and SFG beams. For the IR beam line, the lens needs to be made of IR transparent materials, such CaF2 and ZnSe. CaF2 substrate can work within part of the mid-IR region (from 4000 to 1100 cm−1). ZnSe can work within the whole mid-IR region (from 4000 to 400 cm−1) yet its IR transmittance is lower than that of CaF2.

9.

Monochromator: the resolution of a monochromator is the most important factor that we should consider when selecting it. It depends in the focal length of the monochromator and the groove density of the grating used inside. In addition, when choosing a grating, we need to balance between the groove density and the efficiency of a grating. A grating with more grooves might have a good resolution yet have a low efficiency. In addition, there are two ways to position the monochromator on the laser table: it can be simply put on the laser table horizontally or it can be put onto a tilted platform. In the latter case, the tilt angle platform can be adjusted to match the SFG beam’s incoming direction.

10.