Letter pubs.acs.org/JPCL
Cite This: J. Phys. Chem. Lett. 2017, 8, 6118−6123
Selective Suppression of Stimulated Raman Scattering with Another Competing Stimulated Raman Scattering Doyeon Kim,†,‡,# Dae Sik Choi,§,∥,# Jiwoong Kwon,† Sang-Hee Shim,*,†,‡ Hanju Rhee,*,⊥ and Minhaeng Cho*,†,‡ †
Center for Molecular Spectroscopy and Dynamics, Institute for Basic Science (IBS), Seoul 02841, Republic of Korea Department of Chemistry, Korea University, Seoul 02841, Republic of Korea § Technology Human Resource Support for SMEs Center, Korea Institute of Industrial Technology (KITECH), Cheonan 31056, Republic of Korea ∥ Research Institute, Smart Korea, Daejeon 34141, Republic of Korea ⊥ Seoul Center, Korea Basic Science Institute, Seoul 02841, Republic of Korea ‡
S Supporting Information *
ABSTRACT: A three-beam femtosecond stimulated Raman scattering (SRS) scheme is formulated and demonstrated to simultaneously induce two different SRS processes associated with Raman-active modes in the same molecule. Two SR gains involving a common pump pulse are coupled and compete: As one of the Stokes beam intensities increases, the other SRS is selectively suppressed. We provide theoretical description and experimental evidence that the selective suppression behavior is due to the limited number of pump photons used for both of the two SRS processes when an intense depletion beam induces one SRS process. The maximum suppression efficiency was ∼60% with our experimental setup, where the SR gain of the ring breathing mode of benzene is the target SRS signal, which is allowed to compete with another SRS process, induced by an intense depletion beam, of the CH stretching mode. We anticipate a potential of this new switching-off concept in super-resolution label-free microscopy.
S
for simultaneous detection of a large spectral range.21 Therefore, multiple vibrational modes in one or more molecules can be stimulated simultaneously within the spectral range of FSRS.16,17,20 In general, FSRS is performed in uniform light intensities within the spectral range and the SRS processes are considered as independent.18,20,21 A couple of imaging applications reported simultaneous stimulation of parallel, spectrally distinct SRS processes with multiple light pulses.22,23 Yang et al. used a fiber-based frequency generator to produce one pump and two Stokes pulses for detecting two different molecules simultaneously.22 Lu et al. developed a three-color SRS microscopy with a pulse shaper carving out three narrow bands from a broadband pump pulse.23 PMMA and polystyrene beads were simultaneously imaged with three SRS processes, among which two transitions were resonant with two vibrational modes of polystyrene. The experiment was performed under low light intensities so that the two transitions could be considered as independent processes, and their SRS intensities were linear with the incident beam intensities. However, little is known about what happens when two distant Raman transitions are no longer
pontaneous Raman scattering, induced by the interaction of a molecule with an incident light and a vacuum field, has been widely used in various studies of chemical systems.1 Stimulated Raman scattering (SRS) is a coherent version induced by the interaction of a molecule with two coherent laser fields.2−10 Consequently, SRS signal is proportional to the photon number densities of both the pump and stimulating beams so that the Raman scattering intensity is enhanced by several orders of magnitude.11,12 The SRS emission propagates along the direction of the incident coherent light beams, resulting in spatially confined output photons, thereby increasing the efficiency of detecting Raman signal.10 Furthermore, the number of stimulated Raman-scattered photons is proportional to the number of Raman-active oscillators, which enables a quantitative measurement of a molecular species in various samples.13,14 The amplified SRS signal allows for label-free, chemical-specific microscopy of biological samples.15 Femtosecond stimulated Raman spectroscopy (FSRS) has been applied for real-time structural measurements of chemical reaction dynamics.16−20 The interaction between narrowband picosecond (ps) pump pulse and a broadband femtosecond (fs) probe pulse gives a sharp vibrational signal added to the broad probe spectrum. The bandwidth of the pump pulse determines the frequency resolution, whereas the broadband probe allows © XXXX American Chemical Society
Received: October 25, 2017 Accepted: December 6, 2017 Published: December 6, 2017 6118
DOI: 10.1021/acs.jpclett.7b02752 J. Phys. Chem. Lett. 2017, 8, 6118−6123
Letter
The Journal of Physical Chemistry Letters
Figure 1. (a) Energy-level diagram of two SRS processes with three light pulses of pump, Stokes, and depletion. Frequency differences of pumpStokes and pump-depletion pairs match the frequencies of two Raman-active modes (ν̃, ν̃′). (b) Scheme of the experimental setup in which the pump, Stokes, and depletion pulses are collinearly focused into a sample. The pump beam is modulated by an optical chopper for obtaining SR gain from Stokes spectra. Narrow bandpass filters (NBFs) are used to carve out and shift the pump and depletion bandwidths. (c) Scheme of incident and output pulse spectra with SR gain and loss. (d) Spontaneous Raman spectrum of liquid benzene (blue). The inset displays the depletion beam spectra at resonant (red, peaked at 1026.5 nm) and nonresonant (black, peaked at 1030.5 nm) conditions.
Figure 2. Experimental setup for three-color (pump: 781 nm, Stokes: 850 nm, depletion: 1026.5 nm) SRS spectroscopy. NBF, narrow bandpass filter; DM, dichroic mirror; NF, notch filter; SPF, short pass filter; ND, neutral density filter; CCD, charge-coupled device detector. The tilted NBFs (dotted line) on the optical paths of the pump and depletion beams are used to slightly tune the center wavelengths.
beams are overlapped in time and space to induce two SRS processes (Figures 1 and 2). We used two strong Raman-active modes of benzene: the ring breathing mode (ν̃ = 992 cm−1) and the C−H stretch mode (ν̃′ = 3062 cm−1) (Figure 1d). The pump−Stokes frequency difference is matched to ν̃ (ν̃p − ν̃s = 992 cm−1), whereas the pump−depletion frequency difference is tuned to ν̃′ (ν̃p − ν̃d = 3062 cm−1), leading to two SRS processes (Figure 1a,c). Because both SRS processes require annihilations of the common Raman pump photons, a strong competition arises between them.
independent but coupled. Recently, we presented a detailed theoretical description of three-beam double SRS processes by solving the coupled differential equations for two parallel Stokes gain processes and pump loss in terms of photon number densities (see ref [S1] in the Supporting Information (SI)). It was theoretically shown that the SR gain of one vibration mode can be selectively suppressed by controlling the SRS process of another competing Raman-active mode. In the present work, we carried out three-beam double SRS measurements, where ps pump, fs Stokes, and ps depletion 6119
DOI: 10.1021/acs.jpclett.7b02752 J. Phys. Chem. Lett. 2017, 8, 6118−6123
Letter
The Journal of Physical Chemistry Letters
notably distorted due to cross-phase modulation, thermal lensing effect, and so on.24 The suppression efficiency can be theoretically formulated by using the interaction Hamiltonian and Fermi golden rule (see Supplementary Note I in the SI for more details). Under our experimental conditions [I d (0) > I p (0) > I s (0)], the contribution from ΔIs to the change in the depletion beam intensity can be ignored, and then the approximate rate equations for the increased intensities of the Stokes and depletion beams become ⎛ μp V ⎞ μV dΔIs(z) = GsIs(0)⎜⎜ Ip(0) − d ΔId(z)⎟⎟ ℏωdc dz ⎝ ℏωpc ⎠
(2)
⎛ μp V ⎞ μV dΔId(z) = Gd⎜⎜ Ip(0) − d ΔId(z)⎟⎟(Id(0) + ΔId(z)) dz ℏωdc ⎝ ℏωpc ⎠ (3)
where μ is the refractive index and V is the interaction volume. The auxiliary constants Gs and Gd are related to the Raman gain coefficients (Supplementary Notes I-A and I-B in the SI). By solving the coupled differential equations in eqs 2 and 3, an analytical solution is obtained as ⎛ μp V ⎞ μV Ip(0) + d Id(0)⎟⎟z ΔIs(z) = GsIs(0)⎜⎜ ℏωdc ⎝ ℏωpc ⎠ ⎧ ⎛ μ ω I (0) ⎞ ⎪ Gs p d p Is(0)⎨ln⎜⎜ + + 1⎟⎟ ⎪ ⎝ μd ωpId(0) Gd ⎠ ⎩ μp V μV ⎛ μ ω I (0) ⎞⎫ Ip(0) + d Id(0)]z ⎪ Gd[ p d p c c ℏ ω ℏ ω ⎜ ⎟⎬ p d − ln⎜ +e ⎟⎪ (0) I μ ω p d d ⎝ ⎠⎭
Figure 3. Stimulated Raman gain with depletion pulse. (a) SRG spectra of pure benzene liquid with 0−250 nJ depletion pulse. The pump and Stokes pulse energies are 3 and 0.8 nJ, respectively. (b) Peak intensity of each SRG spectrum in panel a (black squares) plotted with respect to x, the ratio of the incident photon numbers of the depletion beam to those of the pump beam. Red triangles are the depletion efficiencies calculated from the maximum intensities. The fit line of the data using eq 5 and ξ = 0.017 is shown in red. (c) Suppression efficiencies (both experimental values and fitted lines) at two pump energy levels (red: 3 nJ, black: 5 nJ).
(4)
and the depletion efficiency is given as ⎧ 1 ⎛ x + 1 ⎞⎫ η = 1 − ⎨(1 + x) + ln⎜ ξ(1 + x) ⎟⎬ ξ ⎝ xe ⎩ + 1 ⎠⎭
where the dimensionless coordinates ξ and x are defined as ξ ≡ Gdnp(0)z and x ≡ nd(0)/np(0) (np and nd are the numbers of the pump and depletion photons). To compare the theoretical prediction (Figure S2) with experimental results (Figure 3), we calculated the ξ value by using the known gain coefficient (Supplementary Note III in the SI) and compare that with the value obtained by fitting experimental results (Figure 3b) with eq 5. Although a few approximations were used to derive eq 5, our calculated value of 0.07 appears to be in agreement with the experimental value of 0.017. More importantly, the dependence of suppression coefficient on the ratio of the depletion to pump photon numbers is well-described by the theory, eq 5. To further investigate the beam intensity dependence of the efficiency in selectively suppressing the target SRS gain, we increased the pump energy from 3 to 5 nJ and repeated the same measurements (Figure 3c and Figure S3). When the suppression efficiencies with the two pump energy levels were plotted together with respect to the dimensionless x variable (Figure 3c), the curves were consistent with the simulated ones (Figure S2) with eq 5. The difference between the two curves with two pump energy levels becomes significantly reduced when plotted with respect to the depletion beam intensity (Figure S3c). This is again consistent with the theory in which
As varying intensities of the depletion beam acting as a stimulating Stokes beam for C−H stretch SRS, we obtained the SRG spectra of the ring breathing mode (Figure 3). The pump−Stokes SRG spectra were measured while the pump energy per pulse was fixed at 3 nJ, and the depletion pulse energy was increased from 0 to 250 nJ (corresponding to ∼2.1 TW/cm2). Clearly, the SRG intensity of the ring breathing mode is strongly suppressed by the C−H stretch SRS induced by the pump−depletion beam pair. As the depletion beam energy increases, the SR gain decays monotonically (Figure 3b). To quantify how strongly the ring breathing SRG is reduced by the competing C−H stretching SRG, we define the suppression efficiency η as η≡
ΔIs(z , Id(0) = 0) − ΔIs(z , Id(0)) ΔIs(z , Id(0) = 0)
(5)
(1)
where ΔIs is the Stokes Raman gain, Id is the depletion beam intensity, and z is the coordinate along the beam propagation direction. In the case of the depletion beam energy of 250 nJ (∼100 times stronger than the pump energy), the suppression efficiency is ∼60%. Beyond 250 nJ, SRG spectra became 6120
DOI: 10.1021/acs.jpclett.7b02752 J. Phys. Chem. Lett. 2017, 8, 6118−6123
Letter
The Journal of Physical Chemistry Letters
Figure 4. Resonant versus off-resonant conditions. (a,b) Pump pulse spectra transmitted through pure benzene liquid with 0−250 nJ depletion pulse energy at the resonant (a) and off-resonant (b) conditions. (c) Integrated pump beam intensities at the resonant (red) condition in panel a and the off-resonant (black) condition in panel b plotted with respect to the depletion pulse energies. The data in the resonant condition are well-fitted with a single-exponential function with decay constant of 114.5 nJ. (d) SRG suppression efficiencies under the resonant (red) condition in Figure 3b and off-resonant (black) condition.
the 38 cm−1 shift makes the pump−depletion beam pair offresonant to the C−H stretch mode. Because of the offresonance, the amount of pump energy loss becomes much less than the case before shifting the depletion wavelength (Figure 4b,c). When resonant, the maximum pump depletion efficiency was 80% (Figure 4c, red); however, when off-resonant, the maximum suppression efficiency dropped to 11% (Figure 4c, black). The requirement of the resonant condition (ν̃C−H stretch = ν̃p − ν̃d) for an efficient reduction of the pump intensity indicates that the pump depletion is caused by the stimulated Raman loss process involving C−H stretch mode excitations. We also recorded the SRG spectra of the ring breathing mode under the off-resonant condition. The maximum SRG suppression efficiency was 9% (Figure 4d, black), which was much smaller than the maximum suppression efficiency of 57% when the pump-depletion beam pair was resonant with the C− H stretch mode. These results confirm that the 992 cm−1 SRS signal is efficiently suppressed only when the pump−depletion beam pair is resonant with the C−H stretch mode of benzene. Our work provides a new concept of three-color doubleresonance femtosecond SRS spectroscopy that is capable of effectively suppressing the SRS signal of interest by means of another SRS process. We experimentally demonstrated the scheme and described the results with theoretical expressions obtained from the coupled differential equations for the Stokes and depletion beam intensities. By carrying out the double SRS measurements with tunable depletion beam frequency, we elucidated the underlying mechanism in selective suppression of SR gain signal, which is associated with the competition between two different but coupled SRS processes by the same pool of pump photons. This selective suppression of a chosen SRS signal by another SRS channel is analogous to a pair of parallel and competing chemical reactions involving one common reactant.
the suppression efficiency is a function of x (= nd(0)/np(0)) and ξ (= G dn p(0)z). We also carried out the same measurements with the Stokes energy of 1.0 J, which is the maximal level achievable with our setup (Figure S4). The fitted ξ value is 0.016, which is very close to that (0.017) obtained from the data with 0.8 nJ Stokes pulse (Figure 3b), which is also consistent with eq 5 that indicates no dependence of suppression efficiency on the Stokes beam intensity. To elucidate the mechanism of the suppression behavior, we measured the pump spectra with increasing depletion beam energy (Figure 4). The maximum intensity decays exponentially as the energy of depletion beam increases (Figure 4c, red). To theoretically describe the pump intensity loss, we used the rate equations in eqs 2 and 3 and found that the pump intensity change due to the double SRS process is given as Ip(z) = Ip(0) + ΔIp(z) ⎡ ⎛ μ VG I (0) μp VGI̅ p(0) ⎞ ⎤ μ VGdId(0) s s ⎟z ⎥ = Ip(0) exp⎢ −⎜⎜ s + d + ⎢⎣ ⎝ ℏωsc ℏωdc ℏωpc ⎟⎠ ⎥⎦
(6)
In eq 6, the pump intensity decreases exponentially with respect to the depletion intensity with the decay constant of μdVGd/ ℏωdc. The Raman loss of the pump beam (Figure 4c, red) decreases much faster with respect to the depletion beam intensity than the Stokes gain signal intensity does (Figure 3b). This can be explained by eq 4, leading to the approximate linear dependence of the Stokes gain signal on the depletion beam intensity (Supplementary Note I in the SI). To validate the selectivity of this suppression phenomenon, we deliberately changed the wavelength of the depletion beam by tilting the corresponding bandpass filter by 3°, resulting in the center wavelength shifted by ∼4 nm (Figure 1d, inset). Regarding the bandwidth of the C−H stretch mode (∼9 cm−1), 6121
DOI: 10.1021/acs.jpclett.7b02752 J. Phys. Chem. Lett. 2017, 8, 6118−6123
Letter
The Journal of Physical Chemistry Letters
Stokes spectra with and without the pump pulse, which are denoted as Ipump−on(ν̃) and Ipump−off(ν̃), respectively, were obtained to calculate the SRS gain spectrum, ΔIs(ν̃), as ΔIs(ν̃) = Ipump−on(ν̃)/Ipump−off(ν̃).
Our three-beam double SRS scheme may pave a way for a super-resolution label-free Raman microscopy. A doughnutshaped depletion beam for selective suppression of the target SRS signal in the periphery of the focal spot of the pump and Stokes beams may be used to break the diffraction limit of SRS microscopy.25,26 As numerically shown in the paper by one of the authors (see ref [S1] in Supporting Information), similar to the stimulated emission depletion (STED) fluorescence microscopy,27−30 the three-beam double SRS technique can be a novel approach to super-resolution vibrational microscopy. According to the numerical simulation, the width of effective point spread function (PSF) can be arbitrarily narrowed by increasing either the intensity ratio Id/Is or the Raman gain coefficient ratio Gd/Gs. The current experimental scheme for benzene can be applied to label-free super-resolution imaging of proteins containing phenylalanine residues in cells and tissues. The application is not limited to molecules with phenyl groups: Any molecules with one strong (Gd) and another relatively weak (Gs) Raman-active mode can be potential imaging targets for label-free super-resolution microscopy. For instance, CH stretch mode has been used for label-free SRS microscopy of lipids, proteins, and nucleic acids in biological samples.31 When the CH stretch of lipids, protein, and nucleic acids is used for depletion, we expect similar depletion behavior as that of the benzene CH stretch because the depletion efficiency in eq 5 depends on Gd of the CH stretch mode. Recently, Silva et al. combined two-beam SRS microscopy with so-called decoherence beam, whose beam profile is doughnut-shaped, demonstrating depletion of SRS signal and enhancement of the spatial resolution by a factor of ∼2.32 The depletion mechanism was speculated as CARS, whose strong nonresonant background might have significantly distorted the PSF. In contrast, our scheme is fully based on SRS, for both imaging and depletion, which have little nonresonant background. Moreover, two Raman modes can be selected deliberately in a way that the competing SRS process has very large Raman gain coefficient so that the suppression of imaging SRS signal is achievable with moderately strong depletion beam. Currently, a three-beam double SRS imaging setup is under construction and will demonstrate the experimental feasibility in the near future.
■
ASSOCIATED CONTENT
S Supporting Information *
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jpclett.7b02752. Theoretical descriptions of three-beam SRS processes and stimulated Raman loss of the pump, experimental methods, and supporting figures. (PDF)
■
AUTHOR INFORMATION
Corresponding Authors
*E-mail:
[email protected]. Tel: +82-2-3290-4747. (S.H.S.) *E-mail:
[email protected] (H.R.). *E-mail:
[email protected] (M.C.). ORCID
Minhaeng Cho: 0000-0003-1618-1056 Author Contributions #
D. K. and D.S.C. contributed equally to this work.
Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS This work was supported by Samsung Science and Technology Foundation SSTF-BA1501-08 for S.-H.S and the Institute for Basic Science IBS-R023-D1 for M.C. and PB2017044 for H.R. We thank Yugyeong Kim for help on performing the experiment. All SRS measurements were performed in the femtosecond Multidimensional Laser Spectroscopic System (FMLS) at the Korea Basic Science Institute.
■
REFERENCES
(1) Matousek, P. Deep non-invasive Raman spectroscopy of living tissue and powders. Chem. Soc. Rev. 2007, 36, 1292−1304. (2) Hellwarth, R. W. Theory of stimulated Raman scattering. Phys. Rev. 1963, 130, 1850. (3) Shen, Y. R.; Bloembergen, N. Theory of Stimulated Brillouin and Raman Scattering. Phys. Rev. 1965, 137, A1787−A1805. (4) Owyoung, A.; Jones, E. D. Stimulated Raman spectroscopy using low-power cw lasers. Opt. Lett. 1977, 1, 152−154. (5) Laubereau, A.; Kaiser, W. Vibrational dynamics of liquids and solids investigated by picosecond light pulses. Rev. Mod. Phys. 1978, 50, 607−665. (6) Penzkofer, A.; Laubereau, A.; Kaiser, W. High intensity Raman interactions. Prog. Quantum Electron. 1979, 6, 55−140. (7) Butcher, P. N.; Cotter, D. The Elements of Nonlinear Optics; Cambridge University Press, 1991. (8) Bloembergen, N. Nonlinear Optics; World Scientific Press, 1996. (9) Mukamel, S. Principles of Nonlinear Optical Spectroscopy; Oxford University Press, 1999. (10) Boyd, R. W. Nonlinear Optics, 3rd ed.; Academic Press, 2008. (11) Bloembergen, N. The Stimulated Raman Effect. Am. J. Phys. 1967, 35, 989−1023. (12) Levine, B.; Bethea, C. Ultrahigh sensitivity stimulated Raman gain spectroscopy. IEEE J. Quantum Electron. 1980, 16, 85−89. (13) Freudiger, C. W.; Min, W.; Saar, B. G.; Lu, S.; Holtom, G. R.; He, C.; Tsai, J. C.; Kang, J. X.; Xie, S. X. Label-Free Biomedical Imaging with High Sensitivity by Stimulated Raman Scattering Microscopy. Science 2008, 322, 1857−1861.
■
EXPERIMENTAL METHODS Our three-color double SRS setup is fully described in Supplementary Note IV in the SI and Figure 2. In brief, a femtosecond regenerative amplifier (PHAROS; Light Conversion) and two independently tunable optical parametric amplifiers (OPAs) pumped by the PHAROS were used as light sources for pump, Stokes, and depletion beams, respectively. Using a set of narrow bandpass filters for each beam, we obtained ∼1 ps pump (center wavelength: λ = 781 nm, bandwidth: Δν̃ = 16 cm−1), >100 fs Stokes (λ = 850 nm, Δν̃ = 250 cm−1), and ∼1 ps depletion (λ = 1026.5 nm, Δν̃ = 14 cm−1) beams. Neutral density filters varied the energy level of the depletion beam from 0 to 250 nJ. The three beams were collinearly focused at the sample by an objective lens. The transmitted beams were collected by another objective and sent to the monochromator (SpectraPro 2300i; Princeton Instruments) equipped with a CCD (DU401-BV; Andor). The pump beam was modulated with an optical chopper synchronized to the half (25 Hz) of the laser repetition rate (50 Hz). The Stokes spectra were recorded with the CCD spectrometer, which is externally synchronized with the laser system. Two 6122
DOI: 10.1021/acs.jpclett.7b02752 J. Phys. Chem. Lett. 2017, 8, 6118−6123
Letter
The Journal of Physical Chemistry Letters (14) Cheng, J. X.; Xie, X. S. Coherent Raman Scattering Microscopy; CRC Press, 2016. (15) Cheng, X. J.; Xie, X. S. Vibrational spectroscopic imaging of living systems: An emerging platform for biology and medicine. Science 2015, 350, 8870. (16) Kukura, P.; McCamant, D. W.; Mathies, R. A. Femtosecond Stimulated Raman Spectroscopy. Annu. Rev. Phys. Chem. 2007, 58, 461−488. (17) Yoshizawa, M.; Kurosawa, M. Femtosecond time-resolved Raman spectroscopy using stimulated Raman scattering. Phys. Rev. A: At., Mol., Opt. Phys. 1999, 61, 013808. (18) Kukura, P.; McCamant, D. W.; Yoon, S.; Wandschneider, D. B.; Mathies, R. A. Structural observation of the primary isomerization in vision with femtosecond-stimulated Raman. Science (Washington, DC, U. S.) 2005, 310, 1006−1009. (19) Yoon, S.; Kukura, P.; Stuart, C. M.; Mathies, R. A. Direct observation of the ultrafast intersystem crossing in tris (2, 2 -bipyridine) ruthenium (II) using femtosecond stimulated Raman spectroscopy. Mol. Phys. 2006, 104, 1275−1282. (20) Frontiera, R. R.; Mathies, R. A. Femtosecond stimulated Raman spectroscopy. Laser & Photonics Reviews 2011, 5, 102−113. (21) McCamant, D. W.; Kukura, P.; Yoon, S.; Mathies, R. A. Femtosecond broadband stimulated Raman spectroscopy: Apparatus and methods. Rev. Sci. Instrum. 2004, 75, 4971−4980. (22) Yang, W.; Li, A.; Suo, Y.; Lu, F.-K.; Xie, X. Simultaneous twocolor stimulated Raman scattering microscopy by adding a fiber amplifier to a 2 ps OPO-based SRS microscope. Opt. Lett. 2017, 42, 523−526. (23) Lu, F.-K.; Ji, M.; Fu, D.; Ni, X.; Freudiger, C. W.; Holtom, G.; Xie, S. X. Multicolor stimulated Raman scattering microscopy. Mol. Phys. 2012, 110, 1927−1932. (24) Zhang, D.; Slipchenko, M. N.; Leaird, D. E.; Weiner, A. M.; Cheng, J.-X. Spectrally modulated stimulated Raman scattering imaging with an angle-to-wavelength pulse shaper. Opt. Express 2013, 21, 13864−13874. (25) Gong, L.; Wang, H. Breaking the diffraction limit by saturation in stimulated-Raman-scattering microscopy: A theoretical study. Phys. Rev. A: At., Mol., Opt. Phys. 2014, 90, 013818. (26) Liu, W.; Niu, H. Diffraction barrier breakthrough in coherent anti-Stokes Raman scattering microscopy by additional probe-beaminduced phonon depletion. Phys. Rev. A: At., Mol., Opt. Phys. 2011, 83, 023830. (27) Hell, S. W. Strategy for far-field optical imaging and writing without diffraction limit. Phys. Lett. A 2004, 326, 140−145. (28) Hell, S. W.; Dyba, M.; Jakobs, S. Concepts for nanoscale resolution in fluorescence microscopy. Curr. Opin. Neurobiol. 2004, 14, 599−609. (29) Hell, S. W.; Wichmann, J. Breaking the diffraction resolution limit by stimulated emission: stimulated-emission-depletion fluorescence microscopy. Opt. Lett. 1994, 19, 780−782. (30) Harke, B.; Keller, J.; Ullal, C. K.; Westphal, V.; Schönle, A.; Hell, S. W. Resolution scaling in STED microscopy. Opt. Express 2008, 16, 4154−4162. (31) Lu, F. K.; Basu, S.; Igras, V.; Hoang, M. P.; Ji, M.; Fu, D.; Holtom, G. R.; Neel, V. A.; Freudiger, C. W.; Fisher, D. E.; Xie, X. S. Label-free DNA imaging in vivo with stimulated Raman scattering microscopy. Proc. Natl. Acad. Sci. U. S. A. 2015, 112, 11624−11629. (32) Silva, W. R.; Graefe, C. T.; Frontiera, R. R. Toward Label-Free Super-Resolution Microscopy. ACS Photonics 2016, 3, 79−86.
6123
DOI: 10.1021/acs.jpclett.7b02752 J. Phys. Chem. Lett. 2017, 8, 6118−6123
