Open Access
Add to BibSonomyAbstract
Accessing ultrafast photoinduced molecular dynamics on a femtosecond time-scale with vibrational selectivity and at the same time sub-diffraction limited spatial resolution would help to gain important information about ultrafast processes in nanostructures. While nonlinear Raman techniques have been used to obtain highly resolved images in combination with near-field microscopy, the use of femtosecond laser pulses in electronic resonance still constitutes a big challenge. Here, we present our first results on coherent anti-Stokes Raman scattering (fs-CARS) with femtosecond laser pulses detected in the near-field using scanning near-field optical microscopy (SNOM). We demonstrate that highly spatially resolved images can be obtained from poly(3-hexylthiophene) (P3HT) nano-structures where the fs-CARS process was in resonance with the P3HT absorption and with characteristic P3HT vibrational modes without destruction of the samples. Sub-diffraction limited lateral resolution is achieved. Especially the height resolution clearly surpasses that obtained with standard microCARS. These results will be the basis for future investigations of mode-selective dynamics in the near field.
© 2013 Optical Society of America
1. Introduction
The investigation of the influence of the nano-structuring of materials on the ultrafast photo-induced dynamics is of considerable interest. An important example is the charge carrier formation dynamics in organic solar cells [1]. Here, bulk heterojunctions of donor and acceptor molecules, such as poly(3-hexylthiophene) (P3HT) as p–type and [6,]-phenyl C61 butyric acid methyl ester (PCBM) as n–type semiconductor, play a big role [2]. The surface morphology of this system has been investigated by Kilmov et al.[3] using scanning near-field optical microscopy (SNOM) Raman measurements.
In order to investigate the structure dependence of the exciton dynamics, a combination of an initial femtosecond pump excitation followed by a time-delayed femtosecond coherent anti-Stokes Raman scattering (fs-CARS) event would present a powerful tool. Tuning the laser wavelengths to an electronic transition resonance helps to select a specific electronic state and making use of the Raman resonance will yield mode-specific dynamics, which is strongly influenced by structural properties of the system and its environment [4, 5]. For nano-systems, a high spatial resolution is very important. CARS microscopy has emerged as a well-known nonlinear microscopy technique for imaging chemical and biological samples. The non-invasive, chemical specificity inherent to the contrast originating from vibrations of the nuclei and high signal strength due to the coherent nonlinear interaction make this technique special compared to other existing microscope techniques. Duncan et al.[6] first constructed a CARS microscope by employing a non-collinear geometry. The molecular specificity inherent to this technique was demonstrated by differentiating in a CARS image a two immiscible liquid mixture (octane and acetonitrile) contained in an optical cell. The images were recorded by scanning the interface between the two liquids. In the CARS process, the frequency difference between the pump pulse (ωp) and Stokes pulse (ωS) is tuned to be in resonance with a vibrational (Raman) mode of the molecule (ωp − ωS = ΩR). The coherent excitation of the vibrations makes the third probe pulse (ωpr) to scatter at the anti Stokes frequency (ωaS) such that, ωaS = ωp − ωS +ωpr, following the conservation of energy. In a degenerate pump probe CARS, the pump itself acts as the probe (ωp = ωpr), giving the CARS signal at ωaS = 2ωp − ωS. Zumbusch et al.[7] improved the CARS microscope by employing collinear geometry and tight focusing of the laser beams using high NA objectives. This could improve the spatial resolution compared to the non-collinear configuration by (Duncan et al.). The tight focusing helps to reduce the lack of signal generation due to phase mismatch. The main problem affecting the contrast of the CARS images is the unwanted non-resonant background signal from e.g. solvent molecules or the non-vibrational contributions from the analyte itself. A lot of developments have been made in terms of improving image contrast (suppressing the non resonant background) and faster imaging [8]. The modern CARS microscope can record images in vivo at video rate speed [9]. A possible way to minimize the non-resonant background is the use of a time delay between the laser pulses in a femtosecond CARS experiment due to the short life time of the non-resonant contributions [10].
The spatial resolution of an optical microscope is limited by the diffraction limit (0.61 × λ/NA). SNOM techniques [11, 12] can be combined with microscopy to improve the spatial resolution for spectroscopy [13, 14]. The combination of SNOM for recording signals induced by ultrafast laser pulses yields better image resolution and allows for local probing of ultrafast dynamics confined to a nanometer scale. Hess et al.[15] have shown that spatial averaging of spectral information is reduced by local detection using SNOM. The SNOM techniques can be categorized into two subgroups, (i) one using tips with an aperture and (ii) one using aperture-less tips. In aperture SNOM, a tip with an aperture diameter smaller than the excitation wavelength(s) is employed to collect the signal or as a point source for excitation in the near field [16]. The latter is not applicable with femtosecond pulses because of the chirp-induced temporal and intensity-induced spectral broadening effects. Aperture SNOM has been widely used for high resolution imaging applications that are based on the collection of fluorescence and photoluminescence. Recently, we have demonstrated that a femtosecond pump-probe experiment (transient absorption) in combination with the SNOM technique can be applied for chemical specific imaging along with probing the local dynamics [17]. In the apertureless type, a metallic nano particle attached to the tip is used to enhance the signal produced in the near field. Tip-enhanced Raman is a well known technique and applied in imaging and single molecule detection schemes are being developed in this field [18, 19]. Publications presenting examples for a combination of CARS and SNOM techniques are very limited. Schaller et al.[20] combined aperture SNOM techniques to improve the spatial resolution of CARS images. In this experiment the phase-matching condition, which is already relaxed when using high NA objectives, is not playing a role anymore due to the near-field detection under which interferences do not occur. Kawata and co-workers [21, 22] employed tips with metallic nanoparticle to enhance the CARS signal and could obtain CARS images with a high spatial resolution. However, these implementations of tip-enhanced CARS in the near field were restricted to the use of long laser pulses (picoseconds) with excitation far from electronic resonances.
In order to allow for high temporal resolution femtosecond laser pulses have to be used. When electronic state specificity is required (as in most time-resolved experiments), resonance with an absorption or transient absorption transition is mandatory. Both ultrashort pulses and excitation resonant with molecular absorptions is making CARS microscopy and even more near-field CARS microscopy a very demanding task, since samples can be easily destroyed. Here, we demonstrate that fs–CARS with both electronic and vibrational resonance is capable of yielding highly resolve images of P3HT in different environments. This is the first step towards experiments (like pump–CARS) accessing ultrafast mode-specific dynamics of nano-systems. It has to be pointed out, that the goal of our work was not to improve existing CARS imaging techniques, but rather to make an important step towards a time-resolved application in the near field. Better spatial resolution could be obtained under tight-focusing conditions using lasers in the near infrared spectral region and with picosecond duration along with reduced damage of the sample. However, this would not meet our requirements.
2. Experimental
The scheme of the experimental setup is given in Fig. 1. The 150 fs (775 nm, 1 kHz repetition rate, 1 mJ energy/pulse) pulses from the regeneratively amplified Ti:Sapphire laser (CPA 2010, Clark MXR) are used to pump two optical parametric amplifiers (TOPAS, Light Conversion). One of the OPAs serves as source for the pump pulses and the other one yields the Stokes pulse for the CARS experiment. Pump and Stokes pulses are compressed to ≈ 100 fs using prism pair set-ups. The timing of the individual pulses was controlled using computer-controlled delay stages in a Michelson interferometer like set-up. Moreover, the pulses were aligned collinearly and coupled into an inverted microscope (Olympus) equipped with a commercial scanning probe microscope (SPM) system (Nanonics Multiview 2000). The sample was placed on a piezo-controlled XYZ translator. The pulses were focused on to the sample with an objective lens (Olympus PLAN N 10×, NA = 0.25). The far-field CARS signal was collected by a similar objective in the forward direction. After filtering out the pump and Stokes frequencies using a short-pass edge filter (Semrock, SP01-633RS), the signal was detected by an avalanche photodiode (STM1DAPD10, Amplification Technologies, Inc). Noise and background signals were reduced using a boxcar amplifier. The sample was scanned and simultaneously the intensity was recorded at different (x,y) positions. The near-field CARS signal was collected by a SNOM tip attached to a tuning fork (Nanonics design). The signal was guided through a multi mode fiber to the detector and recorded by the avalanche photo-diode after filtering out the pump and Stokes pulse using the short pass filter. Commercially available (Nanonics) cantilevered optical fiber SNOM tips coated with thin films of Cr and Au with an aperture diameter of less then approx. 300 nm were used in the experiments. The height of the tip above the sample was kept constant using a phase feed-back mechanism [23]. Thus, the fiber could also be used to obtain AFM topographies from the sample with a spatial resolution limited by the fiber tip diameter. 128 × 128 pixels images were recorded with 50–100 ms integration time per pixel.
Fig. 1 Experimental scheme of the SNOM-CARS and microCARS experiments. The lasers are focused from the bottom and the transmitted signal is collected in the forward direction with an objective (far-field microCARS) or SNOM tip (near-field SNOM-CARS).
For the experiments, small structures of P3HT were prepared. Solutions in chlorobenzene contained P3HT (regio regularity greater than 98.5%, from Rieke Metals Inc.) at a concentration of 16 mg/ml. These solutions were stirred for 3–4 hours at an elevated temperature of about 60°C. About 25 mm × 25 mm microscopic slides were cut and cleaned thoroughly (first with acetone and then with isopropanol) and dried with a nitrogen gun. A solution of 500 μl was used for spin-coating on the cleaned glass substrate at 2000 rotation per minute (rpm) for 30 seconds. This resulted in an average film thickness of ≈ 100 nm. Afterwards, thermal annealing was performed at 100°C for 1 minute. The thus obtained surface had a considerable roughness and inhomogeneity in crystallinity, which is also known from the former Raman studies [3].
3. Results and discussion
The AFM topography of the P3HT thin film was calculated using WSXM software developed by Horcas et al.[24]; it is shown in Fig. 2(c). In order to guarantee that the AFM topography matches the SNOM image as precisely as possible, the images were recorded using the SNOM tip having an aperture of ≈ 200 – 300 nm and a metal coating of 0.2 μm thickness. Therefore, the lateral resolution of the AFM images is poorer than that expected from measurements with normal AFM tips. The resolution is also less than that of the SNOM-CARS images where only the aperture diameter is relevant, resulting in an (x,y) precision of slightly less than 300 nm (taken from the full width at half maximum of the smallest resolved features in the cross sections taken from the image data). As already mentioned above, the spin coating and annealing process does not offer a uniform homogeneous morphology of the thin film surface. The height of the “hump” on the right side (in the x direction at ≈ 1.5 to 3 μm) is ≈ 145 nm. The CARS images were recorded with the difference between pump and Stokes laser wavenumbers set to approx. 1440 cm−1; this is in resonance mainly with the C=C ring stretching mode of P3HT [25, 26]. Considering the broad bandwidth of the pulses (≈ 100 cm−1) also some neighboring, less intense modes will be excited simultaneously. The pump laser wavelength was set to 650 nm and the Stokes laser was tuned to 717 nm. Thus, the pump laser was resonant with the ground state absorption of the P3HT thin film [27]. The electronically resonant CARS process helps to reduce the non-resonant background in the CARS images, which can be further improved by introducing a time delay between the exciting laser pulse in the order of the pulse length. As is obvious from our CARS images, the non-resonant background signal from the P3HT environment is having a negligible intensity compared to the P3HT signal. Tuning the pump–Stokes wavenumber difference away from the resonance with intense Raman modes clearly reduced the signal just leaving the non-resonant background of the analyte, which is of the same magnitude than the substrate background signal. This demonstrates that next to the chemical specificity due to the electronic resonance also a mode-specific CARS signal was detected in our experiments. The major limiting factor of the aperture SNOM technique is the low signal throughput with reduced aperture sizes. A lot of effort was put into optimizing the SNOM-CARS setup. Finally, the signal strength was high enough to be detected by our filter-detector assembly attached to the SNOM fiber probes without destruction of the sample. A total laser pulse energy of only ≈ 0.75 nJ was focused onto the sample (each pulse ≈ 0.25 nJ). The destruction of the sample was one of the major reasons why we could not use objectives with high numerical apertures. Additionally, the use of the 10x objective also helped to excite the sample layer more homogeneously over its full thickness. The sample was stable during the course of the measurement, but only relatively small pulse energies already chemically changed the P3HT sample as was also observed by others [28]. We have observed that when the focusing of the femtosecond pulses by the microscope objective alone (“microCARS”) is close below the destruction level, the approach of the SNOM fiber tip is resulting in highly localized damages to the sample. This is due to the tip-enhancement effect [21, 22], which also occurs with our SNOM probes due to their metal coating. In the present case SNOM-CARS resulted in an instantaneous damage of the P3HT samples at energies > 1 nJ. The far-field and the near-field CARS images obtained from the same surface area displayed in Fig. 2(c) are shown in panels (a) and (b) of Fig. 2, respectively. Please note that the intensities were normalized, but the color code used was chosen to cover the full intensity range seen in the respective images. The SNOM-CARS images are correlated well with the AFM topography. The lateral resolution is very high in the SNOM-CARS images and is – as was already pointed out above – even better than that of the corresponding AFM image where the influence of the metal film coating has to be taken into account. The focal spot diameter (full width at half maximum, FWHM) of the laser pulses was determined to be ≈ 2 μm. The microCARS image (far field) of the same area is poorly resolved, which is also due to the use of an objective with relatively small numerical aperture (NA). In panels (d) of Fig. 2 and 3, the intensity profiles along a horizontal section (shown in the CARS images) are shown for both cases. The variation in intensity with the thickness well resolved in the near-field profile. As mentioned above, an interesting finding is that a local enhancement of the laser and CARS signal fields due to the thin coating of the gold film occurs. Unfortunately, it is rather difficult to quantify this in our experiment. The pump pulse frequency is close to the plasmon absorption of the gold nano structures [14]. Cheng et al.[29] have performed FDTD simulations on the near-field effects on CARS imaging and have shown that, the near-field enhancement varies with different parameters such as refractive index, edge effect and surface roughness, etc. The evanescent field is confined to the vicinity of the particles (or here coating of the tapered fiber). This helps to improve the sensitivity and resolution of near-field CARS images.
Fig. 2 MicroCARS image (a), SNOM-CARS image (b), AFM topography (c) and CARS intensity profile (d), along a section parallel to the x direction of the images (3 × 3 μm2 area) are displayed. The color code has been adapted to the full intensity range of the image. The structure seen on the right side of the AFM image has a thickness of ≈ 145 nm.
Fig. 3 MicroCARS image (a), SNOM-CARS image (b), AFM topography (c) and CARS intensity profile (d), along a section parallel to the x direction of the images (20 × 20 μm2 area) are displayed. The color code has been adapted to the full intensity range of the image. The roughness feature on the right side of the AFM has a thickness of ≈ 1.5 μm.
The variation in intensities near the edge of the nano-structure is due to the thickness variation. The thicker the P3HT structure, the more oscillators are included in the probe volume giving rise to the CARS signal. In order to compare the variation of CARS intensity with respect to thickness, we found a region in another sample where the height of the P3HT structure is ≈ 1.5 μm. Figure 3, shows the near-field image (b) and the corresponding AFM topography (c) of this structure. The area away from the thick region corresponds to glass (SiO2), which neither contributes to the resonant CARS nor gives rise to a big non-resonant background signal, which also demonstrates our chemical selectivity and good suppression of nonresonant background signals. It can be seen that both far-field (a) and the near-field image (b) correlate with the AFM topography, with the SNOM-CARS image having a better lateral resolution. As the structure is rather thick, relative intensity changes are seen in both far-field and near-field images. When the thickness is down to a nanometer scale like in the first sample discussed above (compare Fig. 2), the far-field field intensity is almost flat and SNOM-CARS can differentiate the topographic feature much better than the microCARS. Since the SNOM probe only collects light in the near field, the SNOM–CARS does not only provide a very good lateral resolution, but clearly surpasses microCARS (also when high NA objectives are used where the lateral resolution becomes rather good). Using the axial resolution formula 1.5 × λ × n/NA2[30], the maximum resolution is 742 nm for a wavelength of λ = 650 nm and a refractive index of n = 1.6 [31] for high NA (1.4) objectives (even much more for the objective with NA = 0.25 used in our work), and the SNOM CARS data clearly surpass this value.
The cross correlation between the pump and Stokes pulses are shown in Fig. 4. The full width at half maximum (≈ 150 fs) is equal for the near-field and the far-field experiments, showing that the temporal resolution (instrument response function) is not influenced by the SNOM tip. This is not self-evident and has to do with the fact that the pulses used in our experiments are not showing a considerable spatial chirp, which otherwise would result in different contributions at each point of the focal area.
Fig. 4 In order to estimate the temporal resolution of the CARS interaction, the cross correlation between the femtosecond pump and Stokes pulses has been measured using the microCARS and the SNOM-CARS setup. The full width at half maximum (≈ 150 fs) of the near-field and the far-field cross correlation traces reflects the respective instrument response functions, which are equal for microCARS and SNOM-CARS.
4. Conclusion
In conclusion, we have demonstrated that a femtosecond time-resolved CARS experiment is feasible with sub-diffraction limited spatial precision in both vibrational and electronic resonance without destruction of the investigated sample. Nano-structures of poly(3-hexylthiophene) (P3HT) have been imaged with molecular specificity due to electronic resonance with the transition from the P3HT ground state to its excitonic S1 state as well as vibrational resonance with the characteristic C=C ring-stretching mode of this molecule. Besides a sub-diffraction limited lateral resolution an extremely good height resolution results from the near-field optical probing. The fs-SNOM-CARS images are compared to fs-microCARS data obtained in the far field as well as the AFM topography, which is automatically obtained during the SNOM-CARS scan. The ultrashort CARS interaction will facilitate time-resolved experiments where the CARS process probes the dynamics with vibrational mode selectivity. The high temporal resolution allows for the suppression of non-resonant background signal. For future investigations we have started to use fs-SNOM-CARS in a pump-CARS experiment where exciton relaxation dynamics are probed in the near field.
Acknowledgments
Financial support by the German Research Foundation DFG ( MA 1564/17-1) is gratefully acknowledged. The authors thank Prof. Veit Wagner and Dr. Torsten Balster for access to their laboratories and help with the sample preparation.
References and links
1. P. G. Nicholson and F. A. Castro, “Organic photovoltaics: principles and techniques for nanometre scale characterization,” Nanotechnol. 21, 492001 (2010). [CrossRef]
2. K. Yonezawa, M. Ito, H. Kamioka, T. Yasuda, L. Han, and Y. Moritomo, “Carrier formation dynamics of organic photovoltaics as investigated by time-resolved spectroscopy,” Adv. Opt. Tech.Doi [CrossRef] (2012).
3. E. Kilmov, X. L. W. Yang, G. G. Hoffmann, and J. Loos, “Scanning near-field and confocal Raman microscopic investigation of P3HT–PCBM systems for solar cell applications,” Macromolecules 39, 4493–4496 (2006). [CrossRef]
4. T. Siebert, R. Maksimenka, A. Materny, V. Engel, W. Kiefer, and M. Schmitt, “The role of specific normal modes during non-Born-Oppenheimer dynamics: the S1-S0 internal conversion of β-carotene interrogated on a femtosecond time-scale with coherent anti-Stokes Raman scattering,” J. Raman Spectrosc. 33, 844–854 (2002). [CrossRef]
5. V. Namboodiri, A. Scaria, M. Namboodiri, and A. Materny, “Investigation of molecular dynamics in β-carotene using femtosecond pump-FWM spectroscopy,” Laser Phys. 19, 154–161 (2009). [CrossRef]
6. M. Duncan, J. Reintjes, and T. J. Manuccia, “Scanning coherent anti-Stokes Raman microscope,” Opt. Lett. 7, 350–352 (1982). [CrossRef] [PubMed]
7. A. Zumbusch, G. R. Holtom, and X. S. Xie, “Three-dimensional vibrational imaging by coherent anti-Stokes Raman scattering,” Phys. Ref. Lett. 82, 4142–4145 (1999). [CrossRef]
8. C. L. Evans and X. S. Xie, “Coherent anti-Stokes Raman scattering microscopy: chemical imaging for biology and medicine,” Annu. Rev. Anal. Chem. 1, 883–909 (2008). [CrossRef]
9. C. L. Evans, E. O. Potma, M. Puorishaag, D. Cote, C. P. Lin, and X. S. Xie, “Chemical imaging of tissue in vivo with video-rate coherent anti-Stokes Raman scattering microscopy,” Proc. Natl. Acad. Sci. USA 102, 16807–16812 (2005). [CrossRef] [PubMed]
10. A. Volkmer, L. D. Book, and X. S. Xie, “Time-resolved coherent anti-Stokes Raman scattering microscopy: imaging based on Raman free induction decay,” Appl. Phys. Lett. 80, 1505–1507 (2002). [CrossRef]
11. E. Betzig and J. K. Trautman, “Near-field optics: Microscopy, spectroscopy, and surface modification beyond the diffraction limit,” Science 257, 189–195 (1992). [CrossRef] [PubMed]
12. E. Betzig, P. L. Finn, and J. S. Weiner, “Combined shear force and near-field scanning optical microscopy,” Appl. Phys. Lett. 60, 2484–2487 (1992). [CrossRef]
13. M. Lucas and E. Riedo, “Combining scanning probe microscopy with optical spectroscopy for applications in biology and materials science,” Rev. Sci. Instrum. 83, 0611011 (2012). [CrossRef]
14. L. Novotny and J. S. Stranick, “Near-field optical microscopy and spectroscopy with pointed probes,” Annu. Rev. Phys. Chem. 57, 303–331 (2006). [CrossRef] [PubMed]
15. H. F. Hess, E. Betzig, T. D. Harris, L. N. Pfeiffer, and K. W. West, “Near-field spectroscopy of the quantum constituents of a luminescent system,” Science 264, 1740–1745 (1994). [CrossRef] [PubMed]
16. B. Hecht, B. Sick, U. P. Wild, V. Deckert, R. Zenobi, O. J. F. Martin, and D. W. Pohl, “Scanning near-field optical microscopy with aperture probes: fundamentals and applications,” J. Chem. Phys. 112, 7761–7774 (2000). [CrossRef]
17. K. Karki, M. Namboodiri, T. Z. Khan, and A. Materny, “Pump-probe scanning near-field optical microscopy: sub-wavelength resolution chemical imaging and ultrafast local dynamics,” Appl. Phys. Lett. 100, 1531031 (2012). [CrossRef]
18. V. Deckert, “Tip-enhanced Raman spectroscopy,” J. Raman Spectrosc. 40, 1336–1337 (2009). [CrossRef]
19. R. C. Dunn, “Near-field scanning optical microscopy,” Chem. Rev. 99, 2891–2927 (1999). [CrossRef]
20. R. D. Schaller, J. Ziegelbauer, L. F. Lee, L. H. Haber, and R. J. Sayakally, “Chemically selective imaging of sub-cellular structure in human hepatocytes with coherent anti-Stokes Raman scattering (CARS) near-field scanning optical microscopy (NSOM),” J. Phys. Chem. B 106, 8489–8492 (2002). [CrossRef]
21. T. Ichimura, N. Hayazawa, M. Hashimoto, Y. Inouye, and S. Kawata, “Subnanometric near-field Raman investigation in the vicinity of a metallic nanostructure,” Phys. Rev. Lett. 92, 2208011 (2004).
22. K. Furusawa, N. Hayazawa, F. C. Catalan, T. Okamoto, and S. Kawata, “Tip-enhanced broadband CARS spectroscopy and imaging using a photonic crystal fiber based broadband light source,” J. Raman Spectrosc . 43, 656–661 (2012). [CrossRef]
23. B. Hecht, H. Bielefeldt, Y. Inouye, D. W. Pohl, and L. Novotny, “Facts and artifacts in near-field optical microscopy,” J. Appl. Phys. 81, 2492–2498 (1997). [CrossRef]
24. I. Horcas, R. Fernandez, J. M. Gomez-Rodriguez, J. Colchero, J. Gomez-Herrero, and A. M. Baro, “WSXM: a software for scanning probe microscopy and a tool for nanotechnology,” Rev. Sci. Instrum. 78, 0137051 (2007). [CrossRef]
25. S. Falke, P. Eravuchira, A. Materny, and C. Lienau, “Raman spectroscopic identification of fullerene inclusions in polymer/fullerene blends,” J. Raman Spectrosc. 42, 1897–1900 (2011). [CrossRef]
26. X. Feng and X. Wang, “Thermophysical properties of free-standing micrometer-thick poly(3-hexylthiophene) films,” Thin Solid Films 519, 5700–5705 (2011). [CrossRef]
27. S. Cook, A. Furube, and R. Katoh, “Analysis of the excited states of regioregular polythiophene P3HT,” Energy Environ. Sci. 1, 294–299 (2008). [CrossRef]
28. Y. Fu, H. Wang, R. Shi, and J. X. Cheng, “Characterization of photodamage in coherent anti-Stokes Raman scattering microscopy,” Opt. Express 14, 3942–3951 (2006). [CrossRef] [PubMed]
29. L. Chen, H. Zhiwei, L. Fake, Z. Wei, W. H. Dietmar, and S. Colin, “Near-field effects on coherent anti-Stokes Raman scattering microscopy imaging,” Opt. Express 15, 4118–4131 (2007). [CrossRef]
30. R. H. Webb, “Confocal optical microscopy,” Rep. Prog. Phys 59, 427–471 (1996). [CrossRef]
31. S. V. Kamat, S. H. Tamboli, V. Puri, R. K. Puri, J. B. Yadava, and O. S. Joo, “Optical and electrical properties of polythiophene thin films: effect of post deposition heating,” J. Optelectron. Adv. M. 12, 2301–2305 (2010).
