Open Access
Add to BibSonomyAbstract
Multi-photon microscopy operating at 1550 nm is employed as a rapid characterization tool for studying the photostability of three well-known electro-optical materials. Different nonlinear optical responses such as multi-photon excitation fluoresence, second-, and third-harmonic generation can be used as detection probes to reveal the degradation mechanisms. This technique is rapid, accurate, and can be used to study the photostability of a broad range of materials.
© 2014 Optical Society of America
1. Introduction
The photochemical properties of materials are of crucial importance for their acceptance in commercial optical systems. Nowadays, pulsed lasers with high peak intensities are widely used in different optical devices, where the high laser intensities can cause materials to undergo photochemical reactions and remove molecules from the population. This process is called photobleaching and is caused by transitions between excited states of different multiplicity as well as chemical reactions, interaction with the environment and interaction with the excitation light [1, 2]. In general, the photostability of a material dictates its suitability for a particular application and molecules with diverse photochemical properties are required for different optical systems. Highly photo-stable molecules are needed for applications such as optical modulators and switches while materials with a large photobleaching rate are needed for three-dimensional read-write devices [3, 4]. Thus, for successful device fabrication it is crucial to characterize the materials’ photostability properties to ensure the best device performance and yield.
Standard optical characterization techniques such as optical microscopy and Raman spectroscopy [5] are time-consuming and not suitable for rapid characterization in large scale samples. Furthermore, these techniques only indicate a change in materials shape and molecular structure and do not elucidate the bleaching mechanisms. From the discussion above, the need for a better method of understanding the probability and mechanisms of photobleaching of various materials at a specific applied excitation irradiances is clear.
Multi-photon microscopy (MPM) is a powerful technique that allows mapping of thin film samples with any kind of optical response such as multi-photon fluorescence (MPFL), second-, and third-harmonic generation (SHG, THG). There are several reports that discuss the measurement of the photobleaching of materials by observing the decrease of the fluorescence intensity using two-photon imaging [6–8]. However, these studies are limited only to fluorescent materials. The number of photons involved in the bleaching process can be determined by investigating the bleaching rate versus the excitation irradiance. Many researchers have worked to gain a better understanding of photobleaching in the case of one- and two-photon excitations [9, 10]. However, only a few reports are available on photobleaching in the case of three- and more photon excitation [3]. Further, most previous works have been carried out at the wavelengths in the visible range of spectrum which makes it difficult to detect two-, and three-photon excitation signals due to the lack of optical components such as lenses and objectives with high sensitivity in the UV regime.
We have in this article successfully employed a multi-photon microscope with a compact femtosecond fiber laser at 1550 nm to explore electro-optical (EO) materials suitability in terms of photostability for optical devices such as waveguides. Excitation in the 1550 nm band is not only a necessity for telecommunication devices, but also simplifies the detection of second-, and third-harmonic signals due to the existence of high quality optical elements and highly sensitive PMTs in the visible regime. The high photon densities provided from the pulsed laser together with the detection capabilities enable us to utilize various nonlinear optical responses of materials such as SHG, THG, or fluorescence induced by multi-photon absorption for photobleaching studies. This capability significantly broadens the range of materials that can be investigated. Compared to fluorescence detection, THG detection has a large dynamic range and thus larger probing sensitivity. Further, SHG is a beneficial probe for non-centrosymmetric materials with a relatively large second-order susceptibility (χ(2)) as well as for studying the interfaces. To show the capability of our technique, we investigate the photobleaching mechanisms of three well-known EO materials: Disperse Red 1 (DR1), lithium niobate (LiNbO3), and the recently developed EO polymer, SEO250. This method is fast and convenient and can be used for rapid characterization of large scale samples. To our knowledge, this is the first detailed study of the photochemical properties of EO materials in the telecommunications range near 1550 nm.
2. Experimental setup
The experimental configuration is depicted in Fig. 1(a). The light source in the MPM system is a femtosecond laser beam at 1550 nm. The repetition rate and duration of the pump source is approximately 75 MHz and 150 fs, respectively. In [11], Kieu et al. have reported the design and performance of the fiber laser source in more detail. The input femtosecond laser beam is raster-scanned on the sample using a 2D galvo scanner system. The beam is relayed and expanded with a scan lens and a tube lens arranged in a telescopic scheme. The back aperture of the objective is fully illuminated to make use of the full NA (NA=0.5, 20× aspheric objective lens) and create the smallest possible laser spot size on the sample (confocal parameter=7.2 μm, spot size=1.4 μm). The backscattered signals are detected simultaneously using two highly sensitive high gain PMTs through a series of dichroic beam splitters and filters. A narrow bandpass filter (∼20nm FWHM bandwidth) is used before each of the PMTs to detect the SHG (∼780 nm) and THG (∼520 nm) generated from the laser focal spot. The fluorescence due to two- or three-photon excitation can be measured by removing the 780 nm bandpass filter from the transmission path of dichroic mirror. The generated multi-photon signals can also be detected with a sensitive spectrometer with the use of a rotating dichroic mirror at 870 nm.
Fig. 1 a) The schematic of the multiphoton microscope, and b) diagram showing various sources and probes on materials photodegradation study.
The diagram shown in Fig. 1.b demonstrates the main mechanisms (single-, two-, and three-photon excitation, photo-oxidation and temperature increase) involved in material degradation when pulsed lasers with high peak intensities are used. It further shows the various probes that can be utilized by our MPM system to study material photobleaching, among which the FL emission is the most studied probe so far. As explained above, with the various probes available, we have the capability to study the photostability of a broad range of materials needed for various applications.
3. Results and discussion
3.1. Lithium niobate
In this section, we use lithium niobate (LiNbO3) which is a highly stable (thermally, chemically, and mechanically) ferroelectric crystal to understand the uncertainty of our experimental measurements. LiNbO3 has many applications in photonic devices due to its high nonlinear optical coefficients and a broad visible and infrared transparency. Properties and applications of LiNbO3 have been widely studied resulting in several excellent review papers on this material [12, 13]. One of the significant applications of LiNbO3 is for electro-optic modulation. Further, LiNbO3 has a significant SHG coefficient due to its non-centrosymmetric crystal structure. Thus, we use the SHG signal of LiNbO3 as a probe to explore its photostability and the uncertainty of our multi-photon imaging system.
The multi-photon spectrum of the LiNbO3 sample excited at 1550 nm is shown in Fig. 2(a). As expected, the sample shows a strong SHG signal around ∼775 nm. Figure 2(b) demonstrates the normalized SHG signal of the sample continuously irradiated with various powers of the laser (0.5, 1.5, and 2.5 mW average powers) for 3 minutes. As shown, the decay in the SHG signal is less than 2% over 3 minutes of the exposure time. The potential mechanisms that can contribute to the observed decay are material photobleaching, stage drift which moves the sample from the focal plane, laser pump power and polarization drift, among which the stage drift has the largest contribution. The independence of the slight decay in the SHG signal of the input power suggests that the drift is not due to material photobleaching. Further measurements show that the drift in laser power is on average 0.5% of the set power. Other mechanisms such as dispersion of lenses in the system and the drift in the laser polarization have negligible effects on the measurement results.
Fig. 2 a) LiNbO3 multiphoton spectrum excited at 1550 nm, and b) normalized SHG intensity versus laser exposure timing for various laser powers.
3.2. Disperse Red 1
Disperse Red 1 (DR1) is an EO chromophore well-known for its nonlinear optical properties and applications in EO modulators and switches [14]. It is usually doped in an amorphous polymer to enable the formation of uniform amorphous thin films. In this work, polymethylmethacrylate (PMMA) was used as the host material. PMMA is a common material often used in the fabrication of waveguide devices due to its high optical damage threshold and good stability. DR1 was doped in PMMA as follows: a 6% wt. DR1 solution in chlorobenzene was mixed with 6% wt. PMMA solution in chlorobenzene (1:1 volume). The mixture was stirred at 50°C overnight and filtered with 0.2 μm PTFE syringe filters. The solution was spin coated onto the fused silica substrate to make a 150 nm thin film.
The multiphoton spectrum of the sample excited at 1550 nm is shown in Fig. 3(a). Since the sample shows no fluorescence but a strong THG signal (sharp peak at ∼517 nm), we use THG as a probe to explore the photobleaching mechanism of DR1 (the cubic dependence of the output power on the input pump shown in the inset also indicates that the probe is THG signal). Figure 3(b) demonstrates the normalized THG signal reflected back from the sample when it is continuously irradiated with the focused laser beam at 1550 nm for 3 minutes. The THG profile of the sample is shown in Fig. 3(c) where several areas of the sample (26 μm × 26 μm squares, 512×512 pixels) each continuously exposed by various powers and timings of the laser are shown. It can be seen that at very low irradiation power (below 8 mW), photobleaching is negligible. However, for excitation peak power values greater than 8 mW, the process of photobleaching becomes significant and a decay of the THG is observed. In order to fit our experimental data for THG signal, we tried three different functions: single-, double-, and stretched-exponential. The results of Fig. 4(a) demonstrate that both double- and stretched-exponential functions could fit the data with good agreement. It should be noted that several relaxation phenomena in complex condensed-matter systems have been found to follow the stretched-exponential decay law. Specially, dielectric relaxation in polymers are usually described as being a stretched-exponential function [15]. Although a large weight percentage of polymer is used in DR1 samples, we chose to fit the experimental results with a double exponential since that would make the interpretation of the two well-known degradation phenomena in DR1 (photo-oxidation and tran-cis isomerization) easier. A double-exponential decay function can be written as [16]
In this expression T is the THG photon flux, k1 and k2 are the rates of decrease of THG signal, called here photobleaching rates (1/s), T1 and T2 are the initial contributions to the THG signal from each exponential decay (photon/s), and T0 is a noise constant. The two exponential decay rates in the bleaching function represents two different mechanisms involved in the photobleaching process of DR1. The photobleaching rates k1 and k2 versus irradiation power are depicted in Fig. 4(b). As can be seen, the slopes of logarithmic plots are 0.61 and 1.26 for k1 and k2, respectively. Hence, the slopes of the photobleaching rates suggest that a one-photon process is the bleaching source for both mechanisms. It is shown in Fig. 4(c) that the first mechanism associated with k1 is a much faster process compared to the second mechanism (k1 ≫k2). On the other hand, the contribution to the THG signal from the two mechanisms is comparable. As mentioned earlier, the dominant degradation mechanisms in DR1 are photo-oxidation and tran-cis isomerization [17,18]. The photo-oxidation (described by the first mechanism with decay rate of k1) is due to triplet ground state oxygen molecules that absorb infrared photons and become singlet oxygen free-radical species that aggressively attack the chromophores. Further, tran-cis isomerization which leads to statistical photo-orientation of the DR1 molecules, is a relatively slow process (described by the second mechanism with decay rate of k2).
Fig. 3 a) Multiphoton spectrum of DR1 on fused silica, the inset shows the log-log plot of the output THG signal versus input power, b) normalized THG intensity versus exposure timing for various laser powers, and c) THG profile of the sample where each square is continuously exposed by a different laser power and exposure timing.
Fig. 4 a) DR1 decay curve with 30 mW power (circle-marked line), best fitted one- (dash-dotted line), double-(dotted line), and stretched-exponential (solid line) functions, b) loglog plots of the photobleaching rates k1 and k2 vs irradiation power, and c) the ratio between parameters k1 and k2 (left axis) and A1 and A2 (right axis) vs irradiation power.
3.3. SEO250
SEO250 is a popular EO polymer, which also has a relatively large χ(3) and negligible fluorescence when it is unpoled (the emission spectrum is shown in Fig. 5(a) with a strong signal at ∼517 nm and no fluorescence). A prime application of this polymer is in EO channel waveguide devices. SEO250 is highly photostable with negligible photobleaching in the ultraviolet regime of the spectrum and the ability to transmit on the order of 50 mW of cw 1550 nm for hundreds of hours well beyond what is needed for most optical communication applications. The sample was fabricated by dissolving SEO250 3% wt. in dibromomethane and leaving it to stir overnight, then a 350 nm thick film was fabricated by spin coating the solution on a fused silica substrate.
Fig. 5 a) Multiphoton spectrum of SEO250 on fused silica, the inset shows a UV/Vis absorption spectrum of SEO250 (top) and the log-log plot of the output THG signal versus input power (bottom), b) normalized THG intensity versus laser exposure timing for various laser powers, and c) THG profile of the sample where each square is continuously exposed by a different laser power and exposure timing.
Figure 5(b) shows the normalized photobleached THG signal reflected back from the sample versus exposure time for different excitation powers. The THG profile of the sample is demonstrated in Fig. 5(c) with 16 squares (26 μm × 26 μm, 512 × 512 pixels) each continuously exposed by laser powers ranging from 9–25 mW and 15, 30, 60, and 120 sec exposure times (∼0.5 frame/sec). As can be seen in this figure, the observed bleaching has a clear correlation to the amount of laser power as well as to the total duration of the exposure. Figure 6(a) demonstrates that the THG decay of SEO250 versus time could be fitted by both a double-exponential and a stretched-exponential. However, consideration of the relative decay rate and amplitude of the two introduced mechanisms when a double-exponential decay law is used, indicates that there is likely a single process for the photobleaching. Figure 6(b) demonstrates a log-log plot of the photobleaching rates (k1 and k2) versus laser power and the ratio of the two mechanisms contribution (shown in the inset) when a double-exponential is used. The slopes of the logarithmic plots (2.27 for k1 and 1.95 for k2) show that bleaching for both mechanisms is due to a two-photon excitation source since both photobleaching rates depend on square of the excitation power. Further, it can be seen that the two mechanisms are comparable in their speed and contribution. Thus, we chose to use a stretched-exponential function to describe the bleaching mechanism of SEO250:
where T is the THG photon flux, k is the photobleaching rate (1/s), and 0 < β < 1 is a dispersive factor. The slope 2.11 of the plot shown in Fig. 6(c) demonstrates that as expected from the double-exponential model, only two photons are involved in the bleaching process and that the degradation mechanism is two-photon excitation. This is consistent with the fact that the SEO250 EO chromophore has intense one-photon absorption at 775 nm and the noncentrosymmetric nature of the chromophore relaxes the usual parity driven one and two-photon absorption selection rules.Fig. 6 a) Fitting SEO250 decay curve with 19.8 mW power (circle-marked line) with 1-(dash-dotted line), 2-(dotted line), and stretched-(solid line) exponential functions. b) loglog plots of the photobleaching rates k1 and k2 for SEO250, inset: the ratio between parameters A1 and A2 vs irradiation power, and c) the photobleaching rate versus irradiation power when the experimental data was fitted with a stretched-exponential function.
4. Conclusion
In summary, the photobleaching of three electro-optical materials (lithium niobate, Disperse Red 1, and SEO250) was explored under ultrashort pulsed illumination at 1550 nm using multi-photon microscopy. One-photon and two-photon excitation have been identified as the main bleaching mechanisms for DR1 and SEO250, respectively. Further, the large chemical stability of LiNbO3 has been confirmed. This method is designed to be a rapid screening test to assist in the development of new materials and to properly target them for applications. Moreover, it can be realized on a broad range of materials using detection probes such as multi-photon excitation FL, SHG, and THG. Furthermore, the involvement of two, three, or even more photons in the bleaching process can be explored. The main application of this work is in the design of various optical devices such as channel waveguides, EO modulators, etc., where by choosing suitable conditions for device performance, such as excitation threshold and the type of material, fabrication with optimum performance and yield is feasible.
Acknowledgments
This work was supported in part by Air Force Office of Scientific Research COMAS MURI FA9550-10-1-0558, Office of Naval Research NECom MURI, and National Science Foundation CIAN ERC #EEC-0812072. Authors are grateful to Dr. Jingdong Luo at Soluxra for proving the SEO250.
References and links
1. I. Polyzos, G. Tsigaridas, M. Fakis, V. Giannetas, P. Persephonis, and J. Mikroyannidis, “Substituent effect on the photobleaching of pyrylium salts under ultrashort pulsed illumination,” J. Phys. Chem. B 110, 2593–2597 (2006). [CrossRef] [PubMed]
2. M. A. van Dijk, L. C. Kapitein, J. van Mameren, C. F. Schmidt, and E. J. G. Peterman, “Combining optical trapping and single-molecule fluorescence spectroscopy: enhanced photobleaching of fluorophores,” J. Phys. Chem. B 108, 6479–6484 (2004). [CrossRef]
3. I. Polyzos, G. Tsigaridas, M. Fakis, V. Giannetas, and P. Persephonis, “Three-photon induced photobleaching in a three-dimensional memory material,” Opt. Lett. 30, 2654–2656 (2005). [CrossRef] [PubMed]
4. D. A. Parthenopoulos and P. M. Rentzepis, “Three-dimensional optical storage memory,” Science 245, 843–845 (1989). [CrossRef] [PubMed]
5. D. N. Bratashov, A. Masic, A. M. Yashchenok, M. F. Bedard, O. A. Inozemtseva, D. A. Gorin, T. Basova, T. K. Sievers, G. B. Sukhorukov, M. Winterhalter, H. Möhwald, and A. G. Skirtach, “Raman imaging and photodegradation study of phthalocyanine containing microcapsules and coated particles,” J. Raman Spectrosc. 42, 1901–1907 (2011). [CrossRef]
6. M. Gu and D. Day, “Use of continuous-wave illumination for two-photon three-dimensional optical bit data storage in a photobleaching polymer,” Opt. Lett. 24, 288–290 (1999). [CrossRef]
7. C. Eggeling, J. Widengren, R. Rigler, and C. A. M. Seidel, “Photobleaching of fluorescent dyes under conditions used for single-molecule detection: evidence of two-step photolysis,” Anal. Chem. 70, 2651–2659 (1998). [CrossRef] [PubMed]
8. J. Mertz, “Molecular photodynamics involved in multi-photon excitation fluorescence microscopy,” Eur. Phys. J. D. 3, 53–66 (1998). [CrossRef]
9. I. Polyzos, G. Tsigaridas, M. Fakis, V. Giannetas, P. Persephonis, and J. Mikroyannidis, “Two-photon absorption properties of novel organic materials for three-dimensional optical memories,” Chem. Phys. Lett. 369, 264–268 (2003). [CrossRef]
10. S. Gavrilyuk, S. Polyutov, P. C. Jha, Z. Rinkevicius, H. gren, and F. Gel’mukhanov, “Many-photon dynamics of photobleaching,” J. Phys. Chem. A 111, 11961–11975 (2007). [CrossRef] [PubMed]
11. K. Kieu and M. Mansuripur, “Femtosecond laser pulse generation with a fiber taper embedded in carbon nanotube/polymer composite,” Opt. Lett. 32, 2242–2244 (2007). [CrossRef] [PubMed]
12. E. Wooten, K. Kissa, A. Yi-Yan, E. Murphy, D. Lafaw, P. Hallemeier, D. Maack, D. Attanasio, D. Fritz, G. McBrien, and D. Bossi, “A review of lithium niobate modulators for fiber-optic communications systems,” IEEE J. Sel. Topics Quantum Electron. 6, 69–82 (2000). [CrossRef]
13. M. M. Abouellell and F. J. Leonberger, “Waveguides in lithium niobate,” J. Am. Ceram. Soc. 72, 1311–1321 (1989). [CrossRef]
14. H. Mahdavi, E. Mohajerani, M. Mohseni, and Y. Aghili, “A novel approach to analyze photoinduced effects in a DR1 dye-doped hybrid nanocomposite film,” J. Sol-Gel. Sci. Techn. 52, 82–87 (2009). [CrossRef]
15. R. Chen and P.L. Leung, “The decay of OSL signals as stretched-exponential functions,” Radiat. Meas. 37, 519–526 (2003). [CrossRef]
16. G.H. Patterson and D.W. Piston, “Photobleaching in two-photon excitation microscopy,” Biophys. J. 78, 2159–2162 (2000). [CrossRef] [PubMed]
17. A. Galvan-Gonzalez, M. Canva, G. I. Stegeman, R. Twieg, K. P. Chan, T. C. Kowalczyk, X. Q. Zhang, H. S. Lackritz, S. Marder, and S. Thayumanavan, “Systematic behavior of electro-optic chromophore photostability,” Opt. Lett. 25, 332–334 (2000). [CrossRef]
18. A. Galvan-Gonzalez, M. Canva, G. I. Stegeman, R. Twieg, T. C. Kowalczyk, and H. S. Lackritz, “Effect of temperature and atmospheric environment on the photodegradation of some Disperse Red 1-type polymers,” Opt. Lett. 24, 1741–1743 (1999). [CrossRef]
