This study investigates, for the first time, an all-optically controllable random laser based on a dye-doped liquid crystal (DDLC) cell added with a photoisomerizable dye. Experimental results indicate that the lasing intensity of this random laser can be all-optically controlled to decrease and increase sequentially with a two-step exposure of one UV and then one green beam. All-optically reversible controllability of the random lasing emission is attributed to the isothermal nematic(N)→isotropic(I) and I→N phase transitions for LCs due to the UV-beam-induced trans→cis and green-beam-induced cis→trans back isomerizations of the photoisomerizable dye, respectively. The former and the latter can decrease and increase the spatial fluctuations of the order and thus of the dielectric tensor of LCs, respectively, subsequently increasing and decreasing the diffusion constant (or transport mean free path), respectively, and thus decaying and rising the scattering strength for the fluorescence photons in their recurrent multi-scattering process, respectively. The consequent decrease and increase of the lasing intensity for the random laser and thus the rise and descent of its energy threshold are generated, respectively.
©2010 Optical Society of America
Random lasers based on disordered media have received considerable interest in research for their potential applications in photonics and bio-medicine in recent years [1–20]. Such interest is owing to their unique mechanisms and capabilities for lasing emission and superiority to regular lasers, e. g., economic fabrication technology, large-scale production with miniaturization, and high emission efficiency. Many randomly dispersive materials, including semiconductor powders [1–4,7], polymers , human tissues , silver nanopowders , and liquid crystal(LC)-based media [10–19], have been used to develop random lasers. A random laser can generally be created based on a recurrent multi-scattering mechanism with or without a coherent feedback. Notably, a sufficiently long traveling time for the fluorescence photons in the multi-scattering process in an active medium causes the gain for the fluorescence to exceed the optical loss, leading to the occurrence of a random lasing emission [1,4,20].
In above mentioned media, only those LC-based random lasers can be exploited to control the random lasing features with the externally flexible controllability of LC orientation and, thus, of either the refractive index or dielectric property of LCs [10–19]. In particular, Strangi et al. found a surprisingly unexpected random lasing action based on a homogenously-aligned dye-doped LC (DDLC) system with a thick cell-gap (~100μm) . Generation of the random lasing is attributed to the recurrent multi-scattering effect from the fluctuation (i.e. spatial nonuniformity) of the order and, thus, of the dielectric tensor of LCs in the space of the LC bulk.
To realize an applicable LC random laser in photonics, this study describes for the first time an all-optically controllable random laser based on a DDLC cell added with a photoisomerizable dye (azo dye). Experimental results indicate that the lasing intensity of the random laser can be controlled to decrease and increase with increasing the irradiation times of one UV and one green beam, respectively. This all-optically reversible controllability of the LC random laser is attributed to the UV-beam-induced isothermal nematic→isotropic (N→I) and green-beam-induced isothermal I→N phase transitions of the LCs by trans→cis and cis→trans back isomerizations of the azo dye, respectively. The former and the latter mechanisms may decrease and increase the spatial nonuniformity of the order and thus of the dielectric tensor of the LCs, respectively, thereby increasing and decreasing the diffusion constant (or transport mean free path), respectively, and thus decreasing and increasing the scattering strength for the fluorescence photons in their recurrent multi-scattering process, respectively. Subsequently the decay and rise of the random lasing and thus the increase and decrease of the corresponding energy threshold can be obtained, respectively.
2. Sample preparation and experimental setups
The materials used in this experiment include 91.7wt% NLC E7 (ne = 1.7462 and no = 1.5216 at 20°C for λ = 589nm; ni≅(ne + 2no)/3 = 1.5965 in the isotropic phase, from Merck), 0.3wt% laser dye P650 (1,2,3,5,6,7-hexamethyl-8-cyanopyrromethene-difluoroborate, from Exciton), and 8wt% photoisomerizable dye 4MAB (4-Methoxyazobenzene, from Fluka). An empty cell is pre-fabricated with two PVA-coated glass slides separated by two 188μm-thick plastic spacers, in which one of the two slides is pre-rubbed but the other not. Following infusion into the empty cell, the uniform mixture fills the entire cell via the capillary effect to form a homogeneously-aligned DDLC cell added with a photoisomerizable dye.
Figure 1 displays the experimental setup for examining the all-optically controllable random lasing emission of DDLC. One pumped laser beam, derived from a Q-switched Nd:YAG second harmonic generation (SHG) pulse laser (wavelength, 532 nm) with a pulse duration of 8ns, repetition rate of 10Hz and pumped energy, E, is focused by a lens (focal length: 20cm) on the cell with an included angle of ~20° with respect to the cell normal (N). A fiber-optic probe of a fiber-based spectrometer (Jaz-Combo-2, Ocean Optics, resolution: ~0.9nm) is placed to face N with a 6cm-distance from the cell to record the random lasing output. A half-wave plate (λ/2 WP, for 532nm) and a polarizer (P) with a transmission axis along the rubbing direction of the cell (R) are placed in front of the lens for varying the incident pulse energy by rotating the direction of the optical axis of the half-wave plate. Between the lens and the polarizer, a nonpolarizing beam splitter (NBS) is inserted to split the incident beam with half the energy into the detector of the energy meter in order to measure the incident pulse energy. One randomly-polarized UV beam with a fixed intensity of 300mW/cm2 and a variable irradiated time tUV and one CW circularly polarized green beam (from a diode-pumped solid-state laser, wavelength: 532nm) with a fixed intensity of 300mW/cm2 and a variable irradiated time tG are installed to pre-illuminate the pumped spot of the cell when performing the all-optically controllable random lasing experiment. The incident angles relative to N for the green and UV beam are 40° and 180°, respectively.
3. Results and discussion
The following experimental data of associated random lasing emission are analyzed by pre-completing an auxiliary experiment associated with determination of spectral features of the laser dye and azo dye. Figures 1(a) and 1(b) show the absorption and fluorescence emission spectra of the laser dye (red and blue curves, respectively) based on an 0.3wt% P650-doped E7 cell and the absorption spectra of the azo dye based on an 8wt% 4MAB-doped E7 cell, respectively (the phase of the two cells is in isotropic state in this measurement). In Fig. 1(a), peaks of the measured absorption and fluorescence spectra of the laser dye are located at around 600 and 638 nm, respectively. When the wavelength exceeds 640nm, absorption of the laser dye can be neglected. Since absorbance at the wavelength of the pumped pulses is roughly one-third of the peak value, the laser dye can be excited efficiently. In Fig. 1(b), the black and grey curves represent the measured absorption spectra of the azo dye before and after the cell is irradiated by the UV light with 300mW/cm2 for around 2min. The 4MAB has two absorption bands appearing in UV (around 374nm) and visible (around 445nm) regions, associated with π-π* and n-π* transitions, respectively. As shown in the inset, the azo dyes generally sustain themselves at the stable rod-like trans-state in the dark. Following UV irradiation, massive trans-isomers are photoisomerized to curve cis-isomers such that absorption peaks at 374 and 445nm drop and rise, respectively. Moreover, cis-4MAB can convert back to the trans-state rapidly under irradiation by light with a long wavelength (such as a green-blue light) or slowly via a thermal reaction. Consequently, absorption curve of the dye recovers to the initial condition.
Before experimental results of the all-optical controllability of the obtained random lasing emission are discussed, the energy threshold of incident pumped pulses to generate the random lasing based on the DDLC cell added with the azo dye must be pre-determined. During this phase, both UV and green beams are blocked from irradiating the DDLC cell. Figure 3(a) illustrates the variation of the measured fluorescence emission spectrum of the DDLC with the pumped energy E = 13.5–16μJ/pulse. Figure 3(b) summarizes the experimental data in Fig. 3(a), which display variations in peak intensity of the fluorescence output and the corresponding full widths at half-maximum (FWHM) with the pumped energy. The peak intensity of the fluorescence output increases nonlinearly by increasing the pumped energy. Additionally, an energy threshold (Eth) ~14.8μJ/pulse can be obtained, which is indicative of a random lasing emission. Figure 3(c) shows the emission pattern of the random lasing on the screen placed behind the cell at E = 16μJ/pulse. In Fig. 3(a), the FWHM of the random lasing spikes at E = 16μJ/pulse can be obtained as narrow as ≤1nm. In comparison with the data in Fig. 2(a) , the random lasing peaks can occur around 640nm because it is near the wavelength (638nm) of the peak of the fluorescence emission. To elucidate the underlying mechanism of the random lasing presented in Fig. 3, this study also performs the coherent backscattering (CBS) experiment  by probing the DDLC cell with the use of a weak 633nm laser beam. The measured coherent cone width at half-maximum (θ1/2) of the cone-angle-dependent intensity distribution of the backscattering light from probing the DDLC is ~7mrad. According to the previous studies that investigated the coherent backscattering of light by disordered media [22–24], the coherent cone width depends on the transport mean free path ℓ* (defined as the average distance a photon travels before its propagation direction is randomized entirely) and the wavelength of the scattering light by ℓ*≅λ/(2πθ1/2). If substituting λ = 633nm and θ1/2 = 7mrad into this equation, the transport mean free path can be estimated as ℓ*~14μm. Because the condition kℓ* = 2πℓ*/λ~139 > 1 is sufficient, the random lasing observed in this study (Fig. 3) originates from the weak localization of fluorescence photons via recurrent multiple scattering due to the spatial nonuniformity of the dielectric property of the LCs in our DDLC cell [12,14,17–20].
The following experiment assesses the all-optically reversible controllability of the LC random laser with a fixed E = 16μJ/pulse by successive irradiation of one UV and one green beams on the DDLC cell added with the azo dye. Figure 4(a) displays the random lasing obtained once the DDLC is irradiated by the UV beam at various irradiation time periods tUV = 0, 5, 10, 15, and 20s and a fixed irradiation intensity IUV = 300mW/cm2. Obviously, the random lasing intensity can be controlled to decrease by increasing tUV, with those results indicating that the energy threshold to generate the random lasing increases with an increasing tUV. Figure 4(b) summarizes the experimental results associated with the influence of the irradiation of the green beam on the obtained DDLC random laser. First, the UV beam with IUV = 300mW/cm2 is turned on to irradiate the cell for 2 min and then turned off. Meanwhile, the green beam with IG = 300mW/cm2 is turned on at tG = 0s. The cell is then excited by pumped pulses successively at tG = 0, 10, 20, 30, and 40s. Several random lasing signals are subsequently obtained, as displayed by the blue, green, orange, red, and black dotted curves, respectively, in Fig. 4(b). The intensity of the obtained random lasing appears to be controlled to increase back the original value by increasing tG. This finding suggests that the energy threshold can decrease back with an increasing tG. Based on the experimental results in Fig. 4, the DDLC random laser thereby possesses the all-optically (reversible) controllable feature, i.e. the random lasing can gradually decay and rise back (and thus the energy threshold can gradually increase and decrease back) as the cell is irradiated successively by the UV and green beams with increasing their irradiation durations.
Azo dye plays a major role in the above all-optical controllability of random lasing since no similar experimental results shown in Fig. 4 can be obtained based on a 4MAB-free DDLC cell. Therefore, the mechanism for the all-optical controllability of the decrease and increase of the random lasing intensity (and subsequently the increase and decrease of the energy threshold for the random lasing) are attributed to UV- and green-beam-induced isothermal N→I and I→N phase transitions of the LCs via trans→cis and cis→trans back isomerizations of the 4MAB dyes, respectively . According to Fig. 5(a) , the 4MAB dyes are generally in a stable trans-state in the darkness. The rod-like trans-4MAB dyes are aligned with LC molecules via the guest-host effect in the DDLC cell. The 4MAB dyes may absorb UV light and rapidly convert to curve cis-state via trans-cis isomerization and then disturb the order of the LC host. With an increasing tUV, the concentration of azo dyes transforming to cis-state increases such that the LCs gradually change from N to I phase isothermally. This process causes the spatial nonuniformity of the order [δS = S(r + δr)–S(r)] and thus of the dielectric tensor [δε = ε(r + δr)–ε(r)] of the LCs in bulk to gradually decrease from a nonzero (δS≠0 and δε≠0 at nematic phase) to a zero (δS = 0 and δε = 0 at isotropic phase) value. Consequently, the local multiple micro-domain of LCs experienced by the propagating fluorescence photons gradually disappears (as presented in the model of the recurrent multi-scattering in Fig. 5(b)), where δr, δS and δε denote the differential displacement, differential order and differential dielectric tensor of LCs between two adjacent local micro-domains of LCs in bulk, respectively. The diffusion constant of photons (D) in a disordered medium can be expressed with D = (υℓ*)/3 , where ℓ* refers to the transport mean free path and υ is the transport velocity of photons in a disordered LC sample. The decrease of the spatial nonuniformity of the order and thus of the dielectric tensor of the LCs may increase the diffusion constant (or transport mean free path), thereby decreasing the scattering strength of the photons in the DDLC cell. Subsequently, the random lasing intensity can decrease and thus the corresponding energy threshold can increase as tUV increases. As mentioned previously, most 4MAB dyes are transformed to the cis-state as they absorb UV light, which can decay the random lasing of the cell [Fig. 4(a)]. Once the UV beam is turned off and the cell is subsequently irradiated by one green beam, the cis-4MAB dyes may transform rapidly back to the trans-state via cis-trans back isomerisation . The increase of the concentration of the trans-4MAB via consecutive cis-trans back isomerisation may give rise to the reverse phase transition of LCs from I to N if the irradiation time of the green beam gradually increases. The spatial nonuniformity of the order and thus of the dielectric tensor of the LCs then gradually increases back, causing the diffusion constant to decrease back and, thus, the scattering strength to increase back. Consequently, intensity of the random lasing emission increases back and thus the energy threshold decreases back with an increasing tG .
Figures 6(a) and 6(b) display the variations in the DDLC cell pattern, observed under a transmitting polarizing optical microscope (POM) with crossed polarizers, with an increasing tUV from 0 to 50s at IUV = 300mW/cm2 (IG = 0) and then increasing tG from 0 to 50s at IG = 300mW/cm2 (IUV = 0), respectively. The angle between the transmission axis of the polarizer and R in the cell is set at 45°. Obviously, the cell transmission decreases from bright to dark state with an increasing tUV (IG = 0) [Fig. 6(a)], and then increases reversely from dark to bright state with an increasing tG (IUV = 0) [Fig. 6(b)]. These results coincide with the mechanisms to cause the reversible variation of the random lasing intensity and thus of the associated energy threshold that the UV- and green-beam-irradiation induce isothermal N→I and I→N phase transitions of LCs via trans-cis and cis→trans back isomerizations of azo dye, respectively.
This study also elucidates a situation in which controllability of the random lasing presented in Fig. 4 is not attributed to the thermal-induced phase transition of the DDLC under the photo-irradiation. The variation of the temperature on the UV-beam-irradiated spot of the cell is determined with an increasing tUV from 0 to 40s (IUV = 300mW/cm2) using a thermal imager (Fluke, Ti10). Figure 7 summarizes the experimental results in which the measured temperature on the irradiated spot increases from 23.8 to 27.1°C with an increasing tUV = 0 to 40s. Notably, no phase transition occurs because the final temperature of 27.1°C is far from the clearing point of the DDLC (~52°C). Therefore, the all-optically reversible controllability of the LC random laser is not caused by the photo-induced thermal effect but by the photoisomerization effect-induced isothermal phase transition.
This study investigates, for the first time, an all-optically controllable random laser based on a 188μm-thick homogeneously-aligned DDLC cell added with a photoisomerizable dye. Experimental results indicate that the lasing intensity of the random laser can be all-optically controlled to decrease and then increase back in sequence with a two-step exposure of one UV and then one green beam when increasing the individual irradiated time and same fixed intensity (300mW/cm2). The all-optically reversible controllability of the random lasing is attributed to the isothermal N→I and I→N phase transitions of LCs, respectively, due to the UV-beam-induced trans→cis and green-beam-induced cis→trans back isomerizations of the photoisomerizable dye. The former and latter mechanisms can decrease and increase, respectively, the spatial nonuniformity of the order and thus of the dielectric property of LCs, leading to the increase and decrease of the diffusion constant, respectively, and thus the decay and rise of the scattering strength of the fluorescence photons in their recurrent multiple scattering process, respectively. Consequently, the decrease and increase of the lasing intensity of the random laser and thus the increase and decease of its energy threshold can be obtained, respectively. Such an all-optically controllable LC random laser is highly promising for integrated photonic applications.
The authors would like to thank the National Science Council of the Republic of China, Taiwan (Contract numbers: NSC 97-2112-M-040-001-MY2 and NSC 97-2112-M-006-013-MY3) and the Advanced Optoelectronic Technology Center, National Cheng Kung University, under projects from the Ministry of Education for financially supporting this research. We greatly appreciate Ted Knoy for editorial assistance.
References and links
1. N. M. Lawandy, R. M. Balachandran, A. S. L. Gomes, and E. Sauvain, “Laser action in strongly scattering media,” Nature 368(6470), 436–438 (1994). [CrossRef]
2. H. Cao, Y. G. Zhao, S. T. Ho, E. W. Seelig, Q. H. Wang, and R. P. H. Chang, “Random laser action in semiconductor powder,” Phys. Rev. Lett. 82(11), 2278–2281 (1999). [CrossRef]
3. C. A. Vutha, S. K. Tiwari, and R. K. Thareja, “Random laser action in ZnO doped polymer,” J. Appl. Phys. 99(12), 123509 (2006). [CrossRef]
6. S. Frolov, Z. Vardeny, K. Yoshino, A. Zakhidov, and R. Baughman, “Stimulated emission in high-gain organic media,” Phys. Rev. B 59(8), R5284–R5287 (1999). [CrossRef]
7. Q. H. Song, L. Wang, S. M. Xiao, X. C. Zhou, L. Y. Liu, and L. Xu, “Random laser emission from a surface-corrugated waveguide,” Phys. Rev. B 72(3), 035424 (2005). [CrossRef]
8. R. C. Polson and Z. V. Vardeny, “Random lasing in human tissues,” Appl. Phys. Lett. 85(7), 1289–1291 (2004). [CrossRef]
9. G. D. Dice, S. Mujumdar, and A. Y. Elezzabi, “Plasmonically enhanced diffusive and subdiffusive metal nanoparticle-dye random laser,” Appl. Phys. Lett. 86(13), 131105 (2005). [CrossRef]
10. D. S. Wiersma, M. Colocci, R. Righini, and F. Aliev, “Temperature-controlled light diffusion in random media,” Phys. Rev. B 64(14), 144208 (2001). [CrossRef]
11. S. M. Morris, A. D. Ford, M. N. Pivnenko, and H. J. Coles, “Electronic control of nonresonant random lasing from a dye-doped smectic A* liquid crystal scattering device,” Appl. Phys. Lett. 86(14), 141103 (2005). [CrossRef]
12. G. Strangi, S. Ferjani, V. Barna, A. De Luca, C. Versace, N. Scaramuzza, and R. Bartolino, “Random lasing and weak localization of light in dye-doped nematic liquid crystals,” Opt. Express 14(17), 7737–7744 (2006). [CrossRef] [PubMed]
13. Q. H. Song, S. M. Xiao, X. C. Zhou, L. Y. Liu, L. Xu, Y. G. Wu, and Z. S. Wang, “Liquid-crystal-based tunable high-Q directional random laser from a planar random microcavity,” Opt. Lett. 32(4), 373–375 (2007). [CrossRef] [PubMed]
15. Q. H. Song, L. Y. Liu, L. Xu, Y. G. Wu, and Z. S. Wang, “Electrical tunable random laser emission from a liquid-crystal infiltrated disordered planar microcavity,” Opt. Lett. 34(3), 298–300 (2009). [CrossRef] [PubMed]
16. S. Gottardo, S. Cavalieri, O. Yaroshchuk, and D. S. Wiersma, “Quasi-two-dimensional diffusive random laser action,” Phys. Rev. Lett. 93(26), 263901 (2004). [CrossRef]
17. Y. J. Liu, X. W. Sun, H. I. Elim, and W. Ji, “Gain narrowing and random lasing from dye-doped polymer dispersed liquid crystals with nanoscale liquid crystal droplets,” Appl. Phys. Lett. 89(1), 011111 (2006). [CrossRef]
18. C.-R. Lee, S.-H. Lin, C.-H. Guo, S.-H. Chang, T.-S. Mo, and S.-C. Chu, “All-optically controllable random laser based on a dye-doped polymer-dispersed liquid crystal with nano-sized droplets,” Opt. Express 18(3), 2406–2412 (2010). [CrossRef] [PubMed]
19. S. Ferjani, L. Sorriso-Valvo, A. De Luca, V. Barna, R. De Marco, and G. Strangi, “Statistical analysis of random lasing emission properties in nematic liquid crystals,” Phys. Rev. E Stat. Nonlin. Soft Matter Phys. 78(1), 011707 (2008). [CrossRef] [PubMed]
20. D. S. Wiersma, “The physics and applications of random lasers,” Nat. Phys. 4(5), 359–367 (2008). [CrossRef]
23. areM. van Albada, M. van der Mark, and A. Lagendijk, “Observation of weak localization of light in a finite slab: Anisotropy effects and light path classification,” Phys. Rev. Lett. 58(4), 361–364 (1987). [CrossRef] [PubMed]
25. H.-K. Lee, A. Kanazawa, T. Shiono, T. Ikeda, T. Fujisawa, M. Aizawa, and B. Lee, “All-optically controllable polymer/liquid crystal composite films containing the azobenzene liquid crystal,” Chem. Mater. 10(5), 1402–1407 (1998). [CrossRef]