This work demonstrates, for the first time, an all-optically controllable distributed feedback (DFB) laser based on a dye-doped holographic polymer-dispersed liquid crystal (DDHPDLC) grating with a photoisomerizable dye. Intensity of the lasing emission can be reduced and increased by raising the irradiation intensity of one CW circularly-polarized green beam and the irradiation time of one CW circularly-polarized red beam, respectively. The all-optical controllability of the lasing emission is owing to the green-beam-induced isothermal nematic→isotropic and red-beam-induced isothermal isotropic→nematic phase transitions of the LCs via trans→cis and cis→trans back isomerizations of the azo-dye, respectively, in the LC-droplet-rich regions of the grating. The former (latter) mechanism can reduce (increase) the index modulation and thereby the coupling strength in the DFB grating, resulting in the decay (rise) of the lasing emission. Thermal effect is excluded from possible mechanisms causing such an optical controllability of the lasing emission.
©2010 Optical Society of America
Laser developments based on holographic polymer-dispersed liquid crystal (HPDLC) gratings have attracted substantial interest in recent years because of their unique photonic band property and related potential applications in photonic areas [1–9,14]. Generally, a spatially periodic structure with alternating polymer-rich and LC-droplet-rich regions in an HPDLC grating can be easily fabricated by exposing the photoreactive LC-monomer composite film to an interfering field that is formed by two overlapping coherent optical beams . The formation of index modulation through the grating stripes results in the band structure of the grating. The propagation of optical waves in such a band structure may yield an extremely small group velocity and a rather large density of photonic state (DOS) at the edges of the photonic gap . When fluorescence dyes in such a grating are pre-doped, the coupling of the emitted fluorescence spectrum of the dye with the overlapping band gap of the grating may generate a low-threshold lasing emission at the edges via distributed feedback (DFB) process [3–8].
The lasing features of the HPDLC grating laser can be controlled or switched precisely by varying the LC orientation and thereby the refractive index in the LC-droplet-rich regions of the grating using a thermal  or electrical [2,3,5,7] method, or by altering the LC concentration  or the included angle of incidence of the two writing beams [4,5]. No work has yet been conducted on the development of a DFB laser that is optically controllable using an HPDLC grating cell. Therefore, this work describes, for the first time, an all-optically controllable DFB laser that is based on a dye-doped HPDLC (DDHPDLC) grating cell with a photoisomerizable dye. Experimental results reveal that the intensity of the lasing emission from the formed grating can be reduced and increased by increasing the irradiation intensity of one CW circularly-polarized (CP) green beam and the irradiation time of one CW CP red beam, respectively. This all-optically controllable feature of the laser is attributable to the green-beam-induced isothermal nematic→isotropic (N→I) and red-beam-induced isothermal I→N phase transitions of the LCs by the trans→cis and cis→trans back isomerizations of the azo-dye in the LC-droplet-rich regions of the grating, respectively. These two mechanisms can reduce and increase, respectively, the index modulation, and thereby the coupling strength in the DFB grating, resulting in a decay and rise of the intensity of the lasing output. Thermal effect is excluded from the possible causes inducing such an optical controllability of the laser.
2. Preparation of sample and experimental setup
The materials that were adopted in this work are 34wt% nematic LC (NLC), E7 (ne=1.7462 and no=1.5216 at 20°C for λ=589nm; ni≅(ne+2no)/3=1.5965 at isotropic phase) (from Merck), 39.8wt% monomer, trimethylolpropane triacrylate (TMPTA), 16wt% cross-linking monomer, N-vinylpyrrollidone (NVP), 0.70wt% photo-initiator, rose Bengal (RB), 1.30wt% coinitiator, N-phenylglycine (NPG), 7.00wt% surfactant, octanoic acid, 0.20wt% azo dye, D2 (all from Aldrich), and 1.00wt% laser dye, 4-dicyanmethylene-2-methyl-6-(p-dimethylaminostyryl)-4H-pyran (DCM) (from Exciton). These materials are uniformly mixed and injected into the empty cell which is pre-fabricated by combing two indium-tin-oxide-coated glass slides that are separated by two 30μm-thick plastic spacers. After the mixture fills the cell by the capillary effect, the cell is then exposed symmetrically about the cell normal to an interference pattern that is generated by two overlapping coherent green laser beams (from an Ar+ laser, wavelength: 514.5nm) with an included angle of about 81°, each with an intensity of 40mW/cm2, for 6min to produce a DDHPDLC transmission grating with an azo dye. Figure 1(a) displays an SEM image of the formed DDHPDLC grating with LCs pre-removed from all droplets: the dark stripes in the grating represent the LC-droplet-rich regions; most of the LC droplets are estimated to be nano-sized (<100nm). The grating spacing (Λ) is calculated given by the following formula [4,7];Eq. (1), the estimated value of grating spacing, approximately 0.26μm, is highly consistent with the experimental value obtained from the grating image in Fig. 1(a). The refractive index n p in the polymer-rich regions of the grating is around 1.54-1.55 .
Figure 1(b) depicts the experimental setup for the investigation of the all-optically controllable lasing emission from the pre-formed DDHPDLC grating cell. 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, a repetition rate of 10Hz, and a pulse energy, U, is focused by a lens (focal length: f=20cm) on the DDHPDLC grating at an incident angle of +45° from the cell normal (N). A half-wave plate (λ/2 for 532 nm) and a polarizing beam splitter (PBS) are placed in front of the lens to vary the incident pulse energy. A fiber-optic probe of a fiber-based spectrometer (USB2000, Ocean Optics, optical resolution:~1.4nm) is placed to face the edge surface of the cell to record the lateral fluorescence (or lasing) output along the grating vector G. An analyzer (A) is placed in front of the spectrometer probe to divide the optical field of the fluorescence output into two mutually perpendicular in-plane and out-of-plane components (E IP and E OP), which are parallel and perpendicular to the cell plane, respectively. One CW CP green beam (from a diode-pumped solid-state laser, λG: 532nm, power≤1W) and one CW CP red beam (from an He-Ne laser, λR=633nm, power≤35mW) are installed to illuminate or not illuminate the pulse-excitation region in the grating at incident angles of roughly 180° and −45°, respectively, from N (Fig. 1(b)) to provide the all-optical controllability of the generated lasing emission from the grating cell. The experimental absorption and fluorescence emission spectra of the cell (not shown herein) show that the maxima of the absorption and fluorescence emission are at about 487 and 595nm, respectively. When the wavelength exceeds 610 (720) nm, the absorption (fluorescence emission) is so weak that it is negligible. Since the absorbance of the cell at 532nm is about a half of the maxima at 487nm, the incident pulses can efficiently excite the laser dyes in the cell.
3. Results and discussion
Before the all-optically controllable lasing experiments is performed, the energy threshold of the incident pulses required to generate the lasing emission from the pre-formed DDHPDLC grating must be determined. In this stage, both CW green and red beams are pre-shut off and the A is pre-removed. Figure 2(a) plots the nonlinear variations of the measured lasing intensity and the corresponding full widths at half-maxima (FWHM) with incident pulse pumped energy. An energy threshold (UTH) exists at UTH=13μJ/pulse, which is indicative of lasing. Figure 2(b) presents the obtained lasing signal (black peak) at almost 620nm with U=25μJ/pulse and the corresponding reflection spectrum (red curve) of the grating cell. Clearly, the lasing occurs at nearly the long-wavelength edge of the reflection band, which finding is consistent with photonic band-edge lasing theory for a DFB resonator of a one-dimensional photonic crystal . The FWHM of the lasing signal in Fig. 2(b) is ~1.4nm, which is close to a value found elsewhere  for a similar DDHPDLC system (without azo dye). The inset in Fig. 2(b) presents the photographed pattern of obtained lasing emission at U=25μJ/pulse on a screen.
In the subsequent all-optically controllable lasing experiments, the pumped energy of the incident pulses is fixed at U=25μJ/pulse. The CW green beam with a variable intensity (IG) is turned on to pre-illuminate the cell for tG=1min and then turned off. After five minutes, the grating cell is excited by the incident pumped pulses to generate the lasing output. This sequence is repeated at different IG. The black, red, blue, and green spectra in Fig. 3(a) (Fig. 3(b)) are the obtained lasing spectra for E IP (E OP) at IG=0, 100, 200, and 300mW/cm2, respectively. Clearly, these experimental spectra demonstrate that the pre-illumination of the CW green beam markedly influences the intensity of the generated lasing emission from the grating cell. Additionally, the lasing emission is highly anisotropic at IG=0 and the lasing intensity for E IP (and therefore the lasing anisotropy) declines significantly as IG increases. Such a high anisotropy of lasing emission has been observed elsewhere [6,8] and explained elsewhere . The most likely cause of this anisotropy is the anisotropic orientation of the LC-droplets in the LC-droplet-rich regions of the grating. The LCs in the droplets tend to align parallel to the grating stripes due to the shrinkage of the polymer matrix during the photopolymerization of the grating formation [11,12]. When the laser dyes have been excited by the pumped pulses, the spontaneously emitted fluorescence that propagates along G “sees” a DFB resonator with a spatially periodic distribution of refractive index through alternating droplet-rich and polymer-rich regions. The fluorescence associated with E IP and E OP experiences various differences between the refractive indices in the droplet-rich and polymer-rich regions, |n e-n p| and |n o-n p| (ideally), respectively, at IG=0. In a DFB grating laser with small index and gain modulations (n 1 and α1, respectively), the coupling strength κL can be defined, according to the coupling-wave model, as 
The coupling strength represents the measured degree of feedback of light with a wavelength of λ that is provided by the DFB structure of length L. The coupling caused by gain modulation is negligible (α1≈0 in Eq. (2)) in this study, based on the fact that the scattering from the nano-sized droplets is extremely weak and the assumption that the laser dyes in the grating are uniformly distributed. In the low-gain approximation (α<<κ), the threshold condition for the first resonances, which occur at the edges of the stop band, are given by Eqs. (2) and (3), as the index modulation n 1 and thus the coupling strength κL increases, the threshold gain αL falls. In this work, since |n e-n p|≅0.20 >> |n o-n p|≅0.03 at IG=0, the coupling strength (and therefore the threshold gain) for E IP is much larger (much lower) than that for E OP, leading to the strong anisotropy of lasing emission at IG = 0, displayed in Fig. 3.
The significant decline in the lasing intensity for E IP with increasing IG (Fig. 3(a)) can be attributed to the green-beam-induced isothermal N→I phase transition of LCs by the trans→cis isomerization of D2-dyes in the droplet-rich regions of the grating [14–16], which are shown in the grating model in Fig. 4 . Generally, in the dark, the D2-isomers are stable in the rod-like trans-state, in which the chemical structure is as elucidated elsewhere . The guest-host effect causes the dyes in nematic LC droplets to become aligned with the LCs. The absorption spectra obtained in the authors’ earlier investigation  indicate that trans-dyes absorb no light in the red region. Once the cell is pre-excited by the CW green beam, the D2-dyes may transform to curve cis-isomers and then disturb the order of the LCs, reducing the orientation anisotropy of the LCs in droplets. As IG increases from 0 to 300mW/cm2, the concentration of the cis D2-dye increases to a sufficiently high value to cause the original nematic phase of LC-droplets in the droplet-rich regions to enter the isotropic phase isothermally. During this phase transition, the index modulation for E IP declines remarkably from |n e-n p|≅0.20 to |n i-n p|≅0.05, where n i (~1.6) is the refractive index of the LCs in the isotropic phase in the droplet-rich regions. Hence, the coupling strength is substantially reduced, causing considerable decay in the lasing intensity for E IP. For E OP, the index modulation varies over a range of small values from |n o-n p|≅0.03 to |n i-n p|≅0.05 during the phase transition, in which the coupling is very weak, such that the lasing intensity remains almost zero as IG increases from 0 to 300mW/cm2. Notably, the inset in Fig. 3(b) plots the variation in the obtained lasing spectrum of the grating with IG when analyzer A is removed. The optically controllable lasing signal (as shown in the inset in Fig. 3(b)) comprises almost entirely that for the E IP-component.
The following three-stage experiment is performed to study the inverse optical controllability of the DDHPDLC laser by illumination with the CW red beam. Figure 5 plots the associated experimental results. First, the CW green and red beams are both pre-turned off (IG=IR=0). After the grating cell has been excited by the pumped pulses with U=25μJ/pulse, a strong lasing signal is generated (black curve in Fig. 5). Second, the CW green beam with IG = 300mW/cm2 is turned on to illuminate the cell for tG=1min (IR=0) and then turned off. After five minutes, the pumped pulses are used to excite the cell to generate weak fluorescence (not lasing) emission (red curve in Fig. 5). Third, the CW green beam is kept off (IG=0), and the CW red beam with IR=1500mW/cm2 is turned on to irradiate the grating cell for tR. After five minutes, the grating is excited by pumped pulses with the same U=25μJ/cm2. The third stage is repeated at different tR= 4, 8, and 16min; various lasing spectra are thus produced (blue, green, and pink curves, respectively, in Fig. 5). Experimental results in Fig. 5 show that the intensity of the lasing output can be increased with increasing tR. The mechanism of the inverse optical control of the lasing emission is attributable to the red-beam-induced isothermal I→N phase transition of LCs via the cis→trans back isomerization of D2-dyes in the LC-droplets in the droplet-rich regions of the grating [14–16], as displayed in the grating model in Fig. 4. As stated above, when sufficient D2 dyes absorb green light, causing them to enter cis-state, the LCs in the droplet-rich regions of the grating can be disturbed to become isotropic. As shown in the authors’ earlier work , the D2 cis-isomers exhibit new n-π* type active transitions in the red region. As the cell is, instead, stimulated by the red beam for an increasing irradiation time, the concentration of D2 dyes that transform back to the trans-state may increases, such that the LCs in the LC-droplets in droplet-rich regions of the grating gradually return isothermally to the nematic phase. Such an I→N phase transition significantly increases the index modulation and thereby the coupling strength in the DFB grating for E IP, resulting in the significant recovery of the lasing emission, as revealed by the experimental results in Fig. 5.
The Bragg diffraction efficiency of the grating as a function of time is recorded using a nonpolarized He-Ne laser beam (1mW/cm2) to probe the DDHPDLC grating formation. Experimental results (data not shown) show that the diffraction efficiency increases monotonically with increasing writing time and saturates at a steady value of approximately 50% after the writing time exceeds 2.5 min. Afterwards, the grating cell is irradiated by the CW green beam with IG=300mW/cm2 for 1min. The measured diffraction efficiency decays to roughly 25%. This experimental result indicates that the decay of the diffraction efficiency and thus the effective index contrast of the grating reflect to the decay of the lasing intensity of the grating by the green-beam-induced isothermal N→I phase transition via trans→cis isomerization in the droplet-rich regions of the grating.
The controllability of the lasing intensity based on a DDHPDLC grating cell can also be realized by heating the cell. Associated results can be found in Ref. 8. The mechanism of the controllability of the lasing emission dominated in Ref. 8 is the thermal effect induced phase transition of the LCs in the droplet-rich regions of the grating. When the temperature of the cell increases from room temperature to clearing point of the LCs in the droplets, the index contrast and thus the lasing emission of the grating can significantly decay. To demonstrate that the optical controllability of the lasing emission from the DDHPDLC grating in the present work is not caused by the thermally induced phase transition of the LC-droplets in the droplet-rich regions of the grating when illuminated by the CW green beam, a separate experiment, in which the temperature at the spot of the grating cell that is exposed to the green beam is measured, is performed using a thermal imager (Fluke, Ti20). Before irradiation by the green beam, the measured temperature of the grating cell is around 22.3°C. Next, the green beam with IG=300mW/cm2 irradiates the grating cell for 1min and is then turned off. Thereafter, the temperature of the irradiated spot of the grating is measured to relax back to 22.3°C within 2min. This experimental result indicates that the aforementioned phase transition and thereby the optical controllability of the laser is certainly not caused by the laser-induced thermal effect but by photoisomerization.
To our knowledge, this study is the first to develop and examine an all-optically controllable DFB laser that is based on a DDHPDLC grating with a photoisomerizable dye. Experimental results demonstrate that increasing the irradiation intensity (irradiation time) of one CW CP green (red) beam on the grating reduces (increases) the intensity of the lasing emission. The all-optical controllability of the laser is attributable to the green-beam-induced isothermal N→I phase transition via trans→cis isomerization and the red-beam-induced isothermal I→N phase transition via cis→trans back isomerization in the LC-droplets of droplet-rich regions of the grating. The former (latter) may significantly decrease (increase) the index contrast and thereby the coupling strength in the DFB grating, resulting in the decay (rise) of the lasing output from the grating. The laser-induced thermal effect cannot be responsible for the optical controllability of the DDHPDLC grating laser. Such an all-optically controllable DFB laser device may be utilized in integrated photonics.
The authors would like to thank the National Science Council of the Republic of China, Taiwan (Contract No. 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. Ted Knoy is appreciated for his editorial assistance.
References and links
1. G. S. He, T.-C. Lin, V. K. S. Hsiao, A. N. Cartwright, P. N. Prasad, L. V. Natarajan, V. P. Tondiglia, R. Jakubiak, R. A. Vaia, and T. J. Bunning, “Tunable two-photon pumped lasing using a holographic polymer-dispersed liquid-crystal grating as a distributed feedback element,” Appl. Phys. Lett. 83(14), 2733–2735 (2003). [CrossRef]
2. D. E. Lucchetta, L. Criante, O. Francescangeli, and F. Simoni, “Wavelength flipping in laser emission driven by a switchable holographic grating,” Appl. Phys. Lett. 84(6), 837–839 (2004). [CrossRef]
3. R. Jakubiak, L. V. Natarajan, V. Tondiglia, G. S. He, P. N. Prasad, T. J. Bunning, and R. A. Vaia, “Electrically switchable lasing from pyrromethene 597 embedded holographic-polymer dispersed liquid crystals,” Appl. Phys. Lett. 85(25), 6095–6097 (2004). [CrossRef]
4. S.-T. Wu and A. Y.-G. Fuh, “Lasing in photonic crystals based on dye-doped holographic polymer-dispersed liquid crystal reflection gratings,” Jpn. J. Appl. Phys. 44(2), 977–980 (2005). [CrossRef]
5. V. K. S. Hsiao, C. Lu, G. S. He, M. Pan, A. N. Cartwright, P. N. Prasad, R. Jakubiak, R. A. Vaia, and T. J. Bunning, “High contrast switching of distributed-feedback lasing in dye-doped H-PDLC transmission grating structures,” Opt. Express 13(10), 3787–3794 (2005). [CrossRef] [PubMed]
6. Y. J. Liu, X. W. Sun, P. Shum, H. P. Li, J. Mi, W. Ji, and X. H. Zhang, “Low-threshold and narrow-linewidth lasing from dye-doped holographic polymer-dispersed liquid crystal transmission gratings,” Appl. Phys. Lett. 88(6), 061107 (2006). [CrossRef]
7. S. J. Woltman, M. E. Sousa, H. Zhang, and G. P. Crawford, “Survey of switchable lasing configurations using structures of liquid crystal and polymer dispersions,” Proc. SPIE 6135, 61350B (2006). [CrossRef]
8. Y. J. Liu, X. W. Sun, H. I. Elim, and W. Ji, “Effect of liquid crystal concentration on the lasing properties of dye-doped holographic polymer-dispersed liquid crystal transmission gratings,” Appl. Phys. Lett. 90(1), 011109 (2007). [CrossRef]
9. R. L. Sutherland, V. P. Tondiglia, L. V. Natarajan, T. J. Bunning, and W. W. Adams, “Electrically switchable volume gratings in polymer-dispersed liquid crystals,” Appl. Phys. Lett. 64(9), 1074–1076 (1994). [CrossRef]
10. P. Dowling, M. Scalora, M. J. Bloemer, and C. M. Bowden, “The photonic band edge laser: A new approach to gain enhancement,” J. Appl. Phys. 75(4), 1896–1899 (1994). [CrossRef]
11. 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]
12. J. Qi, M. DeSarkar, G. T. Warren, and G. P. Crawford, “In situ shrinkage measurement of holographic polymer dispersed liquid crystals,” J. Appl. Phys. 91(8), 4795–4800 (2002). [CrossRef]
13. H. Kogelnik and C. V. Shank, “Coupled-wave theory of distributed feedback lasers,” J. Appl. Phys. 43(5), 2327–2335 (1972). [CrossRef]
14. A. Urbas, V. Tondiglia, L. Natarajan, R. Sutherland, H. Yu, J.-H. Li, and T. Bunning, “Optically switchable liquid crystal photonic structures,” J. Am. Chem. Soc. 126(42), 13580–13581 (2004). [CrossRef] [PubMed]
15. H.-C. Yeh, G.-H. Chen, C.-R. Lee, and T.-S. Mo, “Photoinduced two-dimensional gratings based on dye-doped cholesteric liquid crystal films,” J. Chem. Phys. 127(14), 141105 (2007). [CrossRef] [PubMed]
16. H.-C. Yeh, G.-H. Chen, C.-R. Lee, and T.-S. Mo, “Optically switchable biphotonic gratings based on dye-doped cholesteric liquid crystal films,” Appl. Phys. Lett. 90(26), 261103 (2007). [CrossRef]