Wide-angle lasing from photonic crystal nanostructures of a liquid-crystalline blue phase†

Guan-Yu Zhuo ^ab, Shu-Wei Huang ^c and Shih-Hung Lin *^de
aInstitute of New Drug Development, China Medical University, No. 91, Hsueh-Shih Rd., Taichung 40402, Taiwan, Republic of China
^bIntegrative Stem Cell Center, China Medical University Hospital, No. 2, Yude Rd., Taichung 40447, Taiwan, Republic of China
^cDepartment of Physics, National Sun Yatu-sen University, No. 70, Lienhai Rd., Kaohsiung 80424, Taiwan, Republic of China
^dDepartment of Optometry, Chung Shan Medical University, No. 110, Sec. 1, Jianguo N. Rd., Taichung 40201, Taiwan, Republic of China
^eDepartment of Ophthalmology, Chung Shan Medical University Hospital, No. 110, Sec. 1, Jianguo N. Rd., Taichung 40201, Taiwan, Republic of China. E-mail: shihhung@csmu.edu.tw; Tel: +886-4-24730022 ext. 12140

First published on 26th April 2019

---

---

1. Introduction

Liquid crystalline materials with controllable periodic optical nanostructures have received significant attention in photonic applications such as in phase grating systems, tunable reflectors and filters, and small-sized tunable lasers.^1–4 Among these materials, blue-phase liquid crystals (BPLCs) provide a high degree of freedom for structural modifications based on various types of tuning mechanisms. Furthermore, BPLCs have been found to exhibit faster electro-optical responses than nematic liquid crystals (NLC), and no surface rubbing is needed for the LC alignment, opening a new research area in photonic engineering and device fabrication. BPLCs are formed when NLCs are blended with high content of a chiral dopant, wherein the accompanied helical twisting power (HTP) exerted on the NLC molecules forces them to be aligned with two orthogonal helical axes; this results in the formation of a double-twist cylinder (DTC).^5–9 Moreover, the DTCs self-assemble into a three-dimensional (3D) cubic structure with lattice period on the scale of several hundred nanometers. In this unique structure, light waves of certain wavelengths cannot survive inside, showing discontinuities due to Bragg condition, which is identical to the case of a photonic band gap (PBG);^10–15 instead, the group velocity of the light waves at a specific wavelength satisfying the Bragg condition becomes zero at the PBG edge, and therefore, the photons propagate back and forth inside; this results in the generation of standing waves that strengthen the light reflection. Therefore, the BPLC has a structure equivalent to that of a photonic crystal, acting as a nanocavity to reflect visible wavelengths; upon further addition of laser dyes as gain media to BPLCs, lasing emission can be produced at sufficient pumping light energy.^6,15–21

BPLCs exist across the phases of cholesteric LC (CLC) and isotropic LC (ILC) over the limited temperature range of 2–8 K and have at least one of the following fluidic lattice structures: BPI, BPII, or BPIII, which can be easily adjusted by external perturbations, such as temperature and electric and optical fields, via a change in their lattice parameters;^22–25 to date, in most of the applications, BPI and BPII are employed, which have the lattice parameters of approximately 200–400 nm and can produce Bragg reflection in the UV and visible spectral domains. Considering the BPLC displays as an example, the PBG can be manipulated to selectively reflect visible light using an electric field and then exhibit various colors;^26–28 on the other hand, the 3D PBG structure of the BPLC corresponds to three mutually orthogonal axes, from which the lasing emission propagates when a mirror-less liquid crystalline BP laser is used.^{6,7,15–21,29,30}

On reviewing relevant studies reported in the last decade on Bragg light reflection and lasing emission from BPLCs, it was found that Mazzulla et al. used a common BP sample, which was sandwiched between two indium tin oxide (ITO) glasses coated with rubbed polyvinyl-alcohol (PVA) layers for the homogeneous alignment of LC molecules, to demonstrate the adjustment of both the reflection and lasing wavelengths using thermal and electric tuning;¹⁷ in this experiment, the reflection bands of BPI and BPII were separately manipulated at specific temperatures. The total reflection peak shifts were within the range of 70–80 nm. However, the tunability of the lasing emission band could not exceed 30 nm and neither be controlled by temperature nor electric field; the narrow tunable range was ascribed to the intrinsic property of the used BP materials and their cell geometry that only permitted a longitudinal electric field. After this, Choi's group accomplished two important studies, one on BPI¹⁹ and the other on BPII,²¹ mainly focusing on the extension of wavelength tunability. They further included bent-core molecules into the BP sample to stabilize the DCTs that could broaden the temperature range of BPI to 25 °C and that of BPII to 4 °C. Therefore, in the former case, the total reflection peak shift was up to 150 nm, whereas the total lasing peak shift was incomparably wide up to 115 nm upon using two laser dyes, almost covering the range from blue to red. In the latter case, the BP sample treated by surface rubbing (contrary to the former case) showed decreased lasing threshold and a high and constant Bragg reflection; consequently, the total reflection peak shift is 40 nm, whereas the total lasing peak shift increases to 30 nm, which are the best recorded values reported to date for BPII. On the other hand, Lin et al. fabricated a wedge cell to produce a gradient lattice in space, which facilitated the continuous tuning of reflection and lasing emission band by changing the pumped position of the sample at specified temperature.²⁰ The presented results showing a ∼133 nm reflection peak shift and a ∼68 nm lasing peak shift were significantly better than the results of the CLC and DDCLC wedge cells. In addition, this spatial tuning provides various advantages, including fast response time, simultaneous multi-wavelength lasing emission, and high stability without any irreversibility of the tuning mechanisms, as compared to the abovementioned examples. To the best of our knowledge, most of the reported results, are associated with temperature tuning, whereas only a few results have been reported on electric tuning; this may be attributed to the fact that the geometry of parallel-plate glass cell induces a slight birefringence Δn on BPLC (only causes a 20–30 nm total reflection/lasing peak shift) followed by a longitudinal electric field between two ITO glasses. Although temperature tuning can prevent this problem, it is hard to incorporate the BP sample in display technologies due to its long response time that is dependent on the tuning rate of a temperature-controlled stage (∼1 °C min⁻¹). Moreover, the wedge cell is not compatible for display technology due to its geometry and the movement of the pumped spot on the BP sample that cause many complexities in the display or device configuration. The abovementioned problems eventually reduce the possibilities of formation of a high-speed, wavelength-tunable photonic device.

More recently, two pioneering studies reported by Yang's group have addressed the abovementioned issues.^29,30 In an earlier experiment,²⁹ they used a polymer-stabilized BPI film to modulate the PBG structure by the polarity and magnitude of a DC bias and, in turn, control the reflection band shift towards blue (negative bias) or red (positive bias). The main goal of using a cross-linked polymer network was to broaden the temperature range of the BP (∼200 °C) as well as provide flexibility, thermal stability, and wideband tunability of Bragg reflection (∼200 nm) to the used sample. The mechanism of wavelength tuning is referred to as the electromechanical deformation (compression or expansion) of the BP lattice resulting from the displacement of the charged polymer network when subjected to a directional electric force. Furthermore, the results show that the response time is around 5 s in the field on state, which is faster than those in the abovementioned examples. Moreover, the wavelength tunability range of the lasing emission was around 20 nm. After this, they further devised a porous polymer scaffold to template a BP laser,³⁰ which was used to deal with the problems of interference, insolubility of doped compounds and thermal sensitivity of the BP lattice structure when other functional materials were included in the BP sample. Great achievements including strong thermal stability when the temperature was higher than 90 °C, wideband tunability of Bragg reflection (∼150 nm) with high reflectivity, and faster wavelength switching (on time: <5 s; off time: <0.5 s) were achieved. Especially, the wavelength tunability range of the lasing emission was increased to 48 nm. Due to the excellent optical performance of this hybrid intelligent material, it provides vast probabilities for implementation in optoelectronic devices.

Based on the various abovementioned breakthroughs, in this study, we developed a new configuration by combining in-plane switching (IPS) electrodes with a BP sample to demonstrate wide-angle lasing emission. Wavelength tuning was controlled by an electric field at various detection angles (i.e., the angle between the detector and the surface normal of the BP sample, as shown in Fig. 1). The external electric field results in the orientation distribution of the BP lattice, which is stretched by the electrostriction effect along the electric field direction;^1,31 due to the high electro-optical properties of the IPS electrodes that enhance the electric field in the transverse direction, this BP sample exhibits a large reflection/lasing peak shift, fast wavelength tuning, and spatial emission control, which can be a possible fit for integration into tunable photonic devices in the future.

2. Experimental

2.1 Fabrication of the BP sample and IPS electrodes

The sample was prepared following a routine procedure, in which two ITO glass substrates were sandwiched between spacers to provide the cell gap of 27 μm. Regarding the IPS electrodes, the first step was to spin-coat the photoresist onto ITO glasses, and then, a photomask combined with UV light was used for exposure. After this, the ITO glasses were immersed in contrast media, where the illuminated areas on the photoresist were etched. The pattern of the photomask was left on the ITO glasses, which was then processed with ICP (Inductive Couple Plasma)-Etcher to remove the layers of the photoresist and ITO. Finally, the IPS electrodes were formed. The width of the electrode was 25 μm, and the spacing between the sequential positive and negative electrodes in the BP sample was 15 μm. Note that this electrode geometry resulted in a transverse electric field. By contrast, to produce a longitudinal electric field, in the BP sample, the IPS electrode is replaced by a piece of cleaned ITO glass.

2.2 Measurement of the Bragg reflection and lasing spectra of the BP sample

The measurement setup for the Bragg reflection spectrum of the BP sample is illustrated in Fig. 1a. A white light source was coupled into a fiber, which was divided into input and output ports. The input port was further split into two parts, one connected to a spectrometer (USB4000, Ocean Optics SpetraSuite) and another connected to the white light source. The output port of the fiber was treated as a detector. When the white light was launched on the sample, the reflected light was collected by the spectrometer through the fiber. The sample was placed on a temperature controlled stage (LT120, LinkamScientific Instruments) at a certain angle relative to the propagation direction of light, which prevents the reflected light from directly damaging the fiber and spectrometer. Then, the dependency of the reflection bands on applied voltage was determined for different detection angles. To demonstrate an electrically tunable wide-angle lasing from a DDBP sample, the white light source was replaced by a frequency-doubled Nd:YAG pulse laser (532 nm, linear polarized, 6 ns, 10 Hz, 100–140 μJ per pulse for lasing emission), and a neutral density (ND) filter was used for laser power control, as shown in Fig. 1b. Except for a few changes in the optical components, the lasing spectrum of the BP sample was measured following the same procedure as that used for the reflection spectrum.

3. Results and discussion

To find an optimal BP sample, a chiral mixture was selected from three different combinations: the host NLCs were ZTO-5024LA (Chisso), ZTO-5059LA (Chisso), and HTW114200-100 (Fusol), whereas the chiral dopant was S811 (Fusol). Specific mass concentration of the abovementioned chemicals in these mixed systems is described in ESI.† The main reason for choosing the abovementioned NLC candidates is that they easily form BP without the surface rubbing treatment on glass substrates, have large platelet domains on a hundred micrometer scale instead of the temperature range of BP (ZTO-5024LA: ∼2 °C; ZTO-5059LA: ∼7 °C; and HTW114200-100: ∼5.5 °C). The large platelets stabilize the PBG structure, facilitating Bragg light reflection and further enhancing the lasing efficiency (i.e., multiple scattering and random lasing from small platelets are reduced¹⁸). On the other hand, they allow building of suitable short pitch CLC mixtures, wherein BP occurs during the phase evolution process as the temperature of the sample changes. The relationships between the Bragg reflection peak wavelength and temperature for the three mixtures are shown in ESI,† Fig. S1. To investigate the lattice structure of BPLC and the color variations in the multi-platelet domains under an electric field (AC, 1 kHz), the three different samples were separately mounted on the temperature controlled stage of a polarized optical microscope (POM) (Axio Scope.A1, Zeiss). Initially, the temperature was increased for the BPLC to undergo a CLC-ILC phase transition and then gradually lowered down at the rate of 0.01 °C min⁻¹ until BP appeared. Note that this temperature-cycling process is routinely used to confirm the existence of BP before measurements. The POM images were obtained based on the results of ZTO-5024LA+S811, ZTO-5059LA+S811, and HTW114200-100+S811, as shown from left to right in Fig. 2a, which confirm that these systems certainly have BP due to the platelet textures. It was observed that ZTO-5024LA-S811 had the largest platelets of several hundred micrometers with homogeneous colors. Therefore, it was expected to facilitate electric tuning and exhibit a larger peak value in a comparatively uniform spectral profile of the reflection spectrum than the other two mixed systems.

To measure the Bragg reflection peak shift under an electric field at various detection angles, a reflection spectroscopic system was used (Fig. 1a). Fig. 2b illustrates the corresponding peak shifts at different applied voltages, where the peak shifts from ∼570 nm (zero voltage) to 593.81 nm, 589.35 nm, and 591.47 nm on applying 2.6 V μm⁻¹, 3 V μm⁻¹, and 3.4 V μm⁻¹, with the total reflection peak shifts of about 22 nm, 21 nm, and 22 nm, respectively. Fig. 2c shows the reflection peak wavelength plotted as a function of applied voltage for each sample. Note that the applied voltage should be limited at a specific value to prevent unwanted BP-CLC phase transitions. This manifests that the electro-optical properties of ZTO-5024LA are better than those of the other two systems because it requires less electric field strength to obtain the same amount of reflection peak shift, in response to the observation of Fig. 2a; furthermore, the spectral profile of the reflection spectrum of ZTO-5024LA-S811 is sharper than that of the other two systems; this means that ZTO-5024LA-S811 has a stronger band-edge effect on the photonic stop band;¹⁵ this enables a comparatively effective light reflection at the PBG edge (more than 50% reflectivity) and lasing emission when laser dyes are included. Herein, we chose ZTO-5024LA-S811 to demonstrate wide-angle lasing from a liquid crystalline BP.

The BP sample ZTO-5024LA-S811 with IPS electrodes was then prepared for the measurement of Bragg reflection spectra at the detection angles of 0°, 20°, 40°, and 60°. At the starting temperature of BP (50.2 °C), color changes in the BP lattice were observed using a POM according to different applied voltages, as shown in Fig. 3a. The lattice color was pink without an electric field, and it gradually changed to orange and green upon increasing the electric field until the BP collapsed (50 V) and then evolved into the CLC phase. The reflection spectra obtained at the 0° detection angle are shown in Fig. 3b, where the peak wavelength is at 628.9 nm without an electric field and blue-shifts with an increase in the electric field strength. The total reflection peak shift was about 68 nm, and the peak value decreased with an increase in the electric field strength before the BP-CLC phase transition. Under the same experimental condition, the results measured at 20°, 40°, and 60° detection angles are shown in Fig. 3, as well as a summarized table (Table 1) is provided. Note that the stopband of a single lattice plane in BPLC should vary with the observing angle due to the difference in the conditions for Bragg reflection. However, the BP sample in our case does not require surface rubbing for LC alignment; this provides the probability for various lattices to self-assemble in the same lattice plane in an arbitrary direction; therefore, it is possible that they reflect the same wavelength of light at different observing angles when the applied voltage is zero. Fig. 3f demonstrates the dependency of the reflection peak wavelength on the applied voltage for different detection angles. The threshold voltage of 16 V was found at the 0° detection angle, and it became larger with the increasing detection angle. The total reflection peak shifts were nearly similar at all detection angles. In other words, to obtain the same amount of reflection peak shift, the required voltage was positively dependent on the detection angle. A specific lattice plane of BPLC having different responses to the transverse electric field, which is constituted by IPS electrodes, can reflect different optical wavelengths. Hence, the desired wavelength can be extracted by choosing the corresponding detection angle.

---

The abovementioned phenomena can be explained by an intuitive illustration shown in Fig. 4, presenting the cross-section of a BP sample with the IPS electrodes with and without an electric field. In Fig. 4a, the arrows are normal to the BP lattice planes and directed towards the detector at various angles. This demonstrates that by considering the BP sample without surface rubbing for LC alignment, specific and single lattice planes of different DCTs have the chance to be equivalently investigated at different detection angles (i.e., the same Bragg condition). In the experiment, this phenomenon was uniquely observed only when the BP lattice was sufficiently large (several hundreds of micrometers), and a small core detection fiber was used to reduce the disturbance from other unwanted lattice planes. Fig. 2b (for ZTO-5024LA) and Fig. 3b–e show that most of the full-width-at-half maximum (FWHM) values of the spectral profiles are below 25 nm, and some of them have reflectivity higher than 50%. The performance of Bragg reflection in our experimental condition is as good as the previously reported results;^6,7,31 this evidences that our method has the capability of observation of light reflection from a single lattice plane. In Fig. 4b, the distribution of the transverse electric field between electrodes is depicted. The DTCs are reoriented and stretched under the influence of an electric field; this leads to lattice distortion as well as the variation of induced birefringence Δn. In more detail, two mechanisms are dominant in the abovementioned effects: one is the mechanical deformation of the BP lattice in response to the electric field constituted by the opposite charges on both electrodes, which can be elucidated by Maxwell theory and an electrostriction tensor.^24,32–34 Note that the direction of lattice deformation follows the direction of the electric field because the used LC molecules have positive dielectric anisotropy (Δε > 0). On the other hand, the other mechanism refers to an electric field-induced molecular reorientation effect, which can be explained by the Kerr model.^1,35 It is visualized as a virtual effect for electrostriction because the projection of the BP lattice facing the detector is elongated when the DCTs are rotated. Therefore, the Bragg condition is modified via the abovementioned electrostriction effects for the reflection peak shift, utilized for wavelength tuning. According to the orientation distribution of the lattice plane, at the 0° detection angle, the electrostriction effects are more significant in the transverse direction, and therefore, they result in a blue shift of the reflection peak. However, the electrostriction effects are weaker at non-zero detection angles; thus, stronger electric fields are required to achieve the same amount of reflection peak shift measured at the 0° detection angle.

Note that to obtain wide-range wavelength tunability by a transverse electric field, which results in the blue-shift of reflection peak, it is required to choose the lowest temperature point (50.2 °C) of BP, at which the lattice plane reflects the longest wavelength (629 nm), as the starting point to manipulate the reflection band. However, the longitudinal electric field produced by just removing the IPS electrodes results in the red-shift of reflection peak as the BP lattice is stretched in the longitudinal direction. Therefore, to obtain wide-range wavelength tunability, the measurement has been started at the highest temperature point (52 °C) of BP, at which the lattice plane reflects the shortest wavelength (572 nm), as shown in Fig. 2b (for ZTO-5024LA). Consequently, the Bragg reflection peak could be manipulated by the longitudinal electric field to achieve a red shift and by the transverse electric field to achieve a blue shift.

For the demonstration of electrically tunable wide-angle lasing in a DDBP sample, we used a frequency-doubled Nd:YAG pulse laser to excite the sample with IPS electrodes. Typically, 1 wt% of a laser dye (gain medium), PYRROMETH-597 (Exciton), was added for lasing action, and the absorption and emission spectra are shown in ESI,† Fig. S2. Without an electric field, the sample gradually cooled down from the ILC phase at the rate of 0.01 °C min⁻¹ to the BP (50.2 °C). Because the wavelength at the PBG edge is capable of producing lasing emission, we should first verify whether the Bragg reflection spectrum overlaps with the emission spectrum of the laser dye. In our experiments, a neutral density (ND) filter was used to linearly control the strength of the excitation power, and in turn, the lasing threshold was measured. Fig. 5a shows the threshold behavior for the lasing action measured at 0° detection angle, where the threshold energy of the excitation laser is determined to be 90 μJ per pulse. Fig. 5b reveals the lasing phenomenon measured at the 0° detection angle. Note that without an electric field, the reflection peak was at 623 nm, whereas the lasing peak was at 619.5 nm, which was near the PBG edge. The lasing peak was blue-shifted to 602.9 nm upon applying the voltage of 26 V. As expected, the higher the voltage applied, the farther the lasing peak blue-shifted. The total lasing peak shift was about 54 nm when the voltage was increased to 37 V. In reality, the wavelength tunability range can be further extended by increasing the voltage; however, the edge of the fluorescence emission spectrum is too weak to produce lasing emission. In the same manner, the results measured at the detection angles of 20°, 40°, and 60° are shown in Fig. 5c–e, respectively, which are tabulated in Table 2. Fig. 5f shows the wavelength tunability presented as a function of the applied voltage. The trend is in accordance with that measured for the reflection spectra, as shown in Fig. 3f, both manifesting that to obtain the same amount of wavelength peak shift, higher applied voltages are required at larger detection angles. Moreover, the BPG as well as the reflection/lasing peak can be adjusted using the electric field, and the desired wavelength can be extracted at an appropriate detection angle.

---

4. Conclusions

In summary, a host NLC, ZTO-5024LA, with high electro-optical properties and a chiral dopant, S811, were selected for fabricating a BP sample. Because the PBG as well as lattice parameters of the BP sample could be changed by the so-called electrostriction effect, Bragg reflection peak shifts based on the applied voltage measured at various detection angles were produced. After this, by integrating IPS electrodes into the BP sample, the widest tunability range of approximately 70 nm in the reflection spectra was achieved, which was larger than those obtained with a longitudinal electric field (∼22 nm) without using the IPS electrodes. The lattice distortion caused by the transverse electric field had the strongest effect at 0° detection angle, which required less electric field strength to reach the reflection peak shift equivalent to that measured at the non-zero detection angles. Electrically tunable wide-angle lasing was realized in the DDBP sample with the IPS electrodes, from which specific wavelengths were emitted at different detection angles that could be controlled through electric tuning of the PBG structure. The lasing peak was tunable over the 55 nm range for each detection angle, which was comparable to a previous study in which a single laser dye was used.¹⁹ In the near future, to expand the functionalities of the proposed method, we will incorporate polymer materials into the BP sample to extend the temperature range of BP, making it stable to prevent unwanted phase transitions due to overvoltage breakdown and reduce the driving voltage as well.^{15,31,35–37} It is therefore a great opportunity to widen the wavelength tunability range of both the Bragg reflection and the lasing spectra. Furthermore, it is possible to devise the BP sample using a proper electrode width, gap, and organizational structure for the extension of wavelength tunability and reduction of the driving voltage. Electro-responsive lasing at different angles is important in simultaneous multi-wavelength illumination, spectral-resolved measurements, 3D tunable lasers, etc.; the similarity among the abovementioned possibilities is that the lasing emission at specific wavelength can be extracted at an appropriate angle and then guided to the examination site or instrument controlled by the applied voltage. Although this experimental configuration for lasing emission is at the initial stage, it has the potential to be used in integrated photonic applications due to the simplicity of sample fabrication and compatibility to sample geometry.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

The authors gratefully acknowledge the financial support provided by the Ministry of Science and technology, Taiwan (MOST 107-2112-M-039-001, MOST 106-2112-M-040-001-MY2 and MOST 107-2112-M-040-001), and appreciate Prof. Chie-Tong Kuo for the support of useful instruments for the preliminary tests.

References

1. J. Yan, Y. Li and S.-T. Wu, Opt. Lett., 2011, 36, 1404–1406 CrossRef PubMed .
2. T. J. White, R. L. Bricker, L. V. Natarajan, V. P. Tondiglia, L. Green, Q. Li and T. J. Bunning, Opt. Express, 2010, 18, 173–178 CrossRef CAS PubMed .
3. Y.-H. Chen, C.-T. Wang, C.-P. Yu and T.-H. Lin, Opt. Express, 2011, 19, 25441–25446 CrossRef CAS PubMed .
4. J. Beeckman, K. Neyts and P. J. M. Vanbrabant, Liquid-crystal photonic applications, SPIE, 2011 Search PubMed .
5. H.-S. Kitzerow and C. Bahr, Chirality in liquid crystals, Springer, New York, 2001 Search PubMed .
6. W. Cao, A. Muñoz, P. Palffy-Muhoray and B. Taheri, Nat. Mater., 2002, 1, 111–113 CrossRef CAS PubMed .
7. F. Castles, F. V. Day, S. M. Morris, D. H. Ko, D. J. Gardiner, M. M. Qasim, S. Nosheen, P. J. W. Hands, S. S. Choi, R. H. Friend and H. J. Coles, Nat. Mater., 2012, 11, 599–603 CrossRef CAS PubMed .
8. H. Kikuchi, M. Yokota, Y. Hisakado, H. Yang and T. Kajiyama, Nat. Mater., 2002, 1, 64–68 CrossRef CAS PubMed .
9. I. C. Khoo, Liq. Cryst. Rev., 2018, 6, 53–77 CrossRef CAS .
10. E. Yablonovitch, Phys. Rev. Lett., 1987, 58, 2059–2062 CrossRef CAS PubMed .
11. S. John, Phys. Rev. Lett., 1987, 58, 2486–2489 CrossRef CAS PubMed .
12. E. Yablonovitch and T. J. Gmitter, Phys. Rev. Lett., 1989, 63, 1950–1953 CrossRef CAS PubMed .
13. E. Yablonovitch, T. J. Gmitter, R. D. Meade, A. M. Rappe, K. D. Brommer and J. D. Joannopoulos, Phys. Rev. Lett., 1991, 67, 3380–3383 CrossRef CAS PubMed .
14. J. P. Dowling, M. Scalora, M. J. Bloemer and C. M. Bowden, J. Appl. Phys., 1994, 75, 1896–1899 CrossRef CAS .
15. S. Yokoyama, S. Mashiko, H. Kikuchi, K. Uchida and T. Nagamura, Adv. Mater., 2006, 18, 48–51 CrossRef CAS .
16. T. Isomura, H. Yoshida, A. Fujii and M. Ozaki, Mol. Cryst. Liq. Cryst., 2010, 516, 197–201 CrossRef CAS .
17. A. Mazzulla, G. Petriashvili, M. A. Matranga, M. P. De Santo and R. Barberi, Soft Matter, 2012, 8, 4882–4885 RSC .
18. C.-W. Chen, H.-C. Jau, C.-T. Wang, C.-H. Lee, I. C. Khoo and T.-H. Lin, Opt. Express, 2012, 20, 23978–23984 CrossRef CAS PubMed .
19. S.-T. Hur, B. R. Lee, M.-J. Gim, K.-W. Park, M. H. Song and S.-W. Choi, Adv. Mater., 2013, 25, 3002–3006 CrossRef CAS PubMed .
20. J.-D. Lin, S.-Y. Huang, H.-S. Wang, S.-H. Lin, T.-S. Mo, C.-T. Horng, H.-C. Yeh, L.-J. Chen, H.-L. Lin and C.-R. Lee, Opt. Express, 2014, 22, 29479–29492 CrossRef PubMed .
21. K. Kim, S.-T. Hur, S. Kim, S.-Y. Jo, B. R. Lee, M. H. Song and S.-W. Choi, J. Mater. Chem. C, 2015, 3, 5383–5388 RSC .
22. H. J. Coles and H. F. Gleeson, Mol. Cryst. Liq. Cryst. Inc. Nonlinear Opt., 1989, 167, 213–225 CrossRef CAS .
23. P. Etchegoin, Phys. Rev. E: Stat. Phys., Plasmas, Fluids, Relat. Interdiscip. Top., 2000, 62, 1435–1437 CrossRef CAS .
24. G. Heppke, B. Jérôme, H.-S. Kitzerow and P. Pieranski, J. Phys., 1989, 50, 549–562 CrossRef CAS .
25. A. Chanishvili, G. Chilaya, G. Petriashvili and P. J. Collings, Phys. Rev. E: Stat., Nonlinear, Soft Matter Phys., 2005, 71, 051705 CrossRef PubMed .
26. L. Rao, Z. Ge, S.-T. Wu and S. H. Lee, Appl. Phys. Lett., 2009, 95, 231101 CrossRef .
27. Y.-L. Song, G.-P. Li, F. Chu, H. Dou and Q.-H. Wang, Liq. Cryst., 2018, 45, 1535–1539 CrossRef CAS .
28. M. Jiao, Y. Li and S.-T. Wu, Appl. Phys. Lett., 2010, 96, 011102 CrossRef .
29. M. Wang, C. Zou, J. Sun, L. Zhang, L. Wang, J. Xiao, F. Li, P. Song and H. Yang, Adv. Funct. Mater., 2017, 27, 1702261 CrossRef .
30. M. Wang, C. Zou, C. Li, J. Sun, L. Wang, W. Hu, C. Zhang, L. Zhang, W. He and H. Yang, Adv. Opt. Mater., 2018, 6, 1800409 CrossRef .
31. S.-Y. Lu and L.-C. Chien, Opt. Lett., 2010, 35, 562–564 CrossRef CAS PubMed .
32. H. Stark and H. R. Trebin, Phys. Rev. A: At., Mol., Opt. Phys., 1991, 44, 2752–2755 CrossRef .
33. V. E. Dmitrienko, Liq. Cryst., 1989, 5, 847–851 CrossRef .
34. F. Porsch and H. Stegemeyer, Chem. Phys. Lett., 1989, 155, 620–623 CrossRef CAS .
35. Z. Ge, S. Gauza, M. Jiao, H. Xianyu and S.-T. Wu, Appl. Phys. Lett., 2009, 94, 101104 CrossRef .
36. C.-W. Chen, C.-C. Li, H.-C. Jau, L.-C. Yu, C.-L. Hong, D.-Y. Guo, C.-T. Wang and T.-H. Lin, ACS Photonics, 2015, 2, 1524–1531 CrossRef CAS .
37. M. D. A. Rahman, S. Suhana Mohd and S. Balamurugan, Sci. Technol. Adv. Mater., 2015, 16, 033501 CrossRef PubMed .

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c9tc01350d

This journal is © The Royal Society of Chemistry 2019