A broadban-tunable photonic bandgap and thermally convertible laser with an ultra-low lasing threshold from a refilled chiral polymer template†

Jia-De Lin ^a, Hsin-Yu Lin ^a, Guan-Jhong Wei ^a, Yu-Chou Chuang ^a, Lin-Jer Chen ^a, Ting-Shan Mo ^b and Chia-Rong Lee *^a
aDepartment of Photonics, National Cheng Kung University, Tainan 701, Taiwan. E-mail: crlee@mail.ncku.edu.tw
^bDepartment of Electro-Optical Engineering, Kun Shan University of Technology, Tainan 710, Taiwan

First published on 18th March 2019

---

---

Introduction

Cholesteric liquid crystals (CLCs) may self-assemble spontaneously into a helical structure if nematic liquid crystals (NLCs) are doped with chiral materials. This structure possesses a one-dimensional periodic distribution of a director that rotates along a helical axis; thus, a CLC can be regarded as a self-organized 1D photonic crystal with photonic bandgaps (PBGs).¹ Given its helical structure, CLC can selectively reflect a circularly polarized light with the same handedness as its helix. The central wavelength (λ_c) of the reflected light can be determined by multiplying the average refractive index (n_av) with the helical pitch (p) of the CLC. Thus, the PBG of the CLC can be tuned by adjusting the average refractive index or helical pitch under the action of external stimuli.^2,3 CLCs have attracted considerable attention in diverse applications, particularly for wide-band PBG tuning and low-threshold tunable laser devices, on the basis of their self-organization capability and high flexibility in tunability.^4–11

CLC lasers have gained significant attention over the last two decades because of their multiple advantages such as their spontaneously self-assembling helically periodic structure, large coherence area, potential for multidirectional emission, and the high tuning ability of the PBG.⁴ However, the thermal instability of fluidic liquid crystal (LC) molecules in CLCs can significantly distort the resonant structure and increase the energy threshold for lasing emission, which hinders the use of CLC lasers in practical applications. In recent years, LC templates have been adopted to fabricate LC lasers based on their high stability in the mimicked chiral polymeric scaffold, which provides a high-quality cavity for lasing emission.^12–15 Guo et al. proposed research on laser emission based on a CLC template by backfilling a dye-doped CLC (DDCLC) with opposite handedness.¹⁴ Given the hyper-reflectivity of the CLC template, a highly efficient lasing condition compared with conventional DDCLC lasers can be achieved. Muñoz et al. used a CLC template with a pitch (longitudinal) gradient across the cell gap. They were the first to demonstrate the continuous wave (CW) defect-mode lasing lasting for seconds, and its partial success can be attributed mainly to the stable structure provided by the template.¹⁵ Thus, the LC template technique has the potential to realize the fabrication of CW LC tunable lasers, which is the most challenging issue for future commercialization. Although the stable structure of a CLC template intrinsically limits the tunable properties of CLCs, the associated tunings of the PBG of the CLC templates can still be obtained through external stimuli, including temperature,^16,17 optical field,^18,19 and electric field.^20,21 Even so, the CLC templates still exhibit two or more of the following problems in tuning, such as irreversibility or hysteresis, long tuning time, low-quality tuning (including apparent deformation of PBG during tuning), high voltage, and thermal disturbance through the above-mentioned stimuli. The spatial tuning method is superior to the aforementioned methods because of its multiple advantages, such as a high stability from the invariance of the resonant structure and no need for the above-mentioned stimuli during spatial tuning, no irreversibility or hysteresis issues, time-saving capability, and wide spectral tunability.^22–24

This work in turn merges the advantages of the intrinsically stable template and the spatial tuning to develop spatially-tunable PBG devices and lasers based on dye-doped liquid-crystal-refilled (DDLC-refilled) gradient-pitched CLC polymer template samples. Experimental results show that the PBG can be spatially tuned over the entire visible region and the lasing is convertible thermally between single-mode and multi-mode for the DDLC-refilled CLC template sample. In addition, through so-called thermally reverse diffusion (TRD) and rapid annealing processing (RAP) during the washing-out/refilling process, the sample can be refined so as to possess superior properties such as linear spatial tunability and an ultra-low lasing threshold of ∼16 nJ per pulse (under a pumped diameter of ca. 250 μm), the magnitude of which is much lower than those in most DDCLC/quantum-dot-doped CLC (QDCLC) and polymer-stabilized DDCLC systems. Based on multiple merits such as high stability, the wide-band and wide operation region of the linear tunabilities, the ultra-low lasing threshold, and the thermal convertibility of the lasing mode, the refined template laser device has a high potential for applications in photonics and displays.

Experimental

Sample preparation.

The CLC–monomer materials used for fabricating the CLC polymer templates mainly include NLC, E7 (from Fusol-Material), right-handed chiral dopants, R811 and R1011 (both from Fusol-Material), a chiral monomer, RM691 (from Merck), an achiral diacrylate monomer, RM257 (from Merck) and a photoinitiator, Irg184 (from Pufeng). In the E7 host, the helical twisting power (HTP) values of R811, R1011, and RM691 are around 11.24 μm⁻¹, 37.7 μm⁻¹, and 3.73 μm⁻¹, respectively. The photoinitiators can absorb UV light to generate free radicals to trigger the chain polymerization process of RM691 and RM257, where the function of RM257 is to strengthen the crosslinking polymerization. The compositions of the right-handed CLC–monomer mixtures reflecting in the red and blue regions, labelled as the CLC-R and CLC-B mixtures, respectively, are listed in Table 1. The extra addition of 4.0 wt% R1011 into the CLC-B mixture can largely increase the resultant HTP value such that the PBG can blue-shift from the red to the blue region. The CLC-B and CLC-R mixtures are the original materials used to fabricate the gradient-pitched CLC polymer template, whose fabrication method is addressed in the next paragraph. The refilled material into the formed gradient-pitched template is the DDLC mixture, including LC of 99.5 wt% 5CB (from Sigma-Aldrich) and a laser dye of 0.5 wt% P597 (from Exciton). The reason for choosing 5CB, instead of E7, as the refilled material will be explained later.

---

Two kinds of gradient-pitched CLC polymer template samples are prepared with the template-made materials of CLC-R and CLC-B mixtures and the refilling material of DDLC. Each sample is fabricated by using a washing-out/refilling method including four stages as follows.

Experimental setup

A reflective optical fiber (R400-7-VIS-NIR, Ocean Optics) connected to a tungsten halogen lamp (LS-1, Ocean Optics) is employed to guide white light to normally irradiate the sample. The reflected light from the sample is detected by a fiber-optic probe of the same fiber connected to the spectrometer (Jaz-Combo-2, Ocean Optics, optical resolution: ∼0.9 nm) to measure the reflection spectra of the sample. The sample is placed at a hot-stage (PE94, Linkam) and the hot-stage is pre-fixed at a translation stage for being able to change the measured or pumped position on the sample along the horizontal direction (that is, along the x axis from x = 0 mm to x = 15 mm). In addition, a frequency doubled Nd-YAG pulse laser (LAB130-10, Spectra-Physics) with a wavelength of 532 nm, a repetition rate of 10 Hz, a pulse duration of 8 ns, and a pumped energy E is set as the pump source. The energy of the pump source can be adjusted by a combination of a half-wave plate and a polarizer. A non-polarizing beam splitter is employed to divide the pump beam into two sub-beams with identical energy. The transmitted pulse light is detected by the energy meter (PD10-C, from Ophir) for measuring the pumped energy, while the reflected one is focused on the sample by a lens (focal length f = 10 cm) with an incident angle of θ = 15° for generating the lasing emission from the sample. The diameter of the pumped region on the sample is around 250 μm. By moving the translation stage, the lasing emissions from various pumped positions from x = 0 to x = 15 mm along the pitch-gradient direction can be measured by the fiber-based spectrometer.

Results and discussion

Fig. 1(a) and (b) show the reflection spectra measured from x = 0 to x = 15 mm for the unrefined and refined samples at the after-refilling stage, respectively, (the corresponding reflection spectra at the before-curing and after-curing stages can be found in Fig. S2 in the ESI†). For comparison with Fig. 1(a) and (b), the characterized relations between the PBG and the position of the samples, represented by the relations between the central wavelength (λ_c) and the wavelengths at the short- and long-wavelength edges (SWE and LWE, respectively) of the PBG and the position (x) of the unrefined and refined samples, are plotted in Fig. 1(c) and (d), respectively, (the corresponding relations between λ_c and x at the before-curing and after-curing stages can be found in Fig. S3 in the ESI†). Apparently, both the obtained PBG gradients for the unrefined and refined samples at each stage cover the full visible region. The PBG gradients are observable from the corresponding reflective rainbow-like images of the samples which are attached above the PBG curves in Fig. 1. This result indicates that both the PBGs of the two samples can be spatially tuned in the entire visible spectrum. However, the spatial tuning abilities for the two samples are different. First, the PBG gradient [characterized by dλ_c(x)/dx] in the entire visible region and the related pitch gradient dp(x)/dx can be operated linearly in a limited spatial region for the unrefined sample, but they appear nearly uniform in a much wider region of operation for the refined sample in each stage (Fig. 1(c), (d) and Fig. S3, ESI†). Because of the significantly lower viscosity of the refilled LCs in the isotropic phase compared to that in the nematic phase, the chiral dopants during TRD in the refined sample are much easier to diffuse than during NRD in the unrefined sample. This reason causes the above-mentioned discrepancy between the spatial distributions of the pitch and the PBG in the unrefined and refined samples. Second, the texture perfection for the two samples are different. Fig. 2(a) and (b) present the reflective polarized optical microscope (R-POM) images at regions that gradually reflect red to blue light for the unrefined and refined samples at the before-curing stage, respectively. Apparently, the refined sample exhibits a nearly defect-free gradient-pitched CLC texture, in comparison to that lots of defects are found in the unrefined cell. A defect-free texture reflects a structure with a high order parameter of LC molecules in the refined sample. The refined sample undergoing the RAP at the before-curing stage is helpful for effectively eliminating the defects in the sample. These superior properties of the refined sample (e.g., wide linear tuning region and defect-free texture) are significantly beneficial for further applications, such as broadband and tunable low-threshold lasers.

Laser development is highly desirable based on the DDLC-refilled CLC polymer templates because of the merits of this system such as the high stability of the optical resonant cavity provided by the polymer template and the lack of predamage of the fluorescence dyes which are refilled into the template after the UV polymerization. Thus, this work develops a spatially tunable laser based on the refined DDLC-refilled gradient-pitched template laser. The following shows the associated experimental results on the lasing features at 22.4 °C and 44 °C. The refilled DDLC in the nanopores around the polymer networks of the refined template at 22.4 °C and 44 °C are pre-confirmed to be at the nematic and isotropic phases, respectively. Fig. 3(a) and (b) show the lasing emission spectra obtained at T = 22.4 °C and E = 0.40 μJ per pulse and at T = 44 °C and E = 0.12 μJ per pulse, respectively, at various positions (x = 8.0, 9.0, 9.5, 10.0, 10.5, 11.2, 11.7, 12.2, 12.7, and 13.5 mm), attached with the corresponding reflection spectra at those positions. Evidently, the appearances of the lasing emissions at the two temperatures are quite different. The lasing emission at each position of the refined laser has multi-peaks at 22.4 °C but only a single-peak at 44 °C. The reason for the difference of the lasing modes (single and multiple longitudinal modes or, simply, single- and multi-modes, respectively) at the two temperatures is discussed in the following paragraphs. The tunable lasing spectra for the multi- and single-mode lasing emissions is distributed from 561 nm to 656 nm and from 555 nm to 658 nm, respectively. The corresponding tunable spectral ranges are 95 and 103 nm, respectively, which both cover regions from green to red. Apparently, the lasing peak occurring at a certain position of the laser is just located at the SWE of the corresponding PBG at that position. This result is confirmed by comparing the lasing emissions and the reflection spectra of the refined sample presented in Fig. 3(a) and (b) at 22.4 °C and 44 °C, respectively. The correspondence of the spectral locations between the lasing peak and the SWE of the PBG at each position indicates that all lasing emissions occurring at various positions are attributable to the band-edge lasing mechanism, whether at 22.4 °C or 44 °C. Although band-edge lasing from CLCs usually occurs at the LWE of the PBG, the lasing in the present study occurs at the SWE due to two possible reasons. First, the fluorescence peak is better matched with the SWE.⁹ Second, the order parameter of the transition dipole moment in the DDLC (0.5 wt% of P597 in 5CB) of the present study was measured as 0.13 by the polarized fluorescence method.²⁶ The low order parameter of the transition dipole moment implies the orientations of the dye molecules in the nematic host do not prefer parallel or perpendicular positions to the long axes of LC molecules, which is considered as the possible reason for the lasing emissions at the SWE.²⁷ Furthermore, the following tests the handedness of the circular polarization for three lasing emissions with various peak wavelengths of the refined laser at three randomly-selected positions. Fig. 4(a) and (b) shows the obtained spectra of the three multi-mode (single-mode) lasing emissions with peak wavelengths of 591, 597, and 604 nm (592, 601, and 607 nm) at T = 22.5 °C and E = 0.90 μJ per pulse (at T = 44 °C and E = 0.11 μJ per pulse). Those solid and dotted lasing emission curves in Fig. 4 are obtained by placing a left- and a right-circular polarized chiral reflector, respectively, behind the refined laser, respectively, after the laser is excited with the above-mentioned conditions. Apparently, the measured lasing emissions at 22.4 °C or 44 °C all have a right-circular polarization whose handedness is the same as that of the chiral materials used in fabricating the template. This result is a signature indicating the band-edge lasing mechanism for all lasing signals presented in Fig. 3, especially for the abovementioned multi-mode lasing phenomenon at 22.4 °C which is not attributed to the random lasing mechanism.²⁸

To clarify the reason for the discrepancy between the above-mentioned lasing modes occurring at 22.4 °C and 44 °C, we further examine the micro-structure of the template. Fig. 5(a) and (b) present the SEM photos of the refined template at the after-washing-out stage at two various positions, say, x = 3 mm and 11 mm, respectively. Apparently, there is a large number of nanopores formed inside the template. Lots of nanopores can be contained in the pumped region. Based on the observation of the nanoporous morphology of the SEM template, the above-mentioned discrepancy of lasing emissions occurring at 22.4 °C and 44 °C can be explained as follows. When the DDLC molecules are refilled into the nanopores of the template at the birefringent nematic phase, the director and thus the refractive index of the LCs slightly fluctuates at various sub-regions of each pumped region because of the variations in porous shapes and sizes from hole to hole in the template. This effect can cause a slight fluctuation in the effective refractive index of the refilled CLC template, in short, n_LC(₁) ≠ n_LC(₂), where ₁ and ₂ are the position vectors of two various sub-regions of a certain pumped region, resulting in a slight fluctuation in the PBG and its band-edges from one position to another in the same pumped region. Therefore, the multiple lasing peaks in the refilled CLC template sample at 22.4 °C can simultaneously appear at the slightly fluctuated band-edges of the PBG in each pumped region.²⁹ In contrast, when the DDLC in the nanopores is heated to the isotropic state (44 °C), the refractive indices in these nanopores at various sub-regions of the same pumped region become identical; in short, n_i(₁) = n_i(₂) = n_i, inducing the almost identical PBG and thus the single-mode band-edge lasing emission at each pumped region. Fig. S4 in the ESI† gives a schematic illustration for the DDLC-refilled template sample. The narrower selective reflection band at 44 °C also indicates that the residue birefringence of the refilled CLC template is smaller compared to that at 22.4 °C.⁹ In other words, the laser presented herein exhibits a thermal convertibility between the multi-mode and single-mode operated at the nematic and isotropic phases of the DDLC, respectively. In this work, 5CB is selected as the refilled material because of its low clearing point compared to E7. The low clearing point of 5CB is the main factor that the thermal convertibility for the lasing emissions of the refilled template samples is feasibly easier. Fig. 6 shows the emission wavelength of the DDLC-refilled template sample as a function of temperature at a randomly-selected pumped position. Apparently, the lasing wavelength keeps fixed at T ≤ 31 °C and increases in a stepwise way (around 0.34 nm for each step), and then drops to another constant value as the temperature is higher than the clearing point (42 °C). The stepwise variation of the lasing wavelength is due to the limitation of the optical resolution of the spectrometer. The slight red-shift of the peak wavelength below 42 °C is due to the slight increase of the ordinary refractive index of the refilled LC with increasing temperature, rather than the change of the helical pitch,³⁰ when the refilled LC is near the clearing point. The slight variation of the lasing wavelength shows that the thermal stability of the template laser is pretty high in a wide temperature range. Especially, the peak wavelength of the lasing emission is extraordinarily stable at T ≤ 31 °C and at T ≥ 42 °C. Considering the thermal degradation of lasing emission caused by local heating and thus the pitch or phase change in polymer-free CLC lasers,³¹ the high thermal stability exhibited by the present CLC template sample is potentially beneficial for the further development of advanced LC lasers.

The variations of the peak intensity and the corresponding full-width at half maxima (FWHM) for the multi-mode band-edge lasing emissions with the pumped energy measured at x = 8.0, 9.0, 9.5, 10.0, 10.5, 11.2, 11.7, 12.2, 12.7, and 13.5 mm at 22.4 °C are presented in Fig. S5(a)–(j) in the ESI,† respectively. Meanwhile, similar results for the single-mode band-edge lasing emissions measured at those above-mentioned positions at 44 °C are shown in Fig. S6(a)–(j) in the ESI,† respectively. Apparently, the energy threshold and lasing linewidth of lasing emission measured at 44 °C are lower and narrower than those measured at 22.4 °C, respectively. Fig. 7 summarizes the experimental results shown in Fig. S5 and S6 in the ESI† as the variations of the energy threshold with the lasing wavelength. The fluorescence spectrum of the laser dye is presented by a black curve in each figure for reference and the sharp peak at 532 nm is the stray light from the pump source. The energy thresholds are distributed from 63 nJ per pulse to 315 nJ per pulse and from 16 nJ per pulse to 84 nJ per pulse for the multi- and single-mode band-edge lasing emissions, respectively. The energy threshold of the lasing emission is concave upwards with increasing lasing wavelength in each of the two figures; the variation is just contrary to the concave downward tendency of the fluorescence emission with increasing wavelength above 550 nm. However, the energy threshold of the lasing emission monotonically increases with a decrease in the lasing wavelength regardless of the variation of the fluorescence intensity below 580 nm. This result is primarily caused by the increased reabsorption effect of the fluorescence photons by the dyes with decreasing lasing wavelength.²⁴

To realize the influence of the refining process (that is, the TRD and RAP) on the lasing features of the DDLC-refilled gradient-pitched template laser, this work further compares the lasing features of the unrefined and refined lasers. Fig. S7(a) and (b) in the ESI† show the lasing emission spectra of the unrefined laser at T = 22.4 °C (E = 0.97 μJ per pulse) and at T = 44 °C (E = 0.48 μJ per pulse) at the pumped positions of x = 9.8 mm to x = 11.8 mm, attached with the corresponding reflection spectra displayed in the bottom of Fig. S7(a) and (b) (ESI†). These lasing phenomena of the multi- and single-mode lasing emissions are similar to those displayed previously based on the refined laser. Fig. S8(a) and (b) in the ESI† further show the variations of the peak intensity of the strongest multi- and single-mode band-edge lasing emissions and the corresponding FWHMs with pumped energy, respectively, measured at x = 10.2 mm of the unrefined laser. The lowest lasing threshold (corresponding FWHM) obtained for the multi- and single-mode lasing of the unrefined laser are approximately 50 nJ per pulse (3.8 nm) and 56 nJ per pulse (2.1 nm), respectively. Clearly, the optimum lasing threshold of the single-mode band-edge lasing decreases from 56 nJ per pulse (linewidth ≈ 2.1 nm) measured at x = 10.2 mm of the unrefined laser to 16 nJ per pulse (linewidth ≈ 0.7 nm) measured at x = 9.5 mm of the refined laser. That is, both the lasing threshold and linewidth of the DDLC-refilled template laser can be decreased significantly if the fabrication of the laser undergoes the refining process. As mentioned in the Experimental section, the RAP is an effective method to efficiently eliminate massive defects and increase the order parameter of LCs in a CLC cell. The performance of lasing emission in a CLC cell can be significantly improved if the CLC has fewer defects and a high order parameter, which are beneficial to decreasing the scattering loss and increasing the fluorescence efficiency.³²

It is worthwhile to mention that the optimum lasing threshold (16 nJ per pulse under a pumped diameter of ca. 250 μm) of the refined template laser at the isotropic state of the DDLC in the present work is ultra-low compared to the band-edge lasers based on most DDCLC/QDCLC and polymer-stabilized DDCLC systems.^{5–11,29–39} This result is because traditional DDCLC and/or polymer-stabilized DDCLC lasers have some unavoidable drawbacks such as the thermal instability of the DDCLC resonator and the pre-damage of the laser dyes during the photopolymerization for the polymer-stabilized DDCLC, respectively. In contrast, the refined DDLC-refilled CLC template sample, in addition to having a defect-free texture before curing by undergoing a refining process, provides a relatively stable resonator of the polymeric framework and prevents pre-damage of the laser dyes in the refilled stage. Given the effort to further improve the fluorescence efficiency of the gain media and the reduction of heat dissipations, commercialized CW LC tunable lasers based on the presently refined template may be realized. For example, the organic laser dye in the triplet state can be substituted by the gain media behaving as two-level systems, such as quantum dots,⁴⁰ to eliminate the non-radiative process and subsequently lead to the reduction of heat dissipation.⁴¹ Another method of sample rotation can be used to avoid heat accumulation so as to reduce the thermal instability of lasers.⁴²

Conclusions

In conclusion, this study develops two lasers based on refined and unrefined DDLC-refilled gradient-pitched CLC polymer template samples. Both the samples are fabricated by UV-curing and washing-out/refilling processes but the pitch gradients for the unrefined and refined samples are generated by undergoing a NRD and a rapid annealing process following a TRD, respectively. Experimental results show that each template laser has wide-band PBG-tunability in the entire visible region and thermal convertibility between multi- and single-mode band-edge lasing emissions. The PBG and lasing characteristics for the refined laser, however, are much superior to those of the unrefined laser. Due to the multiple advantages, such as high stability, ultra-low lasing threshold, wide-band and linear spatial PBG-tunability, and laser-mode convertibility, the refined template laser has potential for use in photonics and displays. In addition, the refined CLC template may potentially help realize CW LC tunable lasers, considering the improved gain media and emission stability.^41,42

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

The authors would like to thank the Ministry of Science and Technology (MOST) of Taiwan (contract numbers: MOST 106-2628-E-006-007-MY3 and MOST 106-2112-M-006-003-MY3) for financially supporting this research.

Notes and references

1. V. I. Kopp, Z.-Q. Zhang and A. Z. Genack, Prog. Quantum Electron., 2003, 27, 369 CrossRef CAS .
2. P. G. de Gennes and J. Prost, The Physics of Liquid Crystals, Oxford University Press, New York, 1993 Search PubMed .
3. Y. Wang, Z. Zheng, H. K. Bisoyi, K. G. Gutierrez-Cuevas, L. Wang, R. S. Zola and Q. Li, Mater. Horiz., 2016, 3, 442 RSC .
4. H. Coles and S. Morris, Nat. Photonics, 2010, 4, 676 CrossRef CAS .
5. L.-J. Chen, J.-D. Lin, S.-Y. Huang, T.-S. Mo and C.-R. Lee, Adv. Opt. Mater., 2013, 1, 637 CrossRef .
6. J.-D. Lin, M.-H. Hsieh, G.-J. Wei, T.-S. Mo, S.-Y. Huang and C.-R. Lee, Opt. Express, 2013, 21, 15765 CrossRef CAS PubMed .
7. B. Liu, Z. Zheng, X. Chen and D. Shen, Opt. Mater. Express, 2013, 3, 519 CrossRef CAS .
8. Z. Zheng, B. Liu, L. Zhou, W. Wang, W. Hu and D. Shen, J. Mater. Chem. C, 2015, 3, 2462 RSC .
9. A. Varanytsia, H. Nagai, K. Urayama and P. Palffy-Muhoray, Sci. Rep., 2015, 5, 17739 CrossRef CAS PubMed .
10. J. Xiang, A. Varanytsia, F. Minkowski, D. A. Paterson, J. M. D. Storey, C. T. Imrie, O. D. Lavrentovich and P. Palffy-Muhoray, Proc. Natl. Acad. Sci. U. S. A., 2016, 113, 12925 CrossRef CAS PubMed .
11. W. Huang, C. Yuan, D. Shen and Z. Zheng, J. Mater. Chem. C, 2017, 5, 6923 RSC .
12. J.-D. Lin, H.-L. Lin, H.-Y. Lin, G.-J. Wei, Y.-C. Chuang, L.-J. Chen, S.-Y. Huang, C.-Y. Huang, T.-S. Mo and C.-R. Lee, J. Mater. Chem. C, 2017, 5, 3222 RSC .
13. F. Castles, F. Day, S. M. Morris, D. H. Ko, D. J. Gardiner, M. M. Qasim, S. Nosheen, P. J. Hands, S. S. Choi, R. H. Friend and H. J. Coles, Nat. Mater., 2012, 11, 599 CrossRef CAS PubMed .
14. J. Guo, H. Wu, F. Chen, L. Zhang, W. He, H. Yang and J. Wei, J. Mater. Chemp., 2010, 20, 4094–4102 RSC .
15. A. Muñoz, M. E. McConney, T. Kosa, P. Luchette, L. Sukhomlinova, T. J. White, T. J. Bunning and B. Taheri, Opt. Lett., 2012, 37, 2904 CrossRef PubMed .
16. M. E. McConney, V. P. Tondiglia, J. M. Hurtubise, L. V. Natarajan, T. J. White and T. J. Bunning, Adv. Mater., 2011, 23, 1453 CrossRef CAS PubMed .
17. M. E. McConney, T. J. White, V. P. Tondiglia, L. V. Natarajan, D.-K. Yang and T. J. Bunning, Soft Matter, 2012, 8, 318 RSC .
18. A. Ryabchun, I. Raguzin, J. Stumpe, V. P. Shibaev and A. Y. Bobrovsky, ACS Appl. Mater. Interfaces, 2016, 8, 27227 CrossRef CAS PubMed .
19. M. E. McConney, V. P. Tondiglia, J. M. Hurtubise, T. J. White and T. J. Bunning, Chem. Commun., 2011, 47, 505 RSC .
20. S. S. Choi, S. M. Morris, W. T. Huck and H. J. Coles, Adv. Mater., 2010, 22, 53 CrossRef CAS PubMed .
21. S. M. Wood, J. A. J. Fells, S. J. Elston and S. M. Morris, Macromolecules, 2016, 49, 8643 CrossRef CAS .
22. T. Manabe, K. Sonoyama, Y. Takanishi, K. Ishikawa and H. Takezoe, J. Mater. Chem., 2008, 18, 3040 RSC .
23. J.-D. Lin, C.-L. Chu, H.-Y. Lin, B. You, C.-T. Horng, S.-Y. Huang, T.-S. Mo, C.-Y. Huang and C.-R. Lee, Opt. Mater. Express, 2015, 5, 1419 CrossRef .
24. J.-D. Lin, T.-Y. Wang, T.-S. Mo, S.-Y. Huang and C.-R. Lee, Sci. Rep., 2016, 6, 30407 CrossRef CAS PubMed .
25. R. Ozaki, T. Shinpo and H. Moritake, Appl. Phys. Lett., 2008, 92, 163304 CrossRef .
26. K.-C. Shin, F. Araoka, B. Park, Y. Takanishi, K. Ishikawa, Z. Zhu, T. M. Swager and H. Takezoe, Jpn. J. Appl. Phys., 2004, 43, 631 CrossRef CAS .
27. A. H. Gevorgyan, K. B. Oganesyan, R. V. Karapetyan and M. S. Rafayelyan, Laser Phys. Lett., 2013, 10, 1025802 Search PubMed .
28. S.-C. Chen, J.-D. Lin, C.-R. Lee and S.-J. Huang, J. Phys. D: Appl. Phys., 2016, 49, 165102 CrossRef .
29. J. Schmidtke, W. Stille, H. Finkelmann and S. T. Kim, Adv. Mater., 2002, 14, 746 CrossRef CAS .
30. M. F. Moreira, I. C. S. Carvalho, W. Cao, C. Bailey, B. Taheri and P. Palffy-Muhoray, Appl. Phys. Lett., 2004, 85, 2691 CrossRef CAS .
31. A. Varanytsia and P. Palffy-Muhoray, Proc. SPIE, 2013, 8828, 88281F CrossRef .
32. C.-R. Lee, J.-D. Lin, T.-S. Mo, C.-T. Horng, H.-Y. Sun and S.-Y. Huang, Opt. Express, 2015, 23, 10168 CrossRef CAS PubMed .
33. P. V. Shibaev, V. I. Kopp and A. Z. Genack, J. Phys. Chem. B, 2003, 107, 6961 CrossRef CAS .
34. M. Uchimura, Y. Watanabe, F. Araoka, J. Watanabe, H. Takezoe and G. Konishi, Adv. Mater., 2010, 22, 4473 CrossRef CAS PubMed .
35. Y. Watanabe, M. Uchimura, F. Araoka, G. Konishi, J. Watanabe and H. Takezoe, Appl. Phys. Express, 2009, 2, 102501 CrossRef .
36. S. M. Morris, A. D. Ford, M. N. Pivnenko, O. Hadeler and H. J. Coles, Phys. Rev. E: Stat., Nonlinear, Soft Matter Phys., 2006, 74, 061709 CrossRef CAS PubMed .
37. H.-T. Wang, J.-D. Lin, C.-R. Lee and W. Lee, Opt. Lett., 2014, 39, 3516 CrossRef CAS PubMed .
38. L.-J. Chen, J.-H. Dai, J.-D. Lin, T.-S. Mo, H.-P. Lin, H.-C. Yeh, Y.-C. Chuang, S.-A. Jiang and C.-R. Lee, ACS Appl. Mater. Interfaces, 2018, 10, 33307 CrossRef CAS PubMed .
39. H. Yoshida, Y. Shiozaki, Y. Inoue, M. Takahashi, Y. Ogawa, A. Fujii and M. Ozaki, J. Appl. Phys., 2013, 113, 203105 CrossRef .
40. A. Zrenner, E. Beham, S. Stufler, F. Findeis, M. Bichler and G. Abstreiter, Nature, 2002, 418, 612 CrossRef CAS PubMed .
41. J. Ortega, C. L. Folcia and J. Etxebarria, Materials, 2008, 11, 5 CrossRef PubMed .
42. M. C. Normand, P. Chen, C. Can and P. J. W. Hands, Opt. Express, 2018, 26, 26544 CrossRef CAS PubMed .

---

Footnote

† Electronic supplementary information (ESI) available: Pictures and plots to show the improvement via thermally reverse diffusion and rapid annealing processes; plots to show energy thresholds of the refined DDNLC-refilled gradient-pitched template; and characteristics of the unrefined laser sample. See DOI: 10.1039/c9tc00194h

This journal is © The Royal Society of Chemistry 2019