Microfluidic generation of cholesteric liquid crystal droplets with an integrative cavity for dual-gain and controllable lasing

Abstract

The integration of one more gain media in droplet microlasers with morphology-dependent modes, which can be employed in optofluidic systems as multi-wavelength lasing sources, is highly attractive and demands new cavity design and fabrication approaches. Here, cholesteric liquid crystal (CLC) droplets with an integrative triple-emulsion cavity are fabricated via glass-capillary-based microfluidic technologies and dual-gain lasing with variable modes, flexibly configured by the combination and incorporation of gain dyes and CLCs into both the core and shell. The distributed feedback (DFB) mode, formed by the feedback from the self-assembled helix periodic structure of CLCs, the whispering gallery (WG) mode, and the hybrid, is selectively excited by controlling the spatial coupling between the pump beam and the droplet with gain. With the merits of dual-gain and controllable lasing, a prototype dual-wavelength-ratiometric thermometer with self-calibration capability is expected to be developed. Furthermore, the anisotropic CLC core is substituted with an isotropic fluid and the gain from the CLC shell is additionally removed, DFB lasings in both shell and core are absent, and only Bragg-shell reflection-based hybrid modes are excited for lasing. The CLC droplet microlasers with an integrative cavity are expected to provide a new route to future lab-on-chip (LOC) applications.

---

Introduction

Droplet microlasers with morphology-dependent modes are one of the most outstanding examples to be integrated into optofluidic devices, in which the fusion of microfluidics and optics enables unique performance that is not obtainable in traditional optical systems built using solid materials, such as glasses, crystals and metals.^1–12 Nowadays, microlasers based on liquid crystal (LC) droplets have triggered considerable interest and hold promising potential for high-throughput bio/chemical sensing applications in aqueous environments.^6–8,13,14 Multi-wavelength lasing emissions from fluorescently labelled nematic liquid crystal (NLC) droplets have been exploited in terms of wavelength tunability and mode selectivity.^15–18 Apart from the conventional nematic and smectic LC droplets, cholesteric liquid crystal (CLC) droplets with self-assembled helix periodic structure arranged along the radial direction allow for flexible control of the propagation of light in the visible range and have been widely investigated with great capabilities.^19–26 From a broader perspective, the distinct optical configurations with radially arranged and tuneable one-dimensional (1D) photonic crystal (PC) structures in sphere-shape microdroplets are attractive for all-wave optical confinement and lasing emissions in aqueous environments.^27–30 For light with a wavelength inside the photonic band gap (PBG) of 1D PC, the propagation is prohibited and the photonic density of states (DOS) is accordingly low.³¹ If a defect is introduced in a PC structure, light with a wavelength in the PBG is trapped for a long time and a defect mode resonance forms. Except for in defect mode, the group velocity of photons near the band-edge approaches zero and the photon life also increases to be infinitely long.³² The long life of photons enables the significant enhancement of the light–matter interaction, thus, the band-edge mode (also called the DFB mode), defect mode and aforementioned WG mode can be simultaneously excited for lasing oscillations in a CLC droplet with sufficient gain.^7,30

Recently, tremendous progress has been achieved in exploring the controllable outcoupling of laser modes through selectively filtering the resonant modes in coupled cavity structures.^32–35 An alternative strategy was proposed by utilizing the premium benefits of CLC droplets, such as the perfect rotational symmetry of the optical configuration and the sensitive responsiveness of the PBG to external stimuli.³⁶ The modulation and selection of the defect mode, DFB mode and WG mode, which has been achieved in dye-doped CLC (DDCLC) microshells by tuning the spatial coupling of the pump beam and the droplet with gain, could be endowed with high flexibility.^36–38 Besides this, benefiting from the rapid advancements in microfluidic emulsification, great efforts have been devoted to the generation of monodispersed CLC droplets, allowing for the encapsulation of various fluidic components with diverse functionalities.^{28,29,39–42} Nevertheless, the DDCLC lasing and modes from a single-layer gain are constrained by the one gain band and the monotonous cavity structure. Integration of one more gain medium in an individual droplet and the accordant multi-dimensional controls of the lasing, such as lasing position, lasing mode, etc., are anticipated. Designs for new cavity structures and the development of innovative fabrication approach are also demanded.

In this work, we fabricate CLC droplets with an integrative triple-emulsion cavity structure via glass-capillary-based microfluidic technology. We demonstrate dual-gain lasing with both core and shell modes by dissolving appropriate dye dopants and CLCs into the shell and core, and selective excitations by controlling the spatial coupling between the pump beam and the droplet. The temperature-dependent lasing behaviour was then investigated and the discrepant shift of the lasing wavelength in the core and shell enabled the development of a prototype dual-wavelength-ratiometric thermometer with self-calibration capability. In addition, we substituted the anisotropic CLC core with an isotropic fluid and additionally removed the gain from the shell. In addition, advanced control of the single-gain lasing in the triple-emulsion droplet was performed. CLC droplet microlasers with controllable multi-wavelength outputs might offer a new route to optofluidic applications.

Results and discussion

Mode resonances and microfluidic fabrication of the CLC droplets with an integrative triple-emulsion cavity

An outstanding characteristic of CLC is its selective circularly polarized reflection owing to its spontaneous self-assembly of periodic dielectric helical structures with a PBG that can be readily tuned through changing the concentration of a chiral dopant. A typical optical reflection spectrum of a CLC with PBG of between 490 and 560 nm measured at an incidence angle of θ = 0° is presented in Fig. 1(a). If the wavelength of light is inside the PBG, the propagating is suppressed, the density of photon states diverge at the edge of the band gap, and DFB lasing or band-edge lasing can thus be realized productively at a low threshold pumping. Here, the CLC triple-emulsion droplets are employed for dual-gain lasing emission and the cross profile is shown in Fig. 1(b). The CLC shell and core, separated by a layer of a deionized water (DI) solution of 10 wt% polyvinyl acetate (PVA), can be doped with a dye for optical gain. For a designed CLC droplet with an integrative cavity structure, Fig. 1(c) to (e) schematically show the ray optics of mode resonances possibly excited for lasing emission.

Fig. 1 (a) A typical optical refection spectrum (θ = 0°) of CLC with a PBG between 490 nm and 560 nm. The DFB lasing is located at the edges of the photonic band owing to the long photon dwell time. (b) Cross profile of a CLC triple-emulsion droplet containing a CLC core and a shell, between which aqueous PVA solution assists the degenerated planar alignment of CLCs. Ray optic schematics for optical resonances: (c) shell and core DFB resonances, (d) TIR-based WG resonances in the shell and core, and (e) hybrid resonance with optical reflections on the Bragg-shell.

In Fig. 1(c), the light rays in the dashed rectangles present the DFB modes in both the shell and the core. Fig. 1(d) presents the ray trajectories of the shell and core WG modes formed by continuous light total internal reflections (TIRs) on the external and internal boundaries of the shell, and on the boundary of the core, respectively. Fig. 1(e) presents the optical feedback rays of the hybrid modes from the CLC Bragg-shell with a random incident angle. By simply tuning the spatial coupling between the pump beam and the droplet, a certain mode can be excited for lasing oscillation in the dye-doped core or shell.

As shown in Fig. 2(a) and (b), the triple-emulsion droplets are fabricated via a capillary microfluidic device, which comprises three tapered round capillaries aligned coaxially and assembled in a square capillary. The discontinuity of the innermost core-sheath flow causes the concurrent generation of water-in-oil-in-water (W/O/W) double-emulsion droplets with only the outermost CLC shell. By varying the velocity of each flowing phase in the microfluidic device, we can obtain DDCLC triple-emulsion droplets with variable shell thicknesses and core diameters. The interlayer of aqueous PVA solution, which isolates the oil core and oil shell with different spacings, helps to facilitate the degenerated planar alignment of LC molecules in both the shell and core along the oil/water interfaces. The incubation process is a prerequisite to obtaining CLC-based resonators with radial symmetry. PM567 and DCM are incorporated in the shell and the core, respectively, providing dual-gain for stimulated emissions. As shown in the top right inset, the finished droplets with triple-emulsion cavity structure are stored in a bottle.

Fig. 2 (a) Schematic illustration and (b) optical images of the microfluidic capillary device for the generation of oil-in-water-in-oil-in-water (O/W/O/W) triple-emulsion samples with variable core and shell diameters. (c) Simplified diagram of the pumping configuration for laser characterization with beam diameters of 1.2, 4.5, 8.6 and 11.2 mm. The focus spot of the microscope objective is set close to the back part of the droplets. Top right inset: fabricated triple-emulsion droplets stored in a bottle.

Dual-gain and controllable mode lasing

A simplified scheme of the pumping arrangement is presented in Fig. 2(c) and the detailed experimental setup is given in the ESI† in Fig. S1. As depicted in the right inset, the focus spot of the microscope objective was carefully set close to the back part of the isolated droplets. The pump beam shines on the droplet in a bottom-to-up manner and defines an active gain region with the geometrical shape of an optical cylinder. Therefore, the spatial coupling between the pump beam and the CLC triple-emulsion droplet can be controlled by simply varying the diameter (D) of the aperture along the light path. If the pump beam is obliquely incident, the lasing will not be affected as the dominant pump power still irradiates vertically on the surface of droplet since that the droplet has spherical symmetry.

Fig. 3(a) shows the polarized optical microscopic (POM) images of two DDCLC triple-emulsion droplet samples with nearly the same outer diameter of ∼265 μm and shell thickness of 14 μm (denoted as sample A1 and sample A2). The core diameter of sample A1 is obviously significantly larger than that of sample A2 as the aqueous interlayer between the core and the shell in sample A2 is much thicker.

Fig. 3 (a) POM images of two DDCLC triple-emulsion droplets with nearly the same outer diameter of 265 μm and shell thickness of 14 μm. Scale bar: 50 μm. (b) Optical spectra of lasing emission with a varied D for sample A1 and sample A2 denoted in (a). (c) Laser threshold of different modes excited in sample A1 by varying D.

The dual-gain emission spectra of sample A1 were firstly examined by tuning D and the results are shown in Fig. 3(b). For D = 1.2 mm, a small part of the outermost shell of the droplet is covered by the pump beam and only one lasing peak (denoted by “■”) at 562.4 nm is observed. From the measured reflection spectrum of the shell CLC (Fig. S2†), we know that the peak is the shell DFB lasing as the wavelength is located at the long band-edge of the shell CLC. In addition, following the gain absorption of the pump beam in the shell, core DFB lasing without sufficient gain is not excited and observed. Owing to the slight overlap of the mode volume with the pump beam, the WG modes in both the shell and the core are also not excited effectively and higher thresholds are required. As D increases to 4.5 mm, the DFB modes in both the core (●) and the shell (■), as well as the WG modes (▲), are all exited simultaneously. The identification of another lasing peak at 627.5 nm is based on the photonic band of the core CLC as well, while the intermediate peaks around ∼606 nm (▲) are ascribed to WG mode lasings built by continuous reflections on the boundaries of the shell or the core. The polarizations of DFB mode lasings are expected to be left-hand and right-hand circular polarizations owing to the opposite helicities of the chiral dopants. As D is further increased to 8.6 mm, the shell DFB lasing disappears and the WG lasings become stronger. The results indicate that WG mode lasings extract more gain from the pump energy owing to the larger gain coupling area while the DFB mode in the shell is accordingly suppressed by the lasing competition. From the disappearance of the shell DFB lasing and that the DDCLC shell first obtains the gain from the pump beam, we know that the multi-lasing peaks are most likely to be shell WG modes.

To check the gain influences on the lasing modes of CLC droplets, we also pumped sample A2 by fixing D at 4.5 mm. It can be seen that both shell DFB lasing and WG lasing are achieved, while the core DFB lasing is not excited successfully. Compared with the lasing modes in sample A1, we can find that the smaller core cannot obtain sufficient gain as a smaller gain volume is involved for excitation, while the shell gain volume is nearly equal (the shell thickness of the two samples is nearly the same). Fig. 3(c) compares the thresholds of lasing modes excited in sample A1 for the different D. As D = 1.2 mm, the threshold of the shell DFB lasing (2.04 × 10⁻² J cm⁻²) is relatively low, while both the shell WGM and core DFB modes are not excited owing to the small space coupling between the pump beam and mode volume. When D increases to 4.5 mm, the threshold for shell DFB lasing increases slightly to 3.03 × 10⁻² J cm⁻² and is much less than that of core DFB mode lasing (9.76 × 10⁻² J cm⁻²), while the WG mode lasing is most likely be excited since the shell volume covered by the pump beam is expanded and the gain coupling of WGM is accordingly enhanced. When D is further increased to 8.6 mm, the shell DFB lasing disappears owing to the competition with shell WGM lasing and additionally the threshold of the core DFB lasing increase to 20 × 10⁻² J cm⁻². The further increase of pump coverage on the droplet enhances the gain coupling of the shell WGMs and simultaneously weakens that of the core DFB modes. A detailed evaluation of the threshold values can be found in Fig. S3† (D = 4.5 mm). On the whole, the results imply that a larger D simultaneously weakens the lasing excitation of the shell DFB and enhances the lasing possibility of the shell WGM.

Temperature-dependent lasing behaviour of the CLC droplets with dual-gain media

The temperature-dependent lasing behaviours of the CLC droplet microlaser are also investigated with D = 4.5 mm. The evolution of lasing spectra as a function of temperature (T) is given in Fig. 4(a). The emission wavelength of shell DFB lasing (■) is nearly unchanged owing to the reflection band of the Bragg-shell of the DDCLC mixture being thermo-insensitive, while the DFB lasing (●) in the thermo-responsive core shifts from 596.2 nm to 648.8 nm as T decreases gradually from 35 °C to 29 °C. The corresponding reflection spectra of the core DDCLCs, as shown in Fig. 4(b), confirm the redshift of core DFB lasing. At a lower temperature of 27 °C, since the gain profile of the DDCLC core no longer supports the long-band-edge DFB mode, thus the DFB mode at the short-band-edge of the PBG (624.3 nm) is excited successfully and the corresponding lasing peak hops from the long-band-edge to the short-band-edge. Based on the DFB lasing emissions of DDCLCdroplets, we can develop a prototype dual-wavelength-ratiometric thermometer. The self-calibration capability of the CLC droplet dual-gain microlasers is revealed by plotting the ratio values of the core and shell DFB lasing peaks in Fig. 4(c) as T decreases from 35 °C to 29 °C. Besides, the threshold of shell DFB lasing is characterized to be lower than that of core DFB lasing at these temperatures (Fig. S4†). Therefore, we suggest that the pumping energy used to operate the thermometer must exceed the threshold of core DFB lasing. In addition, the WG lasing peaks around ∼606 nm are almost unchanged, indicating that the WG lasings are mainly determined by the profile of the gain spectrum of the shell, which is independent on the temperature.

Fig. 4 (a) Series of emission spectra of sample A1 excited at various temperatures from 27 °C to 35 °C. The DFB lasing lines from the shell and core are denoted by “■” and “●”, respectively. (b) Reflection spectra of the thermo-sensitive CLC core measured from 27 °C to 35 °C. The long and short band edges are denoted by “*” and “*”, respectively. (c) The ratio values of two DFB lasing wavelengths from the core and shell obtained by varying temperatures from 27 °C to 35 °C. The inset shows the temperature-dependent DFB lasing wavelength of the thermosensitive DDCLC core.

Hybrid modes in CLC triple-emulsion microlasers

By substituting the anisotropic CLC mixture with an isotropic 1-bromohexadecane solution of organic dye PM597 in the core and additionally removing the gain from the CLC shell, we can fabricate single-gain CLC triple-emulsion droplets and implement advanced control of the lasing in the integrated microcavity. Fig. 5(a) presents the POM image of a sample (called sample B). The diameters of shell and core are estimated to be ∼590 and ∼390 μm, respectively, and the shell thickness is ∼50 μm. In terms of optical confinement, the shell with a PBG can function as a Bragg reflector for the light emitting from the core if the wavelength is in the PBG, which suggests that the lasing oscillation in a single-gain core could be realized and controlled by tuning the gain coupling between the pump and the according modes constructed by the continuous optical reflections from the Bragg-shell as sufficient gain is given.

Fig. 5 (a) POM image of a CLC triple-emulsion droplet sample (sample B) with outer diameter of ∼590 μm and shell thickness of ∼390 μm. (b) and (c) Schematic display of the coupling between the pumping beam and the droplet as the focus spot shifts from the centre to the edge of the droplet. (d) Emission spectra for pumping locations i to v. (e) Lasing spectra of dye-doped droplet without CLCs as the pump beam shifts from the centre to the edge.

Firstly, as shown in Fig. 5(b), we focussed the pump beam at the centre of the droplet (position i) with a fixed energy of 10.2 × 10⁻² J cm⁻². The emission spectra are shown in Fig. 5(d) and multi-mode lasings near ∼595 nm are clearly observed. Two kinds of modes are possibly responsible for the lasing: The first is the hybrid mode, mainly relying on the optical reflections from the Bragg-shell and additionally the boundary of the core and the PVA solution acts as a weak reflector. The second is the WG mode formed via TIR at the boundary of the core and the PVA solution. For the first kind mode, the lasing wavelength is related to the overlap between the reflection band of the Bragg-shell and the gain fluorescence spectrum of the core (see Fig. S5†). When the incident angle θ increases, the reflection band between λ_0s/cos θ and λ_0l/cos θ of the Bragg-shell would shift toward the short wavelength (λ_0s and λ_0l are defined as the short- and long-wavelength band edges of the Bragg-shell as θ = 0). Accordingly, the lasing wavelength shifts toward the short wavelength for a fixed gain fluorescence spectrum. For the second kind mode, the lasing modes with wavelength covered by the gain spectrum would be excited. Since the free spectral range (FSR) of ∼0.18 nm of the WG modes in the ∼390 μm-diameter core sphere is far less than the band width (∼60 nm) of the fluorescence spectrum, thus the lasing modes will just hop between the adjacent WG modes and λ shifts slightly if the gain spectrum is not changed.

In order to confirm the lasing modes, we then shifted the focus position of the pump beam from the centre (i) to the edge (v); the coupling schematics between the pump beam and the droplet are shown in Fig. 5(b) and (c). The incidence angle of the pump beam on the Bragg-shell increases from 0 to θ₄ accordingly. The modes with similar incident angles or propagating paths to the pump beam are most likely excited owing to the strongest gain coupling. In other words, the emitted lights from the gain core have similar incident angles θ to the pump wave. As described above, for a certain Bragg optical grating, the vertical component of wave vector k_⊥ is fixed in the PBG. If θ increases, the total k = k_⊥/cos θ (k = k_⊥ + k_∥) increases and the reflection band of the Bragg-shell shifts to the shorter wavelength productively. As seen from Fig. 5(d), we found that the shifts of the lasing peaks are consistent with the behaviours of the hybrid mode formed by the continuous reflections from the Bragg-shell. Thus, the lasing modes are confirmed to be hybrid modes. Finally, since the gain area covered by the pump beam in the core becomes very small, the lasings become much weaker at pump position v.

For comparison, WG modes, formed by continuous TIRs on the boundary of the core and the PVA solution, are excited for lasing in the droplets without an outside optical feedback structure. The droplets with a diameter close to that of the core in sample B are picked out for laser characterization. The lasing spectra are shown in Fig. 5(e) as the pump position shifts from the centre to the edge of the droplet (see the insets). The results show that the lasing peaks for the WG modes are barely shifted, indicating that the lasing wavelengths are independent of the pump positions and further confirming that the lasing modes in sample B are the hybrid modes. The results indicate that the flexible combination and incorporation of the CLC Bragg-grating and gain dye into the integrative triple-emulsion droplet cavity allows robust control not only of dual-gain lasing, but also of single-gain lasing.

Experimental

Materials

DDCLC mixtures are produced by doping R-form and S-form chiral dopants into nematic LCs to render right-handed and left-handed helical structures, respectively, and then doping with appropriate fluorescent dyes for stimulated emission gain. The employed nematic LC host was E7 (Xianhua), whose physical properties are as follows: clearing point (T_c) = 59 °C, refractive indices n_e = 1.741, n_o = 1.517 at λ = 633 nm and 20 °C. The R-form chiral dopant was R5011 (HCCH, helical twisting power HTP ≈ 110 μm⁻¹). The S-form chiral dopant was S811 (HTP ≈ 11.1 μm⁻¹). The DDCLC mixture as the shell of samples A1 and A2 was prepared by mixing 2.7 wt% R5011, 96.8 wt% E7 and 0.5 wt% dye PM567 (exciton). The DDCLC mixture as the core of samples A1 and A2 consists of 25.9 wt% S811, 73.6 wt% E7 and 0.5 wt% dye DCM (exciton). The CLC mixture as the shell of sample B was prepared by mixing 2.4 wt% R5011 and 97.6 wt% E7. 1-Bromohexadecane solution was doped with 0.5 wt% dye PM597 (exciton) as the core of sample B. The chiral pitch (P) could be altered by changing the concentration of the chiral dopant, accordingly influencing the wavelength of the stop band. All CLC mixtures were heated to above their clear point and ultrasonically oscillated to ensure uniformity.

Fabrication of CLC triple-emulsion droplets

The capillary microfluidic device consisting of three tapered cylindrical capillaries aligned coaxially and assembled in a square capillary is illustrated in Fig. 2(a) and (b). The tip of the injection tubes is tapered to a 60 μm orifice by using a Narishige PN-30 micropipette puller and then broken to the desired diameter of 150 μm by using a Narishige MF-900 microforge. Taped cylindrical capillaries were sanded to 300 μm orifices as collecting tubes. The inner and outer walls of the tapered capillaries were rendered to be hydrophilic or hydrophobic after being treated with trimethoxysilyl-propoxypolyethylene oxide (Gelest) and n-octadecyltrimethoxy silane (Gelest), respectively. To fabricate samples A1 and A2, the core CLC solution and an aqueous solution for the isolation layer were simultaneously injected through the cylindrical capillary with the 60 μm orifice, guaranteeing that the O/W core-sheath flow formed the innermost CLC cores. The shell DDCLC mixture was injected through the interstice between the cylindrical capillary with the 150 μm orifice and the square capillary, whereas the continuous phase (PVA solution) was injected through the interstice between the collecting tube and the square capillary. Therefore, the outermost CLC stream along the outer wall and the counter water can coaxially flow into the orifice of the collecting tube, forming oil-in-water-in-oil-in-water (O/W/O/W) triple-emulsion droplets. The typical flow rates for the four phases were set to 150, 600, 900, and 3500 μL h⁻¹ from the inside out, using syringe pumps (NE-501, New Era). The generation of triple-emulsion droplets was monitored with an inverted optical microscope (XD202, Jiangnan Novel).

Optical characterization

The detailed experimental setup used for laser pumping and characterization is shown in Fig. S1.† The triple-emulsion droplets were observed under a POM in transmission mode (PM6000, Jiangnan Novel) equipped with a digital camera (DCC1645C, Thorlabs). To characterize the laser emission behaviours, the output beam of a linear polarized frequency-doubling Q-switched Nd:YAG laser (532 nm, 12 ns, 1 Hz) was expanded and collimated to irradiate the samples in an end-pumping manner. Emitted light with an angle of less than 29° can be collected by the same microscope objective, and the output energy and optical spectra were analysed using an energy meter (PD10, Ophir) and a fibre spectrometer (HR2000+, Ocean Optics), respectively.

Conclusions

In conclusion, we have fabricated droplets with an integrative triple-emulsion cavity via glass-capillary-based microfluidic technology and demonstrated dual-gain lasing with modes configured by the combination and incorporation of gain dyes and CLCs into the core and the shell. By simply tuning the spatial coupling of the pump beam and the droplet, lasing excitation control was also demonstrated. For a triple-emulsion droplet with a temperature-insensitive shell and temperature-sensitive core, a dual-wavelength-ratiometric thermometer was developed with self-calibration capability. As well as dual-gain lasing, advanced control of single-gain lasing in the triple-emulsion droplet is also performed and the tunability of the hybrid modes formed by the continuous reflections of the Bragg-shell is demonstrated. The incorporation of gain dyes in both the shell and the core of CLC triple-emulsion droplets allows for dual-gain lasing while the tuning of the gain coupling enables controllable lasing excitation. Such CLC droplet microlasers with advanced lasing modes control could find possible applications in lab-on-chip (LOC) systems.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

We acknowledge the funding and support from the National Natural Science Foundation of China (NSFC) (No. 61675172) and the Natural Science Foundation of Fujian Province, China (No. 2017J01124).

Notes and references

1. K. J. Vahala, Nature, 2003, 424, 839.
2. D. Psaltis, S. R. Quake and C. Yang, Nature, 2006, 442, 381–386.
3. S. K. Y. Tang, Z. Li, A. R. Abate, J. J. Agresti, D. A. Weitz, D. Psaltis and G. M. Whitesides, Lab Chip, 2009, 9, 2767.
4. A. Chiasera, Y. Dumeige, P. Féron, M. Ferrari, Y. Jestin, G. N. Conti, S. Pelli, S. Soria and G. C. Righini, Laser Photonics Rev., 2010, 4, 457–482.
5. S. K. Y. Tang, R. Derda, Q. M. Quan, M. Loncar and G. M. Whitesides, Opt. Express, 2011, 19, 2204–2215.
6. A. Bakal, C. Vannahme, A. Kristensen and U. Levy, Appl. Phys. Lett., 2015, 107, 106.
7. D. McGloin, Rep. Prog. Phys., 2017, 80, 054402.
8. T. Reynolds, N. Riesen, A. Meldrum, X. Fan, J. M. M. Hall, T. M. Monro and A. François, Laser Photonics Rev., 2017, 11, 1600265.
9. Y. Z. Shi, S. Xiong, L. K. Chin, Y. Yang, J. B. Zhang, W. Ser, J. H. Wu, T. N. Chen, Z. C. Yang, Y. L. Hao, B. Liedberg, P. H. Yap, Y. Zhang and A. Q. Liu, Lab Chip, 2017, 17, 2443.
10. Y. Z. Shi, S. Xiong, L. K. Chin, J. B. Zhang, W. Ser, J. H. Wu, T. N. Chen, Z. C. Yang, Y. L. Hao, B. Liedberg, P. H. Yap, D. P. Tsai, C. W. Qiu and A. Q. Liu, Sci. Adv., 2018, 4, eaao0773.
11. Y. Z. Shi, S. Xiong, Y. Zhang, L. K. Chin, Y. Y. Chen, J. B. Zhang, T. H. Zhang, W. Ser, A. Larson, L. S. Hoi, J. H. Wu, T. N. Chen, Z. C. Yang, Y. L. Hao, B. Liedberg, P. H. Yap, D. P. Tsai, C. W. Qiu and A. Q. Liu, Nat. Commun., 2018, 9, 815.
12. V. D. Ta, Y. Wang and H. Sun, Adv. Opt. Mater., 2019, 1900057.
13. M. Humar, Liq. Cryst., 2016, 43, 1–14.
14. M. Urbanski, C. G. Reyes, J. Noh, A. Sharma, Y. Geng, V. S. R. Jampani and J. P. Lagerwall, J. Phys.: Condens. Matter, 2017, 29, 133003.
15. M. Humar, M. Ravnik, S. Pajk and I. Muševič, Nat. Photonics, 2009, 3, 595–600.
16. M. Humar and I. Musevic, Opt. Express, 2011, 19, 19836–19844.
17. T. A. Kumar, M. A. Mohiddon, N. Dutta, N. K. Viswanathan and S. Dhara, Appl. Phys. Lett., 2015, 106, 3–600.
18. K. J. Lee, S. J. Kim, D. Kang and J. H. Kim, Opt. Express, 2015, 23, 24903.
19. M. Humar and I. Musevic, Opt. Express, 2010, 18, 26995.
20. J. Fan, Y. Li, H. K. Bisoyi, R. S. Zola, D. K. Yang, T. J. Bunning, D. A. Weitz and Q. Li, Angew. Chem., Int. Ed., 2015, 54, 2160–2164.
21. H. K. Bisoyi and Q. Li, Chem. Rev., 2016, 116, 15089–15166.
22. Z. G. Zheng, Y. N. Li, H. K. Bisoyi, L. Wang, T. J. Bunning and Q. Li, Nature, 2016, 531, 352–357.
23. L. Wang and Q. Li, Adv. Funct. Mater., 2016, 26, 10–28.
24. L. Wang, D. Chen, K. G. Gutierrez-Cuevas, H. K. Bisoyi, J. Fan, R. S. Zola, G. Q. Li, A. M. Urbas, T. J. Bunning, D. A. Weitz and Q. Li, Mater. Horiz., 2017, 4, 1190.
25. Y. Geng, J. H. Jang, K. G. Noh, J. Noh, J. P. F. Lagerwall and S. Y. Park, Adv. Opt. Mater., 2017, 6, 1700923.
26. M. Schwartz, G. Lenzini, Y. Geng, P. B. Ronne, P. Y. A. Ryan and J. P. F. Lagerwall, Adv. Mater., 2018, 30, 1707382.
27. M. G. Donato, J. Hernandez, A. Mazzulla, C. Provenzano, R. Saija, R. Sayed, S. Vasi, A. Magazzu, P. Pagliusi, R. Bartolino, P. G. Gucciardi, O. M. Marago and G. Cipparrone, Nat. Commun., 2014, 5, 3656.
28. J. H. Kang, S. H. Kim, A. Fernandez-Nieves and E. Reichmanis, J. Am. Chem. Soc., 2017, 139, 5708.
29. S. S. Lee, B. K. Kim, Y. H. Kim and S. H. Kim, Sci. Adv., 2018, 4, eaat8276.
30. Y. Uchida, Y. Takanishi and J. Yamamoto, Adv. Mater., 2013, 25, 3234–3237.
31. E. Yablonovitch, Phys. Rev. Lett., 1987, 58, 2059.
32. V. I. Kopp, B. Fan, H. K. M. Vithana and A. Z. Genack, Opt. Lett., 1998, 23, 1707–1709.
33. C. H. Zhang, C. L. Zou, H. Y. Dong, Y. L. Yan, J. N. Yao and Y. S. Zhao, Sci. Adv., 2017, 3, e1700225.
34. J. Zhao, Y. Yan, C. Wei, W. Zhang, Z. Gao and Y. S. Zhao, Nano Lett., 2018, 18, 1241–1245.
35. H. Dong, C. Zhang and Y. S. Zhao, Adv. Opt. Mater., 2019, 1900037.
36. Y. L. Lin, L. L. Gong, K. J. Che, S. S. Li, C. X. Chu, Z. P. Cai, C. J. Yang and L. J. Chen, Appl. Phys. Lett., 2017, 110, 676–5778.
37. L. J. Chen, Y. N. Li, J. Fan, H. K. Bisoyi, D. A. Weitz and Q. Li, Adv. Opt. Mater., 2014, 2, 904.
38. L. J. Chen, L. L. Gong, Y. L. Lin, X. Y. Jin, H. Y. Li, S. S. Li, K. J. Che, Z. P. Cai and C. J. Yang, Lab Chip, 2016, 16, 1206–1213.
39. A. S. Utada, E. Lorenceau, D. R. Link, P. D. Kaplan, H. A. Stone and D. A. Weitz, Science, 2005, 308, 537–541.
40. S. Y. Teh, R. Lin, L. H. Hung and A. P. Lee, Lab Chip, 2008, 8, 198–220.
41. L. Shang, Y. Cheng and Y. Zhao, Chem. Rev., 2017, 117, 7964.
42. T. Y. Lee, T. M. Choi, T. S. Shim, R. A. Frijns and S. H. Kim, Lab Chip, 2016, 16, 3415.

---

Footnotes

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c9lc00655a

‡ These authors contributed equally.

---