Fluorescent proteins, owing to their excellent optical properties, biocompatibility, high quantum yield, remarkable photostability, and broadband tunability, exhibit significant potential in optical devices. Currently, most fluorescent materials tend to exhibit fluorescence quenching at high concentrations, whereas fluorescent proteins maintain good fluorescence characteristics in solid-state form, thus overcoming the abovementioned limitation. In this study, 2D whispering-gallery-mode (WGM) solid-state fluorescent protein lasers are fabricated via the coffee-ring effect on a silica substrate. Furthermore, to optimize laser performance, we combine solid-state fluorescent proteins with high-quality-factor microbubble cavities, which results in lower lasing thresholds and higher quality factors for microlasers. This study demonstrates the feasibility and practicality of solid-state fluorescent proteins in the fields of photonics and optoelectronics, which allows one to expand their application in solid-state lasers, thereby advancing the research and development of novel lasers.

Solid-state fluorescent proteins are used as gain media to successfully construct biocompatible 2D and 3D WGM lasers. The preparation method is simple and rapid, thus enabling one to effectively control the size and uniformity of geometric shapes. We express and purify mCherry fluorescent protein in Escherichia coli BL21 by transforming the pET30a vector containing the mCherry gene into Escherichia coli. The process involves culturing, inducing expression, cell lysis, and purification using a Ni-NTA gravity column, which results in highly concentrated purified protein. Subsequently, mCherry protein solution is applied to the surface of a 1-mm-diameter silica substrate, which forms an approximately 1-μm-thick 2D solid-state fluorescent protein micro-disk laser via the coffee-ring effect. To further enhance protein-laser performance, we combine solid-state fluorescent proteins with microbubble cavities. We process silica capillaries with a diameter of 140 μm by soaking them in piranha solution, etching them with hydrofluoric acid, and then heating them with hydrogen flame while they are stretched, which results in capillaries with an outer diameter of approximately 35 μm. Subsequently, the treated capillaries are heated with a CO₂ laser to form high-quality (Q) factor microbubbles with a diameter of approximately 100 μm and a wall thickness of approximately 1 μm. After injecting the protein solution, the microbubble cavities are rotated and dried to evaporate the water, thus resulting in the uniform attachment of the protein to the inner surface of the microbubble cavities. The optical modes of the 2D and 3D solid-state protein lasers are systematically investigated; the quality factor, threshold, and high-resolution spectra of the lasers are analyzed; and the corresponding simulations are performed, which shows good agreement between the experimental and simulated results.

The fluorescence lifetime, absorption, and photoluminescence spectra of fluorescent protein mCherry are measured (Fig. 2). The relationship between emission intensity and pump energy density for the 2D micro-disk cavity is determined under a threshold of 35.5 μJ/mm2. High-resolution spectrometer testing shows distinct peaks in the 648?654 nm range, thus indicating multimode lasing above the fluorescence background (Fig. 3). The Lorentzian fitting of the laser peak at 654?655 nm yields a full width at half maximum (FWHM) of 0.061 nm and a Q value of approximately 10⁴ at a pump energy density slightly above the threshold (39.3 μJ/mm2). Finite-element method (FEM) simulations using COMSOL Multiphysics show that the optical modes in the 2D fluorescent protein microdisk laser are strictly confined near the outer wall, thereby demonstrating excellent optical confinement (Fig. 4). Furthermore, 3D microbubble lasers are prepared by combining high-quality factor and small mode-volume microbubble cavities with solid-state fluorescent proteins. Lorentzian fitting performed for the transmission spectrum of the microbubble cavity coupled with a taper fiber yields a quality factor of approximately 7.1×10? (Fig. 5). The threshold for the 3D microbubble laser is measured at 1.33 μJ/mm2, which is only 1/27 of the 2D WGM threshold (Fig. 6). This is attributed to the high-quality factor and smaller WGM of the microbubble cavity. The transition from 2D to 3D provides new insights for exploring novel gain media and for analyzing the interactions between solid-state fluorescent proteins and substrate materials.

We successfully construct biocompatible 2D and 3D WGM lasers using solid-state fluorescent proteins as gain media. The preparation method is simple and rapid, thus enabling the size and uniformity of geometric shapes to be controlled effectively. The solid-state fluorescent proteins provide sufficient optical gain for the lasers, whereas the formation of 2D micro-disks and 3D microbubbles effectively confines light propagation, thus resulting in low-threshold multimode lasing. The systematic study and simulations show good agreement between the experimental and simulated results. The fabrication process for realizing a transition from 2D to 3D solid-state fluorescent protein lasers is a novel approach for developing biocompatible WGM lasers, thus providing a valuable platform for fundamental research in the field of nanophotonics. This methodology can be extended to other types of fluorescent proteins, which enables the customization of laser devices to satisfy specific requirements across different wavelength ranges. By further reducing the mode volume of the microbubble cavities and increasing the quality factor, one can achieve strong coupling mechanisms between solid-state fluorescent protein excitons and the microcavities. This strong coupling is crucial for investigating polariton lasing under room-temperature conditions with nanosecond pumping.