Adaptively laser-tagging optofluidic microcavity for single-molecule hydrogen detection across 9-decade concentration span

Adaptively laser-tagging optofluidic microcavity for single-molecule hydrogen detection across 9-decade concentration span | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Adaptively laser-tagging optofluidic microcavity for single-molecule hydrogen detection across 9-decade concentration span Baicheng Yao, Yanhong Guo, Shuya Yuan, Zihan Liu, Teng Tan, Zeping Wang, and 8 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8576965/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted You are reading this latest preprint version Abstract Optical microcavities are exceptional transducers for gas sensing, yet their performance is fundamentally constrained by the trade-off between sensitivity and dynamic range, degradation of quality factor ( Q ) upon integration with sensitive materials, and pervasive system noises. Here, we introduce a laser-tagging optofluidic microcavity that simultaneously overcomes these challenges for single-molecule hydrogen detection. Our architecture employs a hollow whispering-gallery-mode microcavity, functionally inner-coated with Pt/WO 3 nanofilm. Gas detection is mediated through thermal phonon transfer, which prevents the sensitive film from perturbing the optical field. This mechanism preserves an ultrahigh intrinsic Q factor of 1.89 × 10 9 even during sensing, therefore offers a unique tool to develop a laser-tagging method that dynamically locks the probe laser to the microcavity’s optimal operation point, enabling real-time resonance tracking. This approach suppresses phase noise by over three orders of magnitude and facilitates wide-bandwidth optoelectronic heterodyne demodulation. Consequently, we achieve a Hz-level frequency-shift resolution and a measurable resonance shift range of 1 GHz, breaking the conventional inverse relationship between accuracy and dynamic range. In single-shot measurement, our sensor achieves a minimum detectable concentration of 0.1 ppb and a maximum of 162 ppk, spanning 9 orders of magnitude. Moreover, after lock-in amplification, individual molecule dynamics are resolved. The integrated, centimetre-scale footprint of the device ensures robust ‘plug-and-play’ operation outside the laboratory, providing a universal strategy to advance optical microcavities towards a broad spectrum of ultra-precise metrology applications. Physical sciences/Optics and photonics/Applied optics/Optical sensors Physical sciences/Optics and photonics/Optical materials and structures/Microresonators Physical sciences/Optics and photonics/Other photonics/Optofluidics Physical sciences/Materials science/Materials for devices/Sensors and biosensors Physical sciences/Optics and photonics/Applied optics/Optoelectronic devices and components Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Introduction Advanced gas sensor devices serve as the “nose” of modern Internet of things (IoT) systems, playing a crucial role in contemporary production and daily life 1–4 . For instance, during energy extraction and transportation, accurately detecting gas leaks like hydrogen and alkanes is essential for ensuring the safety of mines, pipelines, and urban networks 5,6 . In power operation and maintenance, effectively identifying specific components such as hydrogen and sulfides is vital to ensuring substation safety 7,8 . In food and agricultural production, detecting significant volatile gases like NH 3 is a critical step in maintaining quality control 9,10 . In environmental monitoring, measuring gas molecules such as SO 2 and NO x is fundamental to evaluating pollution levels 11 . In the medical field, detecting respiratory gas molecules like VOCs is important for disease screening 12,13 . Optical gas sensors have demonstrated great potential in diverse biochemical sensing scenarios due to their unique advantages, including high sensitivity, wide bandwidth, and resistance to electromagnetic interference 14,15 . For example, gas analysis technology based on optical spectrum measurement offers unparalleled advantages for fingerprint recognition and high-precision detection of gas molecules 16–18 . Furthermore, the miniaturization of high-performance optical gas detection technology remains a significant focus for scientific research and a key step toward flexible deployment of optical gas sensors 19,20 . In recent years, the integration of various optical microstructures with nano-sensitive materials has enhanced gas sensor performance, with boosted sensitivities and reduced noise levels 21–23 . For example, the assembly of two-dimensional materials like graphene onto optical microstructures, such as D-shaped fibers 24,25 , fiber Bragg gratings 26 , and whispering-gallery-mode (WGM) microcavities 27,28 , effectively addresses the issue of the natural inertness of gas molecules in optics 29 . Specifically, leveraging functionalized microcavities 30 , scientists have achieved part-per-billion (ppb) level ultrahigh sensitivity measurements of various gases, including nitrogen oxides, nitrogen hydrides, carbon oxides, and even enabled the accurate discrimination of multiple gas components within a single microsensor 31–33 . However, there are still several important scientific challenges in current microcavity based gas sensing technology: 1) materials sensitive to gases often suffer significant optical losses, and thus their combination with microcavities can significantly sacrifice the microcavity Q value. This poses a key constraint on further improving the sensitivity of optical measurements and the accuracy of signal analysis 34–38 . 2) There is a natural contradiction between the response amplitude and measurement range, making it difficult to achieve both high sensitivity and a large dynamic range simultaneously 39–42 . 3) In traditional optical gas sensing devices, the lack of a feedback linkage mechanism between the light source and microcavity probe leads to the superposition of noises, which are difficult to suppress integrally 43–46 . These challenges stem from various aspects of physical mechanisms, microcavity structures, and system devices. By addressing these comprehensively, it is possible to significantly improve gas sensing performance and provide a reliable technical path for extending the feasibility of microcavity metrology to a broader range of biochemical sensing applications. Here, we present a real-time hydrogen molecule sensing scheme utilizing a laser-tagging microfluidic cavity. Through the functionalization of a hollow microfluidic WGM cavity with an inner coated metal oxide nanofilm, we achieve high-sensitivity detection of non-polar H 2 molecules. In this architecture, gas adsorption modifies optical time-frequency parameters via thermal phonon transfer 47–49 instead of carrier interaction 50 , and the nanomaterial film does not interfere with the optical resonant field, therefore addressing the critical issue of Q factor degradation and maintaining an intrinsic optical Q value of 1.89 × 10 9 . Accordingly, we propose a laser tagging method where probe light is locked to the optimal operating point of microcavity resonance through dynamic feedback, monitoring the spectral shift. This operation enables over 3-order phase noise suppression and facilitates wide bandwidth optoelectronic heterodyne demodulation. Therefore, we achieve a resolution for frequency-shift measurement on Hz level, with the measurable range of optical resonance frequency shift reaching 1 GHz, overcoming the traditional trade-off between high accuracy and large dynamic range. In single-shot measurements, while maintaining a minimum detectable concentration of 0.1 ppb, the maximum detectable concentration reaches 162 part-per-kilo (ppk). Moreover, after lock-in amplification, individual molecule dynamics is observed. Additionally, our technical solution demonstrates robust reliability and the practicality of “plug and play”. With the effective integration of on-chip optoelectronic links and microfluidic cavities, the entire sensing device is compact, showcasing a single centimetre-scale footprint, which supports broad applications out-of-lab. For validation, we demonstrate efficient detection of the H 2 component in transformer oil soluble gas detection scenarios in industry, achieving an accuracy rate exceeding 99%. Results Figure 1a illustrates the conceptual design for the hydrogen detection using a laser-tagging optofluidic microcavity. In this setup, the resonance of the Pt/WO 3 inner-coated microresonator (PCMR) locked by a 1550 nm probe laser, which in turn beats with a reference laser. The reference laser is stabilized to the ultra-stable Fabry–Pérot cavity via a side-of-slope feedback loop, ensuring excellent frequency stability. The sensing data is extracted from the beat note of these two lasers. During the laser tagging operation, transmission intensity changes of the probe laser are fed back to adjust the frequency of the itself, maintaining its lock on the slope of the resonance. This ensures that the optical frequency of the probe follows the resonance shifting. This implementation is akin to the stabilization of laser frequency combs 17,51 and optomechanical oscillators 52 , but simpler in operation. Through this operation, the measurement of microcavity resonant frequency drift is translated into the measurement of probe laser frequency, indicating potential precision enhancement using the heterodyne method, as the probe laser can conveniently beat with the reference laser in a photodetector. When hydrogen gas flows through the microfluidic hollow tube, H 2 molecules adsorb onto the Pt/WO 3 nanofilm, causing a corresponding change in the frequency of the probe laser locked on the resonance. Schematics and theoretical analysis are detailed in Supplementary Note S1. Figure 1b presents microscopic images of our microfluidic microcavity. A microrod with a width of 140 μm is formed on the outer wall of a capillary silica tube, which has an outer diameter of 2.8 mm, using CO 2 laser engraving method 44,53–55 . This design ensures few mode characteristics and a clean resonant spectrum. We apply a Pt/WO 3 nanofilm coating on the inner wall of the silica tube to capture H 2 molecules, with a film thickness of approximately 400 nm. The fabrication process of the optofluidic microcavity device is detailed in Supplementary Note S2. Figure 1c illustrates the sensing mechanism: when H 2 gas adsorbs onto the Pt/WO 3 nanofilm, hydrogen molecules dissociate into H atoms on the Pt catalyst surface and form equivalent bonds with oxygen on the WO 3 surface, allowing them to move and diffuse within the WO 3 lattice. During this movement and diffusion, H atoms react with O 2- ions to generate localized water 48,56,57 . The dissociation of H 2 releases heat, causing the microfluidic cavity temperature increasing, which subsequently alters the position of each optical resonance peak in the spectrum. Consequently, as the resonance is tracked, the frequency of the probe laser also shifts. This allows high-resolution detection of changes in the probe-reference laser beat note electronically. In Fig. 1d , we experimentally validate this method. Heating the PCMR from 296.11 K to 298.11 K results in the laser tagged on a resonance at approximately 1550 nm shifting from 0 to 780 MHz. Meanwhile, when cooling the PCMR from 298.11 K to 296.11 K, the laser frequency shifts back linearly. On average, the local thermal sensitivity of our microcavity exceeds 390 MHz/K. In Fig. 2a , we present the simulated electric field distributions within our optofluidic microcavity. The top panel displays the longitudinal mode (top view), while the bottom panel shows the transverse mode (side view). In this simulation, the refractive indices of the air and silica are 1 and 1.462, respectively. By optimizing the polarization of the probe light, our microcavity supports few-mode transmission 58 , which ensures a clean resonant spectrum, facilitating the laser tagging operations. Additional simulated mode distributions are provided in Supplementary Note S1, while related measurements are shown in Supplementary Note S3. Uniquely, in our design, the optical field does not overlap with the Pt/WO 3 film, as the thickness of the optofluidic tube is 400 μm (much larger than the optical mode volume), and the optical energy is confined by the micro-rod geometry. This ensures that the Q factor of our micro-rod is not degraded by the nanomaterial deposition. For verification, we record the spectral transmission in Fig. 1b . Totally we observe three mode families, with free spectral ranges around 21.8 GHz. Among them, the TE 02 mode exhibits the highest resonance intensity due to optimal tapered-fiber-microcavity coupling. We select this mode for laser tagging operation. Given the low intrinsic loss of our microcavity, during the laser-scan measurement, clear ringdowns are observed at each resonance. The zoomed-in trace provides more details of the ring-down curve for the TE 02 mode 44 . Referring to a scan speed of 10 nm/s in measurement, we find that the loaded Q factor reaches 0.94 ×10 9 near the critical coupling state, while the intrinsic Q factor approaches 1.89 × 10 9 . Maintaining a high- Q factor is essential for both boosting signal stabilization and enhancing sensitivity in a microcavity. In Fig. 2c , we compare the measured Q factor of our Pt/WO 3 deposited optofluidic microcavity with other microcavity platforms, which are widely used for chemical sensing. Typically, a WGM microresonator integrated with nanomaterials like graphene on its surface 31 has a loaded Q factor that can reach up to 1.4 × 10 8 , restricted by strong absorption loss. Conversely, an optofluidic bubble microcavity without a micro-rod design 59 can achieve a loaded Q factor of approximately 8 × 10 6 , limited by inevitable scattering losses and intermodal crosstalk. Figure 2d presents the measured probe-reference beat note at 10.8 MHz. Utilizing two separate laser diodes, we maintain a signal-to-noise ratio of the beat note that exceeds 40 dB. Unlike conventional sensing methods that rely on optical spectrum measurements, we determine the linewidth of the beat note using an electrical spectrum analyzer. Experimental findings indicate that the integral linewidth of the probe-reference beat note is 1.9 kHz, which is influenced by the laser linewidths rather than the microcavity resonance linewidth in the MHz range. More interestingly, in Fig. 2e , we present the measured single-sideband phase noises (SSB-PNs) of the dual-laser beat note. In free running operation, the SSB-PN of the dual-laser beat note is measured at 70.8 dBc/Hz at 1 Hz offset while -27.4 dBc/Hz at 1 kHz offset. When the probe laser is stabilized by tracking a resonance of the high- Q optofluidic microcavity through real-time feedback, the SSB-PN of the beat note between the stabilized probe and the reference laser is observed to be 37.2 dBc/Hz at 1 Hz offset and -46.8 dBc/Hz at 1 kHz offset. We estimate instantaneous linewidth of the PCMR stabilized probe laser reaches 10.4 Hz. This approach provides a method to detect frequency shifts with resolution down to the Hz level. The laser stabilization mechanism is further discussed in Supplementary Note S1. Additionally, Fig. 2f showcases the long-term stability measurements. Over 300 seconds, the probe-reference beat note without resonance triggering (free running) exhibits a random frequency drift of up to ±0.6 MHz. In contrast, our laser triggering scheme reduces the random frequency drift of the dual-laser beat note to within ±0.05 MHz. This more than one order improvement highlights the necessity and advantage of locking the probe laser on a high- Q microcavity resonance, as this implementation would significantly enhance sensing accuracy. More characterizations about stabilizing the probe laser and the reference laser are shown in Supplementary Note S3. Figure 3a illustrates the experimental setup for hydrogen detection using the laser-tagging optofluidic microcavity. A tapered fiber with a diameter of 1 μm is employed to launch and collect the probe light in the microcavity, while hydrogen gas is introduced via a soft tube. Gas preparation operations are shown in Supplementary Note S3. Concentration of the H 2 gas is accurately controlled by a gas distributor. We note that the entire optofluidic microcavity device is thermally regulated using a flexible polyimide foil heating element (Thorlabs, HT10K) for avoiding macroscopic temperature interference. Figure 3b presents the beat note spectra as the H 2 concentration increases. Initially, the frequency difference between the probe and reference lasers is set to 0 Hz by finely tuning the temperatures of the two cavities. As the H 2 concentration increases from 0 ppm to 100 ppk, the tracked probe follows the shift in the microcavity resonance, resulting in an increase in the beat note from 0 to 723.6 MHz. Throughout this process, the intensity of the dual laser beat note remains stable at around -40 dB, confirming that the laser tagging feedback based on the microcavity resonance is responsive. Additionally, it is observed that the noise level remains constant across this wide measurement range. Now we assess the sensitivity of our device. Figure 3c illustrates the correlation between the beat note shift and hydrogen concentration (represented by blue dots). The experimental data indicates that this relationship is nonlinear. Specifically, at lower hydrogen concentrations, the frequency shift exhibits a steeper slope. Sensitivity values are shown as red dots. For example, the detection of H 2 gas at a concentration of 20 ppb yields a sensitivity of approximately 30 MHz/ppm; at a concentration of 20 ppm, the sensitivity is around 0.4 MHz/ppm; and at a concentration of 20 ppk, the sensitivity reduces to about 0.013 MHz/ppm. This decreasing sensitivity with higher gas concentrations is primarily attributed to the saturation of hydrogen adsorption on the Pt/WO 3 film. Since integrated linewidth of the probe-reference beat note is limited on 1.9 kHz. In single shot, we estimate that the maximum measurable dynamic range of this device is 162 ppk in approximation. Meanwhile, in single-shot measurement, just based on the sensitivity 30 MHz/ppm, the minimum detectable concentration approaches 0.1 ppb. In Fig. 3d , we present the selectivity analysis of our hydrogen sensor. While the Pt/WO 3 nanofilm exhibits a strong chemical response to hydrogen gas, its microscale non-uniformity can result in numerous nanopores. These nanopores might facilitate the adsorption of other gases onto the film, causing slight variations in the dual laser beat frequency. Nonetheless, the selectivity of our device remains impressive. For instance, when employing the Pt/WO 3 based optofluidic microcavity to detect various gas species such as H 2 O, H 2 S, CO, CO 2 , CH 4 , C 2 H 4 , C 7 H 8 (toluene), CHCl 3 , SO 2 , NO 2 , NH 3 , C 3 H 6 O (acetone), CH 3 OH (methanol), and C 2 H 5 OH (ethanol), the maximum sensitivity does not exceed 0.029 MHz/ppm. Given that the long-term frequency uncertainty of our dual-laser beat note is ±0.05 MHz, the interference from other gases in hydrogen sensing can effectively be disregarded. Finally, in Fig. 3e , we illustrate the hydrogen sensor’s recoverability. By cyclically introducing hydrogen gas at varying concentrations and flushing the gas channel with dry air, we record changes in the frequency shift of the beat note. Experimental results show that the recoverability exceeds 98% in several minutes’ delay. This high degree of recoverability is attributed to the naturally reversible interaction between H 2 molecules and the Pt/WO 3 film, suggesting that our device can be reliably reused in gas sensing applications. The device exhibits that a response time and recovery time, defined as the time required to reach 90% of the steady state value, are 10 s and 15 s @ 1ppm H 2 , respectively. Both the response and recovery times increase with rising concentration, consistent with the characteristics of thermal response processes. More recovery spectral maps are shown in Supplementary Note S3. Since the laser-tagging scheme demonstrates ultrahigh spectral resolution (> 1 dB/kHz), we can employ lock-in amplification to enhance sensitivity further. Figure 4a illustrates the setup and mechanism for measurements based on lock-in amplification. The beat note of the probe and reference lasers is first filtered using an electrical bandpass filter to suppress random noise, then locked with a lock-in amplifier (Stanford SR860). The output from the lock-in amplifier, I L , is a continuous-wave signal proportional to the signal intensity at the locked frequency I S,f . When there is a change in the probe-reference beat note, I L is accordingly modified. Specifically, Δ I L = Δ I S , f I R , where I R is the inner reference intensity. Using the lock-in operation, this intensity modification can be amplified by over 60 dB due to noise cancellation after integration. In Fig. 4b , our scheme is depicted. Initially, we use a crystal oscillator to down-convert the probe-reference laser beat note (at 16.9 MHz) to 64.4 kHz, aligning with the bandwidth requirements of our lock-in amplifier. This signal linewidth suggests that the maximum measurement range for H 2 gas is less than 0.1 ppb when applying lock-in amplification. The down-converted signal exhibits an integrated linewidth of 6.28 kHz and an intensity SNR > 10 dB, with an integration time constant of 1 μs. The lock-in point is set at 68.7 kHz to monitor spectral shifts when sensing H 2 molecules. Figure 4c illustrates the lock-in traces observed during the experiment. Initially, the optofluidic microcavity is filled with dry air from 0 to 15 seconds. At 15 seconds, we introduce 0.1 ppb H 2 into the microtube, then maintaining this concentration until 45 seconds. Subsequently, we cleanse the optofluidic microcavity with dry air again from 60 to 75 seconds. During this process, the I L increases from 3.1 V to 4.2 V and subsequently decreases back, indicating dynamic changes in output intensity due to interactions between H 2 molecules and the intracavity deposited Pt/WO 3 . Upon closer examination of this trace, it becomes evident that when only dry air is present in the microtube, the I L remains stable. However, in the presence of H 2 , the trace exhibits temporal steps, indicating adsorption and desorption behaviors of H 2 molecules. The height of these steps is an integer multiple of 0.02 V, providing significant evidence for single-molecule dynamics. Given that the noise level of the lock-in temporal trace is below 0.005 V, single molecular interactions can be distinctly identified. Lastly, Fig. 4d presents a count of molecular steps occurring over one second (40 ~ 41 s), recording 843 adsorption/desorption events. Notably, the occurrence of single molecule adsorption (+1) or desorption (-1) is predominant. For steps with varying heights, their distribution generally follows a power-law pattern, consistent with the intrinsic nature of gas molecule dynamics 24,31,60,61 . Discussion Based on the laser tagging scheme, our microfluidic cavity hydrogen sensor, featuring both ultrahigh sensitivity and exceptionally wide dynamic range, can be compactly integrated into a fully functional system, resulting in a highly automated microsystem assembly with outstanding practical performance. As illustrated in Fig. 5a , the laser diodes, optofluidic microcavity, feedback electronics, and signal processing module are integrated into a single 30 cm × 30 cm × 10 cm chassis, making the system suitable for deployable applications beyond laboratory settings. Gas can be continuously introduced into and expelled from the optical microfluidic cavity through fluidic channels. We show more technical details in Supplementary Note S3, and an application demo in Supplementary Movie 1. To validate its performance, the sensor was applied to a representative use case: detecting hydrogen content in transformer insulation gas, a capability critical to power safety ( Fig. 5b ). We collected several gas samples extracted from operational transformers, in which the concentrations of common fault marker gases had been pre-calibrated. Our sensor was then used to measure the hydrogen concentrations in these samples. As shown in Fig. 5c , three different gas samples were tested, all with pure nitrogen as the background gas. Specifically, in gas sample #1, there were 0.4 ppm H 2 , 2 ppm CH 4 , 3.9 ppm C 2 H 4 , 15 ppm H 2 S, 2 ppm CO, 23 ppm SO 2 ; in gas sample #2, the gas mixture contained 7 ppm H 2 , 3 ppm CH 4 , 9 ppm C 2 H 2 , 2 ppm C 2 H 4 5 ppm H 2 S, 1 ppm CO; in gas sample #3, the components included 56 ppm H 2 , 13 ppm CH 4 , 6 ppm C 2 H 2 , 18 ppm C 2 H 4 , 1.9 ppm H 2 S, 7 ppm CO, 11 ppm SO 2 . For these samples, our sensor measured H 2 concentrations of 0.405 ppm, 6.92 ppm, and 56.04 ppm, respectively, achieving accuracy rates exceeding 99%. Thanks to the high hydrogen selectivity of the Pt/WO 3 sensing material, the presence of other gases does not interfere with hydrogen detection. Finally, Fig. 5d and 5e compare the performance of our sensor with other state-of-the-art hydrogen sensing devices, including both electrical 62–66 and optical 49,55,67–69 types. With a detection limit maintained at the 0.1 ppb level, the laser tagging approach uniquely enables a wide dynamic range exceeding 160 ppk. This combination of accuracy and range surpasses that of conventional electrical and optical detection methods. Moreover, within the Pt/WO 3 embedded microfluidic cavity architecture, the interaction between hydrogen and the nanomaterial is rapid, occurring within seconds, and remains stable across a broad range of temperature conditions. In this work, we demonstrate a laser-tagging optofluidic microcavity for single-molecule hydrogen detection across 9-decade concentration span. By depositing a layer of Pt/WO 3 nanofilm inside an optofluidic WGM microcavity, we achieve H 2 sensing via the thermal refraction effect caused by gas adsorption exotherm. In this strategy, the optical field does not overlap with the sensitizing material, achieving a high intrinsic Q factor approaching 1.89 × 10 9 . By locking a tunable laser onto this microcavity, we transform the conventional passive measurement method in optics into a highly coherent active measurement based on optoelectronics. This approach offers remarkable noise suppression, demodulation bandwidth expansion, and real-time capability. As a result, our method overcomes the challenge of achieving both ultrahigh sensitivity and an ultra-large dynamic range in chemical sensing. In single-shot H 2 sensing, we demonstrate a limit of detection down to 1 ppb and a measurable range up to 162 ppk. Leveraging lock-in amplification, we also observe single-molecule dynamics. Furthermore, this sensor illustrates outstanding selectivity, excellent recoverability, and plug-and-play practicality for use outside the laboratory, paving the way for various applications. Methods Mechanism of the laser stabilization and tagging . We utilize the slope-locking technique for laser frequency stabilization and dynamic tagging. This method employs a laser servo feedback loop to lock the laser frequency to the resonance peak of a high- Q optical cavity. When the laser frequency detunes from resonance, the reflected or transmitted optical intensity changes correspondingly. As a result, frequency fluctuations are converted into intensity variations, which are then transformed into voltage signals via a photodetector and a transimpedance amplifier. The performance of the slope-locking frequency stabilization scheme is fundamentally governed by the Q factor of the optical resonator. A higher Q factor corresponds to a narrower resonance linewidth and a steeper slope near resonance, thereby amplifying the error signal induced by laser frequency deviations and improving stabilization sensitivity. More details are shown in Supplementary Note S1. Fabrication and characterization of the PCMR . The silica hollow microrod resonator was fabricated using a laser reflow technique, which involves three main steps: polishing, etching, and annealing. A carbon dioxide laser beam, focused through a ZnSe lens, was used to process the microtube surface. Through computer-controlled automation, this 3-step process enables the fabrication of size controllable, highly uniform, few-mode, ultrahigh- Q microrod resonators within minutes. Pt/WO 3 nanoparticles were synthesized via a liquid-phase chemical method. A suspension is prepared by mixing Pt and WO 3 precursors in ethanol. After stirring and subsequent evaporation, Pt/WO 3 nanoparticles are deposited onto the inner wall of the microfluidic cavity. The morphology and composition of the intracavity nanomaterials are characterized using scanning electron microscopy and X-ray diffraction. More details are provided in Supplementary Note S2. Device details and optoelectronic encapsulation. The whole hydrogen sensing microsystem majorly includes a probe laser diode (14 PIN, Accelink), a reference laser diode (14 PIN, Accelink), low noise photodetectors (DET01CFC, Thorlabs), a PID tagging feedback loop and a signal processing core in FPGA (ACAU15, Xilinx). In optics: 1) the probe laser drives the microfluidic cavity; 2) the reference laser is stabilized by using an ultra-stable Fabry-Perot microresonator. In electronics: 1) after photodetection #1, the sensor modulated probe laser intensity alteration offers the signal for tuning the probe frequency. 2) after photodetection #2, the reflected intensity of the reference laser provides the signal for locking the reference frequency. 3) beat note of the probe laser and the reference laser are obtained after photodetection #3, frequency shift of the dual laser beat note tells the sensing response. We note that since the hydrogen detection relies on local temperature alteration, the whole microsystem requires a macroscopic thermal electrical cooler. More detailed performance metrics are provided in Supplementary Note S3. Declarations Acknowledgments The authors acknowledge support from the National Key Research and Development Program of China (Grant 2023YFB2806200), the National Natural Science Foundation of China (Grant U24A20311, 62305050, 62371106) and the Industrial Chain Project of Southern Grid. Author contributions B.C.Y. and Y.H.G. led the general study, and raised the scheme. Y.H.G. and T.T. performed the theoretical analysis. Y.H.G. and S.Y.Y. fabricated the microcavity devices. Y.H.G., Z.H.L. and Y.W. performed the material characterization. S.Y.Y., Z.H.L. and T.T. helped the Q factor measurement and optimization. Y.H.G., T.T., and B.C. contributed the laser stabilization and tagging measurements. Y.H.G., S.Y.Y., B.W.L. and H.D.X. built the setup and conducted the gas sensing experiment. Z.P.W., Z.H.X., Y.Q.Z., and L.P. helped the device package and out-of-lab verification. Y.J.R. supervised this work. All authors processed and analyzed the results. B.C.Y., Y.H.G. and T.T. prepared the manuscript. Competing interests Authors declare that they have no competing interests. Data and materials availability All data are available in the main text or the supplementary materials. References Jo, Y. et al. MOF‐Based Chemiresistive Gas Sensors: Toward New Functionalities. Adv. Mater. 35 , (2023). Tang, Y., Zhao, Y. & Liu, H. Room-Temperature Semiconductor Gas Sensors: Challenges and Opportunities. ACS Sensors 7 , 3582–3597 (2022). Sun, N. et al. Fully printable integrated multifunctional sensor arrays for intelligent lithium-ion batteries. Nat. Commun. 16 , 7361 (2025). Wang, L. et al. Review and Perspective: Gas Separation and Discrimination Technologies for Current Gas Sensors in Environmental Applications. ACS Sensors 8 , 1373–1390 (2023). Buttner, W. J., Post, M. B., Burgess, R. & Rivkin, C. An overview of hydrogen safety sensors and requirements. Int. J. Hydrogen Energy 36 , 2462–2470 (2011). He, T., Wang, W., He, B. G. & Chen, J. Review on Optical Fiber Sensors for Hazardous-Gas Monitoring in Mines and Tunnels. IEEE Trans. Instrum. Meas. 72 , 1–22 (2023). Hübert, T., Boon-Brett, L., Palmisano, V. & Bader, M. A. Developments in gas sensor technology for hydrogen safety. Int. J. Hydrogen Energy 39 , 20474–20483 (2014). Koo, W. T. et al. Chemiresistive hydrogen sensors: Fundamentals, recent advances, and challenges. ACS Nano 14 , 14284–14322 (2020). Timmer, B., Olthuis, W. & Berg, A. van den. Ammonia sensors and their applications—a review. Sensors Actuators B Chem. 107 , 666–677 (2005). Andre, R. S. et al. Recent Progress in Amine Gas Sensors for Food Quality Monitoring: Novel Architectures for Sensing Materials and Systems. ACS Sensors 7 , 2104–2131 (2022). Dhall, S., Mehta, B. R., Tyagi, A. K. & Sood, K. A review on environmental gas sensors: Materials and technologies. Sensors Int. 2 , 100116 (2021). Glöckler, J. et al. Infrared Spectroscopic Electronic Noses: An Innovative Approach for Exhaled Breath Sensing. ACS Sensors 10 , 427–438 (2025). Shin, J. et al. A non-contact wearable device for monitoring epidermal molecular flux. Nature 640 , 375–383 (2025). An, N. et al. Graphene-Fiber Biochemical Sensors: Principles, Implementations, and Advances. Photonic Sensors 11 , 123–139 (2021). Cao, Z. et al. Biochemical sensing in graphene-enhanced microfiber resonators with individual molecule sensitivity and selectivity. Light Sci. Appl. 8 , 107 (2019). Liang, Q., Bisht, A., Scheck, A., Schunemann, P. G. & Ye, J. Modulated ringdown comb interferometry for sensing of highly complex gases. Nature 638 , 941–948 (2025). Tan, T. et al. Biomimetic acoustic perception via chip-scale dual-soliton microcombs. eLight 5 , 22 (2025). Wang, Q. et al. Dual-comb photothermal spectroscopy. Nat. Commun. 13 , 2181 (2022). Yan, Y. et al. Nanoliter‐Scale Light–Matter Interaction in a Fiber‐Tip Cavity Enables Sensitive Photothermal Gas Detection. Laser Photon. Rev. 18 , (2024). Ma, J. et al. Microscale fiber photoacoustic spectroscopy for in situ and real-time trace gas sensing. Adv. Photonics 6 , (2024). Chen, J., Xiong, Y., Xu, F. & Lu, Y. Silica optical fiber integrated with two-dimensional materials: towards opto-electro-mechanical technology. Light Sci. Appl. 10 , 78 (2021). Guo, Y. et al. Optical Microcavities Empowered Biochemical Sensing: Status and Prospects. Adv. Devices Instrum. 5 , 1–28 (2024). Yao, B.-C. et al. Interdisciplinary advances in microcombs: bridging physics and information technology. eLight 4 , 19 (2024). An, N. et al. Electrically Tunable Four-Wave-Mixing in Graphene Heterogeneous Fiber for Individual Gas Molecule Detection. Nano Lett. 20 , 6473–6480 (2020). Wang, Y. et al. Noise canceled graphene-microcavity fiber laser sensor for ultrasensitive gas detection. Photonics Res. 11 , A1 (2023). Guo, Y. et al. Gas detection in a graphene based dual-mode fiber laser microcavity. Sensors Actuators B Chem. 348 , 130694 (2021). An, N. et al. Brillouin lasers in a graphene microresonator for multispecies and individual gas molecule detection. APL Photonics 8 , (2023). Liang, Y. et al. Harnessing sub-comb dynamics in a graphene-sensitized microresonator for gas detection. Front. Optoelectron. 17 , 12 (2024). Tan, T., Jiang, X., Wang, C., Yao, B. & Zhang, H. 2D Material Optoelectronics for Information Functional Device Applications: Status and Challenges. Adv. Sci. 7 , 2000058 (2020). Liu, J. et al. Emerging material platforms for integrated microcavity photonics. Sci. China Physics, Mech. Astron. 65 , 104201 (2022). Tan, T. et al. Multispecies and individual gas molecule detection using Stokes solitons in a graphene over-modal microresonator. Nat. Commun. 12 , 6716 (2021). Guo, Y. et al. A Monolithic Graphene‐Functionalized Microlaser for Multispecies Gas Detection. Adv. Mater. 34 , (2022). Zhang, H. et al. Gas Detection With Switchable Selectivity in a Functionalized-Graphene Integrated Microrod Resonator. Photonic Sensors 15 , 250423 (2025). Zhi, Y., Yu, X. C., Gong, Q., Yang, L. & Xiao, Y. F. Single Nanoparticle Detection Using Optical Microcavities. Adv. Mater. 29 , (2017). Yu, D. et al. Whispering-gallery-mode sensors for biological and physical sensing. Nat. Rev. Methods Prim. 1 , 83 (2021). Altug, H., Oh, S.-H., Maier, S. A. & Homola, J. Advances and applications of nanophotonic biosensors. Nat. Nanotechnol. 17 , 5–16 (2022). Jiang, X., Qavi, A. J., Huang, S. H. & Yang, L. Whispering-Gallery Sensors. Matter 3 , 371–392 (2020). Chen, W., Özdemir, Ş. K., Zhao, G., Wiersig, J. & Yang, L. Exceptional points enhance sensing in an optical microcavity. Nature 548 , 192–195 (2017). Wang, Y., Gao, R. & Xin, X. Hollow-core fiber refractive index sensor with high sensitivity and large dynamic range based on a multiple mode transmission mechanism. Opt. Express 29 , 19703 (2021). Lan, G. & Gao, Y. Surface Plasmon Resonance Sensor With High Sensitivity and Wide Dynamic Range. IEEE Sens. J. 18 , 5329–5333 (2018). Xing, T. et al. Fast Switching Acoustic Sensor With Ultrahigh Sensitivity and Wide Dynamic Response Range Based on Ultrahigh- Q CaF 2 Resonator. J. Light. Technol. 40 , 5775–5780 (2022). Liao, J. & Yang, L. Optical whispering-gallery mode barcodes for high-precision and wide-range temperature measurements. Light Sci. Appl. 10 , 32 (2021). Hu, J., Wan, F., Wang, P., Ge, H. & Chen, W. Application of frequency-locking cavity-enhanced spectroscopy for highly sensitive gas sensing: a review. Appl. Spectrosc. Rev. 57 , 378–410 (2022). Pan, T.-Y. et al. Boosting silica micro-rod Q factor to 8.28 × 109 for fully stabilizing a soliton microcomb. Nat. Commun. 16 , 8878 (2025). Cao, X., Yang, H., Wu, Z. L. & Li, B. B. Ultrasound sensing with optical microcavities. Light Sci. Appl. 13 , (2024). Cheng, H. et al. Harnessing micro-Fabry–Pérot reference cavities in photonic integrated circuits. Nat. Photonics 19 , 992–998 (2025). Si, B. et al. Core-shell nanostructures Pt-WO3 for colorimetric and fiber Bragg grating hydrogen sensors. Int. J. Hydrogen Energy 137 , 775–782 (2025). Dai, J. et al. Performance of fiber Bragg grating hydrogen sensor coated with Pt-loaded WO3 coating. Sensors Actuators B Chem. 190 , 657–663 (2014). Wang, C. et al. Highly sensitive fiber grating hydrogen sensor based on hydrogen-doped Pt/WO3. Sensors Actuators B Chem. 404 , 135250 (2024). Yao, B. et al. Graphene-Enhanced Brillouin Optomechanical Microresonator for Ultrasensitive Gas Detection. Nano Lett. 17 , 4996–5002 (2017). Chang, B. et al. Dispersive Fourier transform based dual-comb ranging. Nat. Commun. 15 , 4990 (2024). Zhang, H. et al. Mode-locked optomechanical frequency combs in a graphene-silica microresonator. Sci. Adv. 11 , (2025). Wang, M. et al. Cascading-Induced Fundamental Linewidth Enhancement of a Microcavity Brillouin Laser. ACS Photonics (2025) doi:10.1021/acsphotonics.5c00127. Cui, W. et al. Ultra-low time jitter transform-limited dissipative Kerr soliton microcomb. Opt. Express 31 , 37154 (2023). Gao, F. et al. Sub-ppb-level sensitive hydrogen gas sensing based on PDMS-Pt enhanced microrod cavity. Opt. Express 33 , 42043 (2025). Liu, L., Liu, Y. & Liu, C. Enhancing the Understanding of Hydrogen Evolution and Oxidation Reactions on Pt(111) through Ab Initio Simulation of Electrode/Electrolyte Kinetics. J. Am. Chem. Soc. 142 , 4985–4989 (2020). Koo, W.-T. et al. Chemiresistive Hydrogen Sensors: Fundamentals, Recent Advances, and Challenges. ACS Nano 14 , 14284–14322 (2020). Zhang, H. et al. Soliton Microcombs Multiplexing Using Intracavity-Stimulated Brillouin Lasers. Phys. Rev. Lett. 130 , 153802 (2023). Yao, B. et al. Graphene-Enhanced Brillouin Optomechanical Microresonator for Ultrasensitive Gas Detection. Nano Lett. 17 , 4996–5002 (2017). Clauset, A., Shalizi, C. R. & Newman, M. E. J. Power-Law Distributions in Empirical Data. SIAM Rev. 51 , 661–703 (2009). Tartaglione, V., Farges, C. & Sabatier, J. Nonlinear dynamical modeling of adsorption and desorption processes with power-law kinetics: Application to CO2 capture. Phys. Rev. E 102 , 052102 (2020). Zhang, X. et al. Ultralow detection limit and ultrafast response/recovery of the H2 gas sensor based on Pd-doped rGO/ZnO-SnO2 from hydrothermal synthesis. Microsystems Nanoeng. 8 , 67 (2022). Agarwal, S. et al. An efficient hydrogen gas sensor based on hierarchical Ag/ZnO hollow microstructures. Sensors Actuators B Chem. 346 , 130510 (2021). Nakate, U. T., Ahmad, R., Patil, P., Yu, Y. T. & Hahn, Y.-B. Ultra thin NiO nanosheets for high performance hydrogen gas sensor device. Appl. Surf. Sci. 506 , 144971 (2020). Luo, N. et al. Ultralow detection limit MEMS hydrogen sensor based on SnO2 with oxygen vacancies. Sensors Actuators B Chem. 354 , 130982 (2022). Mandal, S. et al. A robust organic hydrogen sensor for distributed monitoring applications. Nat. Electron. 8 , 343–352 (2025). Yan, X. et al. High Sensitivity Hydrogen Sensors Based on Optical Micro/Nanofiber Couplers Working at the Dispersion Turning Point. J. Light. Technol. 41 , 4283–4289 (2023). Yue, Y., Ding, H. & Chen, C. Optical hydrogen sensors based on silica self-assembled mesoporous microspheres. Int. J. Hydrogen Energy 46 , 1403–1410 (2021). Boelsma, C. et al. Hafnium—an optical hydrogen sensor spanning six orders in pressure. Nat. Commun. 8 , 15718 (2017). Additional Declarations There is NO Competing Interest. Supplementary Files SupplementaryMovie.mp4 Supplementary Movie 002SupplementaryInformation.docx Supplementary Information of Adaptively laser-tagging optofluidic microcavity for single-molecule hydrogen detection across 9-decade concentration span Cite Share Download PDF Status: Under Review Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8576965","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":585213407,"identity":"63051917-6190-4694-9ab8-4bfe9cfc8596","order_by":0,"name":"Baicheng Yao","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABB0lEQVRIiWNgGAWjYHACNgYGA4YEIIPxAUxIglgtzAYw1URoYQBrYZMgSovB8cPHHvwosMvjl26/Vs1TUVfH38B88DYPg10eTi1n0tINewySiyXnnCm7zXPmsITEAbZkax6G5GKcWg7kmEnwGBxI3HAjJ+02b9sBCQMGHjNpHoYDiQ24tJx/Yyb5B6hlP1BLMe+/OqAW/m/4tdzIAZoJskUi/RgzbwMzyBY2vFokbzxLN5YxSE6ccSOHWXLOscOSMw6zGVvOAYrg0sJ3PvnYwzd/7BL7Z6Q//PCmpo6fv7354Y03FXY4tSgcgDN5DCA0M9jBONQDgTzCLPYHuJWNglEwCkbBiAYAA2FVti7/qaoAAAAASUVORK5CYII=","orcid":"https://orcid.org/0000-0001-8368-5815","institution":"University of Electronic Science and Technology of China","correspondingAuthor":true,"prefix":"","firstName":"Baicheng","middleName":"","lastName":"Yao","suffix":""},{"id":585213408,"identity":"670ae663-1fc8-4b9c-9ba4-115f6455722d","order_by":1,"name":"Yanhong Guo","email":"","orcid":"","institution":"Key Laboratory of Optical Fiber Sensing and Communications (Education Ministry of China), University of Electronic Science and Technology of China","correspondingAuthor":false,"prefix":"","firstName":"Yanhong","middleName":"","lastName":"Guo","suffix":""},{"id":585213409,"identity":"545785e2-6479-48a7-9059-06a1f4b9ae5d","order_by":2,"name":"Shuya Yuan","email":"","orcid":"","institution":"Key Laboratory of Optical Fiber Sensing and Communications (Education Ministry of China), University of Electronic Science and Technology of China","correspondingAuthor":false,"prefix":"","firstName":"Shuya","middleName":"","lastName":"Yuan","suffix":""},{"id":585213410,"identity":"336fc422-2ff5-4233-8da1-169961e5bea6","order_by":3,"name":"Zihan Liu","email":"","orcid":"","institution":"Key Laboratory of Optical Fiber Sensing and Communications (Education Ministry of China), University of Electronic Science and Technology of China","correspondingAuthor":false,"prefix":"","firstName":"Zihan","middleName":"","lastName":"Liu","suffix":""},{"id":585213411,"identity":"7eb13add-b6ae-4081-8aec-10f87ac9c067","order_by":4,"name":"Teng Tan","email":"","orcid":"https://orcid.org/0000-0003-1820-3618","institution":"University of Electronic Science and Technology of China","correspondingAuthor":false,"prefix":"","firstName":"Teng","middleName":"","lastName":"Tan","suffix":""},{"id":585213412,"identity":"131ad71e-8366-45b1-9321-05ba65c47929","order_by":5,"name":"Zeping Wang","email":"","orcid":"","institution":"Key Laboratory of Optical Fiber Sensing and Communications (Education Ministry of China), University of Electronic Science and Technology of China","correspondingAuthor":false,"prefix":"","firstName":"Zeping","middleName":"","lastName":"Wang","suffix":""},{"id":585213413,"identity":"744f0255-983f-4d49-be43-bcca7f634fc0","order_by":6,"name":"Yaqian Zhao","email":"","orcid":"","institution":"China Southern Power Grid Co., Ltd.","correspondingAuthor":false,"prefix":"","firstName":"Yaqian","middleName":"","lastName":"Zhao","suffix":""},{"id":585213414,"identity":"1a632753-8ab6-4d94-bdc6-0348bf76b3a3","order_by":7,"name":"Bing Chang","email":"","orcid":"","institution":"University of Electronic Science and Technology of China","correspondingAuthor":false,"prefix":"","firstName":"Bing","middleName":"","lastName":"Chang","suffix":""},{"id":585213415,"identity":"b2ab0ec8-0773-4988-b138-a3dc607a0f48","order_by":8,"name":"Zhenheng Xu","email":"","orcid":"","institution":"China Southern Power Grid Co., Ltd.","correspondingAuthor":false,"prefix":"","firstName":"Zhenheng","middleName":"","lastName":"Xu","suffix":""},{"id":585213416,"identity":"97940bc0-b387-451e-93db-047db0107768","order_by":9,"name":"Handing Xia","email":"","orcid":"","institution":"China Academic of Engineering Physics","correspondingAuthor":false,"prefix":"","firstName":"Handing","middleName":"","lastName":"Xia","suffix":""},{"id":585213417,"identity":"6bb68651-14ac-4508-a0ac-13008459ae04","order_by":10,"name":"Lei Peng","email":"","orcid":"","institution":"Guang-dong Provincial Key Laboratory of Electric Power Equipment Reliability, Electric Power Research Institute of Guangdong Power Grid Co.,Ltd.","correspondingAuthor":false,"prefix":"","firstName":"Lei","middleName":"","lastName":"Peng","suffix":""},{"id":585213418,"identity":"9adeb4c3-2353-4ba0-bd9b-0ed4acef4cb6","order_by":11,"name":"Yu Wu","email":"","orcid":"","institution":"University of Electronic Science and Technology of China","correspondingAuthor":false,"prefix":"","firstName":"Yu","middleName":"","lastName":"Wu","suffix":""},{"id":585213419,"identity":"af045a40-5347-4426-b3c5-ac73d1181952","order_by":12,"name":"Bowen Li","email":"","orcid":"","institution":"Key Laboratory of Optical Fiber Sensing and Communications (Education Ministry of China), University of Electronic Science and Technology of China","correspondingAuthor":false,"prefix":"","firstName":"Bowen","middleName":"","lastName":"Li","suffix":""},{"id":585213420,"identity":"2fafcedc-4247-44a5-b9f6-ee63d3bc8180","order_by":13,"name":"Yunjiang Rao","email":"","orcid":"https://orcid.org/0000-0003-0717-5586","institution":"University of Electronic Science and Technology of China","correspondingAuthor":false,"prefix":"","firstName":"Yunjiang","middleName":"","lastName":"Rao","suffix":""}],"badges":[],"createdAt":"2026-01-12 03:40:09","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8576965/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8576965/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":101840457,"identity":"8fd8604d-efea-4211-bd68-c56ef5c75277","added_by":"auto","created_at":"2026-02-04 08:27:44","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":1633905,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eConceptual design and principle of the laser-tagging optofluidic microcavity for single-molecule hydrogen detection. a, \u003c/strong\u003eIn the setup, hydrogen gas stream is launched into the optofluidic channel, while frequency of the probe laser follows the shift of the microcavity. PD#1 is used for laser feedback, while probe-reference beat note is generated in PD#2. \u003cstrong\u003eb,\u003c/strong\u003e Microscopic pictures of optofluidic microcavity. In these images, Pt/WO\u003csub\u003e3\u003c/sub\u003e is deposited on the inner wall, while a micro-rod is fabricated on the outer wall to achieve a high-\u003cem\u003eQ\u003c/em\u003e factor. \u003cstrong\u003ec,\u003c/strong\u003e Sensing mechanism. H\u003csub\u003e2\u003c/sub\u003e adsorption on Pt/WO\u003csub\u003e3\u003c/sub\u003e releases heat, changing frequency of the probe-reference beat note. \u003cstrong\u003ed,\u003c/strong\u003e Measured resonance shift of our optofluidic microcavity with temperature increment and decrement.\u0026nbsp;\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-8576965/v1/e7033af5519c0e7f3f7292fd.png"},{"id":101881502,"identity":"1f25ea56-5c3f-49eb-bbfe-5cc4579f77dd","added_by":"auto","created_at":"2026-02-04 15:12:34","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":869694,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eProperties of the optofluidic microcavity and stabilization based on laser-tagging a, \u003c/strong\u003eSimulated (electrical field) distributions of TE\u003csub\u003e02\u003c/sub\u003e mode intracavity. Top view: longitude distribution; bottom view: transverse distribution. \u003cstrong\u003eb,\u003c/strong\u003e Measured transmission spectrum. Top: resonances across 0.4 nm band. Bottom: Ring-down curve of the TE\u003csub\u003e02\u003c/sub\u003e mode. \u003cstrong\u003ec,\u003c/strong\u003e Comparison of the loaded \u003cem\u003eQ\u003c/em\u003e factors, in diverse microcavity platforms for chemical sensing. \u003cstrong\u003ed,\u003c/strong\u003e Beat note of the probe laser and the reference laser. Measured integrated linewidth of the beat note approached 1.9 kHz. \u003cstrong\u003ee-f,\u003c/strong\u003e Measured SSB-PN and long-term frequency stability of the dual laser beat note, in free-running operation and in probe-tagging operation. In \u003cstrong\u003ee,\u003c/strong\u003e the grey dashed line shows the thermal noise limitation.\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-8576965/v1/52cf72e3df193971fa28ecff.png"},{"id":101840460,"identity":"71305d3a-882d-434a-bc33-5d3fda98bb37","added_by":"auto","created_at":"2026-02-04 08:27:46","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":1384822,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003ePerformances for hydrogen detection. a, \u003c/strong\u003ePicture of the sensing setup. \u003cstrong\u003eb,\u003c/strong\u003e Measured dual laser beat notes. With H\u003csub\u003e2\u003c/sub\u003e concentration increasing from 1 ppm to 100 ppk, the beat note shifts from 7.3 MHz to 723.6 MHz. \u003cstrong\u003ec,\u003c/strong\u003e Sensitivity of the device varies with the concentration of H\u003csub\u003e2\u003c/sub\u003e. \u003cstrong\u003ed,\u003c/strong\u003e Selectivity. This device is selectively sensitive to H\u003csub\u003e2\u003c/sub\u003e. \u003cstrong\u003ee,\u003c/strong\u003e Recoverability of the optofluidic microcavity sensor.\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-8576965/v1/88cb43734dc2c826436144f7.png"},{"id":101840478,"identity":"9c7c1961-dcc2-4ad9-a902-5c3844f43c45","added_by":"auto","created_at":"2026-02-04 08:27:48","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":605067,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSingle molecule detection based on lock-in amplification. a, \u003c/strong\u003ePrinciple and experimental setup of the lock-in amplification-based measurement. \u003cstrong\u003eb,\u003c/strong\u003e Measured spectra. Top: we use an electrically generated signal (17.544 MHz) to down-convert the probe-reference laser beat (16.9 MHz). Bottom: the down-converted signal, which is sent into the lock-in amplifier. \u003cstrong\u003ec,\u003c/strong\u003e Top: Output trace of the lock-in amplifier. In 90 s, we launch H\u003csub\u003e2\u003c/sub\u003e into the microfluidic tube and then cleanse it. Bottom: zoomed-in sections. Here the grey curve and the yellow curve shows the cases in N\u003csub\u003e2\u003c/sub\u003e and in H\u003csub\u003e2\u003c/sub\u003e, respectively. \u003cstrong\u003ed,\u003c/strong\u003e Statistics of the steps during one second. Here \u003cem\u003e+N\u003c/em\u003e means \u003cem\u003eN\u003c/em\u003e molecules adsorption, \u003cem\u003e-N\u003c/em\u003e means \u003cem\u003eN\u003c/em\u003e molecules desorption.\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-8576965/v1/7de3887ad80bcb43c566a624.png"},{"id":101881501,"identity":"7b42f599-da13-4889-bf16-2485299ded37","added_by":"auto","created_at":"2026-02-04 15:12:33","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":1605681,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eDiscussion. a, \u003c/strong\u003ePicture of the practically encapsulated hydrogen sensor device.\u003cstrong\u003e b, \u003c/strong\u003eApplication scenario, detecting hydrogen in an electrical transformer.\u003cstrong\u003e c, \u003c/strong\u003eHydrogen measurement in gas mixtures, here H\u003csub\u003e2\u003c/sub\u003e concentration in gas samples 1-3 are 0.4 ppm, 7 ppm and 56 ppm, respectively. This verifies that the sensing of hydrogen can effectively resist interference from other gases. \u003cstrong\u003ed-e, \u003c/strong\u003ePerformance comparisons. We compare limit of detect and dynamic range of advanced H\u003csub\u003e2\u003c/sub\u003e sensors in (d), and compare response time and operating temperature of advanced H\u003csub\u003e2\u003c/sub\u003e sensors in (e). Blue triangles, yellow circle dots, and yellow rhombus dots show performances of electrical sensors, optical WGM microcavity sensors and optical fiber sensors.\u0026nbsp;\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-8576965/v1/8b06efdd891ede74b4a35996.png"},{"id":101882949,"identity":"d436cc5e-ad64-42d3-8975-8a3880c0a135","added_by":"auto","created_at":"2026-02-04 15:26:08","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":8945529,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8576965/v1/e814bc63-3f2f-4362-8767-306d79cba685.pdf"},{"id":101840482,"identity":"6a8775c3-4311-4cb6-9508-0763c28be366","added_by":"auto","created_at":"2026-02-04 08:27:49","extension":"mp4","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":5340433,"visible":true,"origin":"","legend":"Supplementary Movie","description":"","filename":"SupplementaryMovie.mp4","url":"https://assets-eu.researchsquare.com/files/rs-8576965/v1/ff351eff237f00f368bc7cf6.mp4"},{"id":101840461,"identity":"1494c0b4-2c2b-4e8d-a4b4-ce1467803ee5","added_by":"auto","created_at":"2026-02-04 08:27:47","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":9747150,"visible":true,"origin":"","legend":"Supplementary Information of Adaptively laser-tagging optofluidic microcavity for single-molecule hydrogen detection across 9-decade concentration span","description":"","filename":"002SupplementaryInformation.docx","url":"https://assets-eu.researchsquare.com/files/rs-8576965/v1/d0b14b41ff4e8fe67d0ed3f6.docx"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Adaptively laser-tagging optofluidic microcavity for single-molecule hydrogen detection across 9-decade concentration span","fulltext":[{"header":"Introduction","content":"\u003cp\u003eAdvanced gas sensor devices serve as the \u0026ldquo;nose\u0026rdquo; of modern Internet of things (IoT) systems, playing a crucial role in contemporary production and daily life\u0026nbsp;\u003csup\u003e1\u0026ndash;4\u003c/sup\u003e. For instance, during energy extraction and transportation, accurately detecting gas leaks like hydrogen and alkanes is essential for ensuring the safety of mines, pipelines, and urban networks \u003csup\u003e5,6\u003c/sup\u003e. In power operation and maintenance, effectively identifying specific components such as hydrogen and sulfides is vital to ensuring substation safety \u003csup\u003e7,8\u003c/sup\u003e. In food and agricultural production, detecting significant volatile gases like NH\u003csub\u003e3\u003c/sub\u003e is a critical step in maintaining quality control \u003csup\u003e9,10\u003c/sup\u003e. In environmental monitoring, measuring gas molecules such as SO\u003csub\u003e2\u003c/sub\u003e and NO\u003cem\u003e\u003csub\u003ex\u003c/sub\u003e\u003c/em\u003e is fundamental to evaluating pollution levels \u003csup\u003e11\u003c/sup\u003e. In the medical field, detecting respiratory gas molecules like VOCs is important for disease screening \u003csup\u003e12,13\u003c/sup\u003e. Optical gas sensors have demonstrated great potential in diverse biochemical sensing scenarios due to their unique advantages, including high sensitivity, wide bandwidth, and resistance to electromagnetic interference \u003csup\u003e14,15\u003c/sup\u003e. For example, gas analysis technology based on optical spectrum measurement offers unparalleled advantages for fingerprint recognition and high-precision detection of gas molecules \u003csup\u003e16\u0026ndash;18\u003c/sup\u003e. Furthermore, the miniaturization of high-performance optical gas detection technology remains a significant focus for scientific research and a key step toward flexible deployment of optical gas sensors \u003csup\u003e19,20\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eIn recent years, the integration of various optical microstructures with nano-sensitive materials has enhanced gas sensor performance, with boosted sensitivities and reduced noise levels \u003csup\u003e21\u0026ndash;23\u003c/sup\u003e. For example, the assembly of two-dimensional materials like graphene onto optical microstructures, such as D-shaped fibers \u003csup\u003e24,25\u003c/sup\u003e, fiber Bragg gratings \u003csup\u003e26\u003c/sup\u003e, and whispering-gallery-mode (WGM) microcavities \u003csup\u003e27,28\u003c/sup\u003e, effectively addresses the issue of the natural inertness of gas molecules in optics \u003csup\u003e29\u003c/sup\u003e. Specifically, leveraging functionalized microcavities \u003csup\u003e30\u003c/sup\u003e, scientists have achieved part-per-billion (ppb) level ultrahigh sensitivity measurements of various gases, including nitrogen oxides, nitrogen hydrides, carbon oxides, and even enabled the accurate discrimination of multiple gas components within a single microsensor \u003csup\u003e31\u0026ndash;33\u003c/sup\u003e.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eHowever, there are still several important scientific challenges in current microcavity based gas sensing technology: 1) materials sensitive to gases often suffer significant optical losses, and thus their combination with microcavities can significantly sacrifice the microcavity \u003cem\u003eQ\u003c/em\u003e value. This poses a key constraint on further improving the sensitivity of optical measurements and the accuracy of signal analysis \u003csup\u003e34\u0026ndash;38\u003c/sup\u003e. 2) There is a natural contradiction between the response amplitude and measurement range, making it difficult to achieve both high sensitivity and a large dynamic range simultaneously \u003csup\u003e39\u0026ndash;42\u003c/sup\u003e. 3) In traditional optical gas sensing devices, the lack of a feedback linkage mechanism between the light source and microcavity probe leads to the superposition of noises, which are difficult to suppress integrally \u003csup\u003e43\u0026ndash;46\u003c/sup\u003e. These challenges stem from various aspects of physical mechanisms, microcavity structures, and system devices. By addressing these comprehensively, it is possible to significantly improve gas sensing performance and provide a reliable technical path for extending the feasibility of microcavity metrology to a broader range of biochemical sensing applications.\u003c/p\u003e\n\u003cp\u003eHere, we present a real-time hydrogen molecule sensing scheme utilizing a laser-tagging microfluidic cavity. Through the functionalization of a hollow microfluidic WGM cavity with an inner coated metal oxide nanofilm, we achieve high-sensitivity detection of non-polar H\u003csub\u003e2\u003c/sub\u003e molecules. In this architecture, gas adsorption modifies optical time-frequency parameters via thermal phonon transfer \u003csup\u003e47\u0026ndash;49\u003c/sup\u003e instead of carrier interaction \u003csup\u003e50\u003c/sup\u003e, and the nanomaterial film does not interfere with the optical resonant field, therefore addressing the critical issue of \u003cem\u003eQ\u003c/em\u003e factor degradation and maintaining an intrinsic optical \u003cem\u003eQ\u003c/em\u003e value of 1.89 \u0026times; 10\u003csup\u003e9\u003c/sup\u003e. Accordingly, we propose a laser tagging method where probe light is locked to the optimal operating point of microcavity resonance through dynamic feedback, monitoring the spectral shift. This operation enables over 3-order phase noise suppression and facilitates wide bandwidth optoelectronic heterodyne demodulation. Therefore, we achieve a resolution for frequency-shift measurement on Hz level, with the measurable range of optical resonance frequency shift reaching 1 GHz, overcoming the traditional trade-off between high accuracy and large dynamic range. In single-shot measurements, while maintaining a minimum detectable concentration of 0.1 ppb, the maximum detectable concentration reaches 162 part-per-kilo (ppk). Moreover, after lock-in amplification, individual molecule dynamics is observed. Additionally, our technical solution demonstrates robust reliability and the practicality of \u0026ldquo;plug and play\u0026rdquo;. With the effective integration of on-chip optoelectronic links and microfluidic cavities, the entire sensing device is compact, showcasing a single centimetre-scale footprint, which supports broad applications out-of-lab. For validation, we demonstrate efficient detection of the H\u003csub\u003e2\u003c/sub\u003e component in transformer oil soluble gas detection scenarios in industry, achieving an accuracy rate exceeding 99%.\u0026nbsp;\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cstrong\u003eFigure 1a\u003c/strong\u003e illustrates the conceptual design for the hydrogen detection using a laser-tagging optofluidic microcavity. In this setup, the resonance of the Pt/WO\u003csub\u003e3\u003c/sub\u003e inner-coated microresonator (PCMR) locked by a 1550 nm probe laser, which in turn beats with a reference laser. The reference laser is stabilized to the ultra-stable Fabry\u0026ndash;P\u0026eacute;rot cavity via a side-of-slope feedback loop, ensuring excellent frequency stability. The sensing data is extracted from the beat note of these two lasers. During the laser tagging operation, transmission intensity changes of the probe laser are fed back to adjust the frequency of the itself, maintaining its lock on the slope of the resonance. This ensures that the optical frequency of the probe follows the resonance shifting. This implementation is akin to the stabilization of laser frequency combs \u003csup\u003e17,51\u003c/sup\u003e and optomechanical oscillators \u003csup\u003e52\u003c/sup\u003e, but simpler in operation. Through this operation, the measurement of microcavity resonant frequency drift is translated into the measurement of probe laser frequency, indicating potential precision enhancement using the heterodyne method, as the probe laser can conveniently beat with the reference laser in a photodetector. When hydrogen gas flows through the microfluidic hollow tube, H\u003csub\u003e2\u003c/sub\u003e molecules adsorb onto the Pt/WO\u003csub\u003e3\u003c/sub\u003e nanofilm, causing a corresponding change in the frequency of the probe laser locked on the resonance. Schematics and theoretical analysis are detailed in Supplementary Note S1.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFigure 1b\u003c/strong\u003e presents microscopic images of our microfluidic microcavity. A microrod with a width of 140 \u0026mu;m is formed on the outer wall of a capillary silica tube, which has an outer diameter of 2.8 mm, using CO\u003csub\u003e2\u003c/sub\u003e laser engraving method \u003csup\u003e44,53\u0026ndash;55\u003c/sup\u003e. This design ensures few mode characteristics and a clean resonant spectrum. We apply a Pt/WO\u003csub\u003e3\u003c/sub\u003e nanofilm coating on the inner wall of the silica tube to capture H\u003csub\u003e2\u003c/sub\u003e molecules, with a film thickness of approximately 400 nm. The fabrication process of the optofluidic microcavity device is detailed in Supplementary Note S2. \u003cstrong\u003eFigure 1c\u003c/strong\u003e illustrates the sensing mechanism: when H\u003csub\u003e2\u003c/sub\u003e gas adsorbs onto the Pt/WO\u003csub\u003e3\u003c/sub\u003e nanofilm, hydrogen molecules dissociate into H atoms on the Pt catalyst surface and form equivalent bonds with oxygen on the WO\u003csub\u003e3\u003c/sub\u003e surface, allowing them to move and diffuse within the WO\u003csub\u003e3\u003c/sub\u003e lattice. During this movement and diffusion, H atoms react with O\u003csup\u003e2-\u003c/sup\u003e ions to generate localized water \u003csup\u003e48,56,57\u003c/sup\u003e.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cimg src=\"data:image/png;base64,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\"\u003e\u003c/p\u003e\n\u003cp\u003eThe dissociation of H\u003csub\u003e2\u003c/sub\u003e releases heat, causing the microfluidic cavity temperature increasing, which subsequently alters the position of each optical resonance peak in the spectrum. Consequently, as the resonance is tracked, the frequency of the probe laser also shifts. This allows high-resolution detection of changes in the probe-reference laser beat note electronically. In \u003cstrong\u003eFig. 1d\u003c/strong\u003e, we experimentally validate this method. Heating the PCMR from 296.11 K to 298.11 K results in the laser tagged on a resonance at approximately 1550 nm shifting from 0 to 780 MHz. Meanwhile, when cooling the PCMR from 298.11 K to 296.11 K, the laser frequency shifts back linearly. On average, the local thermal sensitivity of our microcavity exceeds 390 MHz/K.\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;In \u003cstrong\u003eFig. 2a\u003c/strong\u003e, we present the simulated electric field distributions within our optofluidic microcavity. The top panel displays the longitudinal mode (top view), while the bottom panel shows the transverse mode (side view). In this simulation, the refractive indices of the air and silica are 1 and 1.462, respectively. By optimizing the polarization of the probe light, our microcavity supports few-mode transmission \u003csup\u003e58\u003c/sup\u003e, which ensures a clean resonant spectrum, facilitating the laser tagging operations. Additional simulated mode distributions are provided in Supplementary Note S1, while related measurements are shown in Supplementary Note S3. Uniquely, in our design, the optical field does not overlap with the Pt/WO\u003csub\u003e3\u003c/sub\u003e film, as the thickness of the optofluidic tube is 400 \u0026mu;m (much larger than the optical mode volume), and the optical energy is confined by the micro-rod geometry. This ensures that the \u003cem\u003eQ\u003c/em\u003e factor of our micro-rod is not degraded by the nanomaterial deposition. For verification, we record the spectral transmission in \u003cstrong\u003eFig. 1b\u003c/strong\u003e. Totally we observe three mode families, with free spectral ranges around 21.8 GHz. Among them, the TE\u003csub\u003e02\u003c/sub\u003e mode exhibits the highest resonance intensity due to optimal tapered-fiber-microcavity coupling. We select this mode for laser tagging operation. Given the low intrinsic loss of our microcavity, during the laser-scan measurement, clear ringdowns are observed at each resonance. The zoomed-in trace provides more details of the ring-down curve for the TE\u003csub\u003e02\u003c/sub\u003e mode \u003csup\u003e44\u003c/sup\u003e. Referring to a scan speed of 10 nm/s in measurement, we find that the loaded \u003cem\u003eQ\u003c/em\u003e factor reaches 0.94 \u0026times;10\u003csup\u003e9\u0026nbsp;\u003c/sup\u003enear the critical coupling state, while the intrinsic \u003cem\u003eQ\u003c/em\u003e factor approaches 1.89 \u0026times; 10\u003csup\u003e9\u003c/sup\u003e.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eMaintaining a high-\u003cem\u003eQ\u003c/em\u003e factor is essential for both boosting signal stabilization and enhancing sensitivity in a microcavity. In \u003cstrong\u003eFig. 2c\u003c/strong\u003e, we compare the measured \u003cem\u003eQ\u003c/em\u003e factor of our Pt/WO\u003csub\u003e3\u003c/sub\u003e deposited optofluidic microcavity with other microcavity platforms, which are widely used for chemical sensing. Typically, a WGM microresonator integrated with nanomaterials like graphene on its surface \u003csup\u003e31\u003c/sup\u003e has a loaded \u003cem\u003eQ\u003c/em\u003e factor that can reach up to 1.4 \u0026times; 10\u003csup\u003e8\u003c/sup\u003e, restricted by strong absorption loss. Conversely, an optofluidic bubble microcavity without a micro-rod design \u003csup\u003e59\u003c/sup\u003e can achieve a loaded \u003cem\u003eQ\u003c/em\u003e factor of approximately 8 \u0026times; 10\u003csup\u003e6\u003c/sup\u003e, limited by inevitable scattering losses and intermodal crosstalk. \u003cstrong\u003eFigure 2d\u003c/strong\u003e presents the measured probe-reference beat note at 10.8 MHz. Utilizing two separate laser diodes, we maintain a signal-to-noise ratio of the beat note that exceeds 40 dB. Unlike conventional sensing methods that rely on optical spectrum measurements, we determine the linewidth of the beat note using an electrical spectrum analyzer. Experimental findings indicate that the integral linewidth of the probe-reference beat note is 1.9 kHz, which is influenced by the laser linewidths rather than the microcavity resonance linewidth in the MHz range.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eMore interestingly, in \u003cstrong\u003eFig. 2e\u003c/strong\u003e, we present the measured single-sideband phase noises (SSB-PNs) of the dual-laser beat note. In free running operation, the SSB-PN of the dual-laser beat note is measured at 70.8 dBc/Hz at 1 Hz offset while -27.4 dBc/Hz at 1 kHz offset. When the probe laser is stabilized by tracking a resonance of the high-\u003cem\u003eQ\u003c/em\u003e optofluidic microcavity through real-time feedback, the SSB-PN of the beat note between the stabilized probe and the reference laser is observed to be 37.2 dBc/Hz at 1 Hz offset and -46.8 dBc/Hz at 1 kHz offset. We estimate instantaneous linewidth of the PCMR stabilized probe laser reaches 10.4 Hz. This approach provides a method to detect frequency shifts with resolution down to the Hz level. The laser stabilization mechanism is further discussed in Supplementary Note S1. Additionally, \u003cstrong\u003eFig. 2f\u003c/strong\u003e showcases the long-term stability measurements. Over 300 seconds, the probe-reference beat note without resonance triggering (free running) exhibits a random frequency drift of up to \u0026plusmn;0.6 MHz. In contrast, our laser triggering scheme reduces the random frequency drift of the dual-laser beat note to within \u0026plusmn;0.05 MHz. This more than one order improvement highlights the necessity and advantage of locking the probe laser on a high-\u003cem\u003eQ\u003c/em\u003e microcavity resonance, as this implementation would significantly enhance sensing accuracy. More characterizations about stabilizing the probe laser and the reference laser are shown in Supplementary Note S3.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003cstrong\u003eFigure 3a\u003c/strong\u003e illustrates the experimental setup for hydrogen detection using the laser-tagging optofluidic microcavity. A tapered fiber with a diameter of 1 \u0026mu;m is employed to launch and collect the probe light in the microcavity, while hydrogen gas is introduced via a soft tube. Gas preparation operations are shown in Supplementary Note S3. Concentration of the H\u003csub\u003e2\u003c/sub\u003e gas is accurately controlled by a gas distributor. We note that the entire optofluidic microcavity device is thermally regulated using a flexible polyimide foil heating element (Thorlabs, HT10K) for avoiding macroscopic temperature interference. \u003cstrong\u003eFigure 3b\u003c/strong\u003e presents the beat note spectra as the H\u003csub\u003e2\u003c/sub\u003e concentration increases. Initially, the frequency difference between the probe and reference lasers is set to 0 Hz by finely tuning the temperatures of the two cavities. As the H\u003csub\u003e2\u003c/sub\u003e concentration increases from 0 ppm to 100 ppk, the tracked probe follows the shift in the microcavity resonance, resulting in an increase in the beat note from 0 to 723.6 MHz. Throughout this process, the intensity of the dual laser beat note remains stable at around -40 dB, confirming that the laser tagging feedback based on the microcavity resonance is responsive. Additionally, it is observed that the noise level remains constant across this wide measurement range.\u003c/p\u003e\n\u003cp\u003eNow we assess the sensitivity of our device. \u003cstrong\u003eFigure 3c\u003c/strong\u003e illustrates the correlation between the beat note shift and hydrogen concentration (represented by blue dots). The experimental data indicates that this relationship is nonlinear. Specifically, at lower hydrogen concentrations, the frequency shift exhibits a steeper slope. Sensitivity values are shown as red dots. For example, the detection of H\u003csub\u003e2\u003c/sub\u003e gas at a concentration of 20 ppb yields a sensitivity of approximately 30 MHz/ppm; at a concentration of 20 ppm, the sensitivity is around 0.4 MHz/ppm; and at a concentration of 20 ppk, the sensitivity reduces to about 0.013 MHz/ppm. This decreasing sensitivity with higher gas concentrations is primarily attributed to the saturation of hydrogen adsorption on the Pt/WO\u003csub\u003e3\u003c/sub\u003e film. Since integrated linewidth of the probe-reference beat note is limited on 1.9 kHz. In single shot, we estimate that the maximum measurable dynamic range of this device is 162 ppk in approximation. Meanwhile, in single-shot measurement, just based on the sensitivity 30 MHz/ppm, the minimum detectable concentration approaches 0.1 ppb.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eIn \u003cstrong\u003eFig. 3d\u003c/strong\u003e, we present the selectivity analysis of our hydrogen sensor. While the Pt/WO\u003csub\u003e3\u003c/sub\u003e nanofilm exhibits a strong chemical response to hydrogen gas, its microscale non-uniformity can result in numerous nanopores. These nanopores might facilitate the adsorption of other gases onto the film, causing slight variations in the dual laser beat frequency. Nonetheless, the selectivity of our device remains impressive. For instance, when employing the Pt/WO\u003csub\u003e3\u003c/sub\u003e based optofluidic microcavity to detect various gas species such as H\u003csub\u003e2\u003c/sub\u003eO, H\u003csub\u003e2\u003c/sub\u003eS, CO, CO\u003csub\u003e2\u003c/sub\u003e, CH\u003csub\u003e4\u003c/sub\u003e, C\u003csub\u003e2\u003c/sub\u003eH\u003csub\u003e4\u003c/sub\u003e, C\u003csub\u003e7\u003c/sub\u003eH\u003csub\u003e8\u003c/sub\u003e (toluene), CHCl\u003csub\u003e3\u003c/sub\u003e, SO\u003csub\u003e2\u003c/sub\u003e, NO\u003csub\u003e2\u003c/sub\u003e, NH\u003csub\u003e3\u003c/sub\u003e, C\u003csub\u003e3\u003c/sub\u003eH\u003csub\u003e6\u003c/sub\u003eO (acetone), CH\u003csub\u003e3\u003c/sub\u003eOH (methanol), and C\u003csub\u003e2\u003c/sub\u003eH\u003csub\u003e5\u003c/sub\u003eOH (ethanol), the maximum sensitivity does not exceed 0.029 MHz/ppm. Given that the long-term frequency uncertainty of our dual-laser beat note is \u0026plusmn;0.05 MHz, the interference from other gases in hydrogen sensing can effectively be disregarded. Finally, in \u003cstrong\u003eFig. 3e\u003c/strong\u003e, we illustrate the hydrogen sensor\u0026rsquo;s recoverability. By cyclically introducing hydrogen gas at varying concentrations and flushing the gas channel with dry air, we record changes in the frequency shift of the beat note. Experimental results show that the recoverability exceeds 98% in several minutes\u0026rsquo; delay. This high degree of recoverability is attributed to the naturally reversible interaction between H\u003csub\u003e2\u003c/sub\u003e molecules and the Pt/WO\u003csub\u003e3\u003c/sub\u003e film, suggesting that our device can be reliably reused in gas sensing applications. The device exhibits that a response time and recovery time, defined as the time required to reach 90% of the steady state value, are 10 s and 15 s @ 1ppm H\u003csub\u003e2\u003c/sub\u003e, respectively. Both the response and recovery times increase with rising concentration, consistent with the characteristics of thermal response processes. More recovery spectral maps are shown in Supplementary Note S3.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;Since the laser-tagging scheme demonstrates ultrahigh spectral resolution (\u0026gt; 1 dB/kHz), we can employ lock-in amplification to enhance sensitivity further. \u003cstrong\u003eFigure 4a\u003c/strong\u003e illustrates the setup and mechanism for measurements based on lock-in amplification. The beat note of the probe and reference lasers is first filtered using an electrical bandpass filter to suppress random noise, then locked with a lock-in amplifier (Stanford SR860). The output from the lock-in amplifier, \u003cem\u003eI\u003csub\u003eL\u003c/sub\u003e\u003c/em\u003e, is a continuous-wave signal proportional to the signal intensity at the locked frequency \u003cem\u003eI\u003csub\u003eS,f\u003c/sub\u003e\u003c/em\u003e. When there is a change in the probe-reference beat note, \u003cem\u003eI\u003csub\u003eL\u003c/sub\u003e\u003c/em\u003e is accordingly modified. Specifically, \u0026Delta;\u003cem\u003eI\u003csub\u003eL\u003c/sub\u003e\u003c/em\u003e = \u0026Delta;\u003cem\u003eI\u003csub\u003eS\u003c/sub\u003e,\u003csub\u003ef\u0026nbsp;\u003c/sub\u003eI\u003csub\u003eR\u003c/sub\u003e\u003c/em\u003e, where \u003cem\u003eI\u003csub\u003eR\u003c/sub\u003e\u003c/em\u003e is the inner reference intensity. Using the lock-in operation, this intensity modification can be amplified by over 60 dB due to noise cancellation after integration. In \u003cstrong\u003eFig. 4b\u003c/strong\u003e, our scheme is depicted. Initially, we use a crystal oscillator to down-convert the probe-reference laser beat note (at 16.9 MHz) to 64.4 kHz, aligning with the bandwidth requirements of our lock-in amplifier. This signal linewidth suggests that the maximum measurement range for H\u003csub\u003e2\u003c/sub\u003e gas is less than 0.1 ppb when applying lock-in amplification. The down-converted signal exhibits an integrated linewidth of 6.28 kHz and an intensity SNR \u0026gt; 10 dB, with an integration time constant of 1 \u0026mu;s. The lock-in point is set at 68.7 kHz to monitor spectral shifts when sensing H\u003csub\u003e2\u003c/sub\u003e molecules.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFigure 4c\u003c/strong\u003e illustrates the lock-in traces observed during the experiment. Initially, the optofluidic microcavity is filled with dry air from 0 to 15 seconds. At 15 seconds, we introduce 0.1 ppb H\u003csub\u003e2\u003c/sub\u003e into the microtube, then maintaining this concentration until 45 seconds. Subsequently, we cleanse the optofluidic microcavity with dry air again from 60 to 75 seconds. During this process, the \u003cem\u003eI\u003csub\u003eL\u003c/sub\u003e\u003c/em\u003e increases from 3.1 V to 4.2 V and subsequently decreases back, indicating dynamic changes in output intensity due to interactions between H\u003csub\u003e2\u003c/sub\u003e molecules and the intracavity deposited Pt/WO\u003csub\u003e3\u003c/sub\u003e. Upon closer examination of this trace, it becomes evident that when only dry air is present in the microtube, the \u003cem\u003eI\u003csub\u003eL\u003c/sub\u003e\u003c/em\u003e remains stable. However, in the presence of H\u003csub\u003e2\u003c/sub\u003e, the trace exhibits temporal steps, indicating adsorption and desorption behaviors of H\u003csub\u003e2\u003c/sub\u003e molecules. The height of these steps is an integer multiple of 0.02 V, providing significant evidence for single-molecule dynamics. Given that the noise level of the lock-in temporal trace is below 0.005 V, single molecular interactions can be distinctly identified. Lastly, \u003cstrong\u003eFig. 4d\u003c/strong\u003e presents a count of molecular steps occurring over one second (40 ~ 41 s), recording 843 adsorption/desorption events. Notably, the occurrence of single molecule adsorption (+1) or desorption (-1) is predominant. For steps with varying heights, their distribution generally follows a power-law pattern, consistent with the intrinsic nature of gas molecule dynamics \u003csup\u003e24,31,60,61\u003c/sup\u003e.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eBased on the laser tagging scheme, our microfluidic cavity hydrogen sensor, featuring both ultrahigh sensitivity and exceptionally wide dynamic range, can be compactly integrated into a fully functional system, resulting in a highly automated microsystem assembly with outstanding practical performance. As illustrated in \u003cstrong\u003eFig. 5a\u003c/strong\u003e, the laser diodes, optofluidic microcavity, feedback electronics, and signal processing module are integrated into a single 30 cm \u0026times; 30 cm \u0026times; 10 cm chassis, making the system suitable for deployable applications beyond laboratory settings. Gas can be continuously introduced into and expelled from the optical microfluidic cavity through fluidic channels. We show more technical details in Supplementary Note S3, and an application demo in Supplementary Movie 1. To validate its performance, the sensor was applied to a representative use case: detecting hydrogen content in transformer insulation gas, a capability critical to power safety (\u003cstrong\u003eFig. 5b\u003c/strong\u003e). We collected several gas samples extracted from operational transformers, in which the concentrations of common fault marker gases had been pre-calibrated. Our sensor was then used to measure the hydrogen concentrations in these samples.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eAs shown in \u003cstrong\u003eFig. 5c\u003c/strong\u003e, three different gas samples were tested, all with pure nitrogen as the background gas. Specifically, in gas sample #1, there were 0.4 ppm H\u003csub\u003e2\u003c/sub\u003e, 2 ppm CH\u003csub\u003e4\u003c/sub\u003e, 3.9 ppm C\u003csub\u003e2\u003c/sub\u003eH\u003csub\u003e4\u003c/sub\u003e, 15 ppm H\u003csub\u003e2\u003c/sub\u003eS, 2 ppm CO, 23 ppm SO\u003csub\u003e2\u003c/sub\u003e; in gas sample #2, the gas mixture contained 7 ppm H\u003csub\u003e2\u003c/sub\u003e, 3 ppm CH\u003csub\u003e4\u003c/sub\u003e, 9 ppm C\u003csub\u003e2\u003c/sub\u003eH\u003csub\u003e2\u003c/sub\u003e, 2 ppm C\u003csub\u003e2\u003c/sub\u003eH\u003csub\u003e4\u003c/sub\u003e 5 ppm H\u003csub\u003e2\u003c/sub\u003eS, 1 ppm CO; in gas sample #3, the components included 56 ppm H\u003csub\u003e2\u003c/sub\u003e, 13 ppm CH\u003csub\u003e4\u003c/sub\u003e, 6 ppm C\u003csub\u003e2\u003c/sub\u003eH\u003csub\u003e2\u003c/sub\u003e, 18 ppm C\u003csub\u003e2\u003c/sub\u003eH\u003csub\u003e4\u003c/sub\u003e, 1.9 ppm H\u003csub\u003e2\u003c/sub\u003eS, 7 ppm CO, 11 ppm SO\u003csub\u003e2\u003c/sub\u003e. For these samples, our sensor measured H\u003csub\u003e2\u003c/sub\u003e concentrations of 0.405 ppm, 6.92 ppm, and 56.04 ppm, respectively, achieving accuracy rates exceeding 99%. Thanks to the high hydrogen selectivity of the Pt/WO\u003csub\u003e3\u003c/sub\u003e sensing material, the presence of other gases does not interfere with hydrogen detection. Finally, \u003cstrong\u003eFig. 5d\u003c/strong\u003e and \u003cstrong\u003e5e\u003c/strong\u003e compare the performance of our sensor with other state-of-the-art hydrogen sensing devices, including both electrical \u003csup\u003e62\u0026ndash;66\u003c/sup\u003e and optical \u003csup\u003e49,55,67\u0026ndash;69\u003c/sup\u003e types. With a detection limit maintained at the 0.1 ppb level, the laser tagging approach uniquely enables a wide dynamic range exceeding 160 ppk. This combination of accuracy and range surpasses that of conventional electrical and optical detection methods. Moreover, within the Pt/WO\u003csub\u003e3\u003c/sub\u003e embedded microfluidic cavity architecture, the interaction between hydrogen and the nanomaterial is rapid, occurring within seconds, and remains stable across a broad range of temperature conditions.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;In this work, we demonstrate a laser-tagging optofluidic microcavity for single-molecule hydrogen detection across 9-decade concentration span. By depositing a layer of Pt/WO\u003csub\u003e3\u003c/sub\u003e nanofilm inside an optofluidic WGM microcavity, we achieve H\u003csub\u003e2\u003c/sub\u003e sensing via the thermal refraction effect caused by gas adsorption exotherm. In this strategy, the optical field does not overlap with the sensitizing material, achieving a high intrinsic \u003cem\u003eQ\u003c/em\u003e factor approaching 1.89 \u0026times; 10\u003csup\u003e9\u003c/sup\u003e. By locking a tunable laser onto this microcavity, we transform the conventional passive measurement method in optics into a highly coherent active measurement based on optoelectronics. This approach offers remarkable noise suppression, demodulation bandwidth expansion, and real-time capability. As a result, our method overcomes the challenge of achieving both ultrahigh sensitivity and an ultra-large dynamic range in chemical sensing. In single-shot H\u003csub\u003e2\u003c/sub\u003e sensing, we demonstrate a limit of detection down to 1 ppb and a measurable range up to 162 ppk. Leveraging lock-in amplification, we also observe single-molecule dynamics. Furthermore, this sensor illustrates outstanding selectivity, excellent recoverability, and plug-and-play practicality for use outside the laboratory, paving the way for various applications.\u003c/p\u003e\n"},{"header":"Methods","content":"\u003cp\u003e\u003cstrong\u003eMechanism of the laser stabilization and tagging\u003c/strong\u003e. We utilize the slope-locking technique for laser frequency stabilization and dynamic tagging. This method employs a laser servo feedback loop to lock the laser frequency to the resonance peak of a high-\u003cem\u003eQ\u003c/em\u003e optical cavity. When the laser frequency detunes from resonance, the reflected or transmitted optical intensity changes correspondingly. As a result, frequency fluctuations are converted into intensity variations, which are then transformed into voltage signals via a photodetector and a transimpedance amplifier. The performance of the slope-locking frequency stabilization scheme is fundamentally governed by the \u003cem\u003eQ\u003c/em\u003e factor of the optical resonator. A higher \u003cem\u003eQ\u003c/em\u003e factor corresponds to a narrower resonance linewidth and a steeper slope near resonance, thereby amplifying the error signal induced by laser frequency deviations and improving stabilization sensitivity.\u0026nbsp;More details are shown in Supplementary Note S1.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFabrication and characterization of the PCMR\u003c/strong\u003e. The silica hollow microrod resonator was fabricated using a laser reflow technique, which involves three main steps: polishing, etching, and annealing. A carbon dioxide laser beam, focused through a ZnSe lens, was used to process the microtube surface. Through computer-controlled automation, this 3-step process enables the fabrication of size controllable, highly uniform, few-mode, ultrahigh-\u003cem\u003eQ\u003c/em\u003e microrod resonators within minutes. Pt/WO\u003csub\u003e3\u003c/sub\u003e nanoparticles were synthesized via a liquid-phase chemical method. A suspension is prepared by mixing Pt and WO\u003csub\u003e3\u003c/sub\u003e precursors in ethanol. After stirring and subsequent evaporation, Pt/WO\u003csub\u003e3\u003c/sub\u003e nanoparticles are deposited onto the inner wall of the microfluidic cavity. The morphology and composition of the intracavity nanomaterials are characterized using scanning electron microscopy and X-ray diffraction. More details are provided in Supplementary Note S2.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDevice details and optoelectronic encapsulation.\u0026nbsp;\u003c/strong\u003eThe whole hydrogen sensing microsystem majorly includes a probe laser diode (14 PIN, Accelink), a reference laser diode (14 PIN, Accelink), low noise photodetectors (DET01CFC, Thorlabs), a PID tagging feedback loop and a signal processing core in FPGA (ACAU15, Xilinx). In optics: 1) the probe laser drives the microfluidic cavity; 2) the reference laser is stabilized by using an ultra-stable Fabry-Perot microresonator. In electronics: 1) after photodetection #1, the sensor modulated probe laser intensity alteration offers the signal for tuning the probe frequency. 2) after photodetection #2, the reflected intensity of the reference laser provides the signal for locking the reference frequency. 3) beat note of the probe laser and the reference laser are obtained after photodetection #3, frequency shift of the dual laser beat note tells the sensing response. We note that since the hydrogen detection relies on local temperature alteration, the whole microsystem requires a macroscopic thermal electrical cooler. More detailed performance metrics are provided in Supplementary Note S3.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgments\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors acknowledge support from\u0026nbsp;the\u0026nbsp;National Key Research and Development Program of China (Grant\u0026nbsp;2023YFB2806200),\u0026nbsp;the National Natural Science Foundation of China (Grant U24A20311,\u0026nbsp;62305050,\u0026nbsp;62371106) and the Industrial Chain Project of Southern Grid.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eB.C.Y. and Y.H.G. led the general study, and raised the scheme. Y.H.G. and T.T. performed the theoretical analysis. Y.H.G. and S.Y.Y. fabricated the microcavity devices. Y.H.G., Z.H.L. and Y.W. performed the material characterization. S.Y.Y., Z.H.L. and T.T. helped the \u003cem\u003eQ\u003c/em\u003e factor measurement and optimization. Y.H.G., T.T., and B.C. contributed the laser stabilization and tagging measurements. Y.H.G., S.Y.Y., B.W.L. and H.D.X. built the setup and conducted the gas sensing experiment. Z.P.W., Z.H.X., Y.Q.Z., and L.P. helped the device package and out-of-lab verification. Y.J.R. supervised this work. All authors processed and analyzed the results. B.C.Y., Y.H.G. and T.T. prepared the manuscript.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAuthors declare that they have no competing interests.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData and materials availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll data are available in the main text or the supplementary materials.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eJo, Y. \u003cem\u003eet al.\u003c/em\u003e MOF‐Based Chemiresistive Gas Sensors: Toward New Functionalities. \u003cem\u003eAdv. Mater.\u003c/em\u003e \u003cstrong\u003e35\u003c/strong\u003e, (2023).\u003c/li\u003e\n\u003cli\u003eTang, Y., Zhao, Y. \u0026amp; Liu, H. Room-Temperature Semiconductor Gas Sensors: Challenges and Opportunities. \u003cem\u003eACS Sensors\u003c/em\u003e \u003cstrong\u003e7\u003c/strong\u003e, 3582\u0026ndash;3597 (2022).\u003c/li\u003e\n\u003cli\u003eSun, N. \u003cem\u003eet al.\u003c/em\u003e Fully printable integrated multifunctional sensor arrays for intelligent lithium-ion batteries. \u003cem\u003eNat. Commun.\u003c/em\u003e \u003cstrong\u003e16\u003c/strong\u003e, 7361 (2025).\u003c/li\u003e\n\u003cli\u003eWang, L. \u003cem\u003eet al.\u003c/em\u003e Review and Perspective: Gas Separation and Discrimination Technologies for Current Gas Sensors in Environmental Applications. \u003cem\u003eACS Sensors\u003c/em\u003e \u003cstrong\u003e8\u003c/strong\u003e, 1373\u0026ndash;1390 (2023).\u003c/li\u003e\n\u003cli\u003eButtner, W. J., Post, M. B., Burgess, R. \u0026amp; Rivkin, C. An overview of hydrogen safety sensors and requirements. \u003cem\u003eInt. J. Hydrogen Energy\u003c/em\u003e \u003cstrong\u003e36\u003c/strong\u003e, 2462\u0026ndash;2470 (2011).\u003c/li\u003e\n\u003cli\u003eHe, T., Wang, W., He, B. G. \u0026amp; Chen, J. Review on Optical Fiber Sensors for Hazardous-Gas Monitoring in Mines and Tunnels. \u003cem\u003eIEEE Trans. Instrum. Meas.\u003c/em\u003e \u003cstrong\u003e72\u003c/strong\u003e, 1\u0026ndash;22 (2023).\u003c/li\u003e\n\u003cli\u003eH\u0026uuml;bert, T., Boon-Brett, L., Palmisano, V. \u0026amp; Bader, M. A. Developments in gas sensor technology for hydrogen safety. \u003cem\u003eInt. J. Hydrogen Energy\u003c/em\u003e \u003cstrong\u003e39\u003c/strong\u003e, 20474\u0026ndash;20483 (2014).\u003c/li\u003e\n\u003cli\u003eKoo, W. T. \u003cem\u003eet al.\u003c/em\u003e Chemiresistive hydrogen sensors: Fundamentals, recent advances, and challenges. \u003cem\u003eACS Nano\u003c/em\u003e \u003cstrong\u003e14\u003c/strong\u003e, 14284\u0026ndash;14322 (2020).\u003c/li\u003e\n\u003cli\u003eTimmer, B., Olthuis, W. \u0026amp; Berg, A. van den. Ammonia sensors and their applications\u0026mdash;a review. \u003cem\u003eSensors Actuators B Chem.\u003c/em\u003e \u003cstrong\u003e107\u003c/strong\u003e, 666\u0026ndash;677 (2005).\u003c/li\u003e\n\u003cli\u003eAndre, R. S. \u003cem\u003eet al.\u003c/em\u003e Recent Progress in Amine Gas Sensors for Food Quality Monitoring: Novel Architectures for Sensing Materials and Systems. \u003cem\u003eACS Sensors\u003c/em\u003e \u003cstrong\u003e7\u003c/strong\u003e, 2104\u0026ndash;2131 (2022).\u003c/li\u003e\n\u003cli\u003eDhall, S., Mehta, B. R., Tyagi, A. K. \u0026amp; Sood, K. A review on environmental gas sensors: Materials and technologies. \u003cem\u003eSensors Int.\u003c/em\u003e \u003cstrong\u003e2\u003c/strong\u003e, 100116 (2021).\u003c/li\u003e\n\u003cli\u003eGl\u0026ouml;ckler, J. \u003cem\u003eet al.\u003c/em\u003e Infrared Spectroscopic Electronic Noses: An Innovative Approach for Exhaled Breath Sensing. \u003cem\u003eACS Sensors\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, 427\u0026ndash;438 (2025).\u003c/li\u003e\n\u003cli\u003eShin, J. \u003cem\u003eet al.\u003c/em\u003e A non-contact wearable device for monitoring epidermal molecular flux. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e640\u003c/strong\u003e, 375\u0026ndash;383 (2025).\u003c/li\u003e\n\u003cli\u003eAn, N. \u003cem\u003eet al.\u003c/em\u003e Graphene-Fiber Biochemical Sensors: Principles, Implementations, and Advances. \u003cem\u003ePhotonic Sensors\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, 123\u0026ndash;139 (2021).\u003c/li\u003e\n\u003cli\u003eCao, Z. \u003cem\u003eet al.\u003c/em\u003e Biochemical sensing in graphene-enhanced microfiber resonators with individual molecule sensitivity and selectivity. \u003cem\u003eLight Sci. Appl.\u003c/em\u003e \u003cstrong\u003e8\u003c/strong\u003e, 107 (2019).\u003c/li\u003e\n\u003cli\u003eLiang, Q., Bisht, A., Scheck, A., Schunemann, P. G. \u0026amp; Ye, J. Modulated ringdown comb interferometry for sensing of highly complex gases. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e638\u003c/strong\u003e, 941\u0026ndash;948 (2025).\u003c/li\u003e\n\u003cli\u003eTan, T. \u003cem\u003eet al.\u003c/em\u003e Biomimetic acoustic perception via chip-scale dual-soliton microcombs. \u003cem\u003eeLight\u003c/em\u003e \u003cstrong\u003e5\u003c/strong\u003e, 22 (2025).\u003c/li\u003e\n\u003cli\u003eWang, Q. \u003cem\u003eet al.\u003c/em\u003e Dual-comb photothermal spectroscopy. \u003cem\u003eNat. Commun.\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, 2181 (2022).\u003c/li\u003e\n\u003cli\u003eYan, Y. \u003cem\u003eet al.\u003c/em\u003e Nanoliter‐Scale Light\u0026ndash;Matter Interaction in a Fiber‐Tip Cavity Enables Sensitive Photothermal Gas Detection. \u003cem\u003eLaser Photon. Rev.\u003c/em\u003e \u003cstrong\u003e18\u003c/strong\u003e, (2024).\u003c/li\u003e\n\u003cli\u003eMa, J. \u003cem\u003eet al.\u003c/em\u003e Microscale fiber photoacoustic spectroscopy for in situ and real-time trace gas sensing. \u003cem\u003eAdv. Photonics\u003c/em\u003e \u003cstrong\u003e6\u003c/strong\u003e, (2024).\u003c/li\u003e\n\u003cli\u003eChen, J., Xiong, Y., Xu, F. \u0026amp; Lu, Y. Silica optical fiber integrated with two-dimensional materials: towards opto-electro-mechanical technology. \u003cem\u003eLight Sci. Appl.\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, 78 (2021).\u003c/li\u003e\n\u003cli\u003eGuo, Y. \u003cem\u003eet al.\u003c/em\u003e Optical Microcavities Empowered Biochemical Sensing: Status and Prospects. \u003cem\u003eAdv. Devices Instrum.\u003c/em\u003e \u003cstrong\u003e5\u003c/strong\u003e, 1\u0026ndash;28 (2024).\u003c/li\u003e\n\u003cli\u003eYao, B.-C. \u003cem\u003eet al.\u003c/em\u003e Interdisciplinary advances in microcombs: bridging physics and information technology. \u003cem\u003eeLight\u003c/em\u003e \u003cstrong\u003e4\u003c/strong\u003e, 19 (2024).\u003c/li\u003e\n\u003cli\u003eAn, N. \u003cem\u003eet al.\u003c/em\u003e Electrically Tunable Four-Wave-Mixing in Graphene Heterogeneous Fiber for Individual Gas Molecule Detection. \u003cem\u003eNano Lett.\u003c/em\u003e \u003cstrong\u003e20\u003c/strong\u003e, 6473\u0026ndash;6480 (2020).\u003c/li\u003e\n\u003cli\u003eWang, Y. \u003cem\u003eet al.\u003c/em\u003e Noise canceled graphene-microcavity fiber laser sensor for ultrasensitive gas detection. \u003cem\u003ePhotonics Res.\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, A1 (2023).\u003c/li\u003e\n\u003cli\u003eGuo, Y. \u003cem\u003eet al.\u003c/em\u003e Gas detection in a graphene based dual-mode fiber laser microcavity. \u003cem\u003eSensors Actuators B Chem.\u003c/em\u003e \u003cstrong\u003e348\u003c/strong\u003e, 130694 (2021).\u003c/li\u003e\n\u003cli\u003eAn, N. \u003cem\u003eet al.\u003c/em\u003e Brillouin lasers in a graphene microresonator for multispecies and individual gas molecule detection. \u003cem\u003eAPL Photonics\u003c/em\u003e \u003cstrong\u003e8\u003c/strong\u003e, (2023).\u003c/li\u003e\n\u003cli\u003eLiang, Y. \u003cem\u003eet al.\u003c/em\u003e Harnessing sub-comb dynamics in a graphene-sensitized microresonator for gas detection. \u003cem\u003eFront. Optoelectron.\u003c/em\u003e \u003cstrong\u003e17\u003c/strong\u003e, 12 (2024).\u003c/li\u003e\n\u003cli\u003eTan, T., Jiang, X., Wang, C., Yao, B. \u0026amp; Zhang, H. 2D Material Optoelectronics for Information Functional Device Applications: Status and Challenges. \u003cem\u003eAdv. Sci.\u003c/em\u003e \u003cstrong\u003e7\u003c/strong\u003e, 2000058 (2020).\u003c/li\u003e\n\u003cli\u003eLiu, J. \u003cem\u003eet al.\u003c/em\u003e Emerging material platforms for integrated microcavity photonics. \u003cem\u003eSci. China Physics, Mech. Astron.\u003c/em\u003e \u003cstrong\u003e65\u003c/strong\u003e, 104201 (2022).\u003c/li\u003e\n\u003cli\u003eTan, T. \u003cem\u003eet al.\u003c/em\u003e Multispecies and individual gas molecule detection using Stokes solitons in a graphene over-modal microresonator. \u003cem\u003eNat. Commun.\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, 6716 (2021).\u003c/li\u003e\n\u003cli\u003eGuo, Y. \u003cem\u003eet al.\u003c/em\u003e A Monolithic Graphene‐Functionalized Microlaser for Multispecies Gas Detection. \u003cem\u003eAdv. Mater.\u003c/em\u003e \u003cstrong\u003e34\u003c/strong\u003e, (2022).\u003c/li\u003e\n\u003cli\u003eZhang, H. \u003cem\u003eet al.\u003c/em\u003e Gas Detection With Switchable Selectivity in a Functionalized-Graphene Integrated Microrod Resonator. \u003cem\u003ePhotonic Sensors\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, 250423 (2025).\u003c/li\u003e\n\u003cli\u003eZhi, Y., Yu, X. C., Gong, Q., Yang, L. \u0026amp; Xiao, Y. F. Single Nanoparticle Detection Using Optical Microcavities. \u003cem\u003eAdv. Mater.\u003c/em\u003e \u003cstrong\u003e29\u003c/strong\u003e, (2017).\u003c/li\u003e\n\u003cli\u003eYu, D. \u003cem\u003eet al.\u003c/em\u003e Whispering-gallery-mode sensors for biological and physical sensing. \u003cem\u003eNat. Rev. Methods Prim.\u003c/em\u003e \u003cstrong\u003e1\u003c/strong\u003e, 83 (2021).\u003c/li\u003e\n\u003cli\u003eAltug, H., Oh, S.-H., Maier, S. A. \u0026amp; Homola, J. Advances and applications of nanophotonic biosensors. \u003cem\u003eNat. Nanotechnol.\u003c/em\u003e \u003cstrong\u003e17\u003c/strong\u003e, 5\u0026ndash;16 (2022).\u003c/li\u003e\n\u003cli\u003eJiang, X., Qavi, A. J., Huang, S. H. \u0026amp; Yang, L. Whispering-Gallery Sensors. \u003cem\u003eMatter\u003c/em\u003e \u003cstrong\u003e3\u003c/strong\u003e, 371\u0026ndash;392 (2020).\u003c/li\u003e\n\u003cli\u003eChen, W., \u0026Ouml;zdemir, Ş. K., Zhao, G., Wiersig, J. \u0026amp; Yang, L. Exceptional points enhance sensing in an optical microcavity. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e548\u003c/strong\u003e, 192\u0026ndash;195 (2017).\u003c/li\u003e\n\u003cli\u003eWang, Y., Gao, R. \u0026amp; Xin, X. Hollow-core fiber refractive index sensor with high sensitivity and large dynamic range based on a multiple mode transmission mechanism. \u003cem\u003eOpt. Express\u003c/em\u003e \u003cstrong\u003e29\u003c/strong\u003e, 19703 (2021).\u003c/li\u003e\n\u003cli\u003eLan, G. \u0026amp; Gao, Y. Surface Plasmon Resonance Sensor With High Sensitivity and Wide Dynamic Range. \u003cem\u003eIEEE Sens. J.\u003c/em\u003e \u003cstrong\u003e18\u003c/strong\u003e, 5329\u0026ndash;5333 (2018).\u003c/li\u003e\n\u003cli\u003eXing, T. \u003cem\u003eet al.\u003c/em\u003e Fast Switching Acoustic Sensor With Ultrahigh Sensitivity and Wide Dynamic Response Range Based on Ultrahigh- Q CaF 2 Resonator. \u003cem\u003eJ. Light. Technol.\u003c/em\u003e \u003cstrong\u003e40\u003c/strong\u003e, 5775\u0026ndash;5780 (2022).\u003c/li\u003e\n\u003cli\u003eLiao, J. \u0026amp; Yang, L. Optical whispering-gallery mode barcodes for high-precision and wide-range temperature measurements. \u003cem\u003eLight Sci. Appl.\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, 32 (2021).\u003c/li\u003e\n\u003cli\u003eHu, J., Wan, F., Wang, P., Ge, H. \u0026amp; Chen, W. Application of frequency-locking cavity-enhanced spectroscopy for highly sensitive gas sensing: a review. \u003cem\u003eAppl. Spectrosc. Rev.\u003c/em\u003e \u003cstrong\u003e57\u003c/strong\u003e, 378\u0026ndash;410 (2022).\u003c/li\u003e\n\u003cli\u003ePan, T.-Y. \u003cem\u003eet al.\u003c/em\u003e Boosting silica micro-rod Q factor to 8.28 \u0026times; 109 for fully stabilizing a soliton microcomb. \u003cem\u003eNat. Commun.\u003c/em\u003e \u003cstrong\u003e16\u003c/strong\u003e, 8878 (2025).\u003c/li\u003e\n\u003cli\u003eCao, X., Yang, H., Wu, Z. L. \u0026amp; Li, B. B. Ultrasound sensing with optical microcavities. \u003cem\u003eLight Sci. Appl.\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, (2024).\u003c/li\u003e\n\u003cli\u003eCheng, H. \u003cem\u003eet al.\u003c/em\u003e Harnessing micro-Fabry\u0026ndash;P\u0026eacute;rot reference cavities in photonic integrated circuits. \u003cem\u003eNat. Photonics\u003c/em\u003e \u003cstrong\u003e19\u003c/strong\u003e, 992\u0026ndash;998 (2025).\u003c/li\u003e\n\u003cli\u003eSi, B. \u003cem\u003eet al.\u003c/em\u003e Core-shell nanostructures Pt-WO3 for colorimetric and fiber Bragg grating hydrogen sensors. \u003cem\u003eInt. J. Hydrogen Energy\u003c/em\u003e \u003cstrong\u003e137\u003c/strong\u003e, 775\u0026ndash;782 (2025).\u003c/li\u003e\n\u003cli\u003eDai, J. \u003cem\u003eet al.\u003c/em\u003e Performance of fiber Bragg grating hydrogen sensor coated with Pt-loaded WO3 coating. \u003cem\u003eSensors Actuators B Chem.\u003c/em\u003e \u003cstrong\u003e190\u003c/strong\u003e, 657\u0026ndash;663 (2014).\u003c/li\u003e\n\u003cli\u003eWang, C. \u003cem\u003eet al.\u003c/em\u003e Highly sensitive fiber grating hydrogen sensor based on hydrogen-doped Pt/WO3. \u003cem\u003eSensors Actuators B Chem.\u003c/em\u003e \u003cstrong\u003e404\u003c/strong\u003e, 135250 (2024).\u003c/li\u003e\n\u003cli\u003eYao, B. \u003cem\u003eet al.\u003c/em\u003e Graphene-Enhanced Brillouin Optomechanical Microresonator for Ultrasensitive Gas Detection. \u003cem\u003eNano Lett.\u003c/em\u003e \u003cstrong\u003e17\u003c/strong\u003e, 4996\u0026ndash;5002 (2017).\u003c/li\u003e\n\u003cli\u003eChang, B. \u003cem\u003eet al.\u003c/em\u003e Dispersive Fourier transform based dual-comb ranging. \u003cem\u003eNat. Commun.\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, 4990 (2024).\u003c/li\u003e\n\u003cli\u003eZhang, H. \u003cem\u003eet al.\u003c/em\u003e Mode-locked optomechanical frequency combs in a graphene-silica microresonator. \u003cem\u003eSci. Adv.\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, (2025).\u003c/li\u003e\n\u003cli\u003eWang, M. \u003cem\u003eet al.\u003c/em\u003e Cascading-Induced Fundamental Linewidth Enhancement of a Microcavity Brillouin Laser. \u003cem\u003eACS Photonics\u003c/em\u003e (2025) doi:10.1021/acsphotonics.5c00127.\u003c/li\u003e\n\u003cli\u003eCui, W. \u003cem\u003eet al.\u003c/em\u003e Ultra-low time jitter transform-limited dissipative Kerr soliton microcomb. \u003cem\u003eOpt. Express\u003c/em\u003e \u003cstrong\u003e31\u003c/strong\u003e, 37154 (2023).\u003c/li\u003e\n\u003cli\u003eGao, F. \u003cem\u003eet al.\u003c/em\u003e Sub-ppb-level sensitive hydrogen gas sensing based on PDMS-Pt enhanced microrod cavity. \u003cem\u003eOpt. Express\u003c/em\u003e \u003cstrong\u003e33\u003c/strong\u003e, 42043 (2025).\u003c/li\u003e\n\u003cli\u003eLiu, L., Liu, Y. \u0026amp; Liu, C. Enhancing the Understanding of Hydrogen Evolution and Oxidation Reactions on Pt(111) through Ab Initio Simulation of Electrode/Electrolyte Kinetics. \u003cem\u003eJ. Am. Chem. Soc.\u003c/em\u003e \u003cstrong\u003e142\u003c/strong\u003e, 4985\u0026ndash;4989 (2020).\u003c/li\u003e\n\u003cli\u003eKoo, W.-T. \u003cem\u003eet al.\u003c/em\u003e Chemiresistive Hydrogen Sensors: Fundamentals, Recent Advances, and Challenges. \u003cem\u003eACS Nano\u003c/em\u003e \u003cstrong\u003e14\u003c/strong\u003e, 14284\u0026ndash;14322 (2020).\u003c/li\u003e\n\u003cli\u003eZhang, H. \u003cem\u003eet al.\u003c/em\u003e Soliton Microcombs Multiplexing Using Intracavity-Stimulated Brillouin Lasers. \u003cem\u003ePhys. Rev. Lett.\u003c/em\u003e \u003cstrong\u003e130\u003c/strong\u003e, 153802 (2023).\u003c/li\u003e\n\u003cli\u003eYao, B. \u003cem\u003eet al.\u003c/em\u003e Graphene-Enhanced Brillouin Optomechanical Microresonator for Ultrasensitive Gas Detection. \u003cem\u003eNano Lett.\u003c/em\u003e \u003cstrong\u003e17\u003c/strong\u003e, 4996\u0026ndash;5002 (2017).\u003c/li\u003e\n\u003cli\u003eClauset, A., Shalizi, C. R. \u0026amp; Newman, M. E. J. Power-Law Distributions in Empirical Data. \u003cem\u003eSIAM Rev.\u003c/em\u003e \u003cstrong\u003e51\u003c/strong\u003e, 661\u0026ndash;703 (2009).\u003c/li\u003e\n\u003cli\u003eTartaglione, V., Farges, C. \u0026amp; Sabatier, J. Nonlinear dynamical modeling of adsorption and desorption processes with power-law kinetics: Application to CO2 capture. \u003cem\u003ePhys. Rev. E\u003c/em\u003e \u003cstrong\u003e102\u003c/strong\u003e, 052102 (2020).\u003c/li\u003e\n\u003cli\u003eZhang, X. \u003cem\u003eet al.\u003c/em\u003e Ultralow detection limit and ultrafast response/recovery of the H2 gas sensor based on Pd-doped rGO/ZnO-SnO2 from hydrothermal synthesis. \u003cem\u003eMicrosystems Nanoeng.\u003c/em\u003e \u003cstrong\u003e8\u003c/strong\u003e, 67 (2022).\u003c/li\u003e\n\u003cli\u003eAgarwal, S. \u003cem\u003eet al.\u003c/em\u003e An efficient hydrogen gas sensor based on hierarchical Ag/ZnO hollow microstructures. \u003cem\u003eSensors Actuators B Chem.\u003c/em\u003e \u003cstrong\u003e346\u003c/strong\u003e, 130510 (2021).\u003c/li\u003e\n\u003cli\u003eNakate, U. T., Ahmad, R., Patil, P., Yu, Y. T. \u0026amp; Hahn, Y.-B. Ultra thin NiO nanosheets for high performance hydrogen gas sensor device. \u003cem\u003eAppl. Surf. Sci.\u003c/em\u003e \u003cstrong\u003e506\u003c/strong\u003e, 144971 (2020).\u003c/li\u003e\n\u003cli\u003eLuo, N. \u003cem\u003eet al.\u003c/em\u003e Ultralow detection limit MEMS hydrogen sensor based on SnO2 with oxygen vacancies. \u003cem\u003eSensors Actuators B Chem.\u003c/em\u003e \u003cstrong\u003e354\u003c/strong\u003e, 130982 (2022).\u003c/li\u003e\n\u003cli\u003eMandal, S. \u003cem\u003eet al.\u003c/em\u003e A robust organic hydrogen sensor for distributed monitoring applications. \u003cem\u003eNat. Electron.\u003c/em\u003e \u003cstrong\u003e8\u003c/strong\u003e, 343\u0026ndash;352 (2025).\u003c/li\u003e\n\u003cli\u003eYan, X. \u003cem\u003eet al.\u003c/em\u003e High Sensitivity Hydrogen Sensors Based on Optical Micro/Nanofiber Couplers Working at the Dispersion Turning Point. \u003cem\u003eJ. Light. Technol.\u003c/em\u003e \u003cstrong\u003e41\u003c/strong\u003e, 4283\u0026ndash;4289 (2023).\u003c/li\u003e\n\u003cli\u003eYue, Y., Ding, H. \u0026amp; Chen, C. Optical hydrogen sensors based on silica self-assembled mesoporous microspheres. \u003cem\u003eInt. J. Hydrogen Energy\u003c/em\u003e \u003cstrong\u003e46\u003c/strong\u003e, 1403\u0026ndash;1410 (2021).\u003c/li\u003e\n\u003cli\u003eBoelsma, C. \u003cem\u003eet al.\u003c/em\u003e Hafnium\u0026mdash;an optical hydrogen sensor spanning six orders in pressure. \u003cem\u003eNat. Commun.\u003c/em\u003e \u003cstrong\u003e8\u003c/strong\u003e, 15718 (2017). \u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"nature-portfolio","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"","title":"Nature Portfolio","twitterHandle":"","acdcEnabled":false,"dfaEnabled":false,"editorialSystem":"ejp","reportingPortfolio":"","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-8576965/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8576965/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Optical microcavities are exceptional transducers for gas sensing, yet their performance is fundamentally constrained by the trade-off between sensitivity and dynamic range, degradation of quality factor (\u003ci\u003eQ\u003c/i\u003e) upon integration with sensitive materials, and pervasive system noises. Here, we introduce a laser-tagging optofluidic microcavity that simultaneously overcomes these challenges for single-molecule hydrogen detection. Our architecture employs a hollow whispering-gallery-mode microcavity, functionally inner-coated with Pt/WO\u003csub\u003e3\u003c/sub\u003e nanofilm. Gas detection is mediated through thermal phonon transfer, which prevents the sensitive film from perturbing the optical field. This mechanism preserves an ultrahigh intrinsic Q factor of 1.89 × 10\u003csup\u003e9\u003c/sup\u003e even during sensing, therefore offers a unique tool to develop a laser-tagging method that dynamically locks the probe laser to the microcavity’s optimal operation point, enabling real-time resonance tracking. This approach suppresses phase noise by over three orders of magnitude and facilitates wide-bandwidth optoelectronic heterodyne demodulation. Consequently, we achieve a Hz-level frequency-shift resolution and a measurable resonance shift range of 1 GHz, breaking the conventional inverse relationship between accuracy and dynamic range. In single-shot measurement, our sensor achieves a minimum detectable concentration of 0.1 ppb and a maximum of 162 ppk, spanning 9 orders of magnitude. Moreover, after lock-in amplification, individual molecule dynamics are resolved. The integrated, centimetre-scale footprint of the device ensures robust ‘plug-and-play’ operation outside the laboratory, providing a universal strategy to advance optical microcavities towards a broad spectrum of ultra-precise metrology applications.","manuscriptTitle":"Adaptively laser-tagging optofluidic microcavity for single-molecule hydrogen detection across 9-decade concentration span","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-02-04 08:27:38","doi":"10.21203/rs.3.rs-8576965/v1","editorialEvents":[],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"nature-photonics","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"nphoton","sideBox":"Learn more about [Nature Photonics](https://www.nature.com/nphoton/)","snPcode":"41566","submissionUrl":"https://mts-nphot.nature.com/cgi-bin/main.plex","title":"Nature Photonics","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature Research","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"8f049112-9281-412e-a0a3-fb5c2663fd84","owner":[],"postedDate":"February 4th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":62259028,"name":"Physical sciences/Optics and photonics/Applied optics/Optical sensors"},{"id":62259029,"name":"Physical sciences/Optics and photonics/Optical materials and structures/Microresonators"},{"id":62259030,"name":"Physical sciences/Optics and photonics/Other photonics/Optofluidics"},{"id":62259031,"name":"Physical sciences/Materials science/Materials for devices/Sensors and biosensors"},{"id":62259032,"name":"Physical sciences/Optics and photonics/Applied optics/Optoelectronic devices and components"}],"tags":[],"updatedAt":"2026-05-05T10:11:13+00:00","versionOfRecord":[],"versionCreatedAt":"2026-02-04 08:27:38","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8576965","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8576965","identity":"rs-8576965","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}