Wanhua Zheng
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Wanhua Zheng has a total of 430 co-authors. Here are the top co-authors by number of shared publications:
Yufei Wang
Institute of Semiconductors, Chinese Academy of Sciences
53 shared publicationsHongwei Qu
Institute of Semiconductors, Chinese Academy of Sciences
47 shared publicationsXuyan Zhou
Weifang Academy of Advanced Opto-electronic Circuits
41 shared publicationsAiyi Qi
Key Laboratory of Solid-State Optoelectronics Information Technology, Institute of Semiconductors, Chinese Academy of Sciences, Beijing, China
32 shared publicationsJingxuan Chen
Laboratory of Solid State Optoelectronics Information Technology, Institute of Semiconductors, Chinese Academy of Sciences, Beijing 100083, People’s Republic of China
29 shared publicationsMingjin Wang
Laboratory of Solid State Optoelectronics Information Technology, Institute of Semiconductors, Chinese Academy of Sciences, Beijing 100083, People’s Republic of China
27 shared publications
Latest Publications
- 2 November 2026
- Published by Optica Publishing Group in Optics Express
- Vol. 34 (23) , 43207
Distributed emitter design for high-power semiconductor laser bars delivering 600 W under CW operation
We systematically investigate the influence of emitter distribution on the performance of the laser bar through both simulations and experiments. Experimental results show that by optimizing the structure and number of emitters, the bar, featuring a cavity length of 6 mm and a periodic distribution of 24 emitters, achieves the highest output power of 650 W at 730 A under CW operation. Its maximum electro-optical conversion efficiency is 67.19%, obtained at an output power of 301 W. Meanwhile, by adopting a non-periodic emitter distribution, the bar with 12 emitters achieves 643 W at 720 A under CW operation. Its maximum electro-optical conversion efficiency is 68.07%, obtained at an output power of 263 W. Its thermal resistance is reduced by 6.3% compared with that of the conventional structure.
- 2 November 2026
- Published by Optica Publishing Group in Optics Express
- Vol. 34 (23) , 43207
Distributed emitter design for high-power semiconductor laser bars delivering 600 W under CW operation
We systematically investigate the influence of emitter distribution on the performance of the laser bar through both simulations and experiments. Experimental results show that by optimizing the structure and number of emitters, the bar, featuring a cavity length of 6 mm and a periodic distribution of 24 emitters, achieves the highest output power of 650 W at 730 A under CW operation. Its maximum electro-optical conversion efficiency is 67.19%, obtained at an output power of 301 W. Meanwhile, by adopting a non-periodic emitter distribution, the bar with 12 emitters achieves 643 W at 720 A under CW operation. Its maximum electro-optical conversion efficiency is 68.07%, obtained at an output power of 263 W. Its thermal resistance is reduced by 6.3% compared with that of the conventional structure.
- 1 November 2026
- Published by IOP Publishing in Journal of Optics
- Vol. 28 (11) , 115203
Pulse symmetry modulation in BBO/Cr 4+ :YAG doubly Q-switched Nd:YAG laser
Symmetric pulses are of great importance for applications related to the temporal pulse profile, where temporal symmetry directly affects energy deposition, processing accuracy, and ranging performance. Nd:YAG is a mature gain medium widely used for acquiring high energy, high peak power nanosecond pulses because of its excellent thermal conductivity, moderate upper-state lifetime, and other advantages. However, the pulse symmetry of Q-switched Nd:YAG laser has received relatively little attention. Active Q-switching of Nd:YAG usually produces a clearly asymmetric pulse. In this paper we reported a diode-pumped, doubly Q-switched Nd:YAG laser operating at 1 kHz with highly symmetric pulse profiles, realized by simultaneously incorporating an electro-optic (EO) modulator and a Cr 4+ :YAG saturable absorber into the resonator. The effects of pump energy and output coupler transmittance on pulse symmetry were systematically investigated. At a pump energy of 1.83 mJ and 50% transmittance, the doubly Q-switched laser achieved a symmetry factor of 1.00, representing a relative increase of 81.8% compared with the purely EO Q-switched Nd:YAG. The obtained highly symmetric pulses exhibited a spectral linewidth of 0.24 nm at a central wavelength of 1064.2 nm, together with beam quality factors of M 2 X = 1.26 and M 2 Y = 1.23.
- 1 November 2026
- Published by IOP Publishing in Journal of Optics
- Vol. 28 (11) , 115203
Pulse symmetry modulation in BBO/Cr 4+ :YAG doubly Q-switched Nd:YAG laser
Symmetric pulses are of great importance for applications related to the temporal pulse profile, where temporal symmetry directly affects energy deposition, processing accuracy, and ranging performance. Nd:YAG is a mature gain medium widely used for acquiring high energy, high peak power nanosecond pulses because of its excellent thermal conductivity, moderate upper-state lifetime, and other advantages. However, the pulse symmetry of Q-switched Nd:YAG laser has received relatively little attention. Active Q-switching of Nd:YAG usually produces a clearly asymmetric pulse. In this paper we reported a diode-pumped, doubly Q-switched Nd:YAG laser operating at 1 kHz with highly symmetric pulse profiles, realized by simultaneously incorporating an electro-optic (EO) modulator and a Cr 4+ :YAG saturable absorber into the resonator. The effects of pump energy and output coupler transmittance on pulse symmetry were systematically investigated. At a pump energy of 1.83 mJ and 50% transmittance, the doubly Q-switched laser achieved a symmetry factor of 1.00, representing a relative increase of 81.8% compared with the purely EO Q-switched Nd:YAG. The obtained highly symmetric pulses exhibited a spectral linewidth of 0.24 nm at a central wavelength of 1064.2 nm, together with beam quality factors of M 2 X = 1.26 and M 2 Y = 1.23.
- Open Access
- 22 October 2026
- Published by Elsevier in SSRN Electronic Journal
Consensus anchoring and bounded refinement for infrared and visible image fusion
Infrared and visible images are complementary. Infrared images highlight thermal targets. Visible images preserve texture and structure. Recent fusion networks improve adaptivity with attention or high-order interaction. Their parameter and computation costs grow at the same time. However, infrared and visible images are registered dual-source observations of the same scene. Their geometric correspondence is complete at the input. The residual discrepancy is small and bounded. Based on this observation, we propose ConRef-Fusion, a two-stage latent fusion framework. Stage I establishes a decoder-compatible shared reference with an equal-weight anchor. The shared decoder is trained only with this anchor, so cross-modal consistency is established during training. Stage II freezes the encoders. A lightweight consensus residual adapter (CRA) predicts a bounded correction. We provide three groups of evidence. First, the anchor must exist: removing it causes a collapse. Second, the anchor must stay balanced: any asymmetric ratio severely degrades fusion quality, and 0.5/0.5 is best on entropy, PSNR, and edge preservation. Third, the refinement should stay conservative: the gate converges to low values and self-limits the refinement. Experiments on the TNO, M3FD, RoadScene, and MSRS benchmarks show balanced fusion quality and the highest downstream detection accuracy. A controlled single-stage comparison confirms that the staged schedule is essential. The code is available at https://github.com/HuangRZzzz/ConRef-Fusion.
- 22 October 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessConsensus anchoring and bounded refinement for infrared and visible image fusion
Infrared and visible images are complementary. Infrared images highlight thermal targets. Visible images preserve texture and structure. Recent fusion networks improve adaptivity with attention or high-order interaction. Their parameter and computation costs grow at the same time. However, infrared and visible images are registered dual-source observations of the same scene. Their geometric correspondence is complete at the input. The residual discrepancy is small and bounded. Based on this observation, we propose ConRef-Fusion, a two-stage latent fusion framework. Stage I establishes a decoder-compatible shared reference with an equal-weight anchor. The shared decoder is trained only with this anchor, so cross-modal consistency is established during training. Stage II freezes the encoders. A lightweight consensus residual adapter (CRA) predicts a bounded correction. We provide three groups of evidence. First, the anchor must exist: removing it causes a collapse. Second, the anchor must stay balanced: any asymmetric ratio severely degrades fusion quality, and 0.5/0.5 is best on entropy, PSNR, and edge preservation. Third, the refinement should stay conservative: the gate converges to low values and self-limits the refinement. Experiments on the TNO, M3FD, RoadScene, and MSRS benchmarks show balanced fusion quality and the highest downstream detection accuracy. A controlled single-stage comparison confirms that the staged schedule is essential. The code is available at https://github.com/HuangRZzzz/ConRef-Fusion.
- 6 October 2026
- Published by Optica Publishing Group in Optics Letters
- Vol. 51 (20) , 5725-5728
Epitaxial strategy for high-polarization semiconductor lasers with process robustness
We demonstrate an epitaxial strategy for improving the degree of polarization (DoP) of semiconductor lasers under packaging-induced stress by increasing the intrinsic TE/TM effective refractive-index difference, n diff 0 . Three distinct epitaxial structures with n diff 0 values of 4.75 × 10 −4 , 9.09 × 10 −4 , and 1.24 × 10 −3 were designed and fabricated. At 14 A, the measured DoP increased from 90.2% to 97.3% with increasing n diff 0 . Structures with larger n diff 0 also exhibited lower sensitivity to packaging-induced stress variations. These results show that engineering n diff 0 through epitaxial design provides a practical route toward process-robust high-polarization semiconductor lasers.
- 6 October 2026
- Published by Optica Publishing Group in Optics Letters
- Vol. 51 (20) , 5725-5728
Epitaxial strategy for high-polarization semiconductor lasers with process robustness
We demonstrate an epitaxial strategy for improving the degree of polarization (DoP) of semiconductor lasers under packaging-induced stress by increasing the intrinsic TE/TM effective refractive-index difference, n diff 0 . Three distinct epitaxial structures with n diff 0 values of 4.75 × 10 −4 , 9.09 × 10 −4 , and 1.24 × 10 −3 were designed and fabricated. At 14 A, the measured DoP increased from 90.2% to 97.3% with increasing n diff 0 . Structures with larger n diff 0 also exhibited lower sensitivity to packaging-induced stress variations. These results show that engineering n diff 0 through epitaxial design provides a practical route toward process-robust high-polarization semiconductor lasers.
- Open Access
- 2 September 2026
- Published by Elsevier in SSRN Electronic Journal
Consensus-guided latent refinement for infrared andvisible image fusion
Infrared and visible images provide complementary cues for scene interpre-tation: infrared measurements highlight thermally salient objects, whereasvisible measurements preserve fine structures and textures. Their fusionis challenging because adaptive modality selection can amplify one sourcebefore the two modalities have formed a compatible representation. Wepropose ConRef-Fusion, a two-stage latent fusion network that first learnsa shared-decoder-compatible representation and then performs local refine-ment around this representation. In the first stage, modality-specific en-coders and a shared decoder are optimized with an equal-weight latent an-chor. In the second stage, the encoders are fixed and a lightweight consensusresidual adapter predicts a bounded correction from the latent discrepancybetween the two modalities. This design preserves a stable cross-modal ref-erence while allowing scene-dependent enhancement of thermal targets andvisible details. Experiments on the TNO, M3FD, RoadScene, and MSRSbenchmarks show that the proposed method provides competitive entropy,complementary-information, fidelity, and edge-preservation scores. Objectdetection experiments on M3FD further indicate that the fused images retaincues useful for high-level perception. Ablation and latent-decoding analyses clarify how the consensus anchor and residual refinement contribute to theobserved performance
- 2 September 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessConsensus-guided latent refinement for infrared andvisible image fusion
Infrared and visible images provide complementary cues for scene interpre-tation: infrared measurements highlight thermally salient objects, whereasvisible measurements preserve fine structures and textures. Their fusionis challenging because adaptive modality selection can amplify one sourcebefore the two modalities have formed a compatible representation. Wepropose ConRef-Fusion, a two-stage latent fusion network that first learnsa shared-decoder-compatible representation and then performs local refine-ment around this representation. In the first stage, modality-specific en-coders and a shared decoder are optimized with an equal-weight latent an-chor. In the second stage, the encoders are fixed and a lightweight consensusresidual adapter predicts a bounded correction from the latent discrepancybetween the two modalities. This design preserves a stable cross-modal ref-erence while allowing scene-dependent enhancement of thermal targets andvisible details. Experiments on the TNO, M3FD, RoadScene, and MSRSbenchmarks show that the proposed method provides competitive entropy,complementary-information, fidelity, and edge-preservation scores. Objectdetection experiments on M3FD further indicate that the fused images retaincues useful for high-level perception. Ablation and latent-decoding analyses clarify how the consensus anchor and residual refinement contribute to theobserved performance
- 1 September 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Electron Devices
- Vol. 73 (9) , 5652-5658
Doping-Engineered Performance Tuning of Si-SPADs in a 40-nm BSI Technology
- 1 September 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Electron Devices
- Vol. 73 (9) , 5652-5658
Doping-Engineered Performance Tuning of Si-SPADs in a 40-nm BSI Technology
- 18 June 2026
- Published by Optica Publishing Group in Optics Express
- Vol. 34 (13) , 23817-23829
Fano resonance in the SWG-DBR metastructure
We study a metastructure formed by vertically integrating a subwavelength grating (SWG) with a distributed Bragg reflector (DBR). Fano resonance occurs in the reflection spectrum of the SWG-DBR metastructure under normal incidence. The Fano resonance arises from the coupling between a broadband reflection background (continuum state) provided by the top reflector (SWG and top DBR) and a narrowband Fabry-Pérot cavity mode (discrete state) formed by the top reflector and the bottom DBR, corresponding to the continuum and discrete states, respectively. The multipole decomposition verifies that the Fano resonance of the proposed SWG-DBR metastructure is dominated by the toroidal dipole. The electric field intensity of the Fano resonance has a maximum at the SWG surface, making it sensitive to the change of the interfacial refractive index. The SWG-DBR metastructure is fabricated, and the reflection spectra under different ambient refractive-index conditions are measured. The Fano resonance peak shift is observed under different liquids covering the surface of the SWG, consistent with the simulated results. The proposed SWG-DBR metastructure is a promising platform for various applications in sensing, lasing, and polariton condensation.
- 18 June 2026
- Published by Optica Publishing Group in Optics Express
- Vol. 34 (13) , 23817-23829
Fano resonance in the SWG-DBR metastructure
We study a metastructure formed by vertically integrating a subwavelength grating (SWG) with a distributed Bragg reflector (DBR). Fano resonance occurs in the reflection spectrum of the SWG-DBR metastructure under normal incidence. The Fano resonance arises from the coupling between a broadband reflection background (continuum state) provided by the top reflector (SWG and top DBR) and a narrowband Fabry-Pérot cavity mode (discrete state) formed by the top reflector and the bottom DBR, corresponding to the continuum and discrete states, respectively. The multipole decomposition verifies that the Fano resonance of the proposed SWG-DBR metastructure is dominated by the toroidal dipole. The electric field intensity of the Fano resonance has a maximum at the SWG surface, making it sensitive to the change of the interfacial refractive index. The SWG-DBR metastructure is fabricated, and the reflection spectra under different ambient refractive-index conditions are measured. The Fano resonance peak shift is observed under different liquids covering the surface of the SWG, consistent with the simulated results. The proposed SWG-DBR metastructure is a promising platform for various applications in sensing, lasing, and polariton condensation.
- Open Access
Mode Discrimination in Quasi-PT-Symmetric Surface-Emitting DFB Semiconductor Lasers with Separated Gain and Radiating-Grating Sections
Surface-emitting distributed-feedback (SE-DFB) semiconductor lasers based on second-order gratings face a fundamental triple constraint: the spatial co-location of gain, grating feedback, and vertical radiation functions limits single-mode selectivity, surface extraction efficiency, and far-field beam quality simultaneously. We propose a quasi-parity-time (PT)-symmetric SE-DFB laser with separated gain and radiating-grating sections. In this design, the electrically injected gain section and the passive second-order grating section are placed in different regions along the cavity axis, thereby separating electrical injection from surface emission without epitaxial regrowth. Coupled-mode theory and two-dimensional finite-element simulations demonstrate that the resulting longitudinal non-Hermitian gain–loss asymmetry produces spatial-overlap-dependent threshold discrimination, enabling an isolated low-threshold lasing branch that remains separated from competing cavity modes over the investigated pump-parameter range. Under the HR–AR boundary condition, the proposed design achieves a threshold gain margin of , more than six times that of a conventional HR–AR DFB benchmark considered here, together with an upward surface extraction efficiency of 23.4% obtained from 2D FEM simulations. A simplified steady-state rate-equation estimate further suggests that the increased threshold margin can support strong side-mode suppression. The design imposes no regrowth requirement and is fully compatible with standard single-growth InP ridge-waveguide fabrication.Open AccessMode Discrimination in Quasi-PT-Symmetric Surface-Emitting DFB Semiconductor Lasers with Separated Gain and Radiating-Grating Sections
Surface-emitting distributed-feedback (SE-DFB) semiconductor lasers based on second-order gratings face a fundamental triple constraint: the spatial co-location of gain, grating feedback, and vertical radiation functions limits single-mode selectivity, surface extraction efficiency, and far-field beam quality simultaneously. We propose a quasi-parity-time (PT)-symmetric SE-DFB laser with separated gain and radiating-grating sections. In this design, the electrically injected gain section and the passive second-order grating section are placed in different regions along the cavity axis, thereby separating electrical injection from surface emission without epitaxial regrowth. Coupled-mode theory and two-dimensional finite-element simulations demonstrate that the resulting longitudinal non-Hermitian gain–loss asymmetry produces spatial-overlap-dependent threshold discrimination, enabling an isolated low-threshold lasing branch that remains separated from competing cavity modes over the investigated pump-parameter range. Under the HR–AR boundary condition, the proposed design achieves a threshold gain margin of , more than six times that of a conventional HR–AR DFB benchmark considered here, together with an upward surface extraction efficiency of 23.4% obtained from 2D FEM simulations. A simplified steady-state rate-equation estimate further suggests that the increased threshold margin can support strong side-mode suppression. The design imposes no regrowth requirement and is fully compatible with standard single-growth InP ridge-waveguide fabrication. - Open Access
High-Stability Thulium-Doped All-Fiber Laser at 2050 nm
High-power thulium-doped fiber lasers (TDFLs) operating near 2050 nm are of great interest for applications including atmospheric gas sensing and free-space optical communication owing to the favorable atmospheric transmission and the strong absorption bands of carbon dioxide (CO2). Here, we report an all-fiber high-power TDFL based on a 793 nm-pumped master oscillator power amplifier (MOPA) architecture. The system comprises a custom-built linear-cavity seed laser and two amplification stages. With a maximum pump power of 818 W, the final amplifier delivers 501 W at 2050 nm with a slope efficiency of 51%. Stable operation is confirmed over two hours at full power, with an RMS power fluctuation of only 0.47%. The measured beam quality factors are 1.31 and 1.27 in the horizontal and vertical directions, respectively, indicating near-diffraction-limited performance. The demonstrated system combines high output power, excellent stability, and good beam quality, and thus provides a promising laser source for 2 μm high-performance applications.Open AccessHigh-Stability Thulium-Doped All-Fiber Laser at 2050 nm
High-power thulium-doped fiber lasers (TDFLs) operating near 2050 nm are of great interest for applications including atmospheric gas sensing and free-space optical communication owing to the favorable atmospheric transmission and the strong absorption bands of carbon dioxide (CO2). Here, we report an all-fiber high-power TDFL based on a 793 nm-pumped master oscillator power amplifier (MOPA) architecture. The system comprises a custom-built linear-cavity seed laser and two amplification stages. With a maximum pump power of 818 W, the final amplifier delivers 501 W at 2050 nm with a slope efficiency of 51%. Stable operation is confirmed over two hours at full power, with an RMS power fluctuation of only 0.47%. The measured beam quality factors are 1.31 and 1.27 in the horizontal and vertical directions, respectively, indicating near-diffraction-limited performance. The demonstrated system combines high output power, excellent stability, and good beam quality, and thus provides a promising laser source for 2 μm high-performance applications. - 4 May 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Photonics Technology Letters
- Vol. 38 (18) , 1299-1302
Si-SPAD Based on 40-nm BSI Technology With Wide Spectrum Response
This paper presents a wide spectral response single-photon avalanche diode (SPAD) designed and fabricated using an advanced 40 nm CMOS Image Sensor technology. To achieve superior electrical and optical performance, the N-well and High-Voltage P-well forming the PN junction were deeply optimized via a customized CMOS process. The proposed structure, based on an N-well/HVPW junction, significantly expands the spectral response range, maintaining a photon detection probability(PDP) exceeding 10% across the 440~940 nm spectrum at an excess bias of 2.5 V. Specifically, the device achieves a peak PDP of 54% at 620 nm and 10% at the near-infrared wavelength of 940 nm at an excess bias of 2.5 V. At room temperature (27 °C), the dark count rate is approximately 39 cps/μm2, with an afterpulsing probability below 1%. Additionally, the full width at half maximum timing jitter is measured at approximately 203 ps at 940 nm. These results demonstrate a viable solution for the design of high-performance, wide-spectral-response SPAD.
- 4 May 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Photonics Technology Letters
- Vol. 38 (18) , 1299-1302
Si-SPAD Based on 40-nm BSI Technology With Wide Spectrum Response
This paper presents a wide spectral response single-photon avalanche diode (SPAD) designed and fabricated using an advanced 40 nm CMOS Image Sensor technology. To achieve superior electrical and optical performance, the N-well and High-Voltage P-well forming the PN junction were deeply optimized via a customized CMOS process. The proposed structure, based on an N-well/HVPW junction, significantly expands the spectral response range, maintaining a photon detection probability(PDP) exceeding 10% across the 440~940 nm spectrum at an excess bias of 2.5 V. Specifically, the device achieves a peak PDP of 54% at 620 nm and 10% at the near-infrared wavelength of 940 nm at an excess bias of 2.5 V. At room temperature (27 °C), the dark count rate is approximately 39 cps/μm2, with an afterpulsing probability below 1%. Additionally, the full width at half maximum timing jitter is measured at approximately 203 ps at 940 nm. These results demonstrate a viable solution for the design of high-performance, wide-spectral-response SPAD.