Recent developments in light orbital angular momentum (OAM) detectors based on the orbital photogalvanic effect (OPGE) of semimetallic materials provide a promising route toward on-chip direct electric readout of the OAM of light, as well as large-scale integration of focal-plane array OAM detection devices.1^–3 Compared with the parallel technology route of on-chip OAM detection, which is based on surface plasmon polaritons (SPPs), OPGE-based OAM detection has a broader operation wavelength range, higher responsivity, and simpler device structure.4^–8 This OAM detection scheme is based on the OPGE driven by the helical phase gradient of light through the electric quadrupole and magnetic dipole response of materials, which is related to the fourth-rank nonlinear tensor of the detection material. The crystal symmetry of the detection material needs to fulfill the symmetry requirement to have a nonvanishing OPGE response. The derivation of the expression for the OPGE response has been described in detail in Refs. 1–3. In general, under the excitation of a Laguerre Gaussian (LG) beam carrying OAM, the photocurrent response arises from the electric quadrupole and magnetic dipole effects corresponding to the response term ${\mathit{J}}_{qp}$, which can be divided into four terms according to its dependence on the SAM (${\sigma }_i$) and OAM ($m$) as follows: $${\mathit{J}}_{qp}(\rho ,\theta ,z)=m·{\sigma }_i{\mathit{J}}_{(1)}(\rho ,\theta ,z)+m{\mathit{J}}_{(2)}(\rho ,\theta ,z)+{\sigma }_i{\mathit{J}}_{(3)}(\rho ,\theta ,z)+{\mathit{J}}_{(4)}(\rho ,\theta ,z).$$(1)The first term $m·{\sigma }_i{\mathit{J}}_{(1)}(\rho ,\theta ,z)$ is proportional to the product of SAM (${\sigma }_i$) and OAM (m) and can be used for OAM detection. Experimentally, the OAM-dependent photocurrent component must be extracted experimentally from a circular polarization-dependent measurement, commonly denoted as the circular photogalvanic effect (CPGE), and the extracted component $(m·{\mathit{J}}_{(1)})$ is proportional to the OAM order m if the power and ring radius of LG beams remain the same for different OAM orders 1 and 2. Then, the topological charge of light OAM can be directly distinguished by the quantized plateau of the CPGE component.

However, to extract the CPGE response, the incoming light must be modulated between the left and right circular polarizations, which was realized previously through continuously rotating a quarter waveplate (QWP), as illustrated in Fig. 1(a).1^–3 The CPGE component is then extracted via the Fourier transform of the QWP angle-dependent photocurrent response. Limited by the speed of mechanical polarization modulation, the operation speed of all previous works is at the minute level, which cannot fulfill the speed requirements of most applications.9^–14 Currently, the slow operation speed is the major drawback of OPGE-based OAM detectors compared with parallel OAM detection technology based on SPPs, which can reach operation speeds on the order of tens of microseconds.4

To achieve high-speed direct detection of OAM, the key is to renovate the polarization modulation technology to increase the circular polarization modulation speed and supplement it with a fast readout technique to extract the CPGE component from the modulated response simultaneously. For fast circular polarization modulation, the use of high-speed electrical polarization modulation techniques, such as photoelastic modulators (PEMs) and electro-optic modulators (EOMs), to replace traditional mechanical modulation could promote the polarization modulation speed up to the megahertz (MHz) and gigahertz (GHz) levels for PEMs and EOMs, respectively.15^–20 The generated polarization-modulated photocurrent response can be extracted via a phase-sensitive detection technique with a lock-in amplifier locked to the polarization modulation frequency, as illustrated in Fig. 1(b). In this work, we demonstrate this electric modulation scheme experimentally via the PEM-based polarization modulation method to replace traditional mechanical modulation to achieve a 50-kHz modulation speed, and the fast-modulated CPGE response can be directly extracted from a lock-in amplifier that is locked to the modulation signal of the PEM. The fast electrical modulation scheme is demonstrated on an OPGE-based OAM detector made from multilayer graphene (MLG). The topological charge of light OAM can be clearly distinguished by the quantized plateau of the CPGE response, which is directly extracted by a lock-in amplifier. Moreover, we compare the OPGE responsivity and OAM resolution capability under mechanical and PEM modulations. We find that the difference in the OPGE responsivity arises from the different polarization modulation and signal readout techniques. Considering that the polarization modulation frequency of the PEM is 50.14 kHz, the mini-second operation speed of the detector is jointly limited by the requirement of the integration time of multiple modulation periods in the readout process using a lock-in amplifier and the microsecond-level response time of the device. Our work solves the major obstacle of slow operation speed caused by the traditional mechanical polarization modulation technique in OPGE detectors, and the new modulation and readout scheme is directly applicable for large scale integration of a focal plane array device.21^–25

The direct OAM detector used in this work is made with few-layer exfoliated graphene, as shown in Fig. 1(c). The graphene flakes are exfoliated from the graphite and transferred onto a $300\mathrm{nm}/500\mu \mathrm{m}$${\mathrm{SiO}}_2/\mathrm{Si}$ substrate. A standard electron-beam lithography technique is used to pattern the $\text{U}$-shaped electrodes to collect the radial photocurrent response of the MLG, as the OPGE response survives along the radial direction according to the symmetry of the MLG.1 Then, the electrodes were deposited by an electron-beam evaporator with 10 nm Ti and 80 nm Au. The deviation of the OPGE response of the MLG device is fully described in Section 1 in the Supplementary Material. Compared with ${\mathrm{TaIrTe}}_4$, which has also been demonstrated to realize OAM detection in the mid-infrared region,2 MLG has one order of magnitude greater OPGE responsivity and recognition capability.1 Moreover, MLG is already epitaxially growable at the wafer scale through either chemical vapor deposition or epitaxial growth on SiC, and the fabrication process is completely CMOS compatible, which promises large-scale integration of ambient stable, mid-infrared direct OAM photodetection devices and OAM-sensitive focal plane array devices.

For the specific device used in this work, the radii of the inner and outer electrodes are $12\mu \mathrm{m}$ and $20\mu \mathrm{m}$, respectively. Figure 1(d) shows the source–drain current $I_{\mathrm{ds}}$ as a function of the back gate voltage $V_g$ under a source–drain voltage $V_{\mathrm{ds}}=5\mathrm{mV}$, together with the $I_{\mathrm{ds}}-V_{\mathrm{ds}}$ measurement at $V_{g }=0$. In the measurement, the back gate voltage $V_g$ is applied by a low-resistivity silicon substrate. The peak in the $I_{\mathrm{ds}}-V_g$ plot signifies the Fermi level at the Dirac point, whereas the linear result from the $I_{\mathrm{ds}}-V_{\mathrm{ds}}$ measurement [inset of Fig. 1(d)] confirms good ohmic contact. For scanning photocurrent measurement, the output from a $4\text{-}\mu \mathrm{m}$ CW quantum cascade laser source is focused into a spot with a radius of $10\mu \mathrm{m}$ by a 40× reflection objective, and scanning microscopy is obtained by controlling the 2D $(x,y)$ movement of the device placed on a motorized stage. Figures 1(e) and 1(f) show the scanning photocurrent microscopy image of the $\text{U}$-shaped MLG detector under the excitation of a basic mode Gaussian beam at $4\mu \mathrm{m}$ with a power of 0.7 mW, together with the in situ scanning reflection image. The photocurrent response mainly originates from the region near the electrodes and the edge of the material, with a responsivity lower than that of the previously reported device because of the reduced thickness.1 The interface with the curved electrodes also has a response but is one order of magnitude smaller than a response from the straight electrode part. Typically, OAM beams are generated by passing the laser beam through a spiral phase plate specially designed for different OAM orders at $4\mu \mathrm{m}$. Then, the OAM beams are focused on the same ring radius and embedded between the inner and outer electrodes. The radial photocurrent can be effectively collected by the $\text{U}$-shaped electrode, and the topological charge of light OAM can be clearly distinguished by the quantized plateau of the CPGE component of the radial OPGE response, which enables the detection of light OAM.

First, we present the measurement results of the CPGE response via conventional mechanical modulation. For such measurements, a quarter-wave plate (QWP) is placed after a polarizer and rotated to modulate the polarization of the OAM beams following the scheme shown in Fig. 2(a). When the QWP angle ($\theta =2\pi ft$) is rotated with rotation frequency $f$, the polarization state of the OAM beams undergoes a 180-deg periodic change of linear ($\theta =0\mathrm{deg}$)-left circular ($\theta =45\mathrm{deg}$)-linear ($\theta =90\mathrm{deg}$)-right circular ($\theta =135\mathrm{deg}$)-linear ($\theta =180\mathrm{deg}$), as shown in Fig. 2(b). The dependence of the first term of the radial OPGE response $m·{\sigma }_i{J}_{(1)}^{\rho }$ on the QWP angle $\theta$ (or $2\pi ft$) can be written as $$m·{\sigma }_i{J}_{(1)}^{\rho }=-m·C(\overrightarrow{\mathit{r}})\sin 4\pi ft,$$(2)where $C(\overrightarrow{\mathit{r}})$ is a coefficient related to the light field distribution and the rank-4 conductivity tensors of the detection material. The spatial integration of $C(\overrightarrow{\mathit{r}})$, which corresponds to the collected radial OPGE response, is a function of the total power $P$, the ring radius of the focused OAM beam, and the rank-4 conductivity tensors of the detection material. The deviation of Eq. (2) is fully described in Section 2 in the Supplementary Material. According to Eq. (2), by measuring the photocurrent response at different QWP angles ($\theta$), the CPGE component can be obtained by extracting the 180-deg periodic component of the photocurrent response through a Fourier transform, and the extracted CPGE response is written as $m·C(\overrightarrow{\mathit{r}})$, which is proportional to the OAM order $m$. Previous work on OAM photodetectors based on ${\mathrm{WTe}}_2$, ${\mathrm{TaIrTe}}_4$, and MLG employed these mechanical polarization modulation and CPGE extraction approaches.1^–3 The topological charge of light OAM can be clearly distinguished by the quantized plateau of the CPGE response, but the operation speed is limited to the order of minutes.1^–3

The CPGE measurement results with mechanical modulation are presented in Figs. 2(c)–2(e), which are qualitatively the same as the results presented in Ref. 1; however, the measurements in this work are performed on a thinner MLG device for fair comparison between the two modulation schemes. In the measurement, the OAM beams are focused by the 40× reflection objective to the ring with a radius of $16\mu \mathrm{m}$ and embedded between the inner and outer $\text{U}$-shaped electrodes. Figure 2(c) shows the photocurrent as a function of the angle of the QWP for OAM orders of $\pm 4$, $\pm 2$, and $\pm 1$ with an excitation power of 1.5 mW. The difference in the photocurrent response for excitations with left and right circularly polarized beams is marked in the figure, which shows similar magnitudes but opposite signs for the OAM order $\pm m$, and the magnitude increases as $|m|$ increases. Figure 2(d) shows the extracted CPGE component $J_C$ as a function of OAM order $m$. The $J_C$ shows step-like changes with OAM order $m$ and is proportional to $m$ without a background signal. These results are consistent with a previous report on a similar MLG device.1 From the measurement results, we can obtain the major merits of device performance. If we define the OPGE responsivity $K$ as the ratio of the CPGE component $J_C$ to the OAM order $m$ under unit OAM light excitation power (P): $K=J_C/(m·P)$, linear fitting with respect to $m$ in Fig. 2(e) can be used to obtain the responsivity ($K$) of the OPGE and the uncertainty of the responsivity $K$ (${\sigma }_K$) to be $K=74.9\mathrm{nA}/\mathrm{W}$ and ${\sigma }_K=4\mathrm{nA}/\mathrm{W}$. Subsequently, we define the resolution capability $R$ of OAM as $R=K/{\sigma }_K$ to account for the signal-to-noise ratio of the OPGE measurement, and we can obtain $R=18.6$. Although the OPGE responsivity of this device is half of that previously reported for a thicker MLG device, these two devices have comparable OAM resolution capabilities because of the reduced noise level of the thinner device used in this work.1 Nevertheless, the performance of the device used in this work is still five times better than that demonstrated for a ${\mathrm{TaIrTe}}_4$ device.2

Next, we present the measurement results when a photoelastic modulator (PEM) is used to replace the QWP to achieve high-speed electric polarization modulation at a frequency of 50.14 kHz and directly extract the CPGE component of the photocurrent response via the lock-in amplifier. A detailed schematic diagram of the measurement setup is shown in Figs. 3(a) and 3(b). The OAM beams, generated by passing basic mode Gaussian beam through a spiral phase plate, are modulated by the PEM and focused on a ring embedded between the $\text{U}$-shaped electrodes of the OAM photodetector. The excited OPGE response of the OAM photodetector is amplified by a pre-amplifier, and then, a lock-in amplifier extracts the CPGE component and outputs it to the computer through a digital acquisition card. The PEM consists of a piezoelectric transducer and a half-wave resonant bar.26^,27 The transducer changes the birefringence properties of the resonant bar by electrically stretching and compressing the vibration optical element, thereby imparting a periodic phase difference to the polarization components along the two optical axes. In this configuration, the optical axes of the PEM are aligned along the $xy$ direction, the polarizer’s polarization direction is set at a 45-deg angle to the $x$-axis, and the peak phase retardation (${\delta }_0$) of the PEM is set to $\pi /2$ [Fig. 3(d)]. To drive the PEM, a sinusoidal modulating voltage [as shown in Fig. 3(b)] driven by an electronic driver circuit is applied to the quartz piezoelectric transducer to drive the rectangular ${\mathrm{ZnSe}}_2$ optical element and create polarization modulation. The phase difference induced by the PEM for two perpendicular polarizations at different times ($t$) can be written as $\delta ={\delta }_0\sin 2\pi ft$, where $f$ is the modulation frequency. The PEM is a resonant device where the transducer is tuned to the resonance frequency of the optical element, which is determined by its bar length and the speed of sound in the material.26 Here, the precise oscillation frequency is fixed at 50.14 kHz, which is determined by the photoelastic properties of the ${\mathrm{ZnSe}}_2$ element and transducer assembly in the PEM. When the OAM beam passes through the PEM, in a single operational cycle of the PEM, the polarization of the OAM beam undergoes a sequence of transitions—linear, left-circular, linear, right-circular, and linear—exhibiting a variation pattern of polarization modulation similar to that realized by rotating the QWP in a period of 180 deg, as shown in Fig. 3(f).

However, the difference in the specific variation in the polarization sequence leads to different variations in the OPGE signal in one operation cycle. For PEM modulation, after the first-order expansion of $\sin \delta$ through the integer-order Bessel function and keeping the leading first-order term, the dependence of the OPGE response term on the QWP angle $\theta =2\pi ft$ can be written as $$m·{\sigma }_i{\mathit{J}}_{(1)}^{\rho }=m·C(\overrightarrow{\mathit{r}})2J_1({\delta }_0)\sin 2\pi ft,$$(3)where $J_1({\delta }_0)$ is a first-order Bessel function. Compared with the mechanical modulation, the CPGE response extracted via the PEM has an additional $2J_1({\delta }_0)$ coefficient, and for peak phase retardation ${\delta }_0=\pi /2$ used in this work, $2J_1(\pi /2)=1.13365$. The deviation of Eq. (3) is fully described in Section 2 in the Supplementary Material. Another difference lies in the readout method of the CPGE response. When the PEM is used, the CPGE response can be directly extracted by the lock-in amplifier locked to the 50.14-kHz periodic modulation of the left and right circular polarizations, and the lock-in amplifier outputs the root mean square of the CPGE response. For mechanical modulation, the phase is continuously changed by rotating a quarter waveplate, and the CPGE response is extracted via the Fourier transform of the photocurrent response, which is dependent on the QWP angle. To exclude background signals and suppress $1/f$ noise, the OAM beams are also modulated by a mechanical chopper, and the photocurrent response is also extracted by a lock-in amplifier locked to the driving frequency of the chopper. The different CPGE read-out approaches introduce additional coefficient differences to the experimentally measured CPGE responses, which are fully described in Section 3 in the Supplementary Material. Overall, for mechanical and PEM modulations, the extracted CPGE response differs only in terms of the constant coefficient because of different polarization modulations and different CPGE readout methods. Theoretically, the experimentally extracted CPGE response with PEM modulation is $\pi J_1(\pi /2)\approx 1.78$ times greater than that with mechanical modulation. More details about the derivation of the extracted CPGE response are presented in Sections S2 and S3 in the Supplementary Material.

The measurement results with PEM modulation are presented in Fig. 4. The excitation conditions of the OAM beam are the same as those for the mechanical modulation measurement. Figure 4(a) shows the extracted CPGE component $J_C$ at OAM orders of $\pm 4$, $\pm 2$, and $\pm 1$ with a constant excitation power of 1.5 mW. The CPGE signal is measured with a lock-in amplifier that is phase-locked to the 50.14-kHz modulation signal of the PEM, and the integration time constant is set to 300 ms, which corresponds to 15,000 modulation cycles for the measurement results shown in Fig. 4(a). The measurement is performed by comparing the lock-in signal by blocking or unblocking the excitation light. Here, we note that there are clear background signals when the light is turned off. The constant background signal is dominated by the pickup of electromagnetic waves leaked from the electric driving head of the PEM. Because the picked signal has the same frequency as the PEM’s operating frequency, it dominates the background that is picked by the lock-in amplifier at the driving frequency. Taking the difference in the on-off signal provides the absolute CPGE response. The results show similar magnitudes but opposite signs for opposite OAM orders $\pm m$, and the magnitude increases as $|m|$ increases. Figure 4(b) shows the extracted CPGE component $J_C$ as a function of the OAM order $m$. The $J_C$ shows step-like changes with the OAM order $m$ and is proportional to $m$. These results are all consistent with those measured with the mechanical modulation shown in Figs. 2(c)–2(e). Here, we note that the magnitude of the extracted CPGE response from PEM modulation is $\sim 1.1$ times greater than that extracted from mechanical modulation, which is different from the theoretical ratio of 1.78; such deviation is mainly due to the limited response time of the device. The response time for a typical MLG device is measured to be $3.42\mu \mathrm{s}$, as presented in Figure S1 in the Supplementary Material. For a 50.14-kHz polarization modulation, each modulation cycle is only $20\mu \mathrm{s}$, which is close to the response time of the device, so the response magnitude of the device is reduced because of the limited photocurrent response speed of the devices, which leads to a lower CPGE response magnitude ratio of the two different modulation schemes.

From the measurement results, we can obtain the major merits of the figures of device performance: the OPGE responsivity $K=82.6\mathrm{nA}/\mathrm{W}$, the uncertainty of the responsivity ${\sigma }_K=8\mathrm{nA}/\mathrm{W}$, and the resolution capability $R=K/{\sigma }_K=9.8$. Figure 4(c) presents a comparison of the OPGE responsivities and resolution capabilities of the two modulation approaches. The OPGE responsivity under PEM modulation is $\sim 1.1$ times greater than that obtained under mechanical modulation. However, owing to the lower signal-to-noise ratio of the CPGE response, the OAM resolution capability for PEM modulation is approximately one third that for mechanical modulation. The operation speed of PEM modulation is limited mainly by the 300-ms measurement time constant ($T$) of the lock-in amplifier. The measurement time constant of the lock-in amplifier can be reduced to compensate for the elevated noise level. Figure 4(d) shows the measurement results with different $T$ (from 1 to 300 ms) for $\mathit{m}=\pm 4$. As $T$ decreases, the photocurrent response remains constant ($\sim 0.5\mathrm{nA}$), but the noise increases dramatically (∼$\propto 1/\sqrt{T}$), and the signal-to-noise ratio, which reflects the OAM resolution capability, decreases with $T$ ($\propto \sqrt{T}$) as illustrated in Fig. 4(f). When the time constant decreases to 1 ms, the signal can barely be resolved, which limits the final operation speed of the device to 1 kHz. Here, we note the trade-off relationship between the operation speed and the signal-to-noise ratio plotted in Fig. 4(e). However, by providing enough OPGE response of the device, the measurement time of the lock-in amplifier can be further decreased, and the operation speed can be improved further. The time constant is ultimately limited by the 50.14-kHz polarization modulation frequency of the PEM and the $3.54\text{-}\mu \mathrm{s}$ photocurrent response time of the device. In principle, the minimum lock-in time constant must be over several polarization modulation cycles, and the modulation period must be twice the response time of the device through the Nyquist–Shannon sampling theorem.

The 3–4 orders improvement in the operation speed demonstrated in this work overcomes the major technical bottleneck of OPGE-based OAM photodetectors. With the ms level operation speed, the OPGE detector not only exceeds the reaction time of human eyes but also exceeds the state-of-the-art OAM detector recently demonstrated based on SPP,4 which has a much narrow operation bandwidth limited by the SPP resonance. Such speed satisfies the requirement of many applications such as remote sensing and optical monitoring. However, such detection speed is still not sufficient for applications that require faster operation speed, such as optical communications. For the PEM modulation scheme demonstrated in this work, the polarization modulation speed is mainly limited by the resonance frequency of the PEM crystal and the photocurrent response speed of the device. A ZnSe crystal is used in this work to achieve polarization modulation at mid-infrared wavelengths. The resonance frequency of the PEM is determined by the photoelastic properties of the ZnSe element and transducer assembly and varies with the material species and size, which are constrained by specific factors, such as the size of the focal plane array and operation wavelength. In general, the modulation frequency is limited to several MHz with a PEM modulation scheme.19^,20 Alternatively, the operation can be tuned via an electro-optical modulator, which can reach a 100-GHz polarization modulation speed.16^,18 However, compared with PEMs, electro-optical modulators have the drawbacks of a lower numerical aperture, higher driving voltage, and smaller operation wavelength range due to fewer choices of optical materials.27 The reading speed of the OAM detector, however, is further limited by the phase-sensitive reading part. The noise level and requirement of the signal-to-noise ratio of the device constrain the maximum bandwidth of the low-pass filter used in the lock-in amplifier. This imposes a limit on the minimum time constant of the lock-in amplifier, which restricts the operation speed of the device to multiple polarization modulation periods. On the other hand, the operation speed can be further limited by the photoresponse time of the detector. Once the modulation frequency approaches or exceeds the photocurrent response bandwidth, the response of the device is retarded to the modulated signal, which decreases the overall magnitude of the photocurrent response.

When specific applications are taken into account, there are additional limitations for the PEM modulation scheme. Besides the common drawbacks of OPGE-based OAM detectors, such as the detection accuracy being highly sensitive to the radius and position of the focused OAM beam, there are additional issues that are introduced by the PEM modulation scheme compared with the mechanical polarization modulation scheme. Experimentally, a critical issue is that the extraction of CPGE response has background signals due to the electrical pickup of electromagnetic waves leaked from the electric driving head of the PEM. Although the leakage signal is a constant background, it adds noise fluctuation and interfering signal to the photodetection, which reduces the recognition capability of the detector. This becomes a more critical issue for on-chip integration when the integration density increases and such leakage becomes more pronounced and more difficult to shield. Another difference is that PEM can only realize a continuous periodic change of polarization, which is different from a direct switch from the left and right circular polarization that can be realized in mechanical modulation and electro-optical modulation. The continuous periodic change of polarization requires more complicated analytic circuits to perform the Fourier transform to obtain a CPGE response.

Furthermore, the PEM modulation scheme demonstrated in this work is directly applicable to focal plane array devices, as illustrated in the schematic diagram proposed in Fig. 5. Focal plane array devices are critical for anti-interfered detection of high-speed target when a single detector is not sufficient to track the fast-moving project and/or object image has to be incorporated for target perception purpose. The OAM detector based on MLG can be scaled to focal plane arrays with large areas epitaxially or CVD-grown multilayer graphene.28^–30 The photoelastic crystal with the transducer assembly can be placed before the OAM detector arrays, and the phase-sensitive reading circuit can be integrated with the photocurrent readout circuit and locked to the electric driving source to extract the CPGE component of each detector cell. According to the extracted CPGE component of each sensing cell, the OAM order can be extracted from any pixel of the focal plane arrays for anti-interfered detection by the cognition of the OAM information of the target. For on-chip integration, miniaturized electrically driven polarization modulation materials are required for OAM detectors, recent progress in the use of atomically thin topological semimetals for tunable phase retardation may satisfy such demand,31 and the driving voltage can be much lower for atomically thin materials and thus facilitate the on-chip integration of all necessary components.

Dehong Yang received his bachelor’s degree in physics from Peking University, Beijing, China, in 2024. He is currently working toward his PhD in condensed matter physics with the International Center for Quantum Materials, Peking University, Beijing, China. His research interest is mainly in OAM photodetection.

Chang Xu received the bachelor’s degree in physics from Peking University, Beijing, China, in 2024. He is currently working toward his PhD in condensed matter physics with the International Center for Quantum Materials, Peking University, Beijing, China. Their research interests include OAM photodetection and topological semimetals.

Jiawei Lai received his PhD in condensed matter physics from Peking University in 2020. He then received funding from the China Postdoctoral Innovative Talent Support Program to pursue postdoctoral research at Peking University. Since 2023, he has been serving as an associate professor at Xi’an Jiaotong University. His work primarily focuses on fundamental physics research in photoelectric detection based on novel quantum materials, encompassing the physical mechanisms of broadband photodetection and micro/nano-optoelectronic devices.

Zipu Fan received his BS degree in physics from Peking University, Beijing, China, in 2019. He is currently working towards his PhD in physics with the School of Physics, Peking University, Beijing, China. His research interests include magnetic Weyl semimetal and photocurrent.

Delang Liang received his BS degree in materials science and engineering from Hunan University, Changsha, China, in 2019. He is currently working towards his PhD in materials science and engineering with the School of Materials Science and Engineering, Hunan University, Changsha, China. His research interests include second harmonic generation (SHG), light-emitting diodes (LEDs), and optical memory.

Shiyu Wang received her BS degree in applied physics from Hunan University, Changsha, China, in 2022. She is currently working towards her PhD in condensed matter physics with the School of Physics, Peking University, Beijing, China. Her research interests include photoluminescence and electroluminescence mechanisms in two-dimensional materials.

Jinluo Cheng received his MD degree with majors in applied physics and his PhD in condensed matter physics, both from the University of Science and Technology of China, Hefei, China, in 2002 and 2007, respectively. He is currently a professor at the Changchun Institute of Optics, Fine Mechanics and Physics, Chinese Academy of Sciences, China. His current research interests focus on the nonlinear optics and light–matter interaction in 2D materials.

Dong Sun received his PhD in physics from University of Michigan in 2009. After postdoc work in University of Michigan and University of Washington, he joined International Center for Quantum Materials of Peking University as an associate professor in 2012 and later promoted to be full professor. He has over 20 years research experience on light–matter interaction and optoelectronic devices.