A self-driven and high-performance photodetector based on a GeSe/Si van der Waals heterojunction with high-speed photoresponse

Abstract

van der Waals (vdW) heterojunction photodetectors exhibit high performance due to their high-quality interface and high design flexibility, and unique properties of two-dimensional (2D) materials. Particularly, combining 2D semiconductors with technologically mature semiconductors offers a promising pathway toward high-performance photodetection. Herein, we report a high-performance self-driven photodetector based on a vertical GeSe/Si vdW heterojunction, constructed using high-quality GeSe single crystals grown by the chemical vapor transport method, which benefits from the type-II band alignment and the strong built-in electric field at the GeSe/Si interface. As a result, the photodetector exhibits a high responsivity of 29.8 A W⁻¹, a high EQE of 6959.7%, a high detectivity of 2.1 × 10¹² Jones, and a fast rise/decay time of 8.5 µs/23.7 µs under 532 nm laser illumination at zero bias. In addition, the GeSe/Si vdW heterojunction photodetectors demonstrate a stable broadband photoresponse and pronounced photovoltaic behavior under visible-light illumination (405–604 nm). This work highlights the advantages of integrating 2D GeSe with silicon via vdW heterojunction engineering and provides a significant strategy for developing self-driven, high-performance photodetectors toward practical optoelectronic applications.

---

New concepts

van der Waals (vdW) heterojunctions provide a promising route to realize high-performance optoelectronic devices by integrating 2D materials with conventional semiconductors. Here, we introduce a GeSe/Si vdW heterojunction photodetector that demonstrates self-driven operation, ultrafast response, and high responsivity. This concept differs from previous GeSe-based devices that relied mainly on 2D/2D combinations, as the integration with technologically mature Si enables superior interface quality and scalable device fabrication. The type-II band alignment between GeSe and Si efficiently separates photogenerated carriers under zero bias, leading to a responsivity of 29.8 A W⁻¹, an EQE of 6959.7%, a detectivity of 2.1 × 10¹² Jones, and a response time of 8.5/23.7 µs. Beyond device performance, this work highlights a universal strategy to bridge emerging 2D semiconductors with industry-standard Si platforms, offering new insights into designing next-generation, energy-efficient, and integrable photodetectors for imaging and optical communication applications.

1. Introduction

Two-dimensional (2D) materials, owing to their layered structures and thickness-dependent optoelectronic properties, have emerged as a versatile platform for next-generation semiconductor devices.^1–3 Significant progress has been made in 2D-material-based field-effect transistors,^4,5 memory devices,^6,7 light-emitting diodes,^8,9 and photodetectors.^10–12 Among these applications, photodetectors, which convert optical signals into electrical signals, play a critical role in optical communication and imaging technologies. The atomic-scale thickness of 2D semiconductors, combined with their high optical absorption coefficients and tunable band structures, enables photodetectors with high responsivity and a broadband photoresponse.^13,14 Moreover, the absence of dangling bonds on their surfaces facilitates the construction of vdW heterostructures with diverse material combinations, offering freedom in device design.^15–17

Germanium selenide (GeSe), as a representative 2D semiconductor, possesses a direct bandgap of approximately 1.1 eV and exhibits p-type conductivity due to the presence of Ge vacancies.^18,19 It features a high hole mobility of up to 128.6 cm² V⁻¹ s⁻¹ and an optical absorption coefficient on the order of 10⁵ cm⁻¹, making it highly attractive for photodetection applications.²⁰ Accordingly, GeSe-based photodetectors fabricated from liquid-phase-synthesized nanosheets, CVD-grown films, and mechanically exfoliated single crystals have demonstrated impressive responsivities.^21,22 For example, photodetectors fabricated using chemical vapor deposition (CVD)-grown GeSe nanosheets exhibited high responsivities up to 3.5 A W⁻¹,²³ and those prepared from mechanically exfoliated GeSe single-crystalline flakes exhibited ultrahigh responsivities exceeding 10⁴ A W⁻¹.^18,20 Moreover, vdW heterostructure photodetectors constructed with GeSe not only exhibit a self-driven photoresponse but also show rapid response and recovery times. Heterostructures such as GeSe/SnS₂,²⁴ GeSe/MoTe₂,^25,26 GeSe/SnSe,²⁷ and GeSe/MoS₂²⁸ have all demonstrated outstanding photodetection performance.

Despite these promising results, most reported GeSe-based photodetectors rely on all-2D heterostructures based on mechanically exfoliated flakes. The limitation significantly hinders the practical deployment of GeSe-based photodetectors. In this context, integrating 2D GeSe with technologically mature and industrially used semiconductors, such as silicon (Si),^29–31 represents a compelling strategy to bridge the gap between high-performance 2D optoelectronics and scalable device manufacturing.³² However, systematic investigations of GeSe/Si vdW heterojunction photodetectors remain scarce.

Herein, we report a high-performance self-driven photodetector based on a GeSe/Si vdW heterojunction. The performance of the photodetector was investigated and it exhibited self-driven photodetection with a high responsivity of 29.8 A W⁻¹, a high EQE of 6959.7%, a high specific detectivity of 2.1 × 10¹² Jones, and a fast rise/decay time of 8.5 µs/23.7 µs under 532 nm laser illumination, together with a broadband visible-light (405–604 nm) photoresponse and pronounced photovoltaic behavior. The device reported here with high performance will support the development of vdW heterojunction photodetectors.

2. Experimental section

2.1. Synthesis and characterization of crystals

GeSe single crystals were successfully grown using the chemical vapor transport (CVT) method. High-purity germanium (In) powder (5N) and selenium (Se) powder (5N) were used as raw materials, with iodine utilized as the transport agent. These materials were carefully transferred into a quartz ampoule in an argon atmosphere in a glove box. The ampoule was subsequently sealed and placed into a tube furnace. The furnace temperature was increased to 965 K and kept at this temperature for 5 days for crystal growth.

The structure of the as-grown GeSe single crystals was characterized by X-ray Diffraction (XRD, PANalytical Empyrean, Netherlands) with Cu K_α1 radiation (λ = 1.5406 Å) at 40 mA and 40 kV. Raman spectroscopy investigations were carried out using a Raman system (ANDOR iXon Ultra, U.K.) with a 532 nm laser of 2 mW power. The microstructure was characterized using energy-dispersive spectroscopy (EDS) and transmission electron microscopy (TEM, JEOL TEM-F200).

2.2. Device fabrication and characterization

The n-Si (resistivity 4–6 Ω cm) substrates were ultrasonically washed with acetone, isopropyl and deionized water. Then, the n-Si substrates were immersed into 5% HF solution to remove their native oxide layer. The GeSe crystals were repeatedly exfoliated by using blue tape, and the tape was pressed firmly against a clean n-Si substrate. After heating at 120 °C for 5 minutes, the tape was rapidly peeled off from the n-Si substrate. Then, the GeSe/Si vdW heterojunctions were successfully fabricated. And the electrode patterns were designed and fabricated by the photolithographic lift-off process. The Au layers and indium tin oxide (ITO) layers were deposited via direct current magnetron sputtering to use as the top electrodes contacting with GeSe. The diameter of the top electrode is 30 µm. The In–Ga alloys were employed as bottom electrodes contacting with n-Si. The GeSe-FETs with Au electrodes were also fabricated using a similar process. The optoelectronic properties of photodetectors and electrical properties of GeSe-FETs were measured using a Keithley 2400 voltage source, a Keithley 6487 picoammeter, an oscilloscope (Keysight DSOX 3104A), an optical chopper, and a 405 nm, 532 nm and 604 nm laser in a custom-designed vacuum probe station with a pressure of 5.0 Pa.

3. Results and discussion

The GeSe single crystals were grown by the CVT method and shown in the inset of Fig. 1(a). Their high-crystallinity is demonstrated by XRD results in Fig. 1(a). The XRD spectra of the GeSe crystals exhibit four main strong and sharp diffraction peaks, which are consistent with the (200), (400, (600), and (800) crystal planes.³³ The Raman spectra of GeSe flakes, shown in Fig. 1(b), exhibit peaks at 80, 149.7 and 187.5 cm⁻¹, corresponding to the vibrational modes of A_1g, B_3g, and A_3g.³⁴ As shown in Fig. 2(a), the morphology of GeSe nanoflakes was further characterized by TEM. For more detailed microstructural characterization, the HRTEM image in Fig. 2(b) demonstrates the high crystallinity of GeSe nanoflakes, with no obvious lattice defects observed. The fast Fourier transform (FFT) pattern taken from the region marked by the red box (the inset of Fig. 2b) exhibits sharp and well-defined diffraction spots, further confirming the high crystallinity of GeSe nanoflakes. Fig. 2(c) presents the enlarged image extracted from the red-box region in Fig. 2(b), where the interplanar spacing is measured (Fig. S1) to be approximately 0.29 nm, corresponding to the (011) plane. The chemical composition of the GeSe nanoflake was analyzed by DES mapping, as depicted in Fig. 2(d)–(f). The Ge and Se elements are uniformly distributed throughout the nanoflakes. The corresponding EDS spectrum is presented in Fig. S2, and the atomic ratio of Ge and Se is close to the stoichiometric ratio of 1 : 1, as shown in the inset table of Fig. S2.

Fig. 1 The characterization of GeSe. (a) The XRD spectra of GeSe single crystals. (b) The Raman spectra of GeSe nanoflakes.

Fig. 2 Characterization of the GeSe nanoflake. (a) TEM image of a GeSe nanoflake. (b) The HRTEM image of the GeSe nanoflakes, and the inset shows the FFT pattern of single-crystals. (c) Cropped image of confined GeSe from the red box in (b). (d)–(f) EDS mapping images of the Ge element and Se element in (d).

The electrical properties of GeSe-FET were investigated as shown in Fig. S3. An obvious gate–voltage (V_g) tunable source–drain current (I_ds) and typical p-type conductivity are observed, with a hole mobility of 2.4 cm² V⁻¹ s⁻¹ and an on/off ratio of ∼10² under a source–drain voltage (V_ds) of 1 V. To confirm that the performance of the GeSe/Si vdW heterojunction photodetectors does not originate from the Schottky contact between the material and the electrodes, the electrical characteristics of the electrode–material interface were investigated as shown in Fig. S4. The linear relationship of the I–V curves is observed between the Au electrodes and GeSe, and between the In–Ga alloys and n-Si. The photodetectors based on the GeSe/Si vdW heterojunctions were further evaluated. A vertical device structure was employed and shown in Fig. 3(a). The output curve under dark conditions exhibits typical rectification behavior of a p–n junction with a rectification ratio of 60 at V_ds = ±1 V in Fig. 3(b). There is an obvious increase of reverse current, attributed to the generation of photocarriers, with increasing laser power intensity (P). The time-resolved I–T curves, as shown in Fig. 3(c), reveal a self-driven photodetection with an on/off ratio increasing from 39 at 1 mW cm⁻² to 5 × 10² at 69 mW cm⁻². The photocurrent (I_ph = I_light − I_dark) follows a power-law relationship with laser power intensity, which is expressed as I_ph ∝ P^α. At zero bias voltage, the device exhibits a low dark current of 10⁻⁹ A. The exponent α equals 1 in an ideal photodetector.³⁵ As shown in Fig. 3(d), the I_ph of the GeSe/Si vdW heterojunction photodetector exhibits a linear relationship, represented by the orange fitting curve, with P in a logarithmic coordinate system. A derived α value of 0.65, which is below the ideal value of 1, suggests the presence of trap states and the photogating effect.^36–38

Fig. 3 A schematic image and the performance of the GeSe/Si heterojunction device. (a) A schematic image of the GeSe/Si heterojunction device. (b) The output curves with and without the laser irradiation of the 532 nm laser. (c) Time-resolved photodetection under the switch on/off irradiation of the 532 nm laser at 0 V. (d) Photocurrent of the device as a function of the 532 nm laser power intensity at 0 V.

To further evaluate the performance of the GeSe/Si vdW heterojunction photodetector, three key parameters, responsivity (R), external quantum efficiency (EQE), and specific detectivity (D*), are employed and calculated using the following equations:^39,40

where S is the active area, h is Planck's constant, c is the speed of light, e is the electron charge, λ is the incident light wavelength, and I_dark is the dark current. R represents the ability to convert light into electrical signals. As shown in Fig. 4(a), the device demonstrated the highest R value of 29.8 A W⁻¹ at 1 mW cm⁻² when V_ds = 0 V, which is larger than those of the reported GeSe/MoS₂ and Si/MoTe₂ photodetectors.^28,41 EQE is the ratio of the number of charge carriers collected by the device to the number of incident photons at a given wavelength. A peak value of EQE that can reach 6959.7% at 1 mW cm⁻² when V_ds = 0 V shown in Fig. 4(b) could be attributed to extra photocurrent contributions due to the photoconductivity and photogating effect.⁴² D* reflects the capacity to detect weak light signals. Usually, D* is dominated by shot noise, 1/f noise, and thermal noise. If only considering the influence of the shot noise, D* could be calculated through equation (3).^43–46 The maximum value of D* reached 2.0 × 10¹² Jones at 1 mW cm⁻² when V_ds = 0 V as shown in Fig. 4(c). These three parameters decrease with increasing P, which could be attributed to the increasing non-recombination of photogenerated carriers originating form trap states.⁴⁷ The rise/decay time of the GeSe/Si vdW heterojunction photodetector was investigated using a pulsed laser and an oscilloscope. As shown in Fig. 4(d), the rise/decay time is defined as the time interval between 10% and 90% of the peak voltage, which was measured to be 8.5/23.7 µs when V_ds = 0 V. The self-driven photodetection behavior with high-speed photoresponse of the device can be explained using the energy band diagram of the GeSe/Si vdW heterojunction. As shown in Fig. 4(e), The electron affinity values of GeSe and Si are 3.2 eV and 4.05 eV, respectively, and their bandgaps are 1.2 eV and 1.12 eV,^26,48 respectively. A built-in electric field is established in the GeSe/Si vdW heterojunction as electrons diffuse from Si to GeSe and holes diffuse from GeSe to Si and reach equilibrium when the Fermi levels of GeSe and Si align. The photogenerated carriers in depletion regions are separated by the built-in electric field at zero bias voltage. And I_ph is formed due to which electrons at the conduction band minimum (CBM) of GeSe are drifted to the CBM of Si while the photogenerated holes at the valance band maximum (VBM) of Si are drifted to the VBM of GeSe as shown in Fig. 4(f).

Fig. 4 The performance of the InSe/Si heterojunction device. (a)–(c) Plots of the R, EQE and D* ratio of the device varying with the laser power intensity at 0 V. (d) The rise/decay time is estimated to be 40 µs/48 µs at 0 V. Energy band diagram of the InSe/Si vdW heterojunction under (e) dark and (f) laser irradiation.

To further evaluate the broadband photoresponse of the GeSe/Si vdW heterojunction photodetector, lasers with wavelengths of 405, 532, and 604 nm were employed to characterize its output characteristics. As shown in Fig. 5(a), an obvious increase of reverse current is observed under laser irradiation with a power intensity of 69 mW cm⁻². The output characteristics of the photodetector under different wavelengths and varying P are presented in Fig. S5(a)–(c). Significantly, the time-resolved self-driven photoresponse under different wavelengths is demonstrated in Fig. 5(b), confirming the capability of the device to operate at zero external bias. And stable self-driven photodetection lasting up to approximately 12.5 minutes is achieved under 532 nm illumination with 69 mW cm⁻² at 0 V as shown in Fig. S6. As shown in Fig. 5(c), the values of α are 0.46, 0.59, and 0.32 for wavelengths of 405 nm, 532 nm, and 604 nm, respectively, suggesting the presence of trap-assisted recombination processes. Then, when P equals 1 mW cm⁻², R can reach the values of 8.07, 7.55, and 18.40 A W⁻¹ for wavelengths of 405 nm, 532 nm, and 604 nm, respectively. Correspondingly, EQE attains values of 2871%, 1759%, and 3778%, while D* can reach the values of 3.21 × 10¹¹ Jones, 3.38 × 10¹¹ Jones, and 8.74 × 10¹¹ Jones for the three wavelengths, respectively. The dependence of R, EQE, and D* on P under different wavelengths is summarized in Fig. S5(d)–(r). In order to evaluate the repeatability and reliability of our devices, several devices have been analyzed. The R values under different wavelengths for different devices are shown in Fig. 5d. The EQE and D* values under different wavelengths for different devices are shown in Fig. S7. In addition, the fast self-driven photoresponse characteristics, including rise/decay time under different wavelengths, are shown in Fig. S8, further demonstrating the broadband photoresponse performance of the GeSe/Si vdW heterojunction photodetectors.

Fig. 5 Broadband photoresponse of the GeSe/Si vdW heterojunction photodetector. (a) The output curves with laser irradiation of 405 nm, 532 nm, and 604 nm lasers. (b) Time-resolved photodetection under the switch on/off irradiation of 405 nm, 532 nm, and 604 nm lasers at 0 V. (c) Photocurrent of the photodetector as a function of the 405 nm, 532 nm, and 604 nm laser power intensities at 0 V. (d) R, EQE, and D* as a function of the laser wavelength.

Furthermore, the device exhibits an obvious photovoltaic behavior. The open-circuit voltage (V_oc) and short-circuit current (I_sc) were extracted and plotted with the varying laser power intensity. As shown in Fig. 6(a), V_oc shows an increase from 0.065 V to 0.158 V as P increases from 1 mW cm⁻² to 69 mW cm⁻². I_sc increased linearly as P increased, and a similar analysis has been done with I_ph. Then, the output electrical power (P_el), which is defined as P_el = I_ds × V_ds, can be extracted from I–V curves shown in Fig. 3(b).⁴⁹ As shown in Fig. 6(b), P_el increased obviously with increasing P. And a highest P_el value of 124.6 nW was realized. The other two important parameters of the photovoltaic effect are the fill factor (FF) and photoelectric conversion efficiency (PCE). The FF is defined as P_el,m/I_scV_oc, where P_el,m is the maximum output electrical power.⁴⁹ P_el,m represented by an orange rectangle is 0.27 at 3.2 mW cm⁻² in Fig. 6(c). PCE can be calculated using P_el,m/P_i,⁴⁹ and a value of 0.2% was obtained at 3.2 mW cm⁻². The calculated FF and PCE versus P are presented in Fig. 6(d), The decreasing trend in the FF with increasing P is attributed to reduced shunt resistance and increased recombination.⁵⁰ The photovoltaic behavior of the GeSe/Si vdW heterojunction photodetector under different wavelengths is further evaluated in Fig. S9.

Fig. 6 Photovoltaic properties of the GeSe/Si vdW photodetector under 532 nm irradiation. (a) I_sc and V_oc as a function of laser power intensity. (b) The electrical power as a function of V_ds at various laser power intensities. (c) The FF and PCE in the output curve under 532 nm laser irradiation with 3.2 mW cm⁻². (d) FF and PCE under different laser power intensities.

The high performance of GeSe/Si vdW heterojunction photodetectors can mainly be attributed to a high-quality interface originating from the vdW heterojunction, type-II energy band structure of the heterojunction, and built-in electric field, leading to a self-driven and fast photoresponse. The performance comparison between our GeSe/Si vdW heterojunction photodetectors and other reported devices is shown in Table 1. Since D* was calculated in different literature studies by different methods due to the experimental conditions, we marked the results obtained using D* = S^1/2R/(2qI_D)^1/2 as (I), and the results obtained using D* = R(S × Δf)^1/2/i_n as (II).

Table 1 Performance comparison between the GeSe/Si vdW heterojunction photodetector and other devices reported in the literature. The calculation of D* using D* = S^1/2R/(2qI_D)^1/2 is marked as (I) and using D* = R(S × Δf)^1/2/i_n is marked as (II)

Heterojunction	Rectification ratio	Self-driven	Wavelength (nm)	R (mA W^−1)	D* (Jones)	Rise/decay time	Ref.
GeSe/WS2/MoS2	10^2	Yes	405	13	7.5 × 10^8 (I)	2.4/5.2 ms	51
GeSe/VP	26	Yes	435	5	10^8 (I)	166/287 ms	52
InSe/Si	5 × 10^3	Yes	400	37.3	—	32/60 ms	53
WS2/Si	10^2	Yes	980	224	1.5 × 10^12 (I)	16/29 µs	54
T-MoSe2/GeSe/B-MoSe2	1.1 × 10^4	Yes	638	206	6.6 × 10^10 (II)	5.19/1.32 ms	55
MoTe2/Si	10^4	No	800	521.1	1.17 × 10^11 (I)	4/4 ms	41
GeSe/SnSe	10^2	Yes	808	0.12	5.44 × 10^6 (I)	23/61 ms	27
GeSe/SnS2	10^2	No (1 V)	1064	50.4 × 10^3	1.09 × 10^10 (I)	2.1/3 ms	24
GeSe/MoTe2	63	No (−5 V)	680	3.3 × 10^3	6.7 × 10^8 (I)	3/12 µs	25
GeSe/MoTe2	300	Yes	808	51.4	4.14 × 10^11 (I)	80/26 ms	26
GeSe/MoS2	10	Yes	532	105	1.46 × 10^10 (I)	110/750 ms	28
ReS2/Si	10^3	Yes	White	0.1	1 × 10^8 (I)	300/500 µs	56
CNT/Si	10^3	Yes	950	370	—	—	57
Graphene/Si	10^3	No (−2.5 V)	White	2.5 × 10^3	—	—	58
GeSe/Si	60	Yes	405	8.07 × 10^3	3.2 × 10^11 (I)	16/41 µs	This work
			532	29.8 × 10^3	2.1 × 10^12 (I)	8.5/23.7 µs
			604	18.4 × 10^3	8.7 × 10^11 (I)	19/50 µs

---

4. Conclusion

In conclusion, a high-performance self-driven photodetector based on a vertical GeSe/Si vdW heterojunction has been demonstrated by integrating high-quality GeSe nanoflakes with silicon. Owing to the intrinsic type-II band alignment and the resulting built-in electric field, the device enables efficient carrier separation at zero bias, leading to a high R of 29.8 A W⁻¹, a high EQE of 6959.7%, a high D* of 2.1 × 10¹² Jones, and a fast rise/decay time of 8.5 µs/23.7 µs under 532 nm laser illumination. In addition, the GeSe/Si vdW heterojunction exhibits pronounced photovoltaic behavior and a broadband photoresponse under visible-light illumination. This work demonstrates that vdW integration of 2D GeSe with silicon is an effective strategy for realizing self-driven and high-performance photodetectors.

Conflicts of interest

There are no conflicts to declare.

Data availability

Supplementary information (SI): the calculation of interplanar spacing, the EDS spectra of the GeSe nanoflakes, the electrical properties of the GeSe FET, the linear relationship of I–V curves between electrodes and materials, and the broadband photoresponse and photovoltaic behavior of the GeSe/Si vdW heterojunction photodetector. See DOI: https://doi.org/10.1039/d6nh00050a.

Acknowledgements

This work was supported by the Natural Science Research Project of Anhui Educational Committee (2024AH051840, 2023AH051669, and 2023AH040232), the Scientific Research Foundation for High-level Talents of Tongling University (2023tlxyrc18 and 2022tlxyrc35), the University Synergy Innovation Program of Anhui Province (GXXT-2022-022), the Natural Science Foundation of Shandong Province (ZR2024QF213), the Science Foundation for the Youth Scholars of Anhui University of Technology (QD202307 and QZ202422), and the Key Research and Development Program of Zhejiang Province, China (2021C01002).

References

1. C. Song, G. Noh, T. S. Kim, M. Kang, H. Song, A. Ham, M.-K. Jo, S. Cho, H.-J. Chai, S. R. Cho, K. Cho, J. Park, S. Song, I. Song, S. Bang, J. Y. Kwak and K. Kang, ACS Nano, 2020, 14, 16266–16300.
2. J. Miao and C. Wang, Nano Res., 2021, 14, 1878–1888.
3. J. Wang, J. Han, X. Chen and X. Wang, InfoMat, 2019, 1, 33–53.
4. Y. Lv, W. Qin, C. Wang, L. Liao and X. Liu, Adv. Electron. Mater., 2019, 5, 1800569.
5. J. Jiang, L. Xu, C. Qiu and L.-M. Peng, Nature, 2023, 616, 470–475.
6. H. Yang, S. O. Valenzuela, M. Chshiev, S. Couet, B. Dieny, B. Dlubak, A. Fert, K. Garello, M. Jamet, D.-E. Jeong, K. Lee, T. Lee, M.-B. Martin, G. S. Kar, P. Sénéor, H.-J. Shin and S. Roche, Nature, 2022, 606, 663–673.
7. P. Singh, S. Baek, H. H. Yoo, J. Niu, J.-H. Park and S. Lee, ACS Nano, 2022, 16, 5418–5426.
8. S. Hwangbo, L. Hu, A. T. Hoang, J. Y. Choi and J.-H. Ahn, Nat. Nanotechnol., 2022, 17, 500–506.
9. N. Gupta, H. Kim, N. S. Azar, S. Z. Uddin, D.-H. Lien, K. B. Crozier and A. Javey, Nano Lett., 2022, 22, 1294–1301.
10. N. Huo and G. Konstantatos, Adv. Mater., 2018, 30, 1801164.
11. D. Wu, C. Guo, L. Zeng, X. Ren, Z. Shi, L. Wen, Q. Chen, M. Zhang, X. J. Li, C.-X. Shan and J. Jie, Light: Sci. Appl., 2023, 12, 5.
12. A. Dodda, D. Jayachandran, A. Pannone, N. Trainor, S. P. Stepanoff, M. A. Steves, S. S. Radhakrishnan, S. Bachu, C. W. Ordonez, J. R. Shallenberger, J. M. Redwing, K. L. Knappenberger, D. E. Wolfe and S. Das, Nat. Mater., 2022, 21, 1379–1387.
13. O. Lopez-Sanchez, D. Lembke, M. Kayci, A. Radenovic and A. Kis, Nat. Nanotechnol., 2013, 8, 497–501.
14. M. Buscema, D. J. Groenendijk, S. I. Blanter, G. A. Steele, H. S. J. van der Zant and A. Castellanos-Gomez, Nano Lett., 2014, 14, 3347–3352.
15. S.-J. Liang, B. Cheng, X. Cui and F. Miao, Adv. Mater., 2020, 32, 1903800.
16. Y. Liu, N. O. Weiss, X. Duan, H.-C. Cheng, Y. Huang and X. Duan, Nat. Rev. Mater., 2016, 1, 16042.
17. H. Qiao, Z. Huang, X. Ren, S. Liu, Y. Zhang, X. Qi and H. Zhang, Adv. Opt. Mater., 2020, 8, 1900765.
18. R. Chen, H. Shi, Y. Liu, M. Wang, D. Yuan, J. Shu, M. A. S. Akash, M. Tian, Z. Hu, J. Xue, H.-Q. Zhao, F. Li and N. Wan, ACS Appl. Mater. Interfaces, 2025, 17, 10158–10167.
19. X. Song, W. Zhou, X. Liu, Y. Gu and S. Zhang, Phys. B, 2017, 519, 90–94.
20. X. Zhou, X. Hu, B. Jin, J. Yu, K. Liu, H. Li and T. Zhai, Adv. Sci., 2018, 5, 1800478.
21. P. Ramasamy, D. Kwak, D.-H. Lim, H.-S. Ra and J.-S. Lee, J. Mater. Chem. C, 2016, 4, 479–485.
22. D.-J. Xue, J. Tan, J.-S. Hu, W. Hu, Y.-G. Guo and L.-J. Wan, Adv. Mater., 2012, 24, 4528–4533.
23. B. Mukherjee, Y. Cai, H. R. Tan, Y. P. Feng, E. S. Tok and C. H. Sow, ACS Appl. Mater. Interfaces, 2013, 5, 9594–9604.
24. L. Yuan, Z. Xu, J. Li, F. Zhang, S. Liu, H. Shi, Q. Xia and M. Zhong, Appl. Phys. Lett., 2023, 122(24), 241106.
25. P. Chen, L. Pi, Z. Li, H. Wang, X. Xu, D. Li, X. Zhou and T. Zhai, Appl. Phys. Lett., 2022, 121(2), 021103.
26. N. Zhang, L. Wu, W. Gao, Q. Zhao, N. Huo and J. Li, Adv. Mater. Interfaces, 2022, 9, 2200150.
27. Y. Mao, T. Deng, Y. Li and F. He, Appl. Phys. Lett., 2024, 124(18), 181106.
28. Y. Xin, X. Wang, Z. Chen, D. Weller, Y. Wang, L. Shi, X. Ma, C. Ding, W. Li, S. Guo and R. Liu, ACS Appl. Mater. Interfaces, 2020, 12, 15406–15413.
29. C. Liu, J. Guo, L. Yu, J. Li, M. Zhang, H. Li, Y. Shi and D. Dai, Light: Sci. Appl., 2021, 10, 123.
30. M. Xu, Y. Wang, J. Liu and D. Yang, Sci. China Inf. Sci., 2024, 67, 160401.
31. Y. Wang, K. Ding, B. Sun, S.-T. Lee and J. Jie, Nano Res., 2016, 9, 72–93.
32. D. Jariwala, T. J. Marks and M. C. Hersam, Nat. Mater., 2017, 16, 170–181.
33. F. Yan, L. Zhao, A. Patanè, P. Hu, X. Wei, W. Luo, D. Zhang, Q. Lv, Q. Feng, C. Shen, K. Chang, L. Eaves and K. Wang, Nanotechnology, 2017, 28, 27LT01.
34. J. Wang, H. Xu, D. He, H. Wang, Y. Kang, Y. Pei, C. Jiang, W. Li and X. Xiao, Adv. Electron. Mater., 2023, 9, 2300177.
35. D. Wu, J. Guo, C. Wang, X. Ren, Y. Chen, P. Lin, L. Zeng, Z. Shi, X. J. Li, C.-X. Shan and J. Jie, ACS Nano, 2021, 15, 10119–10129.
36. P. Luo, F. Wang, J. Qu, K. Liu, X. Hu, K. Liu and T. Zhai, Adv. Funct. Mater., 2021, 31, 2008351.
37. Y. Wang, J. Chen, T. Hu, Y. Huang, W. Zhu, W. Li, Y. Hu, Z. Wei, Z. Fan, L. Zhao and K. Wang, Small, 2025, 21, 2407473.
38. B. Yang, W. Gao, H. Li, P. Gao, M. Yang, Y. Pan, C. Wang, Y. Yang, N. Huo, Z. Zheng and J. Li, Nanoscale, 2023, 15, 3520–3531.
39. G. H. Shin, C. Park, K. J. Lee, H. J. Jin and S.-Y. Choi, Nano Lett., 2020, 20, 5741–5748.
40. Z. Cao, K. Ni, J. Zhou, K. V. Larionov, S. Gao, C. Wu, D. Sun, Y. Yang, X. Zhu, P. B. Sorokin, L. Li, M. Tian, Y. Jiang, W. Gao and X. Liu, Adv. Opt. Mater., 2025, 13, 2500340.
41. D. Wang, X. Wang, Y. Pan, J. Bian, K. Liu, J. Guo, J. Lin, Z. Sun, S. Gou, C. Sheng, X. Dong, H. Su, Y. Zhu, Q. Sun, Z. Xu, A. Guo, L. Shao, H. Chen and W. Bao, Adv. Opt. Mater., 2024, 12, 2400547.
42. H. Fang and W. Hu, Adv. Sci., 2017, 4, 1700323.
43. X. Xu, J. Chen, H. Wu, D. Guo, J. Li, S. Zhang, Y. Xiao, K. Deng, T. He, H. Wang, Z. Wang, F. Wang, F. Zhong, P. Wang, Q. Li and W. Hu, ACS Nano, 2025, 19, 18747–18756.
44. Y. Fang, A. Armin, P. Meredith and J. Huang, Nat. Photonics, 2019, 13, 1–4.
45. T. Dai, Y. Zhang, F. Wang, Q. Dong, H. Ma, S. Zhu, S. Fang, J. Chang and Q. J. Wang, Adv. Opt. Mater., 2025, 13, e01297.
46. J. Ahn, J.-H. Kyhm, H. K. Kang, N. Kwon, H.-K. Kim, S. Park and D. K. Hwang, ACS Photonics, 2021, 8, 2650–2658.
47. T. Zheng, M. Yang, Y. Pan, Z. Zheng, Y. Sun, L. Li, N. Huo, D. Luo, W. Gao and J. Li, ACS Appl. Mater. Interfaces, 2023, 15, 29363–29374.
48. L. Hao, Z. Wang, H. Xu, K. Yan, S. Dong, H. Liu, Y. Du, Y. Wu, Y. Liu and M. Dong, 2D Mater., 2019, 6, 034004.
49. S. Zhang, Y. Ye, H. Li, X. Kan, S. Wang, S. Sun, T. Han, F. Li, L. Shan and M. Long, Adv. Opt. Mater., 2025, 13, 2500127.
50. J. Ma, S. Chen, L. Zhao, J. Chen, Z. Lan, M. Yang, Y. Sun, Z. Zheng, W. Gao and J. Li, Adv. Opt. Mater., 2024, 12, 2302563.
51. Y. Zhang and X. He, Sens. Actuators, A, 2025, 381, 116080.
52. S. Lv, C. Feng, G. Fu and X. Zhuang, ACS Appl. Electron. Mater., 2024, 6, 5766–5772.
53. X. Yang, Z. Liu, F. Gao, S. Zhang, H. Shang, Y. Hu, Y. Zhang, Z. Fu, Y. Huang, W. Feng and P. Hu, ACS Appl. Nano Mater., 2021, 4, 12932–12936.
54. E. Wu, D. Wu, C. Jia, Y. Wang, H. Yuan, L. Zeng, T. Xu, Z. Shi, Y. Tian and X. Li, ACS Photonics, 2019, 6, 565–572.
55. K. An, Y. Pan, X. Rong, T. Zheng, L. Li, H. Sun, J. Zeng, Y. Sang, F. Huang, D. Yue, W. J. Yoo and H. Chen, Adv. Funct. Mater., 2024, 34, 2409331.
56. K. Intonti, A. Pelella, H. Neill, V. Patil, P. K. Hurley, L. Ansari, F. Gity and A. D. Bartolomeo, Appl. Phys. Lett., 2024, 125(17), 173505.
57. A. Pelella, D. Capista, M. Passacantando, E. Faella, A. Grillo, F. Giubileo, N. Martucciello and A. Di Bartolomeo, Adv. Electron. Mater., 2023, 9, 2200919.
58. G. Luongo, F. Giubileo, L. Genovese, L. Iemmo, N. Martucciello and A. D. Bartolomeo, Nanomaterials, 2017, 7, 158.

---