Home Journals Articles Updates About
CN
image
收藏切换
All-Optically Controlled Memristive Device Based on Cu2O/TiO2 Heterostructure Toward Neuromorphic Visual System
收藏切换
PDF
Jun Xie, Xuanyu Shan*, Ningbo Zou, Ya Lin*, Zhongqiang Wang*, Ye Tao, Xiaoning Zhao, Haiyang Xu, Yichun Liu
Research. Vol 8 Article ID 0580
Less
收藏切换
Research. Vol 8 Article ID 0580
• Research Article •
All-Optically Controlled Memristive Device Based on Cu2O/TiO2 Heterostructure Toward Neuromorphic Visual System
Full
Jun Xie, Xuanyu Shan*, Ningbo Zou, Ya Lin*, Zhongqiang Wang*, Ye Tao, Xiaoning Zhao, Haiyang Xu, Yichun Liu
Affiliations
  • Key Laboratory for UV Light-Emitting Materials and Technology (Ministry of Education), College of Physics, Northeast Normal University, Changchun, China.
Published: 2025-01-10 doi: 10.34133/research.0580
Outline
收藏切换

The optoelectronic memristor integrates the multifunctionalities of image sensing, storage, and processing, which has been considered as the leading candidate to construct novel neuromorphic visual system. In particular, memristive materials with all-optical modulation and complementary metal oxide semiconductor (CMOS) compatibility are highly desired for energy-efficient image perception. As a p-type oxide material, Cu2O exhibits outstanding theoretical photoelectric conversion efficiency and broadband photoresponse. In this work, an all-optically controlled memristor based on the Cu2O/TiO2/sodium alginate nanocomposite film is developed. Optical potentiation and depression behaviors have been implemented by utilizing visible (680 nm) and ultraviolet (350 nm) light. Furthermore, a 7 × 9 optoelectronic memristive array with satisfactory device variation and environment stability is constructed to emulate the image preprocessing function in biological retina. The random noise can be reduced effectively by utilizing bidirectional optical input. Beneficial from the image preprocessing function, the accuracy of handwritten digit classification increases more than 60%. Our work presents a pathway toward high-efficient neuromorphic visual system and promotes the development of artificial intelligence technology.

Jun Xie, Xuanyu Shan, Ningbo Zou, Ya Lin, Zhongqiang Wang, Ye Tao, Xiaoning Zhao, Haiyang Xu, Yichun Liu. All-Optically Controlled Memristive Device Based on Cu2O/TiO2 Heterostructure Toward Neuromorphic Visual System[J]. Research, 2025 , 8 (1) : 0580 . DOI: 10.34133/research.0580
Full Text
Less
Introduction
Less
The conventional machine vision is composed of image sensing, storage, and processing units, which features a physical separation from each other [1]. Thereinto, the analog-to-digital conversion and frequent data transmission between these components result in elevated speed latency and energy consumption [2–4]. In contrast, the human visual system (HVS) enables to process enormous image information simultaneously with high speed and low energy consumption, relying on sophisticated neural networks [5–7]. Inspired by the HVS, developing neuromorphic hardware with in-sensor computing architecture would provide an ideal platform for high-efficient image perception [8–11].
Optoelectronic memristor, as an emerging neuromorphic device, has been considered as promising candidate to construct neuromorphic visual system, due to the functional and structural resemblances to biological optical synapses [12–14]. In particular, several neuromorphic visual functions have been demonstrated in the field, including motion detection [15–18], pattern recognition [19–21], and image encryption [22]. In these results, the hybrid optical-electrical operations are usually indispensable to achieve reversible modulation of device conductance, which is unfavorable for real-time image processing with high efficiency [23]. Therefore, developing novel all-optical controlled memristor represents an alternative strategy to eliminate the hardware redundancy and high-power consumption in complicated optical-electrical operation [24–27]. Furthermore, the reversible modulation characteristics in optical manner established a promising paradigm to emulate the antagonism shunting function of bipolar cells in human retina [28]. On this basis, the image preprocessing function can be implemented by using the all-optical modulation characteristics, which is beneficial to enhance imaging quality and accelerate subsequent processing. For example, Li et al. [29] achieved gate-tunable positive/negative photoconductivity in PtS2/hBN/graphene floating gate memory and demonstrated the high-precision data classification. However, the current physical models and memristive materials for all-optically controlled synaptic devices are still quite limited. Therefore, exploring novel all-optically controlled memristor is highly desired for high-efficient neuromorphic visual perception.
Heterostructures with designable electronic interfaces enables to integrate various material advantages for high-performance optoelectronic behaviors and eliminate the complicated hardware circuits [30–34]. Herein, efficient separation of photogenerated electron-hole pairs and excellent light absorbance can be realized in the heterojunction structure [35–38]. In particular, zero-dimensional materials with ultra-high specific surface area provide abundant contact sites for efficient charge transfer at heterogeneous interface [39–43]. At present, constructing zero-dimensional heterojunction is the general strategy for efficient photocatalysis [44]. Inspired by these advances, the all zero-dimensional heterojunction provides a reference model for all-optically controlled memristor. As an intrinsic p-type semiconductor oxide, Cu2O exhibits a bandgap width of 2.0 eV, ensuring the excellent light absorbance in the ultraviolet (UV)-visible region [45]. On the other side, TiO2 with a wide band gap of ~3.2 eV is regarded as a leading candidate for future optoelectronic device, due to the outstanding properties of chemical stability and high carrier mobility [46–48]. As above, the Cu2O/TiO2 zero-dimensional heterojunction provides an excellent foundation for all-optical modulated memristor. However, to the best of our knowledge, the all-optically controlled memristor based on Cu2O/ TiO2 heterojunction has not been reported.
In this work, we demonstrate an all-optically controlled memristor based on Cu2O/TiO2 heterojunction, in which the conductance can be reversibly modulated by utilizing visible (680 nm) and UV (350 nm) light. Based on the optical potentiation/depression behaviors, versatile synaptic functions have been emulated, including excitatory/inhibitory post-synaptic currents (EPSP/IPSP), paired-pulse facilitation/depression (PPF/PPD), and long-term plasticity (LTP/LTD). Furthermore, image preprocessing function has been implemented to enhance feature information and suppress random noise. Then, we constructed a single-layer artificial neural network and demonstrated the classification of handwritten digits, showing recognition accuracy exceeding 90%. The proposed retinomorphic device provides a feasible strategy to develop high-efficient neuromorphic visual system.
Results
Less
Humans obtain more than 80% of external information through the visual system, which is the most essential sensory organs to perceiving their surroundings. Figure 1A illustrates the motivation for developing multifunctional optoelectronic memristor to emulate biological visual system [49]. As depicted in the schematic diagram of HVS, the photoreceptors in biological retina convert the light signals into electrical signals. Meanwhile, the preprocessing capabilities in retina enable to reduce image noise and filter out redundant information. Then, image information is transmitted to the visual cortex through optic nerve for high-level processing and memorization [50]. In order to emulate the human retina, we developed a TiO2/Cu2O heterojunction memristor with all-optical modulation characteristics. As shown in Fig. 1B, the nanocomposite film consisting of Cu2O–TiO2 nanoparticles and sodium alginate acts as functional layer. Herein, the Cu2O–TiO2 nanoparticles acts as functional unit for optoelectronic memristive behaviors. Meanwhile, the sodium alginate is the base materials for desirable mechanical flexibility. The cross-sectional scanning electron microscopy image indicates that the thickness of the Cu2O–TiO2 film is ~490 nm (Fig. 1C). The root mean square roughness values of the nanocomposite film is 2.61 nm, indicating excellent film smoothness and uniformity (Fig. S1). The optoelectronic memristive array can be attached to the hemispherical substrate without obvious bubbles or wrinkles, indicating a conformal contact (Fig. 1D). It is worth noting that the low temperature is necessary for the device fabrication process due to the flexible substrates [51]. The above nonplanar structure is beneficial to achieve a wide field of view. Meanwhile, the natural polysaccharide material (sodium alginate) contains abundant hydrophilic functional groups, which enables a simple solution fabrication of the nanocomposite film (Fig. 1E and Fig. S2) [52]. Figure 1F shows the transmission electron microscopy (TEM) image of TiO2 and Cu2O nanoparticles. Herein, the average diameters of TiO2 and Cu2O are 6.3 and 39.2 nm, respectively (Fig. S3). The lattice fringes with spacings of 0.254 nm and 0.353 nm correspond to (111) plane of cubic Cu2O and (101) plane of anatase TiO2 [53,54]. The above result has been further confirmed in the selected-area electron diffraction (SAED) and x-ray diffraction (XRD) pattern analyses (Fig. S4). The energy-dispersive spectroscopy indicates that the 2 oxide nanoparticles combined with each other, indicating the formation of zero-dimensional heterojunction. The UV-visible absorption spectra of TiO2 and Cu2O/TiO2 nanoparticles are depicted in Fig. 1G. It can be seen that TiO2 nanoparticles exhibit steep absorption in the UV region, corresponding to the TiO2 bandgap of 3.2 eV. In contrast, a broad optical absorption from UV to visible range can be observed in the Cu2O–TiO2 sample, which can be attributed to the narrow bandgap (2.0 eV) of Cu2O nanoparticles.
The optimization of material component and mechanism investigation are first performed in our Cu2O/TiO2 heterojunction device. As shown in Fig. 2A, the TiO2/sodium alginate device exhibits optical potentiation behaviors under the action of UV light (350 nm). Meanwhile, the optical depression behavior is absent with the illumination of visible light (680 nm). For the Cu2O/sodium alginate device, optical potentiation behaviors are observed under UV/visible light, due to the narrow bandgap, as shown in Fig. 2B. In contrast, both optical potentiation and depression response behaviors have been obtained in the Cu2O/TiO2 heterostructure device, i.e., all-optical modulation characteristics. As shown in Fig. 2C, the device exhibits a transient potentiation current under the irradiation of UV light. After the UV illumination is removed, the heterojunction device shows a stable conductance state lower than the initial state, i.e., optical depression behaviors. On the other side, visible light induces obvious enhancement in transient current, which can be partly maintained after the optical signal is removed. Furthermore, the heterojunction device exhibits a broad spectrum of UV-visible response, as plotted in Fig. S5. As the light wavelength increases from 350 to 680 nm, the photoresponse behaviors gradually switch from depression to potentiation. The optical signals of 680 and 350 nm are selected for subsequent investigation, due to the remarkable photo-induced enhancement/inhibition effect. In order to investigate the underlying mechanism, x-ray photoelectron spectroscopy (XPS) analysis is conducted. Compared with the pure TiO2 sample, both the Ti 2p1/2 and Ti 2p3/2 redshifted for 0.09 eV in Cu2O/TiO2 (Fig. 2D and E) [55]. The redshifted peak can be attributed to the electron transfer between Cu2O and TiO2, which confirms the formation of Cu2O/TiO2 heterojunction. Besides that, photoluminescence (PL) emission spectra show that the emission intensity of Cu2O/TiO2 is significantly lower than pure TiO2 [56]. The transient PL spectra in Fig. 2F and G also show shorter lifetime in the Cu2O–TiO2 sample. The above results indicate the suppressed recombination of photogenerated electron-hole pairs in the oxide heterojunction. On this basis, a general model is proposed to explain the all-optically controlled behaviors, as illustrated in Fig. 2H. Herein, Cu2O/TiO2 heterojunction with type-II band alignment is formed (Fig. S6). When the Cu2O–TiO2-based device is irradiated with visible light (680 nm), the photogenerated electron-hole pairs in Cu2O nanoparticles decrease the barrier width, resulting in the conductance increase. For TiO2 nanoparticles, the electrons in conduction band cannot be excited under visible irradiation, due to the bandgap of 3.2 eV. After the visible light is removed, part of photogenerated electrons will recombine with holes in Cu2O, which corresponds to spontaneous decay in photocurrent. On the other side, UV light induces photogenerated electron-hole pair in both Cu2O and TiO2, resulting in transient potentiation behaviors. When the UV irradiation is removed, the photogenerated electrons in the potential well efficiently recombine with holes generated from Cu2O segment by tunnel through or jump over the interface barrier. The above process will increase the barrier width. Hence, the device exhibits optical depression behaviors under the action of UV light.
All-optically controlled synaptic plasticity is performed to emulate the photoresponse characteristics of bipolar cells in biological retina (Fig. 3A). Herein, the device current is regarded as synaptic weight, which is monitored with a bias voltage of 0.03 V. The optimization of film thickness is depicted in Fig. S7. Figure 3B depicts the photoresponse current response of the heterojunction device under visible light (17.25 mW/cm2, 10 s). Transient current enhancement of ~0.3 nA can be observed, which can be maintained for more than 100 s. The optical potentiation behavior is similar to the excitatory postsynaptic current (EPSC) in biological synapse. The ultra-low photocurrent of our all-optically controlled device is beneficial to reduce power consumption (see more details in Table S1). The device exhibits ultralow power consumption of 0.78 pJ, when the illumination duration reduces to 100 ms (Fig. S8). Furthermore, stable optoelectronic memristive behaviors can be achieved even after placing the device in atmosphere environment for ~300 d (Fig. S9). In contrast, Fig. 3C shows the photoresponse behaviors of our Cu2O–TiO2-based device under UV irradiation (2.32 mW/cm2, 10 s). After spontaneous relaxation, the stable state shows a lower device current than the initial state, indicating a long-term depression behavior. Furthermore, PPF and PPD have been demonstrated by utilizing 2 consecutive spikes, which are the foundation to decode temporary information. Figure 3D and E shows response curves under the stimulation of 2 consecutive visible and UV spikes (internal of 5 s). The photoresponse current induced by the second spike (A 2) is obviously larger/smaller than that by the first one (A 1). The facilitation and depression effects are similar with the PPF and PPD function in biological synapse. The PPF/PPD index can be calculated with the equation below: PPF/PPD index = (A 2 − A 1)/A 1 × 100%. Herein, the time interval is the dominant factor for the PPF/PPD index. As the interval time increases from 5 to 60 s, the PPF index decreases from 15.87% to 1.37% and the PPD index decreases from –7.20% to –40.67% (Fig. 3F). The above indexes dependent on interval time can be fitted by the exponential function [57]
PPF / PPD = 1 + C 1 e − Δ t / τ 1 + C 2 e − Δ t / τ 2
where C 1 and C 2 are facilitation constants and τ 1 and τ 2 represent short-term and long-term relaxation time, respectively. Δt is the interval time between 2 consecutive optical spikes. The extracted τ 1 and τ 2 values of PPF/PPD are 70 ms/21 ms and 3.16 s/0.860 s, respectively. Besides the short-term plasticity, the long-term potentiation/depression behaviors can also be modulated with the pulse frequency (Fig. S10). The above results indicate that our all-optically controlled device exhibits evident temporal correlation, which is beneficial for high-efficient image processing.
The heterojunction device has also shown excellent sensitivity to irradiation duration and intensity, which enables real-time processing of complicated image information. In this section, the device current is measured after the UV/visible light is removed for 100 s. As shown in Fig. 3G and H, the device current can be modulated precisely with visible/UV duration and intensity. The long-term potentiation/depression behaviors of 0.756/0.073 nA can be achieved when the irradiation duration increases from 5 to 25 s. Similar potentiation and depression trend can be obtained by applying optical spikes with higher intensity. Furthermore, reversible conductance modulation has also been implemented by utilizing alternate 30 visible and UV spikes. The linearity (determination R2 in linear fits) of long-term potentiation and depression behaviors is evaluated as 0.886 and 0.995, respectively. The quasi-linear and symmetric conductance evolution is of great significance for high-precision image recognition and classification in artificial neural network.
Image preprocessing of biological retina is a critical capability to enhance feature information and suppress random noise, which promotes subsequent high-level image processing in visual cortex [58,59]. In order to emulate the preprocessing function in biological retina, we fabricated a 7 × 9 optoelectronic memristive array (Fig. 4A). The corresponding statistical result of initial current and EPSC values is plotted in Fig. S11. The negligible current fluctuation indicates satisfactory device uniformity. Each memristive unit in the array corresponds to an image pixel. The gray level (from 0 to 255) of each pixel represents the maximum and minimum device current in the array. As shown in Fig. 4B, random noise has been introduced to the ideal image, emulating the inevitable influence of complicated environment. For the unidirectional input, the digital pattern of “3” is irradiated with visible light, while the background represents the dark state. As shown in Fig. 4C, 5 different input images with random noise are sequentially written into the optoelectronic memristor array. The output result of each pixel was recorded after optical input is paused for 100 s. Corresponding output images are plotted in the bottom panel of Fig. 4C. The output image with unidirectional input exhibits pattern “3” with non-negligible image noise. In contrast, we have also performed the image sensing and preprocessing function with bidirectional input (Fig. 4D). For the bidirectional image input, the digital pattern of “3” was written with visible spikes and the background was irradiated with UV light. It is worth noting that the random noise can be effectively suppressed by utilizing UV light in the background. The criteria of structural similarity (SSIM) have been employed to evaluate the noise suppression efficiency (see more details in Materials and Methods). A higher SSIM between preprocessing and ideal image represents better noise suppression process. The SSIM of bidirectional and unidirectional input is 0.8425 and 0.4431, respectively. The output image with bidirectional input has high degree of similarity to the ideal input. As above, our optoelectronic memristive array with bidirectional optical modulation has demonstrated hardware implementation of real-time image preprocessing, which improves the image qualities obtained in non-ideal surroundings.
Besides the image preprocessing, high-level image processing is also highly required in neuromorphic visual system, including pattern recognition and object classification [60]. In this section, we developed an artificial neural network to perform the classification of handwritten digits using the Modified National Institute of Standards and Technology (MNIST) dataset. As shown in Fig. 5A, the artificial neural network is composed of 784 input neurons, 300 hidden neurons, and 10 output neurons. The schematic diagram of memristive crossbar array is demonstrated in Fig. 5B. The synaptic weight is updated by utilizing the backpropagation (BP) algorithm, which follows the long-term potentiation/depression behaviors in Fig. 3I. Figure 5C and D demonstrates the cumulative distribution function (CDF) of current variation at different current level for potentiation and depression behaviors. The uniform current variation indicates excellent linearity of conductance evolution in our heterojunction device. Then, we prepare training dataset of handwritten digital images with 28 × 28 pixel to train the artificial neural network. As shown in Fig. 5E, the recognition rates of our artificial neural network exceeded 90.0% after 10 training epochs. In contrast, the recognition accuracy without bidirectional preprocessing stabilizes at 31.7% after training. The classification output of handwritten digits is displayed in confusion matrix of Fig. 5F and G. The statistics values in matrix diagonals represent the correct classification of each digit, in which the predicted label is consistent with true label. In can be seen that only 178 digits among the 1797 total samples are mismatched, indicating the high-accuracy classification capability of our all-optically controlled memristor. On the other side, the classification error improves significantly when the bidirectional preprocessing is absent, in which 1197 of 1797 digits (from “0” to “9”) have been classified in error. The quantitative analysis demonstrates that image noise in unidirectional input has a strong impact on classification accuracy. As above, the proposed all-optically controlled device provides an ideal platform to improve imaging qualities for high-performance neuromorphic visual perception.
Conclusion
Less
In this work, we developed an all-optically controlled synaptic device with the structure of Au/Cu2O/TiO2/sodium alginate/indium tin oxide. Optical potentiation and depression behaviors can be performed in the proposed device by utilizing visible (680 nm) and UV (350 nm) stimulus, respectively. Several basic synaptic functions have been emulated with optical signals, including excitatory/inhibitory postsynaptic current and paired pulse facilitation and depression. Besides that, our device exhibits the negligible current fluctuation and stable photoresponse behaviors for ~300 d in atmosphere environment. Furthermore, image preprocessing function has been implemented by using the reversible all-optical modulation characteristics. Compared with the unidirectional signals, the bidirectional input with visible and UV light enables to suppress image noise and enhance feature information. On this basis, the accuracy of handwritten digit classification has exceeded 90% in our simulated artificial neural network.
Materials and Methods
Less
Device fabrication
All-optically controlled memristor based on Cu2O/TiO2/sodium alginate nanocomposite was fabricated on the ITO substrate. First, the Cu2O nanoparticles are synthesized by dissolving CuCl2·2H2O (0.852 g) and NaOH (1.20 g) in deionized water and stirring for 10 min at room temperature. Then, l-ascorbic acid (1.760 g) was added into the above solution. The final product was washed with ethyl alcohol and dried in vacuum environment. The Cu2O/TiO2 dispersion liquid was prepared by adding Cu2O nanoparticles (0.03 g), TiO2 nanoparticles (0.03 g), and sodium alginate in deionized water and stirring for 10 h. The mass fraction of TiO2, Cu2O, and sodium alginate is 13.3, 20.0, and 66.7 wt %, respectively. The Cu2O–TiO2/sodium alginate nanocomposite film was fabricated with the spin-coating method. Finally, the top Au electrodes of 500 μm were deposited by sputtering. SSIM can be expressed as follows [61]:
SSIM ( x , y ) = f ( l ( x , y ) , c ( x , y ) , s ( x , y ) )
where l(x, y), c(x, y), and s(x, y) represent the image brightness, contrast, and structure, respectively.
Measurement and characterization
The photocurrent change is monitored by using a semiconductor parameter analyzer (Keithley 2636A) and probe station (TTPX, Lake Shore). Optical signals were performed with a xenon lamp (LA-410UV, Hayashi). All the measurements were performed in ambient atmosphere and room temperature.
Funding
Less
  • NSFC for Distinguished Young Scholar(No. 52025022)
  • the NSFC Program(U23A20568)
  • the NSFC Program(52372137)
  • the NSFC Program(52272140)
  • the NSFC Program(52072065)
  • the China Postdoctoral Science Foundation(GZB20240135)
  • the “111” Project(B13013)
  • the fund from Jilin Province(20220502002GH)
  • the fund from Jilin Province(20230402072GH)
  • The Fundamental Research Funds for the Central Universities(2412023YQ004)
References
Less
1.
Zou XQ, Xu S, Chen XM, Yan L, Han YH. Breaking the von Neumann bottleneck: Architecture-level processing-in-memory technology. Sci China Inf Sci. 2021;64(6): Article 160404.
2.
Choi M, Bae SR, Hu L, Hoang AT, Kim SY, Ahn JH. Full-color active-matrix organic light-emitting diode display on human skin based on a large-area MoS2 backplane. Sci Adv. 2020;6(28):eabb5898.
3.
Choi S, Kang CM, Byun CW, Cho H, Kwon BH, Han JH, Yang JH, Shin JW, Hwang CS, Cho NS, et al. Thin-film transistor-driven vertically stacked full-color organic light-emitting diodes for high-resolution active-matrix displays. Nat Commun. 2020;11(1):2732.
4.
Yang Y. Multi-tier computing networks for intelligent IoT. Nat Electron. 2019;2(1):4–5.
5.
Akbari MK, Zhuiykov S. A bioinspired optoelectronically engineered artificial neurorobotics device with sensorimotor functionalities. Nat Commun. 2019;10:3873.
6.
Meng Y, Li FZ, Lan CY, Lan CY, Bu XM, Kang XL, Wei RJ, Yip S, Li DP, Wang F. Artificial visual systems enabled by quasi-two-dimensional electron gases in oxide superlattice nanowires. Sci Adv. 2020;6(46):eabc6389.
7.
Wang SY, Chen CS, Yu ZH, He YL, Chen XY, Wan Q, Shi Y, Zhang DW, Zhou H, Wang XR, et al. A MoS2/PTCDA hybrid heterojunction synapse with efficient photoelectric dual modulation and versatility. Adv Mater. 2019;31(3):1806227.
8.
Li G, Xie DG, Zhang ZY, Zhou QL, Zhong H, Ni H, Wang JO, Guo EJ, He M, Wang C, et al. Flexible VO2 films for in-sensor computing with ultraviolet light. Adv Funct Mater. 2022;32(29):2203074.
9.
Chen X, Yang DL, Hwang G, Dong Y, Cui BB, Wang DC, Chen HG, Lin N, Zhang WQ, Li HH, et al. Oscillatory neural network-based Ising machine using 2D memristors. ACS Nano. 2024;18(16):10758–10767.
10.
Pereira ME, Martins R, Fortunato E, Barquinha P, Kiazadeh A. Recent progress in optoelectronic memristors for neuromorphic and in-memory computation. Neuromorph Comput Eng. 2023;3(2): Article 022002.
11.
Pereira ME, Deuermeier J, Martins R, Barquinha P, Kiazadeh A. Unlocking neuromorphic vision: Advancements in IGZO-based optoelectronic memristors with visible range sensitivity. ACS Appl Electron Mater. 2024;6(7):5230–5243.
12.
Qian C, Oh S, Choi Y, Kim JH, Sun J, Huang H, Yang JL, Gao YL, Park JH, Cho JH. Solar-stimulated optoelectronic synapse based on organic heterojunction with linearly potentiated synaptic weight for neuromorphic computing. Nano Energy. 2019;66: Article 104095.
13.
Seo S, Jo SH, Kim S, Shim J, Oh S, Kim JH, Heo K, Choi JW, Choi C, Oh S, et al. Artificial optic-neural synapse for colored and color-mixed pattern recognition. Nat Commun. 2018;9(1):5106.
14.
Sun J, Oh S, Choi Y, Seo S, Oh MJ, Lee M, Lee WB, Yoo PJ, Cho JH, Park JH. Optoelectronic synapse based on IGZO-alkylated graphene oxide hybrid structure. Adv Funct Mater. 2018;28(47):1804397.
15.
Zhang ZH, Wang SY, Liu CS, Xie RZ, Hu WD, Zhou P. All-in-one two-dimensional retinomorphic hardware device for motion detection and recognition. Nat Nanotechnol. 2022;17(1):27–32.
16.
Ni Y, Feng JL, Liu JQ, Yu H, Wei H, Du Y, Liu L, Sun L, Zhou JL, Xu WT. An artificial nerve capable of UV-perception, NIR-Vis switchable plasticity modulation, and motion state monitoring. Adv Sci. 2022;9(1):2102036.
17.
Pan X, Shi JW, Wang PF, Wang S, Pan C, Yu WT, Cheng B, Liang SJ, Miao F. Parallel perception of visual motion using light-tunable memory matrix. Sci Adv. 2023;9(39):eadi4083.
18.
Chen JW, Zhou Z, Kim BJ, Zhou Y, Wang ZQ, Wan TQ, Yan JM, Kang JF, Ahn JH, Chai Y. Optoelectronic graded neurons for bioinspired in-sensor motion perception. Nat Nanotechnol. 2023;18(8):882–888.
19.
Chen Y, Zhang M, Li DW, Tang YJ, Ren HH, Li JY, Liang K, Wang Y, Wen LY, Li WB, et al. Bidirectional synaptic phototransistor based on two-dimensional ferroelectric semiconductor for mixed color pattern recognition. ACS Nano. 2023;17(13):12499–12509.
20.
Ahmed T, Tahir M, Low MX, Ren YY, Tawfik SA, Mayes ELH, Kuriakose S, Nawaz S, Spencer MJS, Chen H, et al. Fully light-controlled memory and neuromorphic computation in layered black phosphorus. Adv Mater. 2021;33(10):2004207.
21.
Jo C, Kim J, Kwak JY, Kwon SM, Park JB, Kim J, Park GS, Kim MG, Kim YH, Park SK. Retina-inspired color-cognitive learning via chromatically controllable mixed quantum dot synaptic transistor arrays. Adv Mater. 2022;34(12):2108979.
22.
Han JQ, Shan XY, Lin Y, Tao Y, Zhao XN, Wang ZQ, Xu HY, Liu YC. Multi-wavelength-recognizable memristive devices via surface plasmon resonance effect for color visual system. Small. 2023;19(23):2207928.
23.
Sun QH, Guo ZC, Zhu XJ, Jiang Q, Liu HY, Liu XR, Sun C, Zhang YJ, Wu L, Li RW. Optogenetics-inspired manipulation of synaptic memory using all-optically controlled memristors. Nanoscale. 2023;15(23):10050–10056.
24.
Hu LX, Yang J, Wang JR, Cheng PH, Chua LO, Zhuge F. All-optically controlled memristor for optoelectronic neuromorphic computing. Adv Funct Mater. 2021;31(4):2005582.
25.
Li DW, Ren HH, Chen YT, Tang YJ, Liang K, Wang Y, Li FF, Liu GL, Meng L, Zhu BW. Bidirectionally photoresponsive optoelectronic transistors with dual photogates for all-optical-configured neuromorphic vision. Adv Funct Mater. 2023;33(42):2303198.
26.
Shan XY, Zhao CY, Wang XN, Wang ZQ, Fu SC, Lin Y, Zeng T, Zhao XN, Xu HY, Zhang XT, et al. Plasmonic optoelectronic memristor enabling fully light-modulated synaptic plasticity for neuromorphic vision. Adv Sci. 2022;9(6):2104632.
27.
Cai BQ, Huang Y, Tang LZ, Wang TY, Wang C, Sun QQ, Zhang DW, Chen L. All-optically controlled retinomorphic memristor for image processing and stabilization. Adv Funct Mater. 2023;33(46):2306272.
28.
Fahey PK, Burkhardt DA. Center-surround organization in bipolar cells: Symmetry for opposing contrasts. Vis Neurosci. 2003;20(1):1–10.
29.
Li C, Chen X, Zhang Z, Wu XS, Yu TZ, Bie RT, Yang DL, Yao YG, Wang ZR, Sun LF. Charge-selective 2D heterointerface-driven multifunctional floating gate memory for in situ sensing-memory-computing. Nano Lett. 2024;24(47):15025.
30.
Liang K, Wang R, Ren HH, Li DW, Tang YJ, Wang Y, Chen YT, Song CY, Li FF, Liu GL, et al. Printable coffee-ring structures for highly uniform all-oxide optoelectronic synaptic transistors. Adv Opt Mater. 2022;10(24):2201754.
31.
Sun J, Choi Y, Choi YJ, Kim S, Park JH, Lee S, Cho JH. 2D-organic hybrid heterostructures for optoelectronic applications. Adv Mater. 2019;31(34):1803831.
32.
Fan C, Dai BB, Liang HK, Xu X, Qi ZD, Jiang HT, Duan HG, Zhang QL. Epitaxial growth of 2D Bi2O2Se nanoplates/1D CsPbBr3 nanowires mixed-dimensional heterostructures with enhanced optoelectronic properties. Adv Funct Mater. 2021;31(16):2010263.
33.
Cheng XH, Han Y, Cui BB. Fabrication strategies and optoelectronic applications of perovskite heterostructures. Adv Opt Mater. 2022;10(5):2102224.
34.
Haimeur AE, Hammi M, Sánchez PF, Bakkali H, Blanco EO, Ouannou A, Laazizi A, de la Vega MD, Nouneh K, Echchelh A. Tuning the TiO2/ZnO heterostructures emissions through nickel doping for intriguing optoelectronic and photonic applications. Opt Quant Electron. 2023;55(13):1190.
35.
Zhou X, Wu J, Li QF, Zeng T, Ji Z, He P, Pan WG, Qi XM, Wang CY, Liang PK. Carbon decorated In2O3/TiO2 heterostructures with enhanced visible light-driven photocatalytic activity. J Catal. 2017;355:26–39.
36.
Lang JY, Li CY, Wang SW, Lv JJ, Su YG, Wang XJ, Li GS. Coupled heterojunction Sn2Ta2O7@SnO2: Cooperative promotion of effective electron hole separation and superior visible-light absorption. ACS Appl Mater Interfaces. 2015;7(25):13905–13914.
37.
Ravindar PT, Choppella VS, Mokshagundam AK, Kiruba M, Babu SG, Babu KR, Berchmans LJ, Sreedhar G. Enhanced visible-light-driven photocatalysis of Bi2YO4Cl heterostructures functionallized by bimetallic RhNi nanoparticles. Front Mater Sci. 2018;12(4):405–414.
38.
Qin JK, Yan H, Qiu G, Si MW, Miao P, Duan YQ, Shao WZ, Zhen L, Xu CY, Ye PDD. Hybrid dual-channel phototransistor based on 1D t-se and 2D ReS2 mixed-dimensional heterostructures. Nano Res. 2019;12(3):669–674.
39.
Zhou YJ, Liao F, Liu Y, Kang ZH. The advanced multi-functional carbon dots in photoelectrochemistry based energy conversion. Int J Extrem Manuf. 2022;4(4): Article 042001.
40.
Tian L, Li Z, Wang P, Zhai XH, Wang X, Li TX. Carbon quantum dots for advanced electrocatalysis. J Energy Chem. 2021;55:279–294.
41.
Huo PP, Zhao P, Wang Y, Liu B, Dong MD. An effective utilization of solar energy: Enhanced photodegradation efficiency of TiO2/graphene-based composite. Energies. 2018;11(3):630.
42.
Sreedhar A, Ta QTH, Noh JS. Role of p-n junction initiated mixed-dimensional 0D/2D, 1D/2D, and 2D/2D BiOX (X = cl, Br, and I)/TiO2 nanocomposite interfaces for environmental remediation application: A review. Chemosphere. 2022;305: Article 135478.
43.
Jiang YH, Jing X, Zhu K, Peng ZY, Zhang JM, Liu Y, Zhang WL, Ni L, Liu ZC. Ta3N5 nanoparticles/TiO2 hollow sphere (0D/3D) heterojunction: Facile synthesis and enhanced photocatalytic activities of levofloxacin degradation and H2 evolution. Dalton Trans. 2018;47(37):13113–13125.
44.
Yin GL, Qi XS, Chen YL, Peng Q, Jiang XX, Wang QL, Zhang WH, Gong X. Constructing an all zero-dimensional CsPbBr3/CdSe heterojunction for highly efficient photocatalytic CO2 reduction. J Mater Chem A. 2022;10(42):22468–22476.
45.
Xu CH, Han Y, Chi MY. Cu2O-based photocatalysis. Prog Chem. 2010;22(12):2290–2297.
46.
Etacheri V, Di Valentin C, Schneider J, Bahnemann D, Pillai SC. Visible-light activation of TiO2 photocatalysts: Advances in theory and experiments. J Photoch Photobio C. 2015;25:1–29.
47.
Aguirre ME, Zhou RX, Eugene AJ, Guzman MI, Grela MA. Cu2O/TiO2 heterostructures for CO2 reduction through a direct Z-scheme: Protecting Cu2O from photocorrosion. ACS Appl Mater Interfaces. 2017;217:485–493.
48.
Liu LM, Yang WY, Sun WZ, Li Q, Shang JK. Creation of Cu2O@TiO2 composite photocatalysts with p-n heterojunctions formed on exposed Cu2O facets, their energy band alignment study, and their enhanced photocatalytic activity under illumination with visible light. ACS Appl Mater Interfaces. 2015;7(3):1465–1476.
49.
Kwon SM, Cho SW, Kim M, Heo JS, Kim YH, Park SK. Environment-adaptable artificial visual perception behaviors using a light-adjustable optoelectronic neuromorphic device array. Adv Mater. 2019;31(52):1906433.
50.
Guo T, Zhang BZ, Wang XY, Xiao Y, Sun B, Zhou YN, Wu YMA. Broadband optoelectronic synapse enables compact monolithic neuromorphic machine vision for information processing. Adv Funct Mater. 2023;33(49):2303879.
51.
Silva C, Deuermeier J, Zhang W, Carlos E, Barquinha P, Martins R, Kiazadeh A. Perspective: Zinc-tin oxide based memristors for sustainable and flexible in-memory computing edge devices. Adv Electron Mater. 2023;9(11):2300286.
52.
Su JY, Li YR, Xie DD, Jiang J. Vertical 0.6 V sub-10 nm oxide-homojunction transistor gated by a silk fibroin/sodium alginate crosslinking hydrogel for pain-sensitization enhancement emulation. Mater Horizons. 2023;10(5):1745–1756.
53.
Qi XC, Jin TY, Liu Y, Tian Y, Liu Y, Chi SW, Zhang JC, Hu Y, Fang DW, Wang J. Construction of a dual Z-scheme Cu|Cu2O/TiO2/CuO photocatalyst composite film with magnetic field enhanced photocatalytic activity. Sep Purif Technol. 2022;301: Article 122019.
54.
Qi L, Wang M, Xue JB, Zhang QY, Chen F, Liu QQ, Li WF, Li XH. Simultaneous tuning band gaps of Cu2O and TiO2 to form S-scheme hetero-photocatalyst. Chem Eur J. 2021;27(59):14638–14644.
55.
Shao MZ, Liu DP, Yan BL, Feng XL, Zhang XJ, Zhang Y. Layer-by-layer electrodeposition of FTO/TiO2/CuxO/CeO2 (1 < x < 2) photocatalysts with high peroxidase-like activity by greatly enhanced singlet oxygen generation. Small Methods. 2021;5(7):2100423.
56.
Zhang XJ, Han DF, Dai MJ, Chen K, Han ZY, Fan YY, He Y, Han DX, Niu L. Enhanced photocatalytic degradation of tetracycline by constructing a controllable Cu2O-TiO2 heterojunction with specific crystal facets. Cat Sci Technol. 2021;11(18):6248–6256.
57.
Ding GL, Yang BD, Chen RS, Mo WA, Zhou K, Liu Y, Shang G, Zhai YB, Han ST, Zhou Y. Reconfigurable 2D WSe2-based memtransistor for mimicking homosynaptic and heterosynaptic plasticity. Small. 2021;17(41):2103175.
58.
Shao H, Li YQ, Yang W, He X, Wang L, Fu JW, Fu MY, Ling HF, Gkoupidenis P, Yan F, et al. A reconfigurable optoelectronic synaptic transistor with stable Zr-CsPbI3 nanocrystals for visuomorphic computing. Adv Mater. 2023;35(12):2208497.
59.
Kwon SM, Kwak JY, Song S, Kim J, Jo C, Cho SS, Nam SJ, Kim J, Park GS, Kim YH, et al. Large-area Pixelized optoelectronic neuromorphic devices with multispectral light-modulated bidirectional synaptic circuits. Adv Mater. 2021;33(45):2105017.
60.
Yang B, Lu Y, Jiang DH, Li ZC, Zeng Y, Zhang S, Ye Y, Liu Z, Ou QQ, Wang Y, et al. Bioinspired multifunctional organic transistors based on natural chlorophyll/organic semiconductors. Adv Mater. 2020;32(28):2001227.
61.
Wang Z, Bovik AC, Sheikh HR, Simoncelliet EP. Image quality assessment: From error visibility to structural similarity. IEEE Trans Image Process. 2004;13(4):600–612.
Appendix
Less
Year 2025 volume 8 Issue 1
PDF
228
123
Cite this Article
BibTeX
Article Info
doi: 10.34133/research.0580
  • Receive Date：2024-10-28
  • Online Date：2025-07-23
  • Published：2025-01-10
Article Data
Affiliations
History
  • Received：2024-10-28
  • Revised：2024-12-14
  • Accepted：2024-12-21
Funding
NSFC for Distinguished Young Scholar(No. 52025022)
the NSFC Program(U23A20568)
the NSFC Program(52372137)
the NSFC Program(52272140)
the NSFC Program(52072065)
the China Postdoctoral Science Foundation(GZB20240135)
the “111” Project(B13013)
the fund from Jilin Province(20220502002GH)
the fund from Jilin Province(20230402072GH)
The Fundamental Research Funds for the Central Universities(2412023YQ004)
Affiliations
    Key Laboratory for UV Light-Emitting Materials and Technology (Ministry of Education), College of Physics, Northeast Normal University, Changchun, China.

Corresponding:

* Address correspondence to: (X.S.); (Y.L.); (Z.W.)
References
Share
https://castjournals.cast.org.cn/joweb/research/EN/10.34133/research.0580
Share to
QR

Scan QR to access full text

Cite this article
BibTeX
Citations
表12种不同金属材料的力学参数
科
Family		 属数
Number of
genus		 种数
Number of
species		 占总种数比例
Percentage of
total species (%)		 属
Genus		 种数
Number of
species		 占总种数比例
Percentage of total
species (%)
鹅膏菌科Amanitaceae 2 11 5.26 鹅膏菌属 Amanita 10 4.78
小菇科 Mycenaceae 2 12 5.74 丝盖伞属 Inocybe 5 2.39
多孔菌科 Polyporaceae 8 14 6.70 蜡蘑属 Laccaria 5 2.39
红菇科 Russulaceae 3 23 11.00 小皮伞属 Marasmius 6 2.87
小菇属 Mycena 11 5.26
光柄菇属 Pluteus 5 2.39
红菇属 Russula 17 8.13
栓菌属 Trametes 5 2.39
关闭全屏
  • BibTeX
  • EndNote
  • RefWorks
  • TxT
Links
Articles
Latest Articles
Most Read
Collections
Updates
Events
News
Multimedia
About
About Us
Contact
No. 86 Xueyuan South Road, Haidian District, Beijing
100081
010-62199257
qkjq@cast.org.cn

Copyright © 2025 China Association for Science and Technology. All rights reserved. For all open access content, the relevant licensing terms apply.
Sponsored by the Office of the Leading Group for Cybersecurity and Informatization of CAST, and supported by Science and Technology Review Publishing House