Open Access
Add to BibSonomy
Abstract
Optical memories are vitally important for the future development of high speed and low cost information technologies. Current optical memory devices still suffer from difficulties such as scaling-down of size, short-life expectancy, and non-volatility without the control of a gate electrode. To resolve these obstacles, a robust photoelectronic memory device is designed and demonstrated based on the integration of amorphous InGaZnO (a-IGZO), GNSs, and nitride multiple-quantum-wells light-emitting diode (MQWs LED). Utilizing the inherent nature of the band alignment between a-IGZO and graphene nanosheets (GNSs), electrons can transfer from a-IGZO to GNSs causing a persistent photoconductivity (PPC). With the long-lasting lifetime of PPC, the signal can be written optically and the encoded signal can be read both electrically and optically. The read and write processes reveal little current degradation for more than 10,000 sec, even repeated for more than hundred times. The device can convert invisible information to visible signal, and the encoded information can be simply erased under a reversed bias without a gate electrode. In addition, the memory device possesses a simple vertically stacked structure for 3D integration, and it is compatible with established technologies.
© 2020 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
Memory device is one of the most important electronic components in information technology. With the increasing demand of human society, the bottlenecks of the conventional electronic chips emerge, such as the limitations of communication speeds and energy costs. These drawbacks have driven the development of optical memories to overcome the limitations [1–5]. Through an intensively collaborated effort for decades, the long-life expectancy and non-volatility of the current optical memories can be achieved based on phase-change materials by using waveguides in device chip, or making a transistor-type device. However, the size of these devices is hard to scale down and the encoded signal maintained by the control of gate bias resulting in a large power consumption [3,5–8]. Additionally, several other shortcomings still remain, such as the current speed of data processing and transferring between the processors and memories is limited due to the imbalanced development of each constituent component [2,9]. To strive for perfection, we present a photoelectronic memory device to overcome the above issues based on the integration of amorphous oxide semiconductor (AOS) and graphene nanosheets (GNSs) hybrid film and nitride multiple quantum wells light-emitting diode (MQWs LED) into a vertically stacked structure.
AOS materials have received great scientific attention in recent years, for they exhibit rich characteristics and have been widely used in a variety of optoelectronic components. Among them, amorphous InGaZnO thin-film transistor (a-IGZO TFT), which is a switching element for driving the display, is regarded as a leading role in next generation display technology. It possesses merits of excellent uniformity of large-scale deposition, simple production, low fabrication cost, and high carrier mobility, comparing with conventional a-Si TFTs [10–13]. The visible transparency of IGZO due to its wide band gap (> 3 eV) is also an attractive feature and makes it a promising candidate for transparent displays [14,15], which allows light passing through the transparent panel and displays a clear image. For AOS materials based devices, a common phenomenon of persistent photoconductivity (PPC) (long relaxation time for hours) has been observed [11,16–18]. Even though during the operation in a memory device, the PPC effect can be controlled and removed by applying a gate bias, but it still causes an additional power consumption [19–22]. Nevertheless, there are several studies using the PPC effect to record signals (the light-induced memory-like behavior), and the photonic synaptic devices have been realized [23–25]. This photonic based system can learn and memorize the information by light-induced effect, which has the advantages of high bandwidth and low cross-talk enabling to raise the computing rate. So far, the achievable PPC effect is still not sufficient for the application of long-life expectancy and non-volatile optical memory.
Recently, it was found that the combination of GNSs and AOS material can improve performance of memory devices due to the excellent electrical and optical properties of GNSs [26–28]. However, the working bias of memory devices is very large and to erase the encoded information needs an additional gate bias. To overcome this drawback, we use the hybrid GNSs and AOS composite to design a photoelectronic memory device by integrating with nitride MQWs LED in a vertically stacked structure. The resultant memory device can convert invisible information into visible signal and provide parallel optical and electrical readouts. The fast photon-electron transition can overcome the interconnection delay and improve the maximum data throughput in comparison with conventional electrical memory arrays in serial sequence [29,30]. The established MQWs LEDs were chosen and implemented to combine with the memory layer (GNSs/a-IGZO) in our device to fulfill visible light communication. GaN-based MQWs LEDs have been widely used and dominated the market of solid state lighting due to the superiority of energy savings, great brightness, high emission efficiency, and broadband emission by changing the thickness and composition of quantum well layers [31–35]. Based on our design, the PPC effect can be erased just by applying a reverse bias because of the rectification characteristics of LED. The rectified behavior can effectively reduce the current flow [36] and cause the accumulation of charge carriers in the active layer (GNSs/a-IGZO), which is equivalent to the effect of removing the PPC by a gate bias. Such a simple way of erasing the PPC effect makes our memory device more space-saving when compared with a transistor-type device. The newly designed photoelectronic memory can be easily incorporated with established solid state lighting technologies, which is useful for the development of next generation information technology with practical application.
2. Methods
2.1 Synthesis of GNSs/a-IGZO composites
A sol-gel solution of a-IGZO was prepared by dissolving 88 mg gallium chloride (GaCl3, Sigma-Aldrich Inc.) into 2 mL of ethylene glycol (C2H6O2, Sigma-Aldrich Inc.) and 8 mL of acetonitrile (ACN, Echo Chemical Co.) at 60 °C and stirring at 400 rpm for 1 min. 110 mg indium chloride (InCl3, Sigma- Aldrich Inc.) and 136.4 mg zinc chloride (ZnCl2, Sigma-Aldrich Inc.) were dissolved into the previous solution at 60 °C and stirred at 400 rpm for 12 hours. The stoichiometric molar ratio of In2O3:Ga2O3:ZnO for IGZO was 1:1:1 in our experiment. The solution of GNSs was prepared by dispersing 2 mg powder of reduced graphene oxide into 2 ml deionized water and 18 mL dimethylformamide (DMF, Echo Chemical Co.) and was sonicated for 3 hrs. The resultant GNSs solution was left to stand for 12 hrs, then the upper layer of the solution was extracted. Finally, the composite solution of GNSs/a-IGZO was prepared by blending different volume fraction of GNSs solution into the a-IGZO solution and vibrating in the ultrasonic cleaning machine for 5 mins.
2.2 Device fabrication
First, the bottom MQWs LED was prepared by Metal-Organic Chemical Vapor Deposition (MOCVD) as described in the previous report [37], which exhibits a very high external quantum efficiency of about 60%, comparable with commercially available blue LEDs. Later the LED chip was cleaned ultrasonically in deionized water, acetone and isopropyl alcohol in sequence with each for 15 minutes, and dried with nitrogen gas. Second, the surface of cleaned LED chip became hydrophilic by oxygen plasma treatment [38] for 10 minutes and the GNSs/a-IGZO solution was then spin-coated on the LED chip at 4500 rpm for 40 sec. Third, the sample was annealed at 500 °C for 1 hr in ambient air to remove residual surfactants. Finally, the patterned top electrode Ti/Au (2/20 nm) was deposited by thermal evaporation with a TEM grid and an indium ball was used as the cathode.
3. Results and discussion
3.1 Structure of the device and characteristic of materials
The schematic of the photoelectronic memory device is shown in Fig. 1(a). The thickness of top electrode Ti/Au was patterned by a TEM grid for 2/20 nm with a device area of 83.3×83.3 µm2. The device structure can be divided into two parts: one is MQWs LED for transmitting the encoded information by an optical way; and the other is the active composite layer consisting of GNSs and a-IGZO to record the optical input signal. In Fig. 1(b), when the light arises from the bottom light emitting device, the transmittance of GNSs/a-IGZO in the visible region is above 90%, and the composite layer can absorb UV light due to the absorption of a-IGZO layer with a band gap of ∼3.2 eV. Hence, a 325 nm laser is chosen to be the excitation light source for encoding the optical signal. The uniformly dispersed GNSs in a-IGZO film reveal an average size of around 80 nm as measured by atomic force microscope (AFM) [Fig. 1(c)]. From the transmission electron microscope image [Fig. 1(d)], it shows that the IGZO film is amorphous and the GNSs are estimated to have an average monolayer thickness of about 0.34 nm.
Fig. 1. (a) Schematic representation of the memory device. The scanning electron microscope (SEM) images show that the size of a memory cell and the thickness of the active layer (GNSs/a-IGZO). (b) Transmittance spectrum of the active layer. (c) Atomic force microscope (AFM) image shows the size of graphene nanosheets (GNSs). (d) Transmission electron microscope image of GNSs in a-IGZO film.
3.2 Persistent photoconductivity effect of GNSs/a-IGZO composite
An appropriate concentration of GNSs in a-IGZO film can achieve several unique features in maintaining highly visible transparency, creating fast conduction paths for raising the carrier mobility [28], trapping photogenerated carriers, reducing the recombination of electrons and holes, and enhancing the lifetime of the PPC effect. In order to explore the effect of GNSs, the devices with different concentrations of GNSs in a-IGZO film were fabricated to measure the photocurrent response with a fixed bias of 2.7 V [Fig. 2(a)]. The fabricated devices have a dark conductivity of about 1.25 × 10−3 S·m-1. It shows that the current increases under 325 nm laser illumination with a power density of 50 mW/cm2, and after removing the light source, there is a fast current drop followed by an extremely slow recovery due to the PPC effect. Figure 2(b) and its inset show the photocurrent and the photoresponsivities of the devices with GNSs concentrations of 0% and 0.5% under different excitation power density. The photoresponsivities were calculated by the values of photocurrent divided by light intensity. The value of photocurrent is linearly dependent on the power density, which is consistent with the efficiency of photogenerated carriers being linearly proportional to the absorbed photon flux. Figures 2(a) and 2(b) show that both photocurrent and photoresponsivity are significantly enhanced by the additive of GNSs in a-IGZO film. The PPC effect can be primarily attributed to the charge trapping of oxygen defects as illustrated in previous reports [27–41]. Briefly, under UV illumination, in addition to the generation of electron-hole pairs due to band-to-band transition, the deep-level traps of a-IGZO film, such as oxygen vacancy (Vo) and oxygen interstitial (Io2−), can be ionized (Vo → Vo2++2e−, Io2− → Io+2e−) and generate electrons to the conduction band of a-IGZO. Thus, the photogenerated holes are localized in the oxygen defect levels of a-IGZO, and the conductivity of a-IGZO increases, in turn the PPC occurs. Additionally, the more GNSs in the a-IGZO matrix, the faster the photocurrent reaches the saturation value and the persistent photocurrent becomes much more pronounced after turning off the UV illumination. The underlying reason of this results is because of the fact that the GNSs in a-IGZO matrix can provide effective carrier transporting paths to enhance the carrier mobility and trap photogenerated electrons [28]. And it enables to reduce the recombination of photogenerated electron-hole pairs to achieve an extremely stable PPC effect due to the spatial separation of electrons and holes. Hence, the GNSs play a very important role for the enhancement of the photocurrent, the conductivity, and the long-lasting PPC effect in the device. Notably, the PPC of the device with 0.5% GNSs concentration is highly stable, which contributes to retention performance. The 0.5% GNSs concentration is the optimized condition for our study. If the concentration exceeds 0.5%, the GNSs/IGZO blended film will become more metallic, which will decrease the ratio of the photocurrent versus the dark current. To examine the UV effect on the device, the I-V characteristics of the device are shown in Fig. 2(c), in which we can see that the device has two states, i.e., the original On state and the Off state due to the PPC effect. The result illustrates that after the illumination of UV light, the PPC effect only exists in the region under a forward bias, and it shows a complete recovery under a reversed bias. Therefore, the working bias of the device has to be chosen at the forward bias region. The inset of Fig. 2(c) shows that the pronounced current difference between the On and Off states around the turned-on voltage of MQWs LED (∼2.7 V). This value is larger enough to light up the device after light illumination due to the occurrence of the PPC effect. The I-T curve after 325 nm laser illumination at 2.7 V and the corresponding photos taken by an optical microscope and a mobile phone under different states of the device are shown in Fig. 2(d). The results illustrate that the encoded information can be read by both the electrical way (electrical signal of lower resistance state) and the optical way (light emission from the LED device) due to the stable and long-lasting PPC effect.
Fig. 2. (a) PPC effects of different concentrations (0%, 0.15%, 0.25%, and 0.5%) of GNSs in the a-IGZO matrix were measured at 2.7 V with 325 nm laser illumination at a power density of 50 mW/cm2. (b) Photocurrent of the device with GNSs concentrations of 0% and 0.5% under different excitation power density. The inset shows the corresponding photon-responsibility. (c) I-V characteristics of the device in different states and the inset shows the zoomed-in data around the operating bias. (d) Photos of the device with 0.5% GNSs/a-IGZO were taken by an optical microscope (top) and a mobile phone (bottom) at different time.
3.3 Working mechanism of the device
The operational characteristic of the device is demonstrated in Fig. 3(a), which shows that the On state of the device can be written by the UV illumination of 325 nm due to the long-lasting PPC effect at a fixed working bias of 2.7 V and the device can be recovered to the original Off state by erasing the encoded information under a reverse bias. Note that the erasing process is very distinct and easy to achieve in our demonstration comparing with other previous reports [22,28,42,43], in which a gate bias (the third electrode) is required to control and erase the PPC effect of amorphous oxide semiconductor. The simple vertically stacked structure of our device is the same as the commercial electric memory cell, which can cost-down and possess the advantage of three dimensional integration with many other optoelectronic devices. Figure 3(a) shows that there are three steps (denoted as red number) in one encoding-erasing cycle, including the process of increasing the photocurrent under UV illumination, recovery after light stress, and erasing the PPC effect. The underlying mechanisms of these three steps can be understood based on the schematic diagrams [Figs. 3(b) and 3(c)]. As shown in Fig. 3(b), under UV illumination and forward bias, the electron-hole pairs by band-to-band excitation and oxygen defect ionization are photogenerated to increase the current. The photogenerated electrons flow away to the GNSs (transporting paths) and the positive charges are trapped in the defects. After light stress, the recovery of current is prohibited by the existence of GNSs and defect levels due to the spatial separation of negative and positive charges, and thus the recombination of electron-hole pairs is reduced. The separation of photogenerated charge carriers therefore causes the long-lasting PPC effect and the enhancement of the conductivity. This lower resistance state (higher conductivity) can convert the device into the On state. In Fig. 3(c), a negative bias (reverse bias for the MQWs LED) can be applied to erase the PPC effect. It is because the rectification characteristics of LED under the reverse bias can effectively reduce the current flow, and electrons accumulated in the active layer (GNSs/a-IGZO) can recombine with the trapped holes. This result is also consistent with the measurement under the reverse bias region as shown in Fig. 2(c). Thereafter, the device is converted to the original Off state. Taking the advantage of both the PPC effect existing only under forward bias and the rectification characteristics of the device, the sneak-path effect for high density memory arrays can be overcome, which is a challenge issue for high density memory arrays. On our designed memory array, only the chosen specific cell under forward bias can record and transmit the optical information, and the cross-talk interference (sneak-path) from the neighboring cells can be avoided [44,45].
Fig. 3. (a) I-T curve shows the rewritable ability of the device with 0.5% GNSs/a-IGZO. The < On > state is written by 325 nm laser illumination at 2.7 V. A revised bias of -15 V was applied to erase the information and make the device go back to the < Off > state. The three steps (denoted as red number) of one cycle are the process of increasing the photocurrent, recovery and erasing the PPC respectively. (b, c) Working mechanism of the device under forward bias in (b) and reverse bias in (c).
3.4 Switching and retention performance for electrical and optical signal
In order to have practical application for the device, the demonstration of switching and retention ability is shown in Fig. 4(a). The steady On and Off states of the device were measured under the same conditions as that of Fig. 3(a), and the result shows that it can be repeated for more than hundred times with a negligible degradation of the device performance. For the retention performance of the device, Fig. 4(b) records the On and Off state readouts at 2.7 V for a long period of time. The result of the On state measured after the light stress for 5 minutes demonstrates that the persistent photocurrent can be maintained at 7 µA up to 10,000 sec. The extremely small slope of the fitting line shows that the value of PPC has almost no attenuation making the device a promising long-term, non-volatile, photoelectronic memory without an additional power consumption. Meanwhile, the value of the PPC is sufficient to light up the device as the electroluminescence (EL) shown in Fig. 4(c). The EL intensity was measured at different time after light stress [Fig. 4(d)], which illustrates that the optical signal is as stable as the electrical signal. When the device is in the Off state, no emission can be detected making the On/Off ratio of EL infinity.
Fig. 4. (a) Input signal can be optically written and electrically erased for more than 100 times without degradation for both On and Off states. (b) Electrical retention ability of the device at a fixed working bias of 2.7 V. (c) Electroluminescence (EL) spectra were measured under different states of the device. (d) Optical retention ability of the device for On and Off states at different time and the same working bias after switching off the UV illumination.
Finally, to stress the unique feature of the device, the schematic of dual-function for optical and electrical detection is shown in Fig. 5, which illustrates that the state of our device can be converted between the On (signal 1) and Off (signal 0) by optical write-in (under UV illumination) and electrical erasing (applying the reverse bias). Also, the parallel readout can be achieved by transmitting the signal 0 and 1 in both optical and electrical ways, simultaneously.
Fig. 5. Schematic of the device operation. The currents of the device can be converted to On and Off states by optically written-in and electrically erased. A distinct aspect of the device shows the dual functions of electrically and optically readable capability.
4. Conclusion
In summary, a photoelectronic memory with unique multiple functionalities, including optically rewritable, long-lasting non-volatile, and both optically and electrically readable capabilities, has been demonstrated. This device was made based on the important fact that GNSs blended in a-IGZO film can significantly enhance the lifetime of PPC effect for more than 10,000 sec without attenuation. The stable and long-lasting PPC effect of the active GNSs/a-IGZO layer stacked on top of nitride MQWs LED can encode the optical information and convert invisible information (UV light) to visible signal (blue light). Also, the encoded signal can be simply erased under a reverse bias instead of using a gate electrode, which will scale down the device and reduce additional power consumption. The optically readable capability enables the process of parallel readout, which can increase the maximum data throughput and improve limited transporting rate in memory serials. Our photoelectronic memory device proves the excellent performance of low programming bias, simple vertically stacked structure for 3D integration and retention ability. In view of the established technologies of nitride LEDs and IGZO and the rapid progress of graphene invention, we can anticipate a prosperous future by utilizing our device.
Funding
Ministry of Science and Technology and Ministry of Education of the Republic of China.
Acknowledgments
This work was supported by the Ministry of Science and Technology and Ministry of Education of the Republic of China.
Disclosures
The authors declare no conflicts of interest.
References
1. M. Wuttig, H. Bhaskaran, and T. Taubner, “Phase-change materials for non-volatile photonic applications,” Nat. Photonics 11(8), 465–476 (2017). [CrossRef]
2. Z. Wang, Z. Pang, P. Yang, J. Xu, X. Chen, R. K. V. Maeda, Z. Wang, L. H. K. Duong, H. Li, and Z. Weng, “MOCA: An Inter/Intra-Chip Optical Network for Memory,” Proc. Des. Autom. Conf., 1–6 (2017).
3. T. Leydecker, M. Herder, E. Pavlica, G. Bratina, S. Hecht, E. Orgiu, and P. Samorì, “Flexible non-volatile optical memory thin-film transistor device with over 256 distinct levels based on an organic bicomponent blend,” Nat. Nanotechnol. 11(9), 769–775 (2016). [CrossRef]
4. N. Pleros, S. Pitris, C. Vagionas, P. Maniotis, T. Alexoudi, A. Miliou, and G. T. Kanellos, “Optical Interconnect and Memory Technologies for Next Generation Computing,” IEEE, 3–6 (2016).
5. C. Ríos, M. Stegmaier, P. Hosseini, D. Wang, T. Scherer, C. D. Wright, H. Bhaskaran, and W. H. P. Pernice, “Integrated all-photonic non-volatile multi-level memory,” Nat. Photonics 9(11), 725–732 (2015). [CrossRef]
6. A. N. Tait, T. F. deLima, E. Zhou, A. X. Wu, M. A. Nahmias, B. J. Shastri, and P. R. Prucnal, “Photonic synaptic device capable of optical memory and logic operations,” CLEO Sci. Innov. SW4K7, SW4K.7 (2016).
7. J. F. Song, X. S. Luo, A. E. J. Lim, C. Li, Q. Fang, T. Y. Liow, L. X. Jia, X. G. Tu, Y. Huang, H. F. Zhou, and G. Q. Lo, “Integrated photonics with programmable non-volatile memory,” Sci. Rep. 6(1), 1–7 (2016). [CrossRef]
8. M. T. Hill, H. J. S. Dorren, T. deVries, X. J. M. Leijtens, J. H. denBesten, B. Smalbrugge, Y. S. Oei, H. Binsma, G. D. Khoe, and M. K. Smit, “A fast low-power optical memory based on coupled micro-ring lasers,” Nature 432(7014), 206–209 (2004). [CrossRef]
9. A. Huang, “Architectural Considerations Involved in the Design of an Optical Digital Computer,” Proc. IEEE 72(7), 780–786 (1984). [CrossRef]
10. J. Socratous, K. K. Banger, Y. Vaynzof, A. Sadhanala, A. D. Brown, A. Sepe, U. Steiner, and H. Sirringhaus, “Electronic structure of low-temperature solution-processed amorphous metal oxide semiconductors for thin-film transistor applications,” Adv. Funct. Mater. 25(12), 1873–1885 (2015). [CrossRef]
11. A. Nathan, S. Lee, S. Jeon, and J. Robertson, “Amorphous oxide semiconductor TFTs for displays and imaging,” J. Disp. Technol. 10(11), 917–927 (2014). [CrossRef]
12. T. Kamiya, K. Nomura, and H. Hosono, “Present status of amorphous In–Ga–Zn–O thin-film transistors,” Sci. Technol. Adv. Mater. 11(4), 044305 (2010). [CrossRef]
13. K. Nomura, H. Ohta, A. Takagi, T. Kamiya, M. Hirano, and H. Hosono, “Room-temperature fabrication of transparent flexible thin-film transistors using amorphous oxide semiconductors,” Nature 432(7016), 488–492 (2004). [CrossRef]
14. M. Mativenga, D. Geng, B. Kim, and J. Jang, “Fully transparent and rollable electronics,” ACS Appl. Mater. Interfaces 7(3), 1578–1585 (2015). [CrossRef]
15. S. S. Lee, S. H. Jeon, R. Chaji, and A. Nathan, “Transparent semiconducting oxide technology for touch free interactive flexible displays,” Proc. IEEE 103(4), 644–664 (2015). [CrossRef]
16. I. Hwang, J. Kim, M. Lee, M. W. Lee, H. J. Kim, H. I. Kwon, D. K. Hwang, M. Kim, and H. Yoone, “Wide-Spectral/Dynamic-Range Skin-Compatible Phototransistors Enabled by Floated Heterojunction Structures with Surface Functionalized SWCNTs and Amorphous Oxide Semiconductors,” Nanoscale 9(43), 16711–16721 (2017). [CrossRef]
17. A. Kiazadeh, H. L. Gomes, P. Barquinha, J. Martins, A. Rovisco, J. V. Pinto, R. Martins, and E. Fortunato, “Improving positive and negative bias illumination stress stability in parylene passivated IGZO transistors,” Appl. Phys. Lett. 109(5), 051606 (2016). [CrossRef]
18. J. F. Conley, “Instabilities in amorphous oxide semiconductor Thin-Film transistors,” IEEE Trans. Device Mater. Reliab. 10(4), 460–475 (2010). [CrossRef]
19. D. H. Kim and J. T. Park, “Investigation on stress induced hump phenomenon in IGZO thin film transistors under negative bias stress and illumination,” Microelectron. Reliab. 55(9-10), 1811–1814 (2015). [CrossRef]
20. J. M. Kwon, J. Jung, Y. S. Rim, D. L. Kim, and H. J. Kim, “Improvement in negative bias stress stability of solution-processed amorphous In-Ga-Zn-O thin-film transistors using hydrogen peroxide,” ACS Appl. Mater. Interfaces 6(5), 3371–3377 (2014). [CrossRef]
21. S. Lee, M. Mativenga, and J. Jang, “Removal of negative-bias-illumination-stress instability in amorphous-InGaZnO thin-film transistors by top-gate offset structure,” IEEE Electron Device Lett. 35(9), 930–932 (2014). [CrossRef]
22. S. Jeon, S. E. Ahn, I. Song, C. J. Kim, U. I. Chung, E. Lee, I. Yoo, A. Nathan, S. Lee, J. Robertson, and K. Kim, “Gated three-terminal device architecture to eliminate persistent photoconductivity in oxide semiconductor photosensor arrays,” Nat. Mater. 11(4), 301–305 (2012). [CrossRef]
23. M. Lee, W. Lee, S. Choi, J. W. Jo, J. Kim, S. K. Park, and Y. H. Kim, “Brain-inspired photonic neuromorphic devices using photodynamic amorphous oxide semiconductors and their persistent Photoconductivity,” Adv. Mater. 29(28), 1700951 (2017). [CrossRef]
24. S. Qin, F. Wang, Y. Liu, Q. Wan, X. Wang, Y. Xu, Y. Shi, X. Wang, and R. Zhang, “A light-stimulated neuromorphic device based on graphene hybrid phototransistor,” 2D Mater. 4(3), 035022 (2017). [CrossRef]
25. D. Kuzum, S. Yu, and H. S. P. Wong, “Synaptic electronics: materials, devices and applications,” Nanotechnology 24(38), 382001 (2013). [CrossRef]
26. H. Pang, T. Chen, G. Zhang, B. Zeng, and Z. M. Li, “An electrically conducting polymer/graphene composite with a very low percolation threshold,” Mater. Lett. 64(20), 2226–2229 (2010). [CrossRef]
27. J. Zhang, P. Dong, Y. Gao, C. Sheng, and X. Li, “Performance enhancement of ZITO thin-film transistors via graphene bridge layer by sol-gel combustion process,” ACS Appl. Mater. Interfaces 7(43), 24103–24109 (2015). [CrossRef]
28. M. K. Dai, Y. R. Liou, J. T. Lian, T. Y. Lin, and Y. F. Chen, “Multifunctionality of giant and long-lasting persistent photoconductivity: semiconductor–conductor transition in graphene nanosheets and amorphous InGaZnO hybrids,” ACS Photonics 2(8), 1057–1064 (2015). [CrossRef]
29. Y. R. Liou, G. Haider, S. Y. Cai, C. L. Wu, T. Y. Lin, and Y. F. Chen, “High-performance light-emitting memories: multifunctional devices for unveiling Information by optical and electrical detection,” Adv. Opt. Mater. 4(11), 1744–1749 (2016). [CrossRef]
30. C. W. Chang, W. C. Tan, M. L. Lu, T. C. Pan, Y. J. Yang, and Y. F. Chen, “Electrically and optically readable light emitting memories,” Sci. Rep. 4(1), 5121 (2015). [CrossRef]
31. R. X. G. Ferreira, E. Xie, J. J. D. Mckendry, S. Rajbhandari, H. Chun, G. Faulkner, S. Watson, A. E. Kelly, E. Gu, R. V. Penty, I. H. White, D. C. O’Brien, and M. D. Dawson, “High bandwidth GaN-based micro-LEDs for multi-Gb/s visible light communications,” IEEE Photonics Technol. Lett. 28(19), 2023–2026 (2016). [CrossRef]
32. Z. Liu, W. C. Chong, K. M. Wong, and K. M. Lau, “GaN-based LED micro-displays for wearable applications,” Microelectron. Eng. 148, 98–103 (2015). [CrossRef]
33. S. Yin, C. Wang, W. Zhu, J. Yao, J. Zou, X. Lin, and C. Luo, “Single chip super broadband InGaN/GaN LED enabled by nanostructured substrate,” Opt. Express 22(S5), A1380–A1388 (2014). [CrossRef]
34. S. Nakamura and M. R. Krames, “History of gallium-nitride-based light-emitting diodes for illumination,” Proc. IEEE 101(10), 2211–2220 (2013). [CrossRef]
35. S. Nakamura, “The roles of structural imperfections in InGaN-based blue light-emitting diodes and laser diodes,” Science 281(5379), 956–961 (1998). [CrossRef]
36. D. A. Neaman, Semiconductor Physics and Devices: Basic Principles, 4th Edition (McGraw Hill, 2012).
37. J. Y. Chen, C. Y. Ho, M. L. Lu, L. J. Chu, K. C. Chen, S. W. Chu, W. Chen, C. Y. Mou, and Y. F. Chen, “Efficient spin-light emitting diodes based on InGaN/GaN quantum disks at room temperature: a new self-polarized paradigm,” Nano Lett. 14(6), 3130–3137 (2014). [CrossRef]
38. S. H. Tan, N. T. Nguyen, Y. C. Chua, and T. G. Kang, “Oxygen plasma treatment for reducing hydrophobicity of a sealed polydimethylsiloxane microchannel,” Biomicrofluidics 4(3), 032204 (2010). [CrossRef]
39. X. Liu, C. Wang, B. Cai, X. Xiao, S. Guo, Z. Fan, J. Li, X. Duan, and L. Liao, “Rational design of amorphous indium zinc oxide/carbon nanotube hybrid film for unique performance transistors,” Nano Lett. 12(7), 3596–3601 (2012). [CrossRef]
40. S. Lee, A. Nathan, S. Jeon, and J. Robertson, “Oxygen defect-induced metastability in oxide semiconductors probed by gate pulse spectroscopy,” Sci. Rep. 5(1), 14902 (2015). [CrossRef]
41. S. Jeon, I. Song, S. Lee, B. Ryu, S. E. Ahn, E. Lee, Y. Kim, A. Nathan, J. Robertson, and U. I. Chung, “Origin of high photoconductive gain in fully transparent heterojunction nanocrystalline oxide image sensors and interconnects,” Adv. Mater. 26(41), 7102–7109 (2014). [CrossRef]
42. H. L. Lu, Z. M. Liao, L. Zhang, W. T. Yuan, Y. Wang, X. M. Ma, and D. P. Yu, “Reversible insulator-metal transition of LaAlO3/SrTiO3 interface for nonvolatile memory,” Sci. Rep. 3(1), 2870 (2013). [CrossRef]
43. S. E. Ahn, S. Jeon, Y. W. Jeon, C. Kim, M. J. Lee, C. W. Lee, J. Park, I. Song, A. Nathan, S. Lee, and U. I. Chung, “High-performance nanowire oxide photo-thin film transistor,” Adv. Mater. 25(39), 5549–5554 (2013). [CrossRef]
44. J. Y. Seok, S. J. Song, J. H. Yoon, K. J. Yoon, T. H. Park, D. E. Kwon, H. Lim, G. H. Kim, D. S. Jeong, and C. S. Hwang, “A review of three-dimensional resistive switching cross-bar array memories from the integration and materials property points of view,” Adv. Funct. Mater. 24(34), 5316–5339 (2014). [CrossRef]
45. Y. Ji, D. F. Zeigler, D. S. Lee, H. Choi, A. K. Y. Jen, H. C. Ko, and T. W. Kim, “Flexible and twistable non-volatile memory cell array with all-organic one diode–one resistor architecture,” Nat. Commun. 4(1), 2707 (2013). [CrossRef]
