Open Access
Add to BibSonomy
Abstract
The removal of a sapphire substrate by laser lift-off, photoluminescence detection technology, and the luminous efficiency of size-dependent devices are very hot issues for the Micro-LED display, which is thoroughly studied in this paper. The mechanism of thermal decomposition of the organic adhesive layer after laser irradiation is analyzed in detail, and the thermal decomposition temperature of 450 °C solved by the established one-dimensional model is highly consistent with the inherent decomposition temperature of the PI material. The spectral intensity of PL is higher, and the peak wavelength is red-shifted by about 2 nm compared to EL under the same excitation condition. The results of size-dependent device optical-electric characteristics show that the smaller the device size, the lower the luminous efficiency under the same display resolution and PPI conditions, and the higher corresponding display power consumption.
© 2023 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
Micro-LED display is considered to be a new next-generation of display technology that may be subvert the existing flexible organic light-emitting diode (OLED) display, and has become a new growth and explosion point in the field of display industry [1–4]. Display is a bridge for information interaction between human beings and all things. Especially after the concept of the new network hot word “meta universe” was put forward in 2021, display is everywhere and the era of interconnection of all things has begun [5]. Micro-LED display technology is a self-luminous display technology, through the array of Micro-LED light emitting device is integrated on the active addressing driving substrate to realize separate control and lighting. Micro-LED display has many advantages, such as self-luminous [6], high efficiency, low power consumption, high integration and high stability [7]. It is small in chip size, flexible and easy to disassemble and merge. It can be applied to any display application from small size to large size [8–10].
However, many difficulties and challenges to product commercialization require to be overcome. The challenges of high resolution and high pixel per inch (PPI) require the chip size to get increasingly smaller [11,12]. At present, many scholars have studied GaN based blue/green Micro-LED very deeply, but there is very little research on red devices as one of the three primary colors of display. The removal of sapphire substrate by laser stripping is the first step in Micro-LED display technology. The removal principle is fundamentally different from that of blue/green devices [13–16]. The method of electroluminescence (EL) detection of chips with slow detection speed and low moving alignment accuracy is no longer used in Micro-LED, Instead, non-contact photoluminescence (PL) rapid detection has been considered as the best solution to predict the wavelength and luminance of the device [17]. However, the correlation of the optical between the PL and EL has not been disclosed in detail. In addition, the photoelectric characteristics of the red LED device and the current efficiency related to the size are not deep enough [18,19].
In this paper, the principle of laser decomposition of organic adhesive layer between red LED chip and sapphire interface is deeply discussed, and the temperature parameters directly related to decomposition are calculated by using one-dimensional heat conduction model, which is highly consistent with the experimental results. Besides, the correlation and difference between PL and EL, and the size-depended optical-electronic characteristics are deeply studied.
2. Experimental procedures
The Micro-LED chip arrays with different sizes and specifications were fabricated by the same process flow. The metal organic chemical vapor deposition method was used to grow different epitaxial functional layers on GaAs substrate, and then the epitaxial functional layers were bonded to sapphire substrate by organic adhesive through secondary bonding. The simplified schematic structure of the LED chip is shown in Fig. 1(a). The epitaxial structure of the wafer is composed of 1.5 µm AlxGa1-xIn0.5P n-type current spreading layer, 2 µm AlyGa1-yIn0.5P n-type ohmic contact layer, 0.2 µm n-AlInP confinement layer, quantum well active layer, further comprising 0.2 µm p-AlInP electron barrier layer and 0.45 µm p-GaP current spreading layer. The physical picture of LED is shown in Fig. 1(b).
3. Results and discussion
3.1 Mechanism of laser lift off sapphire substrate
In order to improve the process stability and the light output efficiency of the red Micro-LED, the epitaxially grown GaAs substrate is usually converted to the sapphire substrate, and the organic adhesive layer with high temperature resistance process needs to be used as the intermediate adhesive layer between the epitaxial layer and the sapphire. The laser stripping technology mainly uses the pulse laser with high energy density to act on the interface between the organic adhesive layer and the sapphire substrate. The laser reaches the interface between the adhesive layer and the sapphire and is absorbed by the organic material, so that the film temperature rises and thermal decomposition occurs, then the purpose of stripping can be achieved.
Fig. 2(a) schematically illustrates the peeling process of the sapphire/PI (polyimide) film interface by ultraviolet laser scanning. Fig. 2(b) is an enlarged view of the partial interface in Fig. 2(a), and the process begins to absorb laser accompanied by heat conduction. Then, due to the local high temperature at the interface, the organic layer near the sapphire side becomes molten. The breaking of chemical bonds induced by heat will result in the decomposition of the organic layer with the further increase of temperature, thereby gas products are generated (Fig. 2(c)). These gas products lead to the fracture of the hydrodynamic layer of the molten PI, so that the micro morphology of the sapphire interface after the re-solidification can be reshaped. The complex reactions in the heat exchange interface include melting and decomposition. In addition, the by-product (gas) generated in the reaction confined space will generate a forward thrust to act on the back of the LED chip (Fig. 2(d)), completing the dissociation between the LED chip and the sapphire substrate.
Fig. 2. (a) Schematic diagram of LED / sapphire by ultraviolet laser scanning, (b) - (d) Process of gas products produced during laser ablation.
The temperature field distribution plays a major role in the dissociation of LED devices. The temperature field distribution of sapphire/organic layer interface is simulated by establishing a mathematical model of heat conduction. Since the area of the laser spot is much larger than the LED area, the laser light source is equivalent to heat source of area arrays acting on the sapphire surface, and the heat conduction in the process of laser lift-off is approximately calculated in one-dimensional direction by establishing the plane rectangular coordinate system as shown in Fig. 3.
Fig. 3. (a) One dimensional heat conduction model of laser lift-off (b) Micro-LED morphology after 200 mJ/cm2 energy stripping.
The one-dimensional conduction equation along the positive direction of the y-axis is:
Assuming that the organic adhesive layer is an isotropic uniform material, the single pulse energy density at the interface is
I0 (t) is the energy pulse density at the initial time 0, and R is the reflectivity of the Sapphire / PI organic adhesive layer interface to the laser. Substituting formula (3) into formula (2),
Table 1. Relevant thermodynamic coefficients of polyimide film
3.2 Relationship between photoluminescence and electroluminescence
The 10 × 10 red micro-LED array is transferred to the PM driven backplane through the mass transfer technology after laser striping the sapphire substrate. An ammeter is connected between the power supply and the LED test array to realize the current monitoring. In addition, the photoluminescence lighting of the test array can also be realized by using 375 nm ultraviolet laser to uniformly cover the test array. The schematic diagram of the test principle is shown in Fig. 4(a). When the external power supply is in the open circuit state, the electrons excited by the energy inside the quantum well transition from the low energy level to the high energy level after the laser irradiates the LED array device. At this moment, the electrons and holes are accumulated on both sides of the p-type semiconductor and the n-type semiconductor through the built-in electric field, generating a photovoltage of about 1.4 V. That is, the device shown by the dotted line is equivalent to a current source, and the voltage difference is 1.4 V. When the applied electric field voltage is the same as 1.4 V, all the photogenerated carrier electron hole pairs excited by the incident light participate in the composite emission. At this time, the test current signal of the external circuit is zero, as shown in point A in Fig. 4(b). If the voltage difference applied by the power supply is less than 1.4 V (for example, 1 V), the total potential in the PN junction will be higher than the applied bias signal, which will lead to an increase in the leakage of carriers in the quantum well. In this case, the electron hole pair generated by the light escapes from the quantum well in the opposite direction, generating a reverse leakage current. As the applied bias electric field continues to decrease, the overall current shows that the reverse leakage current gradually increases. When the applied bias voltage is 0 V (point B), it is equivalent to that the two ends of the LED device are directly short circuited and connected by the external wire circuit, and most of the carriers leak directly through the circuit. It is worth noting that not all the carriers in the quantum well leak out at this time. When the applied bias signal continues to decrease to -4 V in the negative direction, the carriers which fully overcome the constraints of the built-in electric field potential energy can be exhausted and released. After the signal in the continuous direction increases, the reverse current signal in the loop tends to a stable value -0.15 mA (point C). We agree that this value is called the reverse saturation current of the device, and can also be understood as the number of all carriers generated by PL. Further, if the PL test and the EL test are to generate the same excitation to the LED device, the EL test should use the same signal as the reverse saturation current to ensure the consistency of the two emission spectra.
Fig. 4. (a) Schematic diagram of comprehensive action principle of PL and EL (b) I-V curve under the joint action of PL and EL
In order to further study the intrinsic relationship between the spectral characteristics of PL and EL, the I-V curve of EL and the I-V curve combined action of PL + EL are described in Fig. 5(a). Before the threshold opening voltage is 1.6 V, electrons cannot flow from the n-type conduction band layer with lower energy level to the p-type conduction band layer with higher energy level due to the difference in barrier energy level in the PN junction, which shows that the PN junction is not open and the current is zero. When the applied bias voltage is higher than the threshold opening voltage, the PN junction will conduct forward and the current will rise sharply. The reverse saturation current of photogenerated carriers of PL is -0.15 mA (Fig. 5(a)), and the forward current of 0.15 mA is used to excite the array device, and the relationship between the spectral characteristics of the two is compared. The peak wavelength of the PL spectrum is red shifted by about 2 nm relative to the peak wavelength of the EL spectrum, and the radiation intensity of the entire spectrum is stronger than that of the latter (Fig. 5(b)). Fig. 5(c)(d) shows the statistical results of saturation current and wavelength shift distribution respectively. The number of test samples is 20, the mean value of saturation current is 1.501 mA, and the standard deviation is 0.013. The mean value of wavelength shift is 1.997 nm, and the standard deviation is 0.04. The statistical experimental results are completely consistent with the above saturated current 1.5 mA and wavelength shift 2 nm, which is sufficient to illustrate the good repeatability and stability of the results of this research.
Fig. 5. (a) I-V curve (b) spectral characteristics after combined action of EL, PL and EL. (c) Saturation current and (d) Wavelength shift distribution statistics.
The difference analysis of EL spectrum and PL spectrum is analyzed in combination with the following Fig. 6. When a forward current is injected, electrons diffuse from the n-type to p-type layer, and holes diffuse from the p-type to n-type layer (Fig. 6(a)). Since the mobility of holes is less than that of electrons (about 1/40 times) [21] and the recombination of electron hole pairs mainly occurs in the p-side quantum well. Besides, the increase of defect hanging bonds on the side wall of the miniaturized Micro-LED, some carriers are more likely to leak from the device side wall in the process of carrier migration and recombination, which reduces the efficiency of carrier recombination in the quantum well. Secondly, there is a possibility of carrier leakage to a certain extent in the quantum well under high current density.
In the case of photoluminescence (Fig. 6(b)), the generated photogenerated carriers mainly occur in the quantum well, and will not leak in the chip sidewall due to the shielding of the piezoelectric field. Therefore, the number of effective radiation recombination carriers in PL is higher than that EL under the same excitation, so the radiation intensity of PL spectrum is greater than that of EL spectrum. Secondly, the number of carriers in the PL quantum well is more, and the PN junction temperature is relatively higher, which causes the lattice field formed by ionizing impurity ions in the impurity semiconductor to cause the ion energy level fission and affect the lattice vibration. The change in lattice symmetry and the multi-body [22] effect in the higher density carrier system will also cause the band gap narrowing. In short, when the PN junction temperature is high, the band gap will shrink, causing the peak wavelength to shift to the long wavelength.
3.3 Capacitance and voltage characteristics
Fig. 7 shows the relationship between the apparent capacitance and voltage of AlGaInP -MQWs -LED in the voltage range of - 4 V ∼ + 4 V, and the apparent capacitance value shows a trend of first keeping unchanged and then decreasing. The apparent capacitance values are all positive before 2.45 V @ 10 K and 2.6 V@500 K. The capacitance mainly consists of two parts: depletion layer capacitance and diffusion capacitance. When the applied voltage is high, the diffusion capacitance plays a major role. The diffusion capacitance is caused by the transport process of minority carriers in light emitting diodes. The capacitance of the depletion layer can be approximately regarded as the parallel plate capacitance, and the width D of the depletion layer can be regarded as the plate spacing of the parallel plate capacitor. The change value of the junction capacitance is the result of the combined action of the capacitance change value Cd (positive value) caused by diffusion, the capacitance change value Cr (negative value) caused by composite emission, and the capacitance Cb (positive value) caused by the depletion layer capacitance. The depletion layer capacitance plays a dominant role under a small forward voltage. When the voltage continues to increase, the sample begins to conduct and the apparent capacitance will decrease under a large forward bias voltage. The junction capacitance C can be considered as a differential capacitance under the AC signal, namely
The dVj always remains positive with the increase of the applied voltage, and the injected carrier dQ increases with the raise of the voltage before the carrier radiation recombination. Therefore, the junction capacitance increases before the PN junction is turned on. After that, the conduction current carriers of PN junction recombine, and the apparent capacitance decreases gradually. In AC signal test, the apparent capacitance will show a negative value when the applied voltage is high, which is called negative capacitance phenomenon. When the frequency of the external small signal is lower (10 kHz) and the DC voltage is higher, the negative capacitance phenomenon will be more obvious. We believe that the carrier Qr consumed by recombination will be greater than the carrier Qi injected by the applied forward voltage. In addition, under a large forward voltage, the junction voltage Vj will gradually become saturated, that is, when Vj increases to a certain value, there will be no obvious change with the increase of the applied bias voltage. It is a gradually decreasing amount although dVj is positive, which means that smaller dVj and larger dQ will have larger C value. Therefore, once negative capacitance occurs, the higher the applied bias voltage is, the greater the absolute value of negative capacitance becomes.3.4 Size-dependent optical-electrical characteristic
Five kinds of different mesa area chip arrays varied from 60 × 50 µm2 to 30 × 20 µm2 were fabricated according to the same process sequence, and the dimensions specifications of each sample was described in Table. 2. Each sample consisted of 10 × 10 Micro-LED arrays and pixel pitch of 156 µm after graphic etching, and effective active area (AA) on each sample was composed of the same chip size.
Table 2. Size parameters of five different samples
The typical I-V characteristic of the as-fabricated Micro-LED array with different dimensions from A to E is measured by IVL test system. There is a strong size-dependent behavior in I-V characteristics of devices (Fig. 8(a)). The current is negatively correlated with the pixel size under the same forward bias voltage, which means that the resistance of smaller chips is much higher. Although the current value of 60 × 50 µm2 chip is higher at a fixed voltage, it does not mean that the performance of large-sized LED devices is better. The display of Micro-LED products generally requires a normal brightness range of 0-1000 nit, and the range of working voltage corresponding to different sizes is 2.15-2.72 V, as shown by the red line in Fig. 8(b), and the purple line represents the corresponding threshold voltage under the current of 1μA of a single device. When the analog circuit is used to drive the screen to emit light, the 256 gray scale is not easy to expand within the voltage range of 0.53 V, which is more suitable for the display of the Micro-LED chip driven by the digital circuit. The digital driving principle is to set the driving voltage to a fixed value, and adjust the brightness corresponding to the light emission of the display by controlling the light emission time. The current density of different Micro-LED chips is compared with the forward voltage as shown in Fig. 8(c).
Fig. 8. Size-dependent characteristics of (a) injection current versus voltage and (b) the cross-voltage in the normal display range, (c) current density versus voltage and (d) luminance versus current density.
However, the characteristic of J-V and I-V characteristic show opposite trends under the same forward bias. The smaller the chip size is, the higher the current density is. The current density value of sample E is 4.25 A / cm2 when the voltage is 2.8 V. It is speculated that the current expansion of larger chips is relatively weak, which is more likely to cause current crowding effect. The farther away from the mesa edge, the lower the current density distribution. The current crowding effect of large-sized LED devices will become worse and worse, which is more likely to cause local high junction temperature leading to failure after long-time operation. On the contrary, the current of small-sized chips has better expansion performance and uniformity, and the side wall area of small-sized chips accounts for a higher proportion, which makes it easier to diffuse locally generated heat and has better heat dissipation performance. The relationship between brightness and current density is shown in Fig. 8(d). The brightness increases gradually with the rise of current density, and the brightness of larger chip is always at a higher level under the same current density.
The peak wavelength and the full width at half maximum of red sample A-E are 628 ± 1 nm and 10 ± 1 nm in the working current of 0.5 ∼ 3 mA (Fig. 9). The number of effective electron hole recombination decreases with the reduction of chip size under the same current loading condition, and the number of recombination luminescence photons decreases correspondingly, which is shown as the spectral peak decreases gradually from sample A to sample E. It is believed that the thermal effect of PN junction is the main reason for the shift of the emission spectrum of red devices to long wavelength. The increase of the forward current of the red LED makes the FWHM of the device wider by 1 nm with the increasing current. The main reasons include that the junction temperature and the lattice vibration increase with the increase of numbers of carriers. The electrons in the excited state first transition to a higher vibrational state, and then return to the ground state, making the emission spectrum wider. Secondly, the increasing carrier density leads to the band gap contraction effect and the band filling effect, which makes the emission spectrum broaden.
As shown in Fig. 10(a), the current efficiency gradually increases with the increase of current density in the low current density working range for samples of different sizes, but the increasing rate decreases gradually. The current efficiency of 50 × 60 µm2 chip can reach 3.86 cd/A when the current density is 1 A/cm2, which is much higher than 0.71 cd/A of 20 × 30 µm2 chip. That is, the larger the area size of the effective light-emitting area of the chip, the larger the corresponding current efficiency value.
Fig. 10. Size-dependent characteristics of current efficiency versus (a) the current density (b) luminance.
At a lower injection current density, the ratio of the side wall circumference to the mesa area of the 20 × 30 µm2 is 6.3 times than 50 × 60 µm2 chip. The non-radiative recombination of carriers caused by the sidewall defects plays a major role in the current efficiency, so the luminous efficiency of the small-size chip which accounts for a larger proportion is lower under the low current density. As can be seen in Fig. 10, the current efficiency gradually decreases from the light emitting array samples A to E. The traditional ABC model equation is used to describe the carrier recombination in the active region. Each recombination mechanism occurs in multiple quantum wells. Photon radiation efficiency ηRad is defined as the ratio of the number of radiation recombination photons to the number of electrons in the injection region, as shown in formula (8), [23,24],
where A, B and C are the corresponding coefficients of Shockly-Read-Hall (SRH) recombination, radiation recombination and auger recombination. The influence of An term on radiation efficiency is dominant in the case of low carrier concentration, and the term of Cn3 plays a more critical role in the current efficiency after the carrier concentration exceeds a certain value. Both of them are the factors leading to the reduction of radiation efficiency. Theoretically, there is an optimal value nopt of carrier concentration to maximize the current efficiency. Similar to the changes relationship between current density and current efficiency, the changes relationship between brightness and current efficiency still shows the larger the size of the light emitting device, the higher the current efficiency, as shown in Fig. 10(b).3.5 Size-dependent relationship between brightness and power consumption
The power consumption of the array under analog driving and digital driving is compared. The light-emitting display is to control the brightness of the devices by using thin film transistors and capacitive storage signals. For analog driving, different Vdata corresponds to different Vgs, and the driving thin film transistor (TFT) converts Vdata into corresponding different current Ids to drive LEDs with different brightness. In the digital driving scheme, each pixel is connected to a switching TFT, which is only used as an analog switch. The gray scale is generated by duty cycle and area ratio gray scale, or a combination of the two. It can also be simply understood that under the fixed current condition, the driving TFT realizes different brightness by controlling the light emission time (time duty ratio).
Fig. 11(a) shows the power consumption distribution of array devices with different sizes at 150, 500, 1000, 1500 and 2000 nit under the state of analog driving. The power consumption of the device increases with the reduction of the product size under any same brightness. For the sample of the same size, the power consumption of the device increases gradually with the increase of brightness. The higher the voltage under the same current condition for smaller size, and the higher current required by the small-size device under the same brightness requirement, and the greater the total power consumption. In addition, the higher the brightness required by chips of the same size, the higher the corresponding current and voltage values. Fig. 11(b) shows the power consumption of devices with different sizes in the digital driving mode. The same driving current corresponds to a certain brightness of different sizes. According to the highest gray-scale comparison of the time duty ratio of 100%, the power consumption of small-sized devices is higher under any brightness (current), and the change trend is similar to that in the analog driving mode. However, the absolute value of power consumption difference under digital driving is smaller than that of the former. The increase of power consumption of sample E relative to that of sample A is 14.1% under 3 mA, while the power consumption of sample E corresponding to 2000nit is 78.7% higher than that of sample A under the analog driving mode.
4. Summary
In this paper, the red Micro-LED bonded on sapphire by adhesive layer is deeply studied and analyzed. The results show that the laser lift-off technology mainly uses the pulse laser with high energy density to act on the interface between the organic adhesive layer and the sapphire substrate. The mechanism of thermal decomposition of the organic adhesive layer after laser irradiation is analyzed in detail, and the thermal decomposition temperature of 450°C solved by the established one-dimensional model is highly consistent with the inherent decomposition temperature of PI material. In addition, the relationship and difference between EL and PL as two detection methods are discussed in depth. The spectral intensity of photoluminescence is higher, and the peak wavelength is red-shifted by about 2 nm under the same excitation conditions. The results of size dependent device optical-electric characteristics show that the smaller the device size, the lower the luminous efficiency under the same display resolution and PPI conditions, and the higher the corresponding display power consumption.
Funding
National Natural Science Foundation of China joint fund for regional innovation and development (No.U20A6004); National Natural Science Foundation of China (No. 51975594).
Disclosures
There are no conflicts of interest related to this article.
Data availability
Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.
References
1. S. C. Han, C. C. Xu, H. J. Li, S. G. Liu, H. W. Xu, Y. K. Zhu, A. L. Fang, and X. Y. Wang, “AlGaInP-based Micro-LED array with enhanced optoelectrical properties,” Opt. Mater. (Amsterdam, Neth.) 114, 110860 (2021). [CrossRef]
2. E.-L. Hsiang, Z. Yang, Q. Yang, Y.-F. Lan, and S.-T. Wu, “Prospects and challenges of mini-LED, OLED, and micro-LED displays,” J. Soc. Inf. Disp. 29(6), 446–465 (2021). [CrossRef]
3. Y. Liu, K. Zhang, F. Feng, K.-W. Chan, S.-Y. Yeung, H.-S. Kwok, and Z. Liu, “The size and temperature effect of ideality factor in GaN/InGaN multiple quantum wells micro-light-emitting diodes,” J. Soc. Inf. Disp. 29(12), 948–960 (2021). [CrossRef]
4. Y. Boussadi, N. Rochat, J. P. Barnes, B. B. Bakir, and C. J. J. o. L. Licitra, “Investigation of sidewall damage induced by reactive ion etching on AlGaInP MESA for micro-LED application,” J. Lumin. 234, 117937 (2021). [CrossRef]
5. X. Zhang, L. Qi, W. C. Chong, P. Li, C. W. Tang, and K. M. Lau, “Active matrix monolithic micro-LED full-color micro-display,” J. Soc. Inf. Disp. 29(1), 47–56 (2021). [CrossRef]
6. J. Day, J. Li, D. Y. C. Lie, C. Bradford, J. Y. Lin, and H. X. Jiang, “Full-Scale Self-Emissive Blue and Green Microdisplays Based on GaN Micro-LED Arrays,” in Conference on Quantum Sensing and Nanophotonic Devices IX, (Proceedings of SPIE2012).
7. X. Ji, F. Wang, P. Lin, L. Yin, and J. Zhang, “Fabrication and Mechanical Properties Improvement of Micro Bumps for High-Resolution Micro-LED Display Application,” IEEE. T. Electron. Dev. 69(7), 3737–3741 (2022). [CrossRef]
8. L. Chaudet, M. Neil, P. Degenaar, K. Mehran, R. Berlinguer-Palmini, B. Corbet, P. Maaskant, D. Rogerson, P. Lanigan, E. Bamberg, and B. Roska, “Development of Optics with Micro-LED Arrays for Improved Opto-electronic Neural Stimulation,” in Conference on Optogenetics - Optical Methods for Cellular Control, (Proceedings of SPIE2013).
9. C.-M. Kang, D.-J. Kong, J.-P. Shim, S. Kim, S.-B. Choi, J.-Y. Lee, J.-H. Min, D.-J. Seo, S.-Y. Choi, and D.-S. Lee, “Fabrication of a vertically-stacked passive-matrix micro-LED array structure for a dual color display,” Opt. Express 25(3), 2489–2495 (2017). [CrossRef]
10. Z. Liu, J. Li, and X. Liu, “Novel Functionalized BN Nanosheets/Epoxy Composites with Advanced Thermal Conductivity and Mechanical Properties,” ACS Appl. Mater. Interfaces 12(5), 6503–6515 (2020). [CrossRef]
11. S.-C. Hsia, M.-H. Sheu, and B.-Y. Chen, “Area-efficient multi-channel active matrix micro-LED driver chip design,” Analog. Integr. Circ. S. 111(1), 137–152 (2022). [CrossRef]
12. Y. Chen, B. Xie, J. Long, Y. Kuang, X. Chen, M. Hou, J. Gao, S. Zhou, B. Fan, Y. He, Y.-T. Zhang, C.-P. Wong, Z. Wang, and N. Zhao, “Interfacial Laser-Induced Graphene Enabling High-Performance Liquid-Solid Triboelectric Nanogenerator,” Adv. Mater. (Weinheim, Ger.) 33, 2104290 (2021). [CrossRef]
13. W. Tian, Y. Wu, T. Wu, L. Dou, X. Cao, and J. Li, “Mechanisms and Performance Analysis of GaN-Based Micro-LED Grown on Pattern Sapphire Substrate by Laser Lift-Off Process,” Ecs. J. Solid. State. Sc. 11(4), 046001 (2022). [CrossRef]
14. Y. Zhao, J. Liang, Q. Zeng, Y. Li, P. Li, K. Fan, W. Sun, J. Lv, Y. Qin, Q. Wang, J. Tao, and W. Wang, “2000 PPI silicon-based AlGaInP red micro-LED arrays fabricated via wafer bonding and epilayer lift-off,” Opt. Express 29(13), 20217–20228 (2021). [CrossRef]
15. Z. Pan, C. Guo, X. Wang, J. Liu, R. Cao, Y. Gong, J. Wang, N. Liu, Z. Chen, L. Wang, M. Ishikawa, and Z. Gong, “Wafer-Scale Micro-LEDs Transferred onto an Adhesive Film for Planar and Flexible Displays,” Adv. Mater. Technol. 5, 2000549 (2020). [CrossRef]
16. C.-F. Lin, J.-J. Dai, M.-S. Lin, K.-T. Chen, W.-C. Huang, C.-M. Lin, R.-H. Jiang, and Y.-C. Huang, “An AlN Sacrificial Buffer Layer Inserted into the GaN/Patterned Sapphire Substrate for a Chemical Lift-Off Process,” Appl. Phys. Express 3(3), 031001 (2010). [CrossRef]
17. K. Kim, G. Jung, J. Kim, Y. Sung, J. Kang, W.-J. Lee, Y. Moon, T. Jeong, and J.-H. Song, “Correlation between photoluminescence and electroluminescence in GaN-related micro light emitting diodes: Effects of leakage current, applied bias, incident light absorption and carrier escape,” Opt. Mater. (Amsterdam, Neth.) 120, 111448 (2021). [CrossRef]
18. C. Tian, W. Wang, J. Liang, Z. Liang, Y. Qin, and J. Lv, “Theoretical and experimental analysis of AlGaInP micro-LED array with square-circle anode,” AIP Adv. 5(4), 041309 (2015). [CrossRef]
19. H.-K. Sung, C. Wang, H.-K. Lee, and N.-Y. Kim, “Enhanced Performance of Large-Area Vertical Light-Emitting Diodes Treated by Laser Irradiation,” J. Nanosci. Nanotechnol. 17(5), 3443–3448 (2017). [CrossRef]
20. K. Wapenaar, “Retrieving the elastodynamic Green's function of an arbitrary inhomogeneous medium by cross correlation,” Phys. Rev. Lett. 93(25), 254301 (2004). [CrossRef]
21. P. Strak, K. Koronski, K. Sakowski, K. Sobczak, J. Borysiuk, K. P. Korona, P. A. Drozdz, E. Grzanka, M. Sarzynski, A. Suchocki, E. Monroy, S. Krukowski, and A. Kaminska, “Instantaneous decay rate analysis of time resolved photoluminescence (TRPL): Application to nitrides and nitride structures (vol 823, 153791, 2020),” J. Alloys Compd. 823, 153791 (2020). [CrossRef]
22. N. S. Wingreen, “Quantum many-body effects in a single-electron transistor,” Science 304(5675), 1258–1259 (2004). [CrossRef]
23. F. Olivier, A. Daami, C. Licitra, and F. Templier, “Shockley-Read-Hall and Auger non-radiative recombination in GaN based LEDs: A size effect study,” Appl. Phys. Lett. 111(2), 022104 (2017). [CrossRef]
24. J.-Y. Chang, Y.-H. Shih, M.-F. Huang, F.-M. Chen, and Y.-K. Kuo, “Shockley-Read-Hall and Auger Recombination in Blue InGaN Tunnel-Junction Light-Emitting Diodes,” Phys. Status Solidi A 215(21), 1800271 (2018). [CrossRef]
