Browsing by Author "Lin, Albert S."

Now showing 1 - 10 of 10

A Unified Mathematical Framework for Intermediate Band Solar Cells
(2012)
Lin, Albert S.
;
Fu, Sze-Ming
;
Zhong, Yan-Kai
;
電子工程學系及電子研究所
;
Department of Electronics Engineering and Institute of Electronics
The modeling of intermediate band solar cell has been developed since 90's and continued effort is made to facilitate the realization of this novel device. Two formulation has been used to model the generation recombination rate of IBSC including conventionally available modified-Shockley-Reed-Hall formulation or later proposed IBSC formulation (Luque and Marti, PRL 78 5014). This paper proves that these two formulations are actually mathematically equivalent and actually one can be derived from the other. A unified mathematical framework can thus be established and the conventional drift-diffusion model can thus be employed for modeling novel IBSC with the inclusion of new model for intermediate band carrier transport. The debate whether the addition of impurity atoms would decrease the efficiency by shorter recombination lifetime or increase the efficiency by more absorption is studied, and results confirm that the efficient removal of photo-generated carriers from valence and conduction bands and solar concentration is the key to the success of subbandgap photovoltaics.
Analytics-statistics mixed training and its fitness to semisupervised manufacturing
(2019-08-13)
Parashar, Parag
;
Chen, Chun Han
;
Akbar, Chandni
;
Fu, Sze Ming
;
Rawat, Tejender S.
While there have been many studies using machine learning (ML) algorithms to predict process outcomes and device performance in semiconductor manufacturing, the extensively developed technology computer-aided design (TCAD) physical models should play a more significant role in conjunction with ML. While TCAD models have been effective in predicting the trends of experiments, a machine learning statistical model is more capable of predicting the anomalous effects that can be dependent on the chambers, machines, fabrication environment, and specific layouts. In this paper, we use an analytics-statistics mixed training (ASMT) approach using TCAD. Under this method, the TCAD models are incorporated into the machine learning training procedure. The mixed dataset with the experimental and TCAD results improved the prediction in terms of accuracy. With the application of ASMT to the BOSCH process, we show that the mean square error (MSE) can be effectively decreased when the analytics-statistics mixed training (ASMT) scheme is used instead of the classic neural network (NN) used in the baseline study. In this method, statistical induction and analytical deduction can be combined to increase the prediction accuracy of future intelligent semiconductor manufacturing.
Down Shifted Converssion for Enhanceed HIT SSolar Cell Efficiency
(2017-01-01)
Lin, Albert S.
;
Parashar, Parag
;
Huang, Wei-Ming
;
Huang, Yi-Wen
;
Jian, Ding-Rung
In this work, HIT solar cell is incorporated with the luminescent down-shifting (LDS) mechanism for the incident solar spectrum. The heterojunction with intrinsic thin-layer(HIT) solar cell is highly efficient, low temperature (200 C), low cost, and highly symmetrical device. Nevertheless, the carrier collection efficiency is low for the short wavelength solar photons due to the surface recombination. Here, a thin phosphor layer is spray coated on the top of HIT solar cells for the luminescent down-shifting of high energy photons and thus enhanced external quantum efficiency (EQE) for the device for lambda=300nm-450nm is observed.
Experimental and Theoretical Study of Low-Cost Hydrothermally Grown Nanowire Silicon Solar Cell
(2013)
Li, Hung-Hsien
;
Lin, Albert S.
;
Zhong, Yan-Kai
;
Fu, Sze-Ming
;
Lai, Shih-Yun
The low cost and high absorbance is the two most important considerations for thin-film silicon photovoltaics. In this work, aluminum doped zinc oxide(AZO) nanowire is grown by low cost hydrothermal method and nano-structured amorphous silicon(alpha-Si) solar cell is realized on these AZO nanowire substrate. The experimental result shows superior light trapping property compared to conventional planar structure with 18.83% improvement in photocurrent. Simulation result shows the required thickness for full absorption for high aspect ratio AZO nanowire solar cell is similar to 0.3 square m, much thinner than planar type solar cell. The calculation reveals that solar cell absorbance increases with the nanowire length and packing density, indicating material volume is the first order effect for nanowire solar cells. The proposed low-cost AZO nanowire array amorphous silicon solar cell is very promising for future nanostructured silicon photovoltaics.
Intelligent Manufacturing: TCAD-Assisted Adaptive Weighting Neural Networks
(2018-01-01)
Huang, Chien Y.
;
Fu, Sze M.
;
Parashar, Parag
;
Chen, Chun H.
;
Akbar, Chandni
Using machine intelligence on device and process performance prediction is an emerging methodology in the IC industry. While semiconductor technology computer-aided design (TCAD) has been researched and developed for over 30 years, it should contribute to or be used in conjunction with machine learning algorithms in solution finding procedure. Here, we propose an adaptive weighting neural network (AWNN) model that combines the advantages of statistical the machine learning model and the physical TCAD model. Using aspect ratio dependent etching as an example, our proposed AWNN outperforms conventional artificial neural network in terms of mean square errors in the test set where 5-10 times reduction is observed. The effectiveness of the TCAD AWNN model can be especially effective in the case of sampling over a vast sample space since the under-sampling problem can be compensated by the TCAD model. The large and nearly unbounded sample space is very common in IC technology, where cascaded and repeated process steps exist (similar to 150 process steps and similar to 20 masks for 90-nm CMOS process).
Intelligent Photolithography Corrections Using Dimensionality Reductions
(2019-10-01)
Parashar, Parag
;
Akbar, Chandni
;
Rawat, Tejender S.
;
Pratik, Sparsh
;
Butola, Rajat
With the shrinking of the IC technology node, optical proximity effects (OPC) and etch proximity effects (EPC) are the two major tasks in advanced photolithography patterning. Machine learning has emerged in OPC/EPC problems because conventional optical-solver-based OPC is time-consuming, and there is no physical model existing for EPC. In this work, we use dimensionality reduction (DR) algorithms to reduce the computation time of complex OPC/EPC problems while the prediction accuracy is maintained. Also, we implement a pure machine learning approach where the input masks are directly mapped to the output etched patterns. While one photolithographic mask can generate many experimental patterns at once, our pure ML-based approach can shorten the trial-and-error period in the photolithographic correction. Additionally, we demonstrate the automation in SEM images preprocessing using feature detection, and this facilitates intelligent manufacturing in semiconductor processing. The input vector dimensions are effectively reduced by two orders of magnitude while the observed mean squared error is not affected significantly. The computation runtime is reduced from 4804 s of the baseline calculation to 10 s-200 s The MSE values changed from the baseline 0.037 to 0.037 for singular value decomposition (SVD), to 0.039 for independent component analysis (ICA), and to 0.035 for factor analysis (FA).
Lithographically fabricable, optimized three-dimensional solar cell random structure
(2013-10-01)
Lin, Albert S.
;
Zhong, Yan-Kai
;
Fu, Sze-Ming
;
Tseng, Chi-Wei
;
Lai, Shih-Yun
The optimized random reflector is highly preferred for solar cells due to its superiority over an un-optimized totally random surface and its potential to exceed the Lambertian limit. There are some obstacles to overcome for realizing optimized random reflectors, including the feasibility from the process viewpoint and the intensive computational demand for large-scale random reflector design. Here a binary random grating is proposed which can be easily fabricated using common lithographic techniques. By using a global optimization algorithm and three-dimensional (3D) EMW simulation, the solar cell structure with 4 x 4 quasi-random binary grating can provide 23% higher integrated absorbance than its periodic grating counterpart and 103.5% higher integrated absorbance than a planar cell, approaching the Lambertian limit. Broad-band transmission improvement at short wavelength and a broad-band waveguiding effect at long wavelength is observed for the optimized 3D geometry. Additionally, the optimized random grating surpasses the periodic grating at all incident angles. The absorbance of the large-scale, fully optimized binary pattern can potentially exceed the Lambertian limit while its computational demand is shown to be manageable.
Lithographically-definable Solar Cell Random Reflector using Genetic Algorithm Optimization
(2012)
Lin, Albert S.
;
Fu, Sze-Ming
;
Zhong, Yan-Kai
;
電子工程學系及電子研究所
;
Department of Electronics Engineering and Institute of Electronics
Randomly textured Lambertian surface provides broad band cosine emission and thus is suitable for photovoltaic application. Nonetheless, variation of efficiency and non-optimized nature of randomly textured devices are undesirable. Here it is shown that using genetic algorithm, a 4x4 binary quasi-random grating can provide 23% higher absorption than 2D periodic grating and 103.5% higher than planar cells, approaching Lambertian limit. The improvement is attributed to broad band transmission for high energy photon and broad band waveguiding effect for low energy photons. Large scale fully-optimized binary grating can potentially surpass Lambertian limit due to its optimized nature and should be employed for future thin-film photovoltaic devices to reduce film thickness and cost.
Optimization of plasmonic cavity-resonant multijunction cells
(2013-06-01)
Lin, Albert S.
;
Fu, Sze-Ming
;
Zhong, Yan-Kai
;
Tseng, Chi-Wei
;
Lai, Shih-Yun
Improving spectral photon harvesting is important for thin-film multijunction cells. We show that efficient spectral flux management can be achieved using genetic algorithm-optimized surface plasmon (SP) cavity-resonant type multijunction cells. We also observe that the excitation of the SP quasi-guided mode, Fabry-Perot mode, and SP polariton significantly enhance the photocurrent of multijunction cells. Two types of cavity structures are investigated. For the optimized SP intermediate reflector and bottom-grating cavity, the resonant cavity mode efficiently increases the long-wavelength absorption in the bottom cell by 63.27%, resulting in reduced absorbance asymmetry between the top and the bottom cells. Accordingly, the matched integrated absorbance is increased by 14.92%. For the optimized SP top-and bottom-grating (TBG) cavity, the integrated absorbance and current matching are improved due to the higher transmission through the solar cell front surface and the excitation of the quasi-guided mode with a more localized field in the bottom cell. The matched integrated absorbance is improved by 85.68% for the TBG cavity. (C) 2013 Society of Photo-Optical Instrumentation Engineers (SPIE) [DOI:10.1117/1.OE.52.6.064002]
Transparent planar indium tin oxide for a thermo-photovoltaic selective emitter
(2020-09-01)
Chen, Yu-Bin
;
Parashar, Parag
;
Yang, Yi-Hua
;
Rawat, Tejender Singh
;
Chen, Shih-Wei
Designing an efficient emitter design is an important step for achieving a highly efficient TPV conversion process. Wavelength-selective emissivity, spectra match between the emitter and TPV cells, and high thermal stability are three main characteristics that must be considered before implementing the emitter. In this work, an indium tin oxide (ITO)/sapphire emitter structure is investigated for TPV application over the temperature range from 200 degrees C to 1000 degrees C. A 1-mu m-thick ITO layer is deposited on a 650-mu m-thick sapphire substrate. In addition, 50-nm-thick SiO2 is deposited on top of the ITO to enhance the performance of emitter at high temperatures. High-temperature emissivity and absorptivity measurement of the emitter samples are obtained using FTIR and a Hitachi U-4100 spectrophotometer, respectively. The resultant SiO2/ITO/sapphire/stainless-steel planar emitter structure has selective emission with high emissivity of similar to 0.8 in the 1-1.6 mu m wavelength regime at 1000 degrees C. This emission range lies at the bandgap edge of silicon TPV cells and thus can be used to harness the true potential for making a low-cost thermophotovoltaic system. (C) 2020 Optical Society of America under the terms of the OSA Open Access Publishing Agreement