Search for: All records

Abstract This work reports the fabrication and characterization of multifunctional, nanostructured passivation layers formed using a self-assembly process that provide both surface passivation and improved light trapping in crystalline silicon photovoltaic (PV) cells. Scalable block copolymer self-assembly and vapor phase infiltration processes are used to form arrays of aluminum oxide nanostructures (Al 2 O 3 ) on crystalline silicon without substrate etching. The Al 2 O 3 nanostructures are characterized using scanning electron microscopy (SEM), transmission electron microscopy (TEM), and spectroscopic ellipsometry. Injection-level dependent photoconductance measurements are used to determine the effective carrier lifetime of the samples to confirm the nanostructures successfully passivate the Si surface. Finite element method simulations and reflectance measurement show that the nanostructures increase the internal rear reflectance of the PV cell by suppressing the parasitic optical losses in the metal contact. An optimized morphology of the structures is identified for their potential use in PV cells as multifunctional materials providing surface passivation, photon management, and carrier transport pathways. 

In this article, we propose a deep learning based semantic segmentation model that identifies and segments defects in electroluminescence (EL) images of silicon photovoltaic (PV) cells. The proposed model can differentiate between cracks, contact interruptions, cell interconnect failures, and contact corrosion for both multicrystalline and monocrystalline silicon cells. Our model utilizes a segmentation Deeplabv3 model with a ResNet-50 backbone. It was trained on 17,064 EL images including 256 physically realistic simulated images of PV cells generated to deal with class imbalance. While performing semantic segmentation for five defect classes, this model achieves a weighted F1-score of 0.95, an unweighted F1-score of 0.69, a pixel-level global accuracy of 95.4%, and a mean intersection over union score of 57.3%. In addition, we introduce the UCF EL Defect dataset, a large-scale dataset consisting of 17,064 EL images, which will be publicly available for use by the PV and computer vision research communities. 

The intrinsic and doped amorphous silicon layers in silicon heterojunction solar cells parasitically absorb light in the short wavelength region of the solar spectrum, lowering the generation current available to the device. Herein, a promising alternative to the hole‐selective amorphous silicon contact layers using only wide bandgap, transparent oxide materials is presented. Using thermal atomic layer deposition, a 1 nm hydrogenated aluminum oxide layer is deposited followed by a 4 nm molybdenum oxide layer on n‐type crystalline silicon. This contact stack provides an effective carrier lifetime of 1.14 ms. It is shown that the molybdenum oxide layer is successfully deposited with a high work function, which facilitates efficient hole extraction and repels majority carriers from the c‐Si surface. Then the implied open‐circuit voltage, saturation current density, and contact resistivity are recorded as a function of contact annealing temperature and show that they are relatively stable up to 200 °C.