Modification of Mo Back Contact with MoO_3−x Layer and its Effect to Enhance the Performance of Cu₂ZnSnS₄ Solar Cells

Abstract

Kesterite Cu2ZnSnS4 (CZTS) has been investigated intensively as a promising absorber material for thin film solar cells. However, the reported best power conversion efficiency (PCE) is still low due to the intrinsic limitations of CZTS defects and the unfavorable front and back contact interfaces. In this study, an intermediate MoO3−x layer is introduced as a primary back contact prior to the Cd‐doped Cu2ZnSnS4 (CZCTS) absorber layer growth. It has been demonstrated that the insertion of the MoO3−x layer can suppress the decomposition reaction between CZCTS and Mo, reducing secondary phases and voids and improving the rear interface quality. The MoO3−x layer can also accelerate potassium diffusion into the CZCTS layer and interfaces, which facilitates the grain growth, passivates interfaces and hence yields better crystallinity. Moreover, the band alignment at the back contact is modified by the MoO3−x layer, resulting in better minority carrier collection and improvement of open‐circuit voltage (VOC). With the optimal MoO3−x layer, the PCE of CZCTS solar cells has increased from ≈5.43 to ≈7%, largely attributed to the ≈60 mV VOC increment. Through the modification of the CZCTS precursor and optimizing the anti‐reflection coating layer, the best PCE achieved is 8.98%, with an active area efficiency of 9.26%.

Role of ZnS Particles in the Performance of Cu2ZnSnS4 Thin Film Solar Cells: A Comparative Study by Active Control of Zinc Deposition in Coevaporated Precursors

“…The RTP‐treated samples were further annealed to 200 °C on a hotplate in air for 10 min before CdS buffer growth. Using chemical bath deposition (CBD), [ 24 ] an ≈60 nm thick CdS buffer layer was grown onto the CZTS surface, which was subsequently annealed at 270 °C for 10 min in an oven immediately after CdS growth. A 50 nm i‐ZnO window layer and 500 nm AZO front contact layer were finally deposited using RF magnetron sputtering.…”

Enabling fine-grain free 2-micron thick CISe/CIGSe film fabrication via a non-hydrazine based solution processing route

“…Cross-section SEM images of selenized CIS film coated (a) on Molybdenum, (b) directly on glass, (c) on 10 nm MoO3 coated Molybdenum, and (d) on 5 nm MoO3 coated Molybdenum A similar phenomenon was observed in literature for Cu 2 ZnSnSe 4 (CZTSe) thin films fabricated via selenization of sulfide Cu 2 ZnSnS 4 (CZTS) films.A film deposited on Mo substrate grew as 2 layers of grains while the one deposited on glass grew into single grain morphology 20. Other similar studies suggest the use of a thin layer of SiO 2 , TiN, MoO 3 on the Mo surface to change the interface between precursor material and back contact [20][21][22][23]. While no such surface modifications for changing grain morphology were studied in the case of CISe/CIGSe thin films, one study showed that a thin layer of MoO 3 layer between CIGSe absorber and Mo back contact improves the band alignment of CIGSe devices 24.…”

“…20 Other similar studies suggest the use of a thin layer of SiO 2 , TiN, MoO 3 on the Mo surface to change the interface between precursor material and back contact. [20][21][22][23] While no such surface modifications for changing grain morphology were studied in the case of CISe/CIGSe thin films, one study showed that a thin layer of MoO 3 layer between CIGSe absorber and Mo back contact improves the band alignment of CIGSe devices. 24 To study the effect of surface modification on grain morphology and device performance, 10 nm of MoO 3 was deposited on the Mo surface via thermal evaporation at the rate of 0.02-0.03 nm/s.…”

A Universal and Facile Method of Tailoring the Thickness of Mo(Sx,Se1−x)2, Contributing to Highly Efficient Flexible Cu2ZnSn(S,Se)4 Solar Cells

“…[ 28 ] On the other hand, the insertion of a barrier layer on top of Mo is widely adopted to modify the back contact. Barrier layers, such as MoO 2 , [ 29 ] MoO 3 , [ 30,31 ] TiB, [ 32 ] Na 2 S, [ 33 ] carbon, [ 34 ] Al‐doped ZnO (AZO), [ 35 ] TiN, [ 36,37 ] and a temporary alloy layer, [ 38 ] have been demonstrated to be helpful to suppress the Mo(S x ,Se 1− x ) 2 thickness and enhance the device performance. However, these methods unexceptionally need an extra process to introduce the blocking layers and increase the risk of introducing impurity elements into the absorber.…”