The role of oxygen doping on elemental intermixing at the PVD‐CdS/Cu (InGa)Se₂ heterojunction

Abstract

Elemental intermixing at the CdS/CuIn1−xGaxSe2 (CIGS) heterojunction in thin‐film photovoltaic devices plays a crucial role in carrier separation and thus device efficiency. Using scanning transmission electron microcopy in combination with energy dispersive X‐ray mapping, we find that by controlling the oxygen in the sputtering gas during physical vapor deposition (PVD) of the CdS, we can tailor the degree of elemental intermixing. More oxygen suppresses Cu migration from the CIGS into the CdS, while facilitating Zn doping in the CdS from the ZnO transparent contact. Very high oxygen levels induce nanocrystallinity in the CdS, while moderate or no oxygen content can promote complete CdS epitaxy on the CIGS grains. Regions of cubic Cu2S phase were observed in the Cu‐rich CdCuS when no oxygen is included in the CdS deposition process. In the process‐of‐record sample (moderate O2) that exhibits the highest solar conversion efficiency, we observe a ~26‐nm‐thick Cu‐deficient CIGS surface counter‐doped with the highest Cd concentration among all of the samples. Cd movement into the CIGS was found to be less than 10 nm deep for samples with either high or zero O2. The results are consistent with the expectation that Cd doping of the CIGS surface and lack of Zn diffusion into the buffer both enhance device performance.

Review of CIGS-based solar cells manufacturing by structural engineering

“…The CIGS-based solar cells can be fabricated on both rigid and flexible substrates by various vacuum and non-vacuum techniques. For example, co-evaporation (Repins et al, 2008), physical vapor deposition (PVD) (He et al, 2019), pulsed laser deposition (PLD) (Tsai et al, 2013), chemical vapor deposition (CVD) (Park et al, 2003), metalorganic chemical vapor deposition (MOCVD) (Choi and Lee, 2007), electron beam deposition (EBD) (Venkatachalam et al, 2008), molecular beam epitaxy (MBE) (Nakada et al, 1999), and sputtering (Delahoy et al, 2004;Kushiya et al, 2001) are vacuum techniques that can be utilized to fabricate CIGS-based solar cells. CIGS solar cell fabrication starts with the deposition of the electrical back contact layer on the substrates and later finished by the coating of the window layer which is primarily deposited using vacuum deposition methods, while, the n-buffer layer can be coated using both vacuum and non-vacuum techniques (Choi and Lee, 2007;Delahoy et al, 2004;He et al, 2019;Kushiya et al, 2001;Nakada et al, 1999).…”

“…Though high-quality samples can be obtained by vacuum deposition, these approaches have several disadvantages such as high cost and time consumption and poor uniformity over a large area owing to a cosine flux distribution, which results in a sharp change in film composition and lower Se incorporation (Kaelin et al, 2004;Ramanujam and Singh, 2017). Vacuum methods are more efficient if the cells are fabricated without a vacuum break in a full-stack deposition tool (He et al, 2019). However, almost all deposition methods require a post-selenization process (Chen et al, 2017b).…”

“…Indirect doping includes external structural manipulation, which will be discussed in the next section. Direct doping by adding an impurity to CIGS helps develop the CIGS performance; however, this method is less popular than indirect doping facilitated by substrates (i.e., Na doping) (Asaduzzaman et al, 2016;Salom� e et al, 2013;Sun et al, 2017), n-buffer (i.e., Cd doping) (Nanayakkara et al, 2017), and additional layers such as a window layer (i.e., Zn doping) (He et al, 2019). Direct doping can be realized by adding an impurity during or after the CIGS growth.…”

“…Solar cells based on CIGS have high efficiency that is similar to that of crystalline silicon (c-Si) solar cells but are less expensive because CIGS can absorb light using only ~2.0-2.5 mm layer thickness, which decreases the use of raw materials (Guillemoles, 2002;Noufi and Zweibel, 2006;Ramanujam and Singh, 2017). The CIGS-based solar cells are easy to fabricate compared to c-Si based solar cells by growing it on various rigid and flexible substrates by vacuum and non-vacuum techniques; thus, CIGS-based solar cells are promising candidates for the nextgeneration power-efficient solar cells (Adel et al, 2016;Badgujar et al, 2015;Chen et al, 2017aChen et al, , 2014Choi and Lee, 2007;Delahoy et al, 2004;He et al, 2019;Kaelin et al, 2004;Kuo et al, 2016;Kushiya et al, 2001;Lee et al, 2011;Liu et al, 2012;Nakada et al, 1999;Park et al, 2003;Repins et al, 2008;Tsai et al, 2013;Venkatachalam et al, 2008). In addition to CdTe thin films, CIGS is included in the second-generation thin-film solar cells but CIGS is non-toxic compared to CdTe (Noufi and Zweibel, 2006;Ramanujam and Singh, 2017).…”

Alcohol Vapor Post‐Annealing for Highly Efficient Sb2S3 Planar Heterojunction Solar Cells

“…The growing demands on renewable and sustainable energy require to develop cost‐effective, stable, efficient solar cells . Accordingly, a diverse range of semiconductors such as copper zinc tin selenide, copper indium gallium selenide, lead sulfide, organic–inorganic hybrid perovskite, bismuth sulfide, antimony selenide, and antimony sulfide (Sb 2 S 3 ) have been used as solar cell materials and achieved impressive power conversion efficiencies (PCEs). Among these materials, Sb 2 S 3 is regarded as one of the most promising candidates for next‐generation photovoltaic materials due to its high absorption coefficient ( α = 10 5 cm −1 ), suitable band gap (1.50–2.20 eV), abundance of materials, low toxicity, and easy processing …”