Metal production requirements for rapid photovoltaics deployment

Abstract

If global photovoltaics (PV) deployment grows rapidly, the required input materials need to be supplied at an increasing rate. In this paper, we quantify the effect of PV deployment levels on the scale of metals production. For example, we find that if cadmium telluride {copper indium gallium diselenide} PV accounts for more than 3% {10%} of electricity generation by 2030, the required growth rates for the production of indium and tellurium would exceed historically-observed production growth rates for a large set of metals. In contrast, even if crystalline silicon PV supplies all electricity in 2030, the required silicon production growth rate would fall within the historical range. More generally, this paper highlights possible constraints to the rate of scaling up metals production for some PV technologies, and outlines an approach to assessing projected metals growth requirements against an ensemble of past growth rates from across the metals production sector. The framework developed in this paper may be useful for evaluating the scalability of a wide range of materials and devices, to inform technology development in the laboratory, as well as public and private research investment.

The rising star in photovoltaics-perovskite solar cells: The past, present and future

“…Chinese producers started to join in the supply chain since 2008 and played a very important role in the subsequent price sharp decline. It is possible for c-Si PVs to further decrease the price to some extent through commercial competition and technical improvement, however, the energy-intensive manufacturing processes for high purity (>99.9995%) silicon ingots and solar panels will be the serious challenges [4][5][6][7]. The second generation of thin film PVs, including cadmium telluride (CdTe) and copper indium gallium selenide (CIGS) solar cells, shared roughly 10% of the PV market.…”

Supply risks associated with CdTe and CIGS thin-film photovoltaics

“…(The latter have recently been described by Jean et al [3].) According to the study of Kavlak et al [47], the total deployment level of CdTe and CIGS modules could only reach 3% and 10%, respectively, of global electricity generation by 2030, if the historically observed 14.7% annual growth rate for all metals were to be reached. Jean et al [3] estimated that for tellurium in CdTe it would require 1500 years at current production rates to reach a deployment level of 25 TWp (corresponding to 100% electricity production by the year 2050).…”

“…Cd, Te and Mo were not considered as essential for future technologies in that study. Nevertheless, these metals are also characterized by increasing production volumes; a lower boundary for future technology demand can be estimated in accordance with Kavlak et al [47] based on historic production statistics. As the units for the expert opinion on ''substitutability" are arbitrary, the results are displayed on a scale from 0 to 100.…”

“…Moreover, according to the assessment of Jean et al, the host metals considered (Si, Ag, Cu, S, Zn, Pb, Sn) are far less subject to constraints [3]. Kavlak et al [47] go into greater detail on this point, showing that the increase in production of In, Ga, Se, Cd and Te required to match global PV deployment targets (e.g., reaching 8% of global electricity generation by 2030) would vastly exceed historically observed metal production growth rates. In particular, global tellurium production would need to grow by 23% per year, in contrast to an historical annual production rate Table 1 Module production and best module efficiency 2010-2015.…”

“…Several authors have recently considered thin-film CdTe and CIGS photovoltaics from the point of view of technological relevance [3], environmental impacts [41], demand-and supplyside economics or costs [42][43][44][45][46][47], and materials supply risk [20,[48][49][50][51][52][53]. Graedel and Nuss [50] have made a multi-element, multi-indicator study of supply risk for CdTe and CIGS absorber materials based on their extensive ''criticality" data bank of the elements [18,54,55].…”