File Name : fig.s1.tiff

Caption : fig.s1  projections for the aggregated capacity of clean energy technologies added annually based on iams scenarios (table s2). the numbers on top of the bars present the ratio of capacity added in each year over the capacity added in 2020. the error bars present the variations in the capacity added for batteries, depending on the battery characteristics, as discussed in section a.2.2. 

File Name : fig.s2.tiff

Caption : fig.s2 cumulative projected capacity of clean energy technologies based on iams (table s2). for each technology, the projected capacity in 2020 is used as a base to estimate the ratio of projected capacity in 2050 over it, except for the electrolyzers, since their development starts mainly after 2030. in addition, the replaced capacity of each technology is presented in its panel by a dashed area. markers in each panel represent data from additional sources used for comparison: circles show iea projections,14,18,19 plus signs denote irena data,17 and cross signs correspond either to data from the european technology and innovation platform for battery (etipb),7 from statista (for stationary batteries projections),13 or from the energy technology perspectives (etp) (for csp).12 dashed lines present the median of projections. notably, long dash dot lines present the upper bound and lower bound capacity for batteries (more on this can be seen in section a.2.2).

File Name : fig.s3.tiff

Caption : fig.s3  material demands from clean energy technologies. panel (a) compares the demand for materials from cets with their production rate dedicated to all sectors in 2020. panel (b) represents materials demand from cets compared with their estimated demand in 2020. oo: materials primarily used in electrolyzers, does not exhibit demand until 2030 and therefore rely on this year or later years, depending on the scenario, as the basis for calculations. timeframes are indicated as follows: 20: 2020, 30: 2030, 40: 2040, 50: 2050. the results of the continued trend and technological change are distinguished by red and blue colors, respectively. in both cases, the median is depicted with a thicker line. the effect of the reductions in material intensities due to learning curves is indicated with star signs, only for median projections. cets: clean energy technologies.

File Name : fig.s4.tiff

Caption : fig.s4 links between the materials (left-hand side) and clean energy technologies (right-hand side), based on material intensity values in 2050, considering the effect of learning curves. the width of the ribbons is proportional to material intensity values. technologies that penetrate the market after 2020 are distinguished by a star. note that to make all the materials visible and comparable with 2020 values, the materials relative to 2020 intensity values are normalized. for clarity, technology types are enclosed in brackets, with the corresponding acronyms found in table s1. 

File Name : fig.s5.tiff

Caption : fig.s5 materials demand over their reserve capacities. to aid comparison, dashed horizontal lines are included. timeframes are indicated as 20: 2020, 30: 2030; 40: 2040, 50: 2050. the results of the continued trend and technological change are distinguished by red and blue colors, respectively. in both cases, the median is depicted with a thicker line. additionally, the resulting median, accounting for reductions in materials intensity based on learning curves, is indicated by star signs. the stacked bars chart presents the share of different technology types in specific material demands. technology type acronyms as given in table s1.

File Name : fig.s6.tiff

Caption : fig.s6 shortages in technologies annual developed capacity based on different iams scenarios. 

File Name : fig.s7.tiff

Caption : fig.s7: approach used to estimate the replacement energy requirements for each clean energy technology.