Ecodesign of Kesterite Nanoparticles for Thin Film Photovoltaics at Laboratory Scale

This manuscript investigates the efficient synthesis of copper zinc tin sulfide (CZTS) nanoparticles for CZTS thin film solar cell applications with a primary focus on environmental sustainability. Underpinning the investigation is an initial life-cycle assessment (LCA) analysis. This LCA analysis is conducted to evaluate the environmental impact of different synthesis volumes, providing crucial insights into the sustainability of the synthesis process by considering the flows of material and energy associated with the process. Life-cycle assessment results demonstrate that significant reductions to the environmental impact can be made by increasing the synthesis volume of CZTS nanoparticle ink. Using a 5-fold increase in volume can reduce all 11 investigated environmental impacts by up to 35.6%, six of these impacts demonstrating reductions >10% and the amount of global warming potential is reduced by 21.4%. Motivated by the LCA results, COMSOL simulations are employed to understand the fluid flow patterns in large-scale fabrication. Various sizes and speeds of stirrer bars are investigated in these simulations, and it is determined that a 50 mm stir bar at 200 rpm represents the optimal configuration for the synthesis process in a 500 mL flask. Subsequently, large-batch CZTS nanoparticle inks are synthesized using these parameters and compared to small-batch samples. The light absorbers are characterized using Raman spectroscopy and X-ray diffraction, confirming favorable properties with close-to-ideal elemental ratios in large-batch synthesis. Finally, solar cell devices fabricated utilizing CZTSSe absorbers from the large volume synthesis process demonstrate comparable performance to those fabricated using small-batch synthesis, with uniform power conversion efficiencies of around 5% across the substrate. This study highlights the potential of large-volume CZTS nanoparticle synthesis for efficient and environmentally friendly CZTS solar cell fabrication, contributing to the advancement of sustainable renewable energy technologies.

Table S6: LCA Glossary Cradle-to-Gate Considers the life cycle stages associated with the mining of the raw materials, transport to the manufacturing location and the manufacturing of the product -in this case the nanoparticle ink.

Goal and Scope
The introduction to the LCA which provides answers to the who, what, where, when, why and how of the LCA study.

Global market
These are types of inputs taken from the Ecoinvent database which take the average of input and output emissions from all over the world for each input and include the transportation of these inputs.

CML-IA
A type of LCA model which investigates a certain set of environmental impact categories, different models investigate different impact categories.Each of the impact categories associated with this method are explained in more detail below.

Eu25+3 2000
The characterisation method is used to group the impact of different materials and emissions relative to a reference substance.For example, Global warming potential (GWP) includes CO2, CO, CH4, N2O, etc which are all expressed in terms of kg CO2 equivalent.

Abiotic depletion
Measured in kg Sb eq.(kilograms of antimony equivalent), describes how the quantity of raw resources (not fossil fuels) is depleted as a result of the product/process being investigated.More scarce raw materials will have a greater contribution to this impact category than more abundant materials.Abiotic depletion (fossil fuels) Measured in MJ (megajoules), describes how the quantity of fossil fuel resources extracted from the Earth are depleted.

Global warming potential
Measured in kg CO2 eq.(kilograms of carbon dioxide equivalent), describes the quantity of greenhouse gasses released into the atmosphere with a time horizon of 100 years (which is why it is expressed as GWP100a in data tables and figures).Ozone layer depletion Measured in kg CFC-11 eq.(kilograms of trichlorofluoromethane equivalent), relates to the emissions that are released which cause damage and degradation to the stratospheric ozone layer resulting in increased UV-B radiation reaching the Earth's surface.This radiation is known to cause harm to plants, humans and animals.CFCs are one of the most familiar chemicals when discussing ozone layer depletion.

Human toxicity
Measured in kg 1,4-DB eq.(kilograms of 1,4-dichlorobenzene equivalent), describe the fate, exposure and effects of toxic matter on humans from emissions and substances.Freshwater aquatic ecotoxicity Measured in kg 1,4-DB eq.(kilograms of 1,4-dichlorobenzene equivalent), describes the fate, exposure and effects of toxic matter on freshwater ecosystems due to emissions and contamination of substances to the air, soil and water.Marine aquatic ecotoxicity Measured in kg 1,4-DB eq.(kilograms of 1,4-dichlorobenzene equivalent), describes the fate, exposure and effects of toxic matter on marine ecosystems due to emissions and contamination of substances to the air, soil and water.Terrestrial ecotoxicity Measured in kg 1,4-DB eq.(kilograms of 1,4-dichlorobenzene equivalent), describes the fate, exposure and effects of toxic matter on terrestrial ecosystems due to emissions and contamination of substances to the air, soil and water.Photochemical oxidation Measured in C2H4 eq.(kilograms of ethylene equivalents), relates to the quantity of reactive substances forming in the atmosphere as a result of released emissions.An example of this is "summer smog".This can affect humans, plants and animals.

Acidification
Measured in kg SO2 eq.(kilograms of sulphur dioxide equivalent), is related to the fate and increased deposition of acidic matter in air, water and land due to emissions.This can negatively affect materials, land and aquatic ecosystems.Eutrophication (sometimes known as nutrification) Measured in kg PO4-eq.(kilogram of phosphate equivalent), describes the quantity of excess nutrients in the environment due to emissions contaminating air, water and soil.These excess nutrients cause excessive growth of plants and algae resulting in oxygen depletion in the ecosystem and death of organisms making the ecosystem inhabitable.
Table S7: Cost analysis of small batch vs large batch synthesis.

Figure S1 :
Figure S1: Hot Injection schematic showcasing the experimental setup used in both 100 ml and 500 ml volume synthesis.

Figure S2 :
Figure S2: Normalised impact assessment results showing ± 5% variation in input data used in the LCA for small batch and large batch nanoparticle synthesis reactions.

Figure S3 :
Figure S3: Shear rate along a line indicated in red in the fluid of the large 500 ml flask (inset) over a range of magnetic stirrer speeds from 100 to 200 rpm.

Figure
Figure S4: 2-D simulation of the fluid surface in (a) 500 ml flask with 50 mm stirrer and (b) 100 ml flask with 20 mm stirrer.The shear rate is reasonably uniform over the fluid surface in the 500 ml flask and approximately double the magnitude calculated for the 100 ml flask.Corresponding shear rate and turbulent dynamic viscosity are shown as a function of arc length for the (c) 500 ml and (d) 100 ml batches.

Figure
Figure S5: X-ray diffractogram showing the relative peak intensities of the selenised CZTSSe crystalline phases found in small and large batch synthesised.XRD diffractograms show the main peak phases at miller indices (112), (220) and (312) against the reference pattern for pure CZTSe (PDF 052-0868) with comparable intensities between 100 ml and 500 ml synthesised inks.Square root subplots of a) 100 ml synthesis and b) 500 ml synthesis between 2θ angles 25 • and 37 • highlight smaller key peaks, which match with either CZTSSe or MoSe 2 diffraction patterns.It is worth noting that both samples have a low intensity peak at 35.5 • corresponding to the Wurtzite phase of CZTSSe(Zhang et al., 2018, Scientific Reports).

Figure S7 :
Figure S7: Current Voltage characteristic of a champion cell from the large synthesis

Table S1 :
Showing the input materials for the small batch and large batch for the LCA.

Table S2 :
LCA raw results for Small Batch vs Large Batch for the nanoparticle synthesis (part 1).

Table S3 :
LCA raw results for Small Batch vs Large Batch for the nanoparticle ink synthesis (part 1 & 2).

Table S4 :
LCA raw results for different electricity mixes for the nanoparticle ink synthesis.

Table S5 :
XRD psuedo-Voigt fit parameters determined for the 3 main CZTS peaks for small batch and large batch.The fitted parameters are: A -area, Γ L -Lorentzian FWHM, Γ G -Gaussian FWHM, Γ -FWHM, η -mixing parameter (1.0 being a fully Lorentzian profile, 0.0 fully Gaussian profile), β -Integral Breadth, Domain size -Determined with the Scherrer equation, Strain -determined using the Wilson equation.
The disposal of 1 lite of CZTS waste would be £3.00.105 mL waste and 525 mL waste are generated in for small scale and large scale respectively.Hazardous waste is collected monthly.The costs of the monthly collection are as follows: Transport £100; Administration £65;