High-throughput screening of polycrystalline solar absorbers (Ext.)

Lead Research Organisation: University of York
Department Name: Physics


This is an extension of the Fellowship: 'Non-equilibrium electron-ion dynamics in thin metal-oxide films' (EP/K003151/1).

The development of low-cost high-efficiency solar cell devices would allow us to make more use of the vast amount of free and clean energy available in sunlight. Materials which absorb light to generate energetic electrons in solar cells are known as solar absorbers. In current consumer level solar cells the solar absorber is crystalline silicon. Silicon based cells exhibit high efficiencies (~25%) but are relatively expensive to produce. For example, it currently takes about 14 years of operation for a typical 4 kW domestic installation to break even (e.g. see http://www.theecoexperts.co.uk/are-solar-pv-panels-good-investment). Driven by the desire to reduce cost there has been a continued focus on the development of new high-efficiency solar absorber materials that are less expensive to manufacture than silicon to form the basis of next generation solar cell technologies.

A general trend in materials development has been the progression from silicon towards more complex binary, ternary and quaternary compound semiconductors, which offer a wider compositional and structural parameter space within which desired properties can be optimised. Highly performing examples include CuInGaSe2, CdTe, Cu2ZnSn(S,Se)4 (CZTS) and lead-halide perovskites (e.g. CH3NH3PbI3, MAPI). Unlike silicon these emerging materials often contain relevantly high concentrations of point defects since they are almost always non-stoichiometric. They are also usually polycrystalline and grain boundaries (together with associated point defects) are known to affect material performance by contributing to non-radiative electron-hole recombination and reduction of open circuit voltage (both effects that reduce efficiency).

While predictive computational materials screening approaches have proved invaluable in helping to identify promising solar absorber materials there are currently no screening approaches that consider the properties of grain boundary defects. This proposal aims to fill this critical gap in the materials modelling toolbox by developing systematic approaches to screen materials against the thermodynamic and electronic properties of grain boundaries. These approaches will be applied to identify optimal compositions and dopants for CdTe, lead-halide perovskites and CZTS materials to help optimise performance and accelerate innovation. We will work closely with experimental collaborators and our industrial partner (Dyesol) to validate theoretical models and test predictions in order to deliver improvement in solar cell performance. The computational screening approaches we develop will also be made available to the wider materials modelling community and will find application in many other areas where the electronic properties of grain boundaries impact on material performance (including thermoelectrics, batteries, photoelectrochemical cells, varistors, transparent conducting oxides and dielectrics to name a few).

Planned Impact

Economy: The discovery and optimisation of new solar absorber materials is a key challenge in the development of next generation photovoltaic technologies that aim to combine high efficiency with low manufacturing costs. Recent years have seen an explosion of new solar absorber materials, with the lead halide perovskites and Cu2ZnSn(S,Se)4 showing particular promise. However, further improvements in the efficiency and stability of these materials are needed before their potential can be realised. The global market for third generation (i.e. non-silicon based) photovoltaics is predicted to reach $38bn by 2022 making this an area with significant potential for economic impact. For example, Dyesol, one of the world's leading alternative photovoltaics manufacturers, is investing AUS $120M into R&D for lead halide perovskites based solar cells (with a new UK laboratory planned). In this project we will partner with Dyesol to test the performance of new absorber materials that emerge from the first principles screening approach. While the focus of this project is on solar absorber materials, the high-throughput screening approaches we will develop for polycrystalline materials will find applications in other economically important technology areas including thermoelectrics, batteries, transparent conducting oxides and dielectrics for microwave application and microelectronics.

Society: Reducing our dependency on fossil fuels and moving towards a sustainable and clean energy infrastructure is essential to ensure the health, security and productivity of our nation. Solar energy is an important resource we can tap into to achieve this goal but currently represents only 2.5% of our total electricity consumption. Decreasing the cost of solar cells by moving to new absorber materials with lower manufacturing costs would increase solar cell uptake for both domestic and commercial energy generation. It would also contribute to reducing carbon dioxide and other harmful emissions, helping us meet international environmental targets and bringing significant health benefits.

Knowledge: Polycrystalline materials are ubiquitous in nature and in technology. Their roles as sinks for defect segregation and diffusion, paths for enhanced electrical conduction, and sites of charge trapping and recombination is often discussed but rarely taken into account in materials design and optimisation. Therefore, the development of a flexible computational methodology for screening the electronic and thermodynamic properties of polycrystalline materials would find application across many areas outside the immediate scope of this project. This includes applications in energy materials, optoelectronic materials and electronics. The new screening tools will be made publically available allowing the wider materials modelling community to benefit from these developments.

People: The project will nurture the development of a postdoctoral researcher assistant (PDRA) and a PhD student. The transferable skills they will develop (including computational, communication, collaborative and management) will be valuable whether they pursue an academic career, take up skilled positions in the wider economy or move into a technological industry. The PDRA and PhD will be involved in conference organisation, outreach and project management providing numerous opportunities for their continued development. As part of the project we will develop an outreach activity targeted towards the general public on next generation solar cell technologies. The aim is to excite, inform and inspire about cutting edge materials research and its role in technology development.


10 25 50
Description The overall aim of this proposal was to develop high-throughput computational methods to screen the thermodynamic and electronic properties of grain boundary defects in solar absorber materials in order to identify materials with optimal performance and stability. To ensure uptake of this approach more widely in high-throughput materials screening we aimed to develop a code to automate the processing and analysis steps involved. This proposal had three main work-packages linked to the primary objectives. Progress against each is reported below:

WP1. Modelling grain boundaries in CdTe, MAPI and CZTS:
Grain boundaries in all three materials above have been simulated each resulting in publications. For example, we investigated the effect of (111) twin defects in mixed lead halide perovskite solar absorbers. We found that these defects are highly stable and in the mixed phase serve as nucleation sites for phase separation reducing open circuit voltage and hence material performance in solar cells [K. P. McKenna, ACS Energy Letters 3, 2663 (2018)]. We also investigated the role of anti-site boundary defects in another solar absorber material CZTS finding that although very stable they are benign for solar cell performance [B. Mendis, K. P. McKenna, M. Rumsey, G. Gurieva, S. Schorr, Journal of Materials Chemistry A 6, 189 (2018)]. We showed how grain boundaries in CdTe can be passivated by coding with Cl and Na [C. Tong, K. P. McKenna, J. Phys. Chem. C 123, 23882 (2019)].

WP2. Development of high-throughput simulation tools:
We have developed codes for high throughput simulation of grain boundary defects and applied them to model grain boundaries in several emerging solar absorber materials including lead halide perovskites, CdTe, Sb2Se3 and CZTS. The GBMaker code which automates construction of grain boundary models has now been released and is openly and freely available to the wider community (https://doi.org/10.5281/zenodo.6334740).

WP3. Application of grain boundary screening to emerging materials:
The approaches developed above have been applied to Sb2Se3 and Sb2S3 uncovering an unusual self-healing property of this material [R. E. Williams et al, ACS Appl. Mater. Interfaces 12, 21730 (2020) & K. P. McKenna, Adv. Electron. Mater. 7, 2000908 (2021)]. Predictions have also been performed for emerging ferroelectric solar absorbers such as CuPbSbS3 and Cu3AsS4 and well as Na2CuP which is predicted to be an earth-abundant high efficiency absorber.
Exploitation Route Predictions could be tested experimentally. We are already working with a number of experimental groups to do this and we are developing follow-on funding applications to build on this initial work. Ultimately, this insight can be used to design improved materials for solar cell applications.
Sectors Energy

URL http://www-users.york.ac.uk/~km816/index.html
Description Collaboration with Greatcell solar 
Organisation Greatcell Solar Ltd
Country Australia 
Sector Private 
PI Contribution We have been meeting regularly with our project partner to discuss emerging materials of interest for next generation solar cells. Based on these discussions we performed calculations of the impact of grain boundaries on mixed lead halide material they are using in their perovskite solar cells [K. P. McKenna, ACS Energy Letters 3, 2663 (2018)]. We showed how phase separation at these tightly stable twin boundaries can lead to reduced performance - in particular a reduction in open circuit voltage.
Collaborator Contribution They have kept up informed about their device designs and materials developments as well as sharing confidential results and open problems of interest.
Impact K. P. McKenna, ACS Energy Letters 3, 2663 (2018)
Start Year 2018