Periodic Reporting for period 1 - IDOL (Inverse Design of Optoelectronic Phosphosulfides)
Berichtszeitraum: 2023-01-01 bis 2025-06-30
Transitioning to a sustainable energy economy is perhaps the most pressing challenge of our generation. The optoelectronics field includes many types of devices which are crucial for sustainable energy conversion: Photovoltaic cells, photoelectrochemical cells, and light-emitting diodes. Importantly, all these technologies have one common requirement on their constituent materials. They all need a wide band gap semiconductor with high luminescence yield, long minority carrier lifetime, high minority carrier mobility, and high optical absorption coefficient.
From a scientific standpoint, the yet-unanswered question is: How can we find the ultimate wide band gap semiconductor from a pool of (at least) a trillion chemically plausible materials? This can be considered an inverse materials design problem. The goal of the IDOL project is to significantly advance our scientific understanding and actual ability to inversely design high-quality semiconductors for optoelectronics. This problem is tackled by combining experimental and computational methods, as well as “high-throughput” and “deep-dive” approaches depending on the specific tasks. The IDOL project focuses on inorganic phosphosulfides as a pool of candidate semiconductors. This family of materials is relevant, earth-abundant, and almost completely unexplored in the (optoelectronics-relevant) thin-film form.
From a scientific standpoint, the yet-unanswered question is: How can we find the ultimate wide band gap semiconductor from a pool of (at least) a trillion chemically plausible materials? This can be considered an inverse materials design problem. The goal of the IDOL project is to significantly advance our scientific understanding and actual ability to inversely design high-quality semiconductors for optoelectronics. This problem is tackled by combining experimental and computational methods, as well as “high-throughput” and “deep-dive” approaches depending on the specific tasks. The IDOL project focuses on inorganic phosphosulfides as a pool of candidate semiconductors. This family of materials is relevant, earth-abundant, and almost completely unexplored in the (optoelectronics-relevant) thin-film form.
We have designed, purchased, and commissioned a unique glovebox-integrated suite of thin-film growth setups that combine high experimental throughput and chemical breadth, including access to sulfur, phosphorus, and most metals in the periodic table. This setup has enabled us to grow some of the first phosphosulfide thin films ever reported. We have conducted high-throughput combinatorial materials discovery campaigns in three ternary phosphosulfide systems: Cu-P-S, Sb-P-S, and Ba-P-S. On the modeling/computation side, we have conducted high-throughput screening of about 1000 unique ternary phosphosulfides by density functional theory.
We have been able to synthesize a number of single-phase thin-film phosphosulfides, among which Cu3PS4. This compound is a 2.5 eV semiconductor. We measured rather high absorption coefficient, carrier mobilities, and carrier lifetimes in our synthesized samples. Thus, Cu3PS4 could find applications in LEDs and photoelectrochemical cells. Simultaneously, we have studied the properties of Cu3PS4 with high-level first-principles computational methods to understand the origin of the favorable properties found by experiment.
To make our high-throughput experimental and computational data findable, accessible, interoperable, and reusable, we have developed a cloud-based FAIR data infrastructure that will be the foundation of the inverse design process proposed in this project. The infrastructure is based on a local customized version of the public NOMAD database. The data is stored in the local database until it is validated and published in scientific journals, at which stage it can easily be pushed to the public database. The hybrid experimental/computational database will be an essential tool to apply artificial intelligence techniques to establish relationships within the data collected in this project and aid the inverse design process.
We have been able to synthesize a number of single-phase thin-film phosphosulfides, among which Cu3PS4. This compound is a 2.5 eV semiconductor. We measured rather high absorption coefficient, carrier mobilities, and carrier lifetimes in our synthesized samples. Thus, Cu3PS4 could find applications in LEDs and photoelectrochemical cells. Simultaneously, we have studied the properties of Cu3PS4 with high-level first-principles computational methods to understand the origin of the favorable properties found by experiment.
To make our high-throughput experimental and computational data findable, accessible, interoperable, and reusable, we have developed a cloud-based FAIR data infrastructure that will be the foundation of the inverse design process proposed in this project. The infrastructure is based on a local customized version of the public NOMAD database. The data is stored in the local database until it is validated and published in scientific journals, at which stage it can easily be pushed to the public database. The hybrid experimental/computational database will be an essential tool to apply artificial intelligence techniques to establish relationships within the data collected in this project and aid the inverse design process.
The material growth equipment designed for this project - now operational - has several features that go beyond the state of the art. (1) Simultaneous access to both phosphorus and sulfur sources; (2) Combination of reactive sputtering, evaporation, and chemical vapor deposition; (3) Access to highly air-sensitive, high vapor-pressure metals.
We have now shown that it is possible to synthesize single-phase thin-film compounds from the pool of inorganic phosphosulfides, and that some of these compounds have remarkably high optoelectronic quality.
The newly developed data infrastructure that combines high-throughput experimental and computational data is a relatively new direction. If it turns out to be a powerful tool for data sharing and artificial intelligence, it has the potential to be widely adopted in the materials community.
Finally, we published an analytical expression for a figure of merit for photovoltaic materials. This figure of merit allows researchers to evaluate the quality of any optoelectronic material based on eight material properties that can be experimentally and computationally determined. It is a substantial generalization compared to simple figure of merits that have previously been proposed, and it will be instrumental in the inverse design process of the present project.
We have now shown that it is possible to synthesize single-phase thin-film compounds from the pool of inorganic phosphosulfides, and that some of these compounds have remarkably high optoelectronic quality.
The newly developed data infrastructure that combines high-throughput experimental and computational data is a relatively new direction. If it turns out to be a powerful tool for data sharing and artificial intelligence, it has the potential to be widely adopted in the materials community.
Finally, we published an analytical expression for a figure of merit for photovoltaic materials. This figure of merit allows researchers to evaluate the quality of any optoelectronic material based on eight material properties that can be experimentally and computationally determined. It is a substantial generalization compared to simple figure of merits that have previously been proposed, and it will be instrumental in the inverse design process of the present project.