Advancements in Catalyst Discovery for Clean Energy
A new project aims to improve catalyst discovery for clean energy production.
Jehad Abed, Jiheon Kim, Muhammed Shuaibi, Brook Wander, Boris Duijf, Suhas Mahesh, Hyeonseok Lee, Vahe Gharakhanyan, Sjoerd Hoogland, Erdem Irtem, Janice Lan, Niels Schouten, Anagha Usha Vijayakumar, Jason Hattrick-Simpers, John R. Kitchin, Zachary W. Ulissi, Aaike van Vugt, Edward H. Sargent, David Sinton, C. Lawrence Zitnick
― 5 min read
Table of Contents
- The Challenge
- Creating Open Catalyst Experiments 2024
- The Experimental Process
- Chemical Reduction
- Spark Ablation
- Testing the Catalysts
- Hydrogen Evolution Reaction (HER)
- Carbon Dioxide Reduction Reaction (CO2RR)
- Bridging the Gap
- Results and Findings
- The Road Ahead
- Conclusion
- Original Source
- Reference Links
The world is facing a serious problem with climate change, and finding better ways to produce clean energy is super important. One of the promising ways to do this is by creating green Hydrogen through a process called electrolysis. However, to make this process really efficient, we need better Catalysts-materials that speed up chemical reactions without being consumed in the process.
Unfortunately, discovering new catalysts has been like looking for a needle in a haystack. There’s a gap between what scientists think will work based on computer models and what actually works in the lab. To bridge this gap, scientists have come up with a big plan called Open Catalyst Experiments 2024, or OCx24 for short.
The Challenge
The current catalyst discovery process is a bit like a game of trial and error. Scientists try different materials based on their knowledge and experience, but this can be slow and has many ups and downs. Different groups of researchers are often working independently, which leads to a lot of overlapping research and not much progress.
A major part of the problem is that experimental results can be hard to reproduce. If one lab finds a promising catalyst, another lab might not be able to get the same results, making it tougher to build on past findings. This is where OCx24 comes in, aiming to create a clear path between lab experiments and computer predictions.
Creating Open Catalyst Experiments 2024
The OCx24 project aims to create a huge dataset filled with experimental studies that can help train computer models. This dataset is expected to help scientists find out which materials are the best candidates for catalysts. The idea is to gather a wide range of data that includes both successful and unsuccessful tests. This should help models get a better understanding of what to look for in new materials.
To achieve this, researchers are using advanced techniques to synthesize new catalyst materials and test them under conditions that resemble real-world industrial processes. For OCx24, they created a dataset with 572 unique samples of catalysts, each made from various combinations of elements.
The Experimental Process
The scientists used two main techniques to create these catalysts: chemical reduction and spark ablation.
Chemical Reduction
This is a wet chemistry method where metal salts are mixed and then reduced using a chemical agent to create nanoparticles. After making the nanoparticles, they are dried and prepared for testing.
Spark Ablation
In this dry method, researchers use sparks to vaporize metal rods and create tiny particles. These particles are then printed onto a substrate, forming a thin layer of nanoparticles. This technique allows for precise control over the composition of the materials.
Both methods have their own challenges, such as ensuring that the catalysts are the right size and composition. Researchers had to be very careful with their methods to avoid issues like oxidation during shipping or inconsistencies in the materials they created.
Testing the Catalysts
Once the catalysts were created, researchers put them through their paces to see how well they performed in electrochemical reactions. They specifically looked at two reactions:
Hydrogen Evolution Reaction (HER)
This reaction generates hydrogen gas, which is a key part of creating green hydrogen. The scientists tested various conditions to find out how effectively each catalyst could produce hydrogen.
Carbon Dioxide Reduction Reaction (CO2RR)
In this reaction, scientists work on converting CO2 into useful products like carbon monoxide or other multi-carbon molecules. The challenge here is to find catalysts that can produce these products efficiently.
The researchers collected data on how much gas was produced during these reactions and how efficiently the catalysts worked. They also used techniques like X-ray fluorescence (XRF) and X-ray diffraction (XRD) to determine the composition and structure of the catalysts.
Bridging the Gap
As part of OCx24, researchers calculated the adsorption energies of different molecules on various surfaces of the catalysts. This helps the scientists understand how well molecules stick to the catalyst surfaces, which is key for improving their performance.
Using advanced computer methods and machine learning, they created models to predict which materials would work best for both HER and CO2RR. Even though the initial models were based on experimental data, they managed to identify some surprising results. For example, they found that platinum, known to be an effective catalyst for hydrogen production, showed up as a strong candidate in their models despite not being included in their training dataset!
Results and Findings
The findings from OCx24 are promising. The dataset provides a robust foundation for researchers to train better models, which in turn can lead to discovering more effective, low-cost catalysts. Hundreds of potential candidates for both hydrogen and carbon reactions were identified, many of which consist of cheaper materials compared to platinum or palladium.
The Road Ahead
The OCx24 project is just the beginning. With more experimental data and improved models, the future looks bright for finding clean energy solutions. By being more systematic and collaborative, researchers hope to pave the way for better catalysts and, ultimately, a greener planet.
Conclusion
In summary, Open Catalyst Experiments 2024 aims to tackle some of the biggest challenges in catalyst discovery with a solid approach that combines experimental work and computational modeling. While the journey is ongoing, the insights gained will undoubtedly help shape the future of clean energy production.
And who knows? Maybe one day, the next big catalyst will come from some unexpected materials, like your grandma's old silverware! So, keep an eye on those drawer treasures; they might just be the key to our clean energy dreams!
Title: Open Catalyst Experiments 2024 (OCx24): Bridging Experiments and Computational Models
Abstract: The search for low-cost, durable, and effective catalysts is essential for green hydrogen production and carbon dioxide upcycling to help in the mitigation of climate change. Discovery of new catalysts is currently limited by the gap between what AI-accelerated computational models predict and what experimental studies produce. To make progress, large and diverse experimental datasets are needed that are reproducible and tested at industrially-relevant conditions. We address these needs by utilizing a comprehensive high-throughput characterization and experimental pipeline to create the Open Catalyst Experiments 2024 (OCX24) dataset. The dataset contains 572 samples synthesized using both wet and dry methods with X-ray fluorescence and X-ray diffraction characterization. We prepared 441 gas diffusion electrodes, including replicates, and evaluated them using zero-gap electrolysis for carbon dioxide reduction (CO$_2$RR) and hydrogen evolution reactions (HER) at current densities up to $300$ mA/cm$^2$. To find correlations with experimental outcomes and to perform computational screens, DFT-verified adsorption energies for six adsorbates were calculated on $\sim$20,000 inorganic materials requiring 685 million AI-accelerated relaxations. Remarkably from this large set of materials, a data driven Sabatier volcano independently identified Pt as being a top candidate for HER without having any experimental measurements on Pt or Pt-alloy samples. We anticipate the availability of experimental data generated specifically for AI training, such as OCX24, will significantly improve the utility of computational models in selecting materials for experimental screening.
Authors: Jehad Abed, Jiheon Kim, Muhammed Shuaibi, Brook Wander, Boris Duijf, Suhas Mahesh, Hyeonseok Lee, Vahe Gharakhanyan, Sjoerd Hoogland, Erdem Irtem, Janice Lan, Niels Schouten, Anagha Usha Vijayakumar, Jason Hattrick-Simpers, John R. Kitchin, Zachary W. Ulissi, Aaike van Vugt, Edward H. Sargent, David Sinton, C. Lawrence Zitnick
Last Update: 2024-11-18 00:00:00
Language: English
Source URL: https://arxiv.org/abs/2411.11783
Source PDF: https://arxiv.org/pdf/2411.11783
Licence: https://creativecommons.org/licenses/by/4.0/
Changes: This summary was created with assistance from AI and may have inaccuracies. For accurate information, please refer to the original source documents linked here.
Thank you to arxiv for use of its open access interoperability.