Milad Abolhasani is an award-winning researcher and ALCOA Professor of Chemical and Biomolecular Engineering at North Carolina State University. He is making an impact in the field of materials discovery. His latest paper, titled “Flow-Driven Data Intensification to Accelerate Autonomous Materials Discovery,” was published in Nature Chemical Engineering and introduces a groundbreaking dynamic flow system that vastly improves data collection and analysis in self-driving laboratories.
This tech-forward method creates more than 10x the data produced by conventional self-driving labs. Unlike those labs, which require steady-state flow experiments, this approach propels powerful outcomes. Through the constant turnover of samples through the system, this dynamic flow configuration collects data as detailed as every 0.5 seconds. With this model, researchers can go straight to the best material candidates on their first try after training. That’s a big jump, and it’s using a fraction of the time and materials that traditional materials research would require.
Breakthrough in Data Collection
Abholhasani’s dynamic flow system is an exciting new model for how self-driving labs should work. In the past, these labs have focused primarily on static flow experiments. The existing standard limited data gathering to individual dots at one specific reaction time. This key constraint drastically impacted the net productivity of the material discovery process.
Abolhasani says that what they ended up building was a system that’s always on. Rather than test individual samples one at a time, this revolutionary method enables round-the-clock operation. This ongoing development and deployment make possible a much richer dataset to be collected.
“For example, instead of having one data point about what the experiment produces after 10 seconds of reaction time, we have 20 data points—one after 0.5 seconds of reaction time, one after 1 second of reaction time, and so on. It’s like switching from a single snapshot to a full movie of the reaction as it happens. Instead of waiting around for each experiment to finish, our system is always running, always learning.” – Milad Abolhasani
This machine learning-enabled capability streamlines selection of the most promising candidate materials for their intended applications. What used to take years can now be done in just weeks or months.
Environmental Impact and Cost Reduction
Beyond improving velocity, Abolhasani’s dynamic flow system directly confronts urgent environmental issues at the same time. The groundbreaking method saves money and lessens the ecological damage by materials experimentation. For starters, by reducing the amount of experiments necessary, the system reduces the use of chemicals as well as chemical waste.
“By reducing the number of experiments needed, the system dramatically cuts down on chemical use and waste, advancing more sustainable research practices,” said Abolhasani.
The impacts of this research go far beyond efficiency. Abolhasani stresses that speed isn’t the sole consideration in the race to discover new materials. We need to make sure we lead with responsibility.
With fewer chemicals required for experimentation, researchers can adopt more sustainable practices while still delivering timely solutions to pressing challenges faced by society.
The Role of Machine Learning
Central to the success of this dynamic flow system is its blending with machine-learning algorithms. These algorithms are the backbone of many predictive models. They decide what experiments to run next by processing the high-quality data that has been collected from new, ongoing experiments.
Abolhasani noted that the machine-learning algorithm they created is at the heart of any self-driving lab. This new algorithm helps predict which experiment the system should run next. With each experimental input, the accuracy of these predictions increases, leading to faster problem solving ability.
“That’s because the more high-quality experimental data the algorithm receives, the more accurate its predictions become, and the faster it can solve a problem. This has the added benefit of reducing the amount of chemicals needed to arrive at a solution,” he added.
This combination of cutting-edge technology and materials science opens an exciting new frontier for scientists. They now have streamlined avenues to explore exciting new materials.