A technology has been developed that uses robots rather than humans to evaluate the performance of newly developed catalysts. By operating 45 times faster than manual work while also improving precision, it is expected to significantly shorten catalyst development timelines.
A research team led by Dr. Ji Chan Park of the Clean Fuel Research Laboratory at the Korea Institute of Energy Research (KIER; President Yi, Chang-Keun) has developed a system that fully automates complex and repetitive catalyst performance evaluation experiments.
To develop new catalysts, large-scale repetitive experiments are required, with frequent changes to catalyst composition and reaction conditions. However, conventional manual experiments are time-consuming, and results often vary even when the same samples are used, depending on who conducts the experiment.
Accordingly, research has been actively pursued in recent years to theoretically predict catalyst performance and automate experiments using computational science and AI technologies. However, some steps such as replacing and loading samples, and changing consumables for overnight and long-duration continuous experiments still require precise manipulation and real-time response, and have therefore remained outside the scope of automation.
KIER researchers overcame the limitations of existing systems by developing an automated platform that uses two robots. By dividing the catalyst performance evaluation workflow into two stages and designing the robots to carry out each stage in a coordinated manner, the team enabled the entire experiment to be conducted from start to finish without human intervention.
The first robot automatically identifies the samples required for analysis based on a pre-defined experimental scenario (e.g., measurement time, sequence, and sample ID), mounts the container in the exact position, and then performs spectral measurements*. As a result, the steps previously carried out by researchers including sample selection, loading, alignment, initiating measurements, and recording are fully automated, reducing both measurement delays and operator-to-operator variability that can occur during personnel handovers.*
* UV/Vis spectroscopy: A method that uses ultraviolet and visible light to measure a sample's absorbance characteristics and analyze changes in concentration, among other parameters (in this study, changes observed as the reaction proceeds were used to evaluate catalyst performance).
The second robot carries out standardized procedures for sample handling (placement, retrieval, and disposal) and consumables replacement to ensure that high-throughput continuous experiments are not interrupted. By taking over routine maintenance tasks that previously required on-site staff, it enables stable long-duration operation and ensures continuous data collection.
The team applied optimized integrated control logic so that the two robots can operate simultaneously and continuously maintain each step of the workflow. Using the developed system, a catalyst performance evaluation that typically takes about 32 days when conducted manually was completed in approximately 17 hours making the process 45 times faster.
In addition, the team confirmed that variability in experimental results decreased by about 32% compared to manual operation, significantly improving data reliability. This demonstrates that stable operation and high-precision data acquisition are possible even in high-throughput continuous experiment settings.
The team also secured a Korean registered patent for the "catalyst performance evaluation automation system," enhancing the technology's credibility and laying the groundwork for commercialization.
Dr. Ji Chan Park, Principal Researcher, said, "This study demonstrates that we can secure highly reliable data in high-throughput experimental environments, going beyond the full automation of catalyst performance evaluation." He added, "Going forward, we will expand the application to a broader range of catalytic reactions and materials research, strengthen the linkage between theory and experiments, and ultimately advance toward AI-driven catalyst development."
Meanwhile, this work was carried out with support from KIER's institutional R&D program and the Ministry of Trade, Industry and Energy (MOTIE), and was published online in Chemical Science (IF 7.5), a leading international journal in the field of chemistry.
