Our team has turned to AI to accelerate the design and discovery of better polymer membranes to efficiently separate carbon dioxide from flue gases.
Invisible and difficult to capture, carbon dioxide (CO₂) is a great challenge in tackling climate change.
Capturing it at the point of origin is thought to be one of the most effective ways to limit its release into the environment. Once captured, the gas could then be sequestered and stored for centuries.
But capturing and separating CO₂ from exhaust gases in energy production and transportation is tricky. Moving it to a storage site so that it doesn’t enter the atmosphere again is also far from trivial. Researchers have been trying to improve these techniques for decades.
Artificial intelligence (AI) could help.
Our IBM Research team has turned to AI to accelerate the design and discovery of better polymer membranes to efficiently separate carbon dioxide from flue gases — the results that we will present at the upcoming 2021 Meeting of the American Physical Society.
Using molecular generative AI modeling, we have identified several hundred molecular structures that could enable more efficient and cheaper alternatives to existing separation membranes for capturing CO₂ emitted in industrial processes. We are now evaluating these candidate molecules with the help of automated molecular dynamics simulation on high-performance computing (HPC) clusters.
We will also present the initial results of two other essential material discovery projects — dealing with carbon sequestration and storage.
Simulating carbon separation and conversion
Safely and effectively storing CO₂ after it’s been captured is still a challenge. One promising approach is injecting the gas into geological formations. Indeed, experts confirm that “…pore space in sedimentary rocks around the globe is more than enough to sequester all the CO₂ that humanity could ever want to remove from the air”.1 But the physics and chemistry of the process at a reservoir rock’s pore scale is not well understood. And the efficiency of CO₂ conversion and storage also depends on the type of rock and the reservoir conditions.
To tackle the issue, we have created a cloud-based tool that simulates fluid flow of carbon dioxide in specific types of rock, allowing scientists to evaluate CO₂ trapping and, eventually, conversion scenarios at pore scale. Ultimately, the technology could enable researchers and engineers to perform rapid analysis and optimization of the rock-specific requirements for mineralizing and storing CO₂ efficiently, safely and long-term.
Also, we’ll have to accelerate the discovery of CO₂-absorbing materials. It can take years, even decades, to discover a new material, or to determine which existing material is best suited to a particular carbon capture application. With our changing climate, there is no time to lose.
In a bid to speed up the process, we have created a cloud-based screening platform to rapidly sift through millions of potential CO₂ adsorbents at the nanoparticle level. The tool should enable materials engineers to select the best materials for enhancing the absorption of carbon dioxide in a particular application.
The platform allows fast searches through large quantities of known structures, enabling faster discovery. For example, it could be used by a chemist to identify the most promising nanomaterials for an industrial process. Once the most viable candidates are identified, the computational framework could then inform chemical synthesis and material optimization for accelerating the discovery in the lab.
In all of our projects, we have combined AI, HPC and cloud technologies to greatly accelerate the discovery of new materials. Our efforts stem from the recently launched Future of Climate global initiative at IBM Research, which pools materials discovery technology and scientific know-how across IBM’s worldwide network of research labs. The broader portfolio also includes the research and development of strategies to reduce the carbon footprint of cloud computing and within the supply chain, as well as techniques to model the impact of climate change.
Of course, climate change is a global challenge, requiring the collaboration of academia and industry — a joint effort of the global research community. This is why IBM has recently become an inaugural member the MIT Climate and Sustainability Consortium, along with other enterprises including Apple, Boeing, Cargill, Dow, PepsiCo and Verizon.
Only together we can advance and adopt our research outcomes at global scale, use our new solutions to formulate a long-term, sustainable climate strategy — and limit climate change.