CX-100917 Optimized Hydrogen Adsorbents via Machine Learning and Crystal Engineering

Award Number: DE-EE0008093
CX(s) Applied: A9, B3.6
Fuel Cells Technologies Office
Location(s): MI
Office(s): Golden Field Office

The U.S. Department of Energy (DOE) is proposing to provide federal funding to the Regents of the University of Michigan to use machine learning techniques to predict and reverse engineer new metal organic frameworks (MOFs) that show promise for high hydrogen storage capacity. Only Budget Period 1 (BP1) is being negotiated at this time so this NEPA review is for BP1 activities only. Additional NEPA review will be required if DOE proposes to continue funding the project into subsequent budget periods.

The University of Michigan (UM) located in Ann Arbor, Michigan would apply machine learning methods to analyze the database of 476,007 real and hypothetical MOFs. This analysis would guide the discovery of new compounds that could break through the so-called Hydrogen (H2) volumetric storage ceiling. UM would search for those MOF properties that correlate with high H2 storage capacities. The UM would also use machine learning techniques in an attempt to predict the needed MOF properties to achieve the desired H2 storage capacity. The UM would then attempt to synthesize unique MOFs with specific morphology through the addition of additives during crystal growth. Finally, Ford Motor Company (Dearborn, MI) would conduct materials characterization.