US 12,422,099 B1
Hydrogen compression and storage systems
Matthew D. Witman, Oakland, CA (US); and Vitalie Stavila, Oakland, CA (US)
Assigned to National Technology & Engineering Solutions of Sandia, LLC., Albuquerque, NM (US)
Filed by National Technology & Engineering Solutions of Sandia, LLC, Albuquerque, NM (US)
Filed on Oct. 17, 2022, as Appl. No. 17/967,459.
Int. Cl. F17C 5/06 (2006.01); F17C 11/00 (2006.01); F17C 13/04 (2006.01)
CPC F17C 5/06 (2013.01) [F17C 11/005 (2013.01); F17C 13/04 (2013.01); F17C 2221/012 (2013.01); F17C 2227/0157 (2013.01); F17C 2227/0171 (2013.01); F17C 2227/0302 (2013.01); F17C 2250/032 (2013.01); F17C 2250/043 (2013.01); F17C 2250/0447 (2013.01); F17C 2250/0642 (2013.01); F17C 2270/01 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A hydrogen compressor comprising:
an inlet valve;
an outlet valve;
a storage container in fluid communication with the inlet valve and the outlet valve;
a heat transfer device; and
storage media arranged inside the storage container, wherein the storage media is formed from a composition, wherein the composition is selected and made by way of a process comprising:
providing a proposed composition for the storage media as input to a computer-implemented machine learning model, wherein the proposed composition comprises a first element and a second element, and further wherein the proposed composition has a known plateau pressure associated therewith;
providing a target plateau pressure for the storage media as input to the computer-implemented machine learning model; and
identifying, by the computer-implemented machine learning model, the composition for the storage media, wherein the composition for the storage media is different from the proposed composition for the storage media, wherein the composition has a computed plateau pressure that is closer to the target plateau pressure than the known plateau pressure, and further wherein the composition for the storage media comprises:
at least one of the first element or the second element; and
a third element;
wherein the computer-implemented machine learning model identifies the composition based upon properties of the first element, the second element, and the third element, and further wherein the properties comprise a ground state volume per atom of an elemental solid, a covalent radius, a Pauling electronegativity, and a number of valence electrons.