2026/03/09 by Anonymous, Ryuhei Sato, Peter I. C. Cooke +8
Materials Science · Physics and Astronomy · #Dynamics (music) #Feature (linguistics) #Hydrogen Storage and Materials #Machine Learning in Materials Science #Metastability #Molecular dynamics #Process (computing) #Quantum, superfluid, helium dynamics #cond-mat.mtrl-sci
paper · pdf · doi:10.1103/pb7t-5bj1
5 figures with supporting information
arxiv created 2026/03/09 · openalex publication_date 2026/07/15 · openalex created_date 2026/07/16 · arxiv updated 2026/08/06 · openalex updated_date 2026/08/06
The synthesis of the high- <a:math xmlns:a="http://www.w3.org/1998/Math/MathML"> <a:msub> <a:mi>T</a:mi> <a:mi mathvariant="normal">c</a:mi> </a:msub> </a:math> superhydride <c:math xmlns:c="http://www.w3.org/1998/Math/MathML"> <c:mrow> <c:msub> <c:mi>CaH</c:mi> <c:mn>6</c:mn> </c:msub> </c:mrow> </c:math> has stimulated significant interest in understanding synthesis pathways for metastable hydrides. However, the microscopic mechanisms governing such hydrogenation reactions remain poorly understood. Here, we show that machine-learning potential molecular dynamics simulations can reproduce and distinguish competing reaction pathways leading to metastable and stable hydrides. By simulating hydrogenation reactions at <d:math xmlns:d="http://www.w3.org/1998/Math/MathML"> <d:mrow> <d:msub> <d:mi>CaH</d:mi> <d:mn>2</d:mn> </d:msub> <d:mtext>/</d:mtext> <d:msub> <d:mi mathvariant="normal">H</d:mi> <d:mn>2</d:mn> </d:msub> </d:mrow> </d:math> and <f:math xmlns:f="http://www.w3.org/1998/Math/MathML"> <f:mrow> <f:msub> <f:mi>CaH</f:mi> <f:mn>4</f:mn> </f:msub> <f:mtext>/</f:mtext> <f:msub> <f:mi mathvariant="normal">H</f:mi> <f:mn>2</f:mn> </f:msub> </f:mrow> </f:math> interfaces, we identify two distinct pathways that produce clathrate-type <h:math xmlns:h="http://www.w3.org/1998/Math/MathML"> <h:mrow> <h:msub> <h:mi>CaH</h:mi> <h:mn>6</h:mn> </h:msub> </h:mrow> </h:math> and A15-type <i:math xmlns:i="http://www.w3.org/1998/Math/MathML"> <i:mrow> <i:msub> <i:mi>CaH</i:mi> <i:mrow> <i:mn>5.75</i:mn> </i:mrow> </i:msub> </i:mrow> </i:math> , respectively. <j:math xmlns:j="http://www.w3.org/1998/Math/MathML"> <j:mrow> <j:msub> <j:mi>CaH</j:mi> <j:mrow> <j:mn>5.75</j:mn> </j:mrow> </j:msub> </j:mrow> </j:math> lies on the convex hull but requires extensive Ca sublattice rearrangement and therefore forms only at elevated temperatures. In contrast, <k:math xmlns:k="http://www.w3.org/1998/Math/MathML"> <k:mrow> <k:msub> <k:mi>CaH</k:mi> <k:mn>6</k:mn> </k:msub> </k:mrow> </k:math> becomes kinetically accessible when <l:math xmlns:l="http://www.w3.org/1998/Math/MathML"> <l:mrow> <l:msub> <l:mi>CaH</l:mi> <l:mn>2</l:mn> </l:msub> </l:mrow> </l:math> is used as the precursor. The crystallographic compatibility between the Ca sublattice of <m:math xmlns:m="http://www.w3.org/1998/Math/MathML"> <m:mrow> <m:msub> <m:mi>CaH</m:mi> <m:mn>2</m:mn> </m:msub> </m:mrow> </m:math> and the body-centered cubic framework of <n:math xmlns:n="http://www.w3.org/1998/Math/MathML"> <n:mrow> <n:msub> <n:mi>CaH</n:mi> <n:mn>6</n:mn> </n:msub> </n:mrow> </n:math> enables a martensitic-like topotactic transformation that bypasses the reconstructive pathway leading to <o:math xmlns:o="http://www.w3.org/1998/Math/MathML"> <o:mrow> <o:msub> <o:mi>CaH</o:mi> <o:mrow> <o:mn>5.75</o:mn> </o:mrow> </o:msub> </o:mrow> </o:math> . These results reveal how precursor structure and thermodynamic stability compete to determine superhydride formation pathways and demonstrate that machine-learning molecular dynamics can directly capture the kinetic selection of metastable phases in reactive materials systems.