Predictive Model for Catalytic Methane Pyrolysis

Methane pyrolysis provides a scalable alternative to
conventional
hydrogen production methods, avoiding greenhouse gas emissions. However,
high operating temperatures limit economic feasibility on an industrial
scale. A major scientific goal is, therefore, to find a catalyst material
that lowers operating temperatures, making methane pyrolysis economically
viable. In this work, we derive a model that provides a qualitative
comparison of possible catalyst materials. The model is based on calculations
of adsorption energies using density functional theory. Thirty different
elements were considered. Adsorption energies of intermediate molecules
in the methane pyrolysis reaction correlate linearly with the adsorption
energy of carbon. Moreover, the adsorption energy increases in magnitude
with decreasing group number in the d-block of the periodic table.
For a temperature range between 600 and 1200 K and a normalized partial
pressure range for H2 between 10–1 and
10–5, a total of 18 different materials were found
to be optimal catalysts at least once. This indicates that catalyst
selection and reactor operating conditions should be well-matched.
The present work establishes the foundation for future large-scale
studies of surfaces, alloy compositions, and material classes using
machine learning algorithms.

甲烷热解（Methane pyrolysis）为传统制氢方法提供了一种可规模化的替代方案，可避免温室气体排放。然而，过高的操作温度限制了其工业规模应用的经济可行性。因此，一项核心科学目标是研发可降低操作温度的催化剂材料，使甲烷热解具备经济可行性。
本研究推导了一种可对潜在催化剂材料进行定性比较的模型。该模型基于密度泛函理论（Density Functional Theory）计算吸附能，共考量了30种不同元素。甲烷热解反应中中间体分子的吸附能与碳的吸附能呈线性相关；此外，吸附能的绝对值随元素周期表d区元素族数的降低而增大。
在600~1200 K的温度区间，以及氢气归一化分压介于10^–1至10^–5的范围内，共计18种不同材料被发现至少可作为一次最优催化剂。该结果表明，催化剂选型与反应器操作条件需要进行精准匹配。
本研究为未来利用机器学习算法开展表面、合金成分及材料类别的大规模研究奠定了基础。