Plasmon-Enhanced Photothermal NH₃ Decomposition

Background

Hydrogen is difficult to store and move: it is a low-density gas at ambient conditions and needs deep cryogenics or heavy pressure vessels to concentrate it. Ammonia sidesteps that problem. It carries roughly twice the volumetric hydrogen density of liquid hydrogen, stays liquid under modest pressure rather than cryogenic temperature, and can move through decades of shipping, storage, and pipeline infrastructure already built for the fertilizer industry. The tradeoff is that ammonia cannot be used directly in a fuel cell or turbine at meaningful efficiency — it first has to be decomposed back into hydrogen and nitrogen at the point of use, and that decomposition step is where most of the remaining engineering difficulty sits.

My approach treats that decomposition step as a full-stack problem rather than a pure chemistry problem. On one side, I design and build the photothermal reactor system itself: the optical path, the reaction cell, and the gas handling that let concentrated light drive the reaction instead of a resistively heated furnace. On the other side, I develop the catalyst that sits inside it — a plasmon-enhanced photothermal catalyst that concentrates incident light into localized heat directly at the catalytic sites, rather than relying on bulk external heating of the whole reactor. Working across both the reactor and the material lets me tune each side against the other, instead of treating either one as fixed.

A third piece runs underneath both of these. Reactor and catalyst screening generates a steady stream of gas chromatography runs and reactor logs, and processing that by hand does not scale past a handful of experiments. I built two internal tools to automate the pipeline end to end: lat-pipeline, for lab data collection and organization, and gc-analysis, for turning raw GC output into conversion and rate numbers. The same instinct — build the tool instead of repeating the manual process — shows up across the rest of my work (see Code). This is ongoing PhD research, and the study itself is not yet published, so what’s described here is scoped to system design and methodology rather than specific results.

What I did

• Designed and built the photothermal reactor system — optical path, reaction cell, and gas handling.
• Developed the plasmon-enhanced photothermal catalyst used inside the reactor.
• Built lat-pipeline and gc-analysis to automate the experimental data pipeline — collection, classification, and conversion/rate analysis — removing manual GC processing from the critical path.lat-pipelinegc-analysis

Outcomes

The reactor and catalyst study is still in progress, with no publication yet. In parallel, a same-lab collaboration with the Jungwon Park Lab at Seoul National University produced a co-authored paper in Science, where my contribution was in-situ DRIFTS analysis. That work tracks how the catalytic behavior of supported platinum clusters changes atom by atom — the kind of structure-property resolution that speaks to the same mechanistic questions I’m asking of the plasmonic catalyst here.

Related publications