aicodedecode/awesome-muse-skills/catalysis-basics
Claude Code and agent skill by aicodedecode/awesome-muse-skills
- community
Checking that the SKILL.md file is still in the repository.
Heterogeneous and homogeneous catalysis — mechanisms, kinetics, catalyst characterization, and deactivation analysis.
Install
npx vskill@latest install aicodedecode/awesome-muse-skills/catalysis-basicsSKILL.md
The same SKILL.md was published earlier as aicodedecode/awesome-muse-skills/catalysis-basics.
Overview
Catalysts accelerate reactions without being consumed, enabling most of the chemical industry. This skill covers the essential framework: catalytic cycles and mechanisms, rate laws and turnover metrics, the Sabatier principle, standard characterization (chemisorption, TPR/TPD, operando spectroscopy), and diagnosing deactivation (poisoning, sintering, coking, leaching).
When to use
- Screening catalysts: choosing metrics (conversion, selectivity, TOF) and fair test conditions
- Measuring kinetics: determining rate laws, activation energies, and reaction orders
- Characterizing a catalyst: dispersion, oxidation state, acidity, active sites
- Troubleshooting deactivation: distinguishing poisoning from sintering from coking
- Comparing homogeneous vs heterogeneous options for a transformation
Core concepts
- Catalytic cycle: the sequence of elementary steps returning the catalyst to its initial state; the turnover frequency (TOF, s⁻¹) normalizes rate per active site — the only fair activity comparison.
- Sabatier principle: optimal catalysts bind intermediates neither too weakly (no activation) nor too strongly (surface poisoned) — the origin of volcano plots across metals and reactions.
- Rate-determining step: the highest-barrier step controls the rate; degree-of-rate-control analysis identifies it rigorously — intuition about "the slow step" is often wrong in networks.
- Selectivity vs conversion: more forcing conditions raise conversion but usually cost selectivity; report both, plus carbon/mass balance — missing mass means unidentified products.
- Structure sensitivity: some reactions need specific ensembles (facets, particle size); others are insensitive — this determines whether nanoparticle engineering matters.
- Homogeneous vs heterogeneous: molecular catalysts offer tunability and defined sites (ligand design, mechanistic study by NMR); heterogeneous offer separation and stability — the choice hinges on product value and process constraints.
- Degree of rate control (DRC): quantifies how much each elementary step (and intermediate) controls the overall rate — the rigorous replacement for "rate-determining step" intuition in complex networks.
- Scaling relations: adsorption energies of related intermediates scale linearly (e.g., OH vs OOH) — these relations create the volcano plot and fundamentally limit single-site catalysts, motivating bifunctional designs.
- Catalyst–support interactions: supports are not inert — strong metal–support interaction (SMSI), spillover, and acid–base sites on the support participate in the chemistry; changing the support changes the catalyst.
Practical workflow
1. Test catalysts fairly
- Fix the conditions that matter: temperature, pressure, feed composition, space velocity (WHSV/GHSV) — vary the catalyst, not the reactor.
- Report conversion, selectivity, yield, and carbon balance at matched conversion levels (compare selectivity at the same conversion, not at arbitrary conditions).
- Measure TOF using counted active sites (chemisorption, not total metal loading) — a 1% dispersed catalyst and a 50% dispersed one differ by 50× in site count.
- Include a blank (no catalyst) and a reference catalyst in every campaign.
2. Measure kinetics properly
- Work in the kinetic regime: verify absence of mass/heat transfer limits (vary particle size, flow rate, dilution — rate must not change).
- Determine orders by varying one reactant partial pressure at a time; measure E_a from Arrhenius plots over a modest T range (avoid mechanism changes).
- Check for deactivation during the measurement — a drifting rate corrupts every derived parameter; use initial rates or extrapolate to t=0.
3. Characterize the working catalyst
- Dispersion/particle size: CO/H₂ chemisorption, TEM, XRD line broadening — cross-check at least two.
- Oxidation state and coordination: XPS (surface), XAS (bulk-average, operando-capable), TPR (reducibility).
- Acidity: NH₃-TPD and pyridine-IR (Brønsted vs Lewis) for acid-catalyzed reactions.
- Operando when it matters: ex situ characterization of a spent catalyst may show a spectator, not the active site — measure under reaction conditions where feasible.
4. Diagnose deactivation
- Poisoning: activity drops, selectivity often shifts; check feed impurities (S, Cl, CO); sometimes reversible by treatment.
- Sintering: particle growth (TEM/XRD) — irreversible; driven by temperature and atmosphere.
- Coking: carbon deposits (TGA/Raman); burn off regenerates activity unless sintering co-occurred.
- Leaching: metal in the product stream (ICP) — fatal for homogeneous-immobilized and some oxide catalysts.
- Distinguish by regeneration tests: activity recovery after oxidation ⇒ coking/poisoning; no recovery ⇒ sintering/leaching.
5. Benchmark against the literature fairly
- Normalize literature rates to the same basis (per active site, per surface area, or per reactor volume) before comparing — published "record activities" often differ only in normalization.
- Match conditions: temperature, pressure, feed composition, and conversion level — selectivity compared at different conversions is meaningless.
- Reproduce one literature benchmark in your own setup before claiming improvements — it validates your reactor, analytics, and methods simultaneously.
6. Quick-reference checklist
- [ ] Active sites counted (chemisorption), not assumed from metal loading
- [ ] Mass/heat transfer limitations tested (particle size, flow variation)
- [ ] Selectivity compared at matched conversion levels
- [ ] Carbon/mass balance closed (missing mass = unidentified products)
- [ ] Blank and reference catalyst included in every campaign
- [ ] Deactivation monitored during kinetic measurements
- [ ] Spent catalyst characterized (sintering vs coking vs poisoning vs leaching)
- [ ] Literature benchmarks reproduced in your setup before claiming improvements
Common pitfalls
- Normalizing by mass instead of sites: grams of catalyst is not a kinetic quantity — TOF per active site is.
- Ignoring transport limitations: a "structure-insensitive" conclusion from a diffusion-limited test is an artifact; always test for it.
- Single-point comparisons: one temperature, one conversion — selectivity–conversion curves tell the real story.
- Ex situ overinterpretation: the catalyst you characterize after cooling and air exposure is not the catalyst that was working.
- Missing mass balance: 100% conversion with 60% identified products means 40% unknown — find it before publishing.
- Accelerated aging fallacy: deactivation mechanisms change with temperature; high-T aging tests can misrank catalysts for low-T service.
- Hotspot blindness: exothermic reactions create temperature gradients in the bed — the measured "isothermal" rate at the thermocouple temperature misrepresents the hot zone; dilute the bed and check.
- Attributing homogeneous catalysis to a solid: leached metal can catalyze the reaction while the solid takes the credit — hot-filtration tests and ICP of the filtrate are mandatory for "heterogeneous" claims with soluble metals.
Security
- Security scanPassed, 100 of 100Pattern checks on the skill's files.
- Model reviewPassed
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingskills/catalysis-basics/SKILL.md
Popularity
- Repository stars
- 10
- Repository forks
- 1
- 7-day trend score
- 17
Stars and forks are for the whole GitHub repository, not this skill alone. Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/aicodedecode/awesome-muse-skills/catalysis-basics)