Skip to content
cccskills
FreePublished on GitHub

tooluniverse-inorganic-physical-chemistry

Inorganic chemistry, physical chemistry, and materials science — crystal structures, coordination chemistry, lattice parameters, thermodynamic properties, electronic structure. Use for unit cell volume calculations, coordination geometry, materials property estimation, and inorganic-mechanism reasoning. Complementary to tooluniverse-organic-chemistry.

The English summary is being prepared. Showing the original description.

Run score Not tested

See how to install

Files included (2)

  • SKILL.md6.0 KB
  • scripts/equilibrium_solver.py8.2 KB

SKILL.md (original)

Review the instructions the agent will receive before you install.

Inorganic & Physical Chemistry

Reasoning Strategy

1. Crystal Structure Questions

When given crystal structure data, always COMPUTE don't guess:

  1. Calculate unit cell volume for the crystal system:

    • Cubic: V = a^3
    • Tetragonal: V = a^2 * c
    • Orthorhombic: V = a * b * c
    • Monoclinic: V = a * b * c * sin(beta)
    • Triclinic: V = abc * sqrt(1 - cos^2(alpha) - cos^2(beta) - cos^2(gamma) + 2*cos(alpha)*cos(beta)*cos(gamma))
    • Hexagonal: V = a^2 * c * sqrt(3)/2
  2. Verify density: d = (Z * M) / (V * Na * 1e-24) where V in ų, M in g/mol, Na = 6.022e23

  3. Preferred: Use CrystalStructure_validate tool (via MCP/SDK). Fallback: python3 skills/tooluniverse-organic-chemistry/scripts/crystal_validator.py --a X --b Y --c Z --alpha A --beta B --gamma G --Z N --MW M --density D

  4. For batch comparison (find the wrong dataset): Save datasets as JSON array and use --datasets path/to/datasets.json

2. Bonding & Covalency Questions

Key reasoning patterns:

  • Covalency = orbital mixing between metal and ligand. Greater overlap = more covalent.
  • Lanthanide/actinide: 4f orbitals of Ce(IV) typically show ENHANCED covalent mixing vs Ce(III) — more contracted 4f in higher oxidation states increases overlap with ligand orbitals
  • But: Enhanced covalency does NOT always mean stronger bonds — it depends on the specific orbital interactions
  • d-block vs f-block: d-orbitals have more radial extension → stronger covalent bonds than f-orbitals
  • Nephelauxetic effect: Reduced electron-electron repulsion in complexes → indicates covalency. Larger effect = more covalent.

3. Noble Gas Chemistry

  • Xe compounds: XeF2 (linear), XeF4 (square planar), XeF6 (distorted octahedral)
  • XeF4 synthesis: Requires Xe + F2 at elevated temperature (400°C) and pressure. Can also form at lower temperatures with specific methods (UV photolysis, electric discharge)
  • Key: Temperature thresholds matter for synthesis efficiency. LOOK UP DON'T GUESS — search literature for specific synthesis conditions.

4. Symmetry & Point Groups

  1. Identify the molecular shape
  2. Find symmetry elements: C_n axes, mirror planes (σ_h, σ_v, σ_d), inversion center (i), improper rotation (S_n)
  3. Use python3 skills/tooluniverse-organic-chemistry/scripts/chemistry_facts.py point_groups for point group lookup
  4. Optical activity: Requires absence of improper rotation axes (S_n, including σ = S_1 and i = S_2). Chiral point groups: C_1, C_n, D_n, T, O, I
  5. Crystal classes with optical activity: Piezoelectric non-centrosymmetric classes that lack mirror planes and inversion

5. Thermodynamics & Kinetics

COMPUTE DON'T ESTIMATE — write Python code for:

  • Gibbs free energy: ΔG = ΔH - TΔS
  • Equilibrium constant: K = exp(-ΔG/RT)
  • Arrhenius equation: k = A * exp(-Ea/RT)
  • Nernst equation: E = E° - (RT/nF) * ln(Q)
  • Clausius-Clapeyron: ln(P2/P1) = -ΔH_vap/R * (1/T2 - 1/T1)

6. Solubility & Equilibrium Calculations

Preferred: Use EquilibriumSolver_calculate tool (via MCP/SDK) with type, ksp, stoich, and other parameters. Fallback: run equilibrium_solver.py directly.

# Simple Ksp: MaXb(s) <-> aM + bX
python3 skills/tooluniverse-inorganic-physical-chemistry/scripts/equilibrium_solver.py \
  --type ksp_simple --ksp 5.3e-27 --stoich 1:3

# Ksp + complex formation (e.g., Al(OH)3 in water with Al(OH)4- complex)
python3 skills/tooluniverse-inorganic-physical-chemistry/scripts/equilibrium_solver.py \
  --type ksp_kf --ksp 5.3e-27 --kf 1.1e33 --stoich 1:3

# Common ion effect (e.g., AgCl in 0.1M NaCl)
python3 skills/tooluniverse-inorganic-physical-chemistry/scripts/equilibrium_solver.py \
  --type common_ion --ksp 1.77e-10 --stoich 1:1 --common-ion 0.1

Key points:

  • ksp_kf mode solves the full charge-balance system numerically (Newton's method) — accounts for free cation, complex anion, and OH-/H+ simultaneously
  • For MX_b + X- <-> MX_(b+1)-, K_overall = Ksp * Kf
  • common_ion mode uses bisection to solve the exact Ksp expression with extra ion concentration
  • Always specify --stoich a:b matching the salt formula (e.g., 1:3 for Al(OH)3, 1:2 for CaF2, 1:1 for AgCl)

7. Spectroscopy Interpretation

  • UV-Vis: d-d transitions (weak, Laporte forbidden), LMCT/MLCT (strong), π→π* (organic)
  • IR: Functional group region (4000-1500 cm⁻¹), fingerprint (1500-400 cm⁻¹)
  • NMR: Chemical shift indicates electronic environment. For counting peaks, identify symmetry-equivalent protons.
  • For peak counting: Draw the structure, identify all symmetry operations, group equivalent H atoms. Use python3 skills/tooluniverse-organic-chemistry/scripts/chemistry_facts.py for reference data.

Available Tools

ToolUse For
PubChem_get_CID_by_compound_nameGet compound CID from name
PubChem_get_compound_properties_by_CIDDetailed compound data by CID
ChEMBL_search_moleculesBioactive compounds
PubMed_search_articlesLiterature on synthesis conditions, properties
CrystalStructure_validate tool (or crystal_validator.py fallback)Verify crystal structure data consistency
EquilibriumSolver_calculate tool (or equilibrium_solver.py fallback)Ksp, complex formation, common-ion solubility

LOOK UP DON'T GUESS

  • Noble gas compound synthesis conditions vary by method — search literature before answering
  • Crystal structure parameters must be computed, not estimated
  • Bonding descriptions (covalent vs ionic) require specific orbital considerations — don't generalize from one system to another

Reviews

No reviews yet. Tell others what you thought.

Comments

No login needed. Share how it worked for you, ask a question, or note where it did not run.

No comments yet. Be the first to write one.

Comments are published right away. Do not include personal information such as email addresses or phone numbers. Up to 2 URLs.

0 / 1,000 characters

This form is protected by reCAPTCHA, and the Google Privacy Policy and Terms of Service apply.

More skills from this repository

Simplest distributed training API. 4 lines to add distributed support to any PyTorch script. Unified API for DeepSpeed/FSDP/Megatron/DDP. Automatic device placement, mixed precision (FP16/BF16/FP8). Interactive config, single launch command. HuggingFace ecosystem standard.

The English summary is being prepared. Showing the original description.

huang-sh/DeepScience4Updated Jul 15, 2026

ADMET property prediction for drug candidates. Full pharmacokinetic panel (Caco-2, PPB, clearance, CYP), toxicity (hERG, AMES, DILI), drug-likeness (Lipinski, QED), using RDKit descriptors and TDC models.

The English summary is being prepared. Showing the original description.

huang-sh/DeepScience4Updated Jul 15, 2026

Interpretable ADMET analysis with mechanistic reasoning. Maps liabilities to structural causes and biological pathways. Based on CoTox (Park 2025) and DrugR (Liu 2026).

The English summary is being prepared. Showing the original description.

huang-sh/DeepScience4Updated Jul 15, 2026

aeon

Free

This skill should be used for time series machine learning tasks including classification, regression, clustering, forecasting, anomaly detection, segmentation, and similarity search. Use when working with temporal data, sequential patterns, or time-indexed observations requiring specialized algorithms beyond standard ML approaches. Particularly suited for univariate and multivariate time series analysis with scikit-learn compatible APIs.

The English summary is being prepared. Showing the original description.

huang-sh/DeepScience4Updated Jul 15, 2026

Infer gene regulatory networks (GRNs) from gene expression data using scalable algorithms (GRNBoost2, GENIE3). Use when analyzing transcriptomics data (bulk RNA-seq, single-cell RNA-seq) to identify transcription factor-target gene relationships and regulatory interactions. Supports distributed computation for large-scale datasets.

The English summary is being prepared. Showing the original description.

huang-sh/DeepScience4Updated Jul 15, 2026

Comprehensive Python library for astronomy and astrophysics. This skill should be used when working with astronomical data including celestial coordinates, physical units, FITS files, cosmological calculations, time systems, tables, world coordinate systems (WCS), and astronomical data analysis. Use when tasks involve coordinate transformations, unit conversions, FITS file manipulation, cosmological distance calculations, time scale conversions, or astronomical data processing.

The English summary is being prepared. Showing the original description.

huang-sh/DeepScience4Updated Jul 15, 2026

See all skills by huang-sh

Report a problem with this skill