Chemical engineering thermodynamics - readable, validated Python, with an 18-notebook course.
engthermo covers an undergraduate course in chemical engineering thermodynamics, from the first law and
steam power cycles through real fluids, mixtures and phase equilibria to reaction equilibrium, combustion and
electrochemistry - with examples from chemical and materials engineering and energy technologies. Every method
is checked against reference formulations (IAPWS-IF97), independent implementations (CoolProp, thermo) and
textbook examples, and a course of executed notebooks with worked solutions teaches the ideas.
from engthermo import cycles, reaction, steam, vle
steam.state(p=3e6, s=6.7e3) # IAPWS-IF97: any state from two properties
cycles.rankine(p_boiler=15e6, p_condenser=10e3, T_inlet=873.15, reheat=(4e6, 873.15))["efficiency"]
vle.bubble_T(["ethanol", "water"], [0.5, 0.5], p=101325.0) # Raoult's law (add gamma=... for real mixtures)
reaction.equilibrium_constant({"N2": -1, "H2": -3, "NH3": 2}, T=723.15)- Data you can trust. The IAPWS-IF97 coefficient tables and the UNIFAC group tables are transcribed programmatically from their published sources, never typed; critical constants, heat capacities and Antoine vapour pressures are taken or fitted from CoolProp's reference equations, with fit errors recorded; every source is named in the code.
- Validated, not just tested. 150 checks in
docs/VALIDATION.md, rerun in CI: the IAPWS verification tables to nine figures, cubic equations and mixture fugacities against CoolProp to 1e-6, UNIFAC against the thermo package to 1e-8, textbook cycles (Cengel & Boles) and mixtures (Smith, Van Ness & Abbott), exact identities (Gibbs-Duhem, van 't Hoff, element balances), and known azeotropes, eutectics and cell voltages. - Robust where textbooks are silent. Trivial-solution detection and pressure scanning in equation-of-state VLE, spinodal-bracketed liquid-liquid solvers, continuation near critical points, and a Gibbs minimiser that raises rather than returning a composition that violates element balances.
- Lightweight. Only NumPy and SciPy are required; CoolProp, iapws and thermo are used for validation only.
| Module | Methods |
|---|---|
steam |
IAPWS-IF97 regions 1-5: any state from (T, p), (p, h), (p, s), (p, x); saturation; two-phase states |
components, idealgas |
32 species with sourced data; ideal-gas Cp, H, S, isentropic processes, mixtures |
tables, cycles |
refrigerant tables (R134a, ammonia, propane); Carnot, Rankine (reheat, efficiencies), Brayton, vapour-compression |
eos |
virial (Pitzer-Abbott); van der Waals, RK, SRK, Peng-Robinson: roots, departure functions, fugacity; mixtures with k_ij |
vapor |
Clausius-Clapeyron, Antoine (fitted, with ranges), saturation from an equation of state |
mixtures, activity |
partial molar and excess properties; Margules, van Laar, Wilson, NRTL, UNIQUAC; fitting, azeotropes, consistency test |
unifac |
original UNIFAC (Hansen et al. 1991 tables), group assignments for common molecules |
vle |
Raoult and modified Raoult: bubble/dew points, Rachford-Rice flash, diagrams; phi-phi bubble and dew points with an EOS |
lle, sle |
binary liquid-liquid equilibria; Flory-Huggins theory; solubility, eutectics, lens diagrams |
reaction |
Delta H, Delta S, Delta G and K(T); single-reaction extent; Gibbs minimisation for many reactions |
combustion, electrochem |
heating values, flue gas, adiabatic flame temperature, Ellingham diagram; cell potentials, Nernst, fuel-cell and electrolyser limits |
datasets |
11 bundled data files (property tables and course data) with sources |
Eighteen executed notebooks, each built around engineering problems, with learning objectives, an "Inside the
algorithm" section and exercises (including an "Implement it yourself" task); every exercise has a worked
solution in solutions/. They open in Google Colab and install engthermo automatically.
| # | Notebook | Applications |
|---|---|---|
| 00 | Python for thermodynamics | units, tables, root finding, property diagrams |
| 01 | The first law and energy balances | furnaces, compressors with intercooling, turbines, valves |
| 02 | Steam and power cycles | Rankine cycles, reheat, plant sizing |
| 03 | Entropy, refrigeration and heat pumps | exergy, vapour-compression cycles, refrigerants, heat pumps |
| 04 | Real gases and equations of state | CO2 pipelines, hydrogen tanks |
| 05 | Departure functions and fugacity | real-gas compressors, fugacity |
| 06 | Vapour pressure and phase change | Antoine, saturation from an EOS, LPG storage |
| 07 | Mixture fundamentals | partial molar volumes, ideal mixing, Gibbs-Duhem |
| 08 | Raoult's law and flash calculations | T-x-y diagrams, flash drums, relative volatility |
| 09 | Activity-coefficient models | ethanol-water, fitting, consistency, the bioethanol azeotrope |
| 10 | UNIFAC prediction | prediction without data, solvent screening |
| 11 | High-pressure phase equilibria | natural-gas dew points, CO2 mixtures, interaction parameters |
| 12 | Liquid-liquid equilibria and polymer solutions | extraction, Flory-Huggins, UCST |
| 13 | Solid-liquid equilibria and alloy phase diagrams | solubility, antifreeze, Bi-Cd and Cu-Ni diagrams |
| 14 | Chemical reaction equilibrium | ammonia, methanol, hydrogen by steam reforming |
| 15 | High-temperature thermochemistry | fuels and CO2, flame temperatures, Ellingham diagram, green steel |
| 16 | Electrochemical thermodynamics | fuel cells, electrolysers, batteries |
| 17 | From messy laboratory data to a model | data repair, consistency, fitting with uncertainties, reporting |
Every image is computed by the library; python tools/make_images.py regenerates them.
 Steam T-s diagram (IAPWS-IF97) with a reheat Rankine cycle |
 Peng-Robinson isotherms of CO2 through the critical point |
 Ethanol-water: UNIFAC finds the azeotrope Raoult's law cannot |
 Ammonia synthesis: the Haber-Bosch compromise |
 Ellingham diagram: which oxides carbon and hydrogen can reduce |
 Cu-Ni lens diagram from two melting points and two enthalpies |
pip install "engthermo @ git+https://github.com/ktwyw/engthermo"
# or, for development (adds CoolProp, iapws and thermo for the validation suite):
git clone https://github.com/ktwyw/engthermo && cd engthermo && pip install -e ".[dev]"| Reference | What is checked | Agreement |
|---|---|---|
| IAPWS-IF97 verification tables; CoolProp; iapws | steam properties in all five regions, saturation line, two-phase and inverse states | 1e-8 |
| CoolProp reference equations | critical constants, ideal-gas Cp (fitted), Antoine vapour pressures, refrigerant tables | ≤ 1 %; tables 2e-4 |
| CoolProp cubic backends | Z, departure functions, fugacity, saturation (PR, SRK); mixture fugacities and a mixture bubble point (PR) | ≤ 1e-6 |
| thermo package | UNIFAC activity coefficients (binary, ternary, negative deviations) | 1e-8 |
| Cengel & Boles; Smith, Van Ness & Abbott | Rankine, reheat and Brayton cycles; Redlich-Kwong volumes | ≤ 0.3 % |
| Exact identities | van 't Hoff, Gibbs-Duhem, G = H - TS, element balances, Rachford-Rice, energy balances | ≤ 1e-6 |
| Experiment / literature | ethanol-water azeotrope, Bi-Cd eutectic, naphthalene solubility, fuel-cell voltage and its temperature coefficient | within stated tolerances |
If engthermo helps your work, please cite it using CITATION.cff (GitHub shows a
"Cite this repository" button).
Yanwei Wang - personal open-source project. GitHub @ktwyw · ORCID 0000-0002-8488-9833 · wangyanwei@gmail.com
Also by the author: engmath (numerical methods), engstat (engineering statistics), fluidmech (fluid mechanics) and engrheo (rheology).
MIT - see LICENSE. Contributions welcome: see CONTRIBUTING.md.