DWSIM Reactive-Distillation Esterification Study
A chemical-engineering study of methyl-acetate reactive distillation covering stage count, reflux, feed placement, reactive-zone design, conversion, separation, duty, convergence, and uncertainty.
Project definition
Problem statement
Reactive distillation combines chemical reaction and separation inside one column. Conversion, purity, energy demand, feed placement, and numerical convergence therefore change together rather than as independent design choices.
The engineering problem is to screen these coupled choices with visible stage equations, enforce strict material-balance and residual limits, retain failed cases, and compare the transparent model with an independent open process-simulator benchmark.
Project objectives
- Model methyl-acetate esterification in an equilibrium-stage reactive-distillation column.
- Compare 16, 24, and 32 stages across four reflux ratios and two feed configurations.
- Calculate acid conversion, distillate ester fraction, bottoms water fraction, stage profiles, and a reboiler-duty proxy.
- Select a released case using deterministic feasibility and Pareto rules.
- Propagate uncertainty with a fixed-seed ensemble while retaining solver failures.
- Create and audit a DWSIM 10.2.3 NRTL separation benchmark after presumed conversion.
Project structure
Project components
Stage model
Solves component balances, vapor-liquid relations, reaction equilibrium, temperature, and stage-to-stage flow.
Campaign runner
Builds the declared 48-case matrix, retains solver diagnostics, and identifies feasible Pareto cases.
Uncertainty study
Perturbs reaction equilibrium, relative volatility, and feed split with deterministic sampling and retry rules.
DWSIM benchmark
Retains a DWSIM 10.2.3 NRTL separation flowsheet and an audit of solved and rejected cases.
Evidence builder
Writes complete result tables, labelled figures, editable documents, fixed PDFs, and verification evidence.
Methodology
Project workflow
- 01Declare the column
Stage count, reflux, two feed locations, reactive interval, flow, and numerical tolerances are set.
- 02Solve each case
The bounded nonlinear system is solved and checked against stage residual and component-balance limits.
- 03Compare feasible designs
Conversion, product composition, duty, and convergence are compared with deterministic release rules.
- 04Test uncertainty
The released design is rerun under parameter variation, with successful and failed cases retained separately.
- 05Inspect DWSIM
The independent NRTL benchmark is opened and compared without presenting it as a simultaneous reactive-column model.
Demonstration scenario
The released case RD-021 uses 24 stages and reflux ratio 2.0. It predicts acid conversion 0.98020, distillate ester fraction 0.97940, bottoms water fraction 0.96732, and a 121.00 kW duty proxy while meeting strict balance and residual limits. The independent DWSIM benchmark then shows how separation responds after presumed conversion.
Engineering
Tools and method
- Tools
- The project uses Python 3.12, DWSIM 10.2.3, NumPy, SciPy, Pandas, Matplotlib for subject analysis, simulation, and results.
- Engineering model
- Python, NumPy, and SciPy for the bounded stage equations, campaign, Pareto selection, and uncertainty propagation.
- Process simulation
- DWSIM 10.2.3 with the NRTL property method for an independent separation benchmark after presumed reaction conversion.
- Data and figures
- Pandas and Matplotlib for retained tables, stage profiles, convergence diagnostics, tradeoffs, and uncertainty plots.
- Verification
- Automated tests cover equations, bounds, balances, campaign behavior, deterministic seeding, file outputs, and command-line execution.
Testing
Evaluation
Evaluation measures
- Acid conversion and product mole fractions
- Reboiler-duty proxy and design tradeoffs
- Maximum component-balance error and stage residual
- Campaign convergence and Pareto status
- Uncertainty intervals and retained failed-case count
- DWSIM solved and rejected benchmark cases
- Automated tests, coverage, dependency audit, and document audit
Project boundaries
- The Python model uses simplified equilibrium-stage relations and a duty proxy.
- The DWSIM file is an NRTL separation benchmark after presumed conversion, not a native simultaneous reactive-column solve.
- Kinetics, activity-coefficient coupling, enthalpy balances, pressure drop, hydraulics, control dynamics, equipment sizing, and experimental validation remain outside scope.
- The results support an engineering study and comparison, not plant design, safety approval, product certification, or an operating guarantee.
Included
- 01Complete Python engineering source code
- 02DWSIM 10.2.3 NRTL separation benchmark and process-flow image
- 0348-case design campaign with complete convergence evidence
- 04Seeded 60-case uncertainty ensemble with retained failed-case count
- 05Complete CSV and JSON numerical results
- 0616 labelled report figures, including three attributed literature figures
- 0772-page project documentation in PDF and editable Word formats
- 0810-page setup and usage guide in PDF and editable Word formats
- 09Annotated references
- 1047 automated tests with 98.75 percent statement and branch coverage
Project record
No information is collected on this page.
- Permanent project ID
- GP-CH-14ZXLRB
- Catalogued
- 21 Aug 2026
- Completed
- 28 Aug 2026
- Verified
- 28 Aug 2026
- Demonstration
- Included in repository
Handover
After purchase
- 01Payment is confirmed
The project is marked unavailable and cannot be purchased again.
- 02Repository access is granted
The buyer's submitted GitHub account receives access to the private repository.
- 03The purchase record is delivered
The certification sheet is prepared from the reviewed buyer details and sent privately by email.