DFT Screening of Ionic Liquids for Rare-Earth Separation

ISEF Category: Chemistry

Subcategory: Computational Chemistry  ·  Difficulty: Advanced  ·  Setup: University Lab  ·  Time: Full Year

The Hook

Old phone speakers, earbuds, and hard drives hide metals that power modern devices. When those devices die, the magnets inside still hold rare earths like neodymium and dysprosium. The hard part is separating them cleanly, because they behave a lot like twins. You can model which ionic liquids might split them apart best before anyone mixes chemicals in a lab.

What Is It?

This project uses computer chemistry to predict which ionic liquids, a type of salt that stays liquid, might separate rare earth metals from each other. Think of it like testing dozens of key shapes in a lock without ever touching the lock. You use quantum chemistry tools, such as DFT, which means density functional theory, to estimate how the molecules behave. Then you use COSMO-RS style models to predict how strongly each liquid prefers one metal over another.

The main idea is selectivity. If one ionic liquid binds neodymium a little better than dysprosium, or the other way around, that small difference can matter a lot in recycling. Your job is to turn that difference into numbers, then rank candidates. This kind of study helps researchers narrow a huge search space before they spend money on lab work.

Why This Is a Good Topic

This is a strong science fair topic because you can ask a clear question, compare many candidates, and produce real numbers without needing a wet lab. It connects to electronic waste, mining pressure, and rare-earth recycling, which gives the work real-world weight. You can learn quantum chemistry basics, molecular modeling, data ranking, and uncertainty analysis. If you do careful controls and explain your assumptions well, this can become a serious computational project.

Research Questions

• How does ionic liquid cation structure change the predicted selectivity for Nd versus Dy??
• What is the effect of changing the anion on computed activity coefficients for rare-earth complexes??
• Does adding a longer alkyl chain to the ionic liquid increase or decrease predicted separation performance??
• To what extent do different ligand functional groups change the ranking of candidate solvents for Nd/Dy separation??
• Which ionic liquids stay most selective when you compare multiple implicit-solvent models??
• How does the predicted selectivity change when you include different rare-earth complex geometries??

Basic Materials

• Computer with enough RAM for quantum chemistry calculations.
• Access to a university or open-source computational chemistry platform.
• Molecular drawing software.
• Spreadsheet software for ranking and plotting results.
• Journal articles or database records for ionic liquid structures.
• Reference data for rare-earth complex chemistry.
• Notes document for tracking model assumptions.

Advanced Materials

• Access to a workstation or cluster with multiple CPU cores.
• Quantum chemistry software that can run DFT geometry optimizations.
• COSMO-RS or a COSMO-style solvation workflow.
• Molecular conformer search software.
• Scripting environment for batch jobs and data parsing.
• Visualization software for molecular orbitals and surface charge maps.
• Literature data on Nd and Dy coordination chemistry.
• Version control system for tracking input files and analysis scripts.

Software & Tools

• Avogadro: Builds and edits molecular structures before quantum calculations.
• ORCA: Runs DFT geometry optimizations and property calculations for candidate molecules.
• Python: Automates file handling, ranking, and statistical plots for your screening set.
• cclib: Reads quantum chemistry output files and helps extract energies and other results.
• Excel or Google Sheets: Organizes candidate lists and compares predicted selectivity values.

Experiment Steps

1. Define the separation target and choose one metal pair, one family of ionic liquids, and one scoring metric.
2. Build a candidate library that varies one structural feature at a time so your comparisons stay fair.
3. Standardize the molecular forms you will model, including metal complexes, counterions, and charge states.
4. Choose a consistent computational workflow for geometry optimization, energy evaluation, and solvation screening.
5. Rank candidates with one primary metric and at least one secondary check, such as stability or sensitivity to model choice.
6. Plan a validation strategy using literature values, trend checks, or a smaller benchmark set before you scale up.

Common Pitfalls

• Comparing molecules with different charges or coordination states, which makes the selectivity ranking meaningless.
• Changing multiple structural features at once, which hides the cause of any performance difference.
• Ignoring conformer choice, which can flip the predicted order of two ionic liquids.
• Treating a single computed number as proof, which ignores model sensitivity and uncertainty.
• Using inconsistent input geometry cleanup across compounds, which adds noise that looks like chemistry.

What Makes This Competitive

A stronger project goes beyond ranking one short list. You can test how stable the ranking stays when you change the solvent model, the metal geometry, or the level of theory. You can also compare one family of ionic liquids against a second family to see whether your trend holds across design rules. Clear error analysis and a careful discussion of model limits will make the work look much more like real research.

Project Variations

• Screen ionic liquids for selectivity between Nd and Pr instead of Nd and Dy to test a lighter rare-earth pair.
• Compare imidazolium-based ionic liquids with phosphonium-based ionic liquids to see which cation family gives better predicted separation.
• Add a machine learning ranking step after the DFT and COSMO-RS screening to test whether simple molecular descriptors predict selectivity.

Learn More

• PubChem: Search structures and basic properties for ionic liquids and related organic ions in the NIH chemical database.
• NIST Chemistry WebBook: Find thermochemical and spectral reference data for comparison points and sanity checks.
• ORCA Manual: Read the free documentation for DFT setup, basis sets, and output interpretation.
• MIT OpenCourseWare Quantum Chemistry: Review core ideas like molecular orbitals, energy surfaces, and electronic structure.
• Journal of Molecular Liquids: Search review articles on ionic liquids, solvation, and separation chemistry through your school library or public abstracts.
• PubMed: Search review articles on rare-earth separation, ionic liquids, and solvent extraction.