Computation of Uranium Ligand Field Parameters

A global optimization based on Lever's theory

This website accompanies the paper «Non-Aqueous Electrochemistry of Uranium Complexes — A Guide to Structure-Reactivity Tuning» (submitted to Chem. Rev. in Dec. 2022). In the following sections, you can modify the set of ligand measurements to be included in the global optimization of the Lever parameters to study the effect of individual measurements. The source of this website is hosted on GitHub: jud1thr/uranium-lfp.

The process works as follows:

  1. Make a selection of ligand data in the table below by ticking the ignore field to exclude certain rows.
  2. Optionally: Add new ligand data using the form at the bottom of the table.
  3. Run the global optimization. Once finished (it should only take a moment), you will get an indication of the optimization result (typically «solution is optimal»), and a few tabs with results appear:
    • Variables holds the optimized value for each component; err denotes the overall error (cf. paper).
    • CPLEX shows the optimization program in CPLEX format, in case you need it for an external solver.
    • Result shows the detailed optimization result by the glpsol solver; this can provide some insight to the numerical constraints
    • Correlation Analysis provides \(R^2\) and (Pearson's) \(r\) values, as well as a fit plot (hover a point to see the ligand data).
    You will also notice that the ligand table now has a column \(\sum ~^UE_L\) that displays the computed value (opposed to the measured \(E_{1/2} (U^{IV/III})\)).

Feel free to reach out to judith.riedhammer@fau.de if you have any questions!

This website makes heavy use of Javascript (e.g. vue, MathJax, glpk.js) and cutting-edge browser features (html5, es6, svg). Please use a current version of Safari, Chrome, Firefox or Edge and disable any Javascript blocker.

Select ligand data to include in the optimization (→ optimization)

Please contact the authors to have your data included, or make a pull request right here on GitHub.
# Ligand Components \(E_{1/2} (U^{IV/III})\) Ignore
{{i+1}} local {{l.e12}} {{l.sum || ''}}

You can (temporarily) add new ligands to the table, to play around with the optimization problem. You need to write those in JSON format; see the two templates (with single and range \(E_{1/2}\)) below.

Hints:

  • You can also click on a line number in the table above to copy the respective values into the field below.
  • You can use LaTeX/mhchem notation for better readibility (cf. existing ligand data).
  • Double-check the equation components to make sure to spell existing components right!

{"lig": "[(tmp)2U(BH4)2]", "eqn": {"tmp": 2, "BH4": 2}, "e12": -1.50}
{"lig": "[(Cp)3U(BH4)]", "eqn": {"Cp": 3, "BH4": 1}, "e12": [-2.0, -1.88]}
{{input_error}}

Global Optimization

Use the tabs below to learn about the optimization results.

  • Variables holds the optimized value for each component; err denotes the overall error (cf. paper).
  • CPLEX shows the optimization program in CPLEX format, in case you need it for an external solver.
  • Result shows the detailed optimization result by the glpsol solver; this can provide some insight to the numerical constraints
  • Correlation Analysis provides \(R^2\) and (Pearson's) \(r\) values, as well as a fit plot (hover a point to see the ligand data).
Variable Component Value
{{u.var}} {{u.val}} 
{{cplex}}
{{res}}
Note that for those ligands with a range of E1/2, the bound closer to the sum is used. Hover a point on the plot to see the ligand data.

R^2 = {{rsquared}}

Pearson r = {{corr}}