Computational design of molecules for an all-quinone redox flow battery

Süleyman Er, Changwon Suh, Michael P. Marshak, Alán Aspuru‐GuzikChemical 2014 · DOI

TL;DR

This paper presents a high-throughput computational screening of over 1700 quinone-based molecules to identify promising candidates for an all-quinone aqueous redox flow battery. By developing quantitative structure-property relationships (QSPRs), the researchers provide a roadmap for designing new molecules with tunable redox potentials and solubilities, paving the way for cheaper and more efficient energy storage.

What Problem Does This Paper Address?

Stationary energy storage is crucial for integrating renewable energy sources like solar and wind into the grid. Redox flow batteries (RFBs) are a promising technology for this purpose, but current metal-based systems are expensive. Organic molecules, particularly quinones, offer a cheaper alternative due to their abundance and tunable properties. However, the vast chemical space of potential quinone derivatives makes experimental screening impractical. This paper addresses the challenge of efficiently discovering and designing new quinone-based molecules with suitable electrochemical properties (redox potential and solubility) for high-performance aqueous RFBs.

Methods

The researchers employed a high-throughput (HT) computational screening approach. They generated a virtual library of 1710 quinone (Q) and hydroquinone (QH2) redox couples by systematically functionalizing parent quinone structures (benzoquinones, naphthoquinones, and anthraquinones) with 18 different R-groups, considering both single and full substitution patterns.

Key computational techniques included:

  • Density Functional Theory (DFT): Used to optimize molecular structures and calculate total energies. Specifically, the Perdew-Burke-Ernzerhof (PBE) functional with the projector augmented wave (PAW) method and a plane-wave basis set was employed.
  • Force Field Calculations: Used to generate initial low-energy conformers for each molecule.
  • Poisson-Boltzmann Solvation Model: Used in conjunction with DFT to estimate solvation free energies (ΔGsolv) as a descriptor for aqueous solubility.
  • Quantitative Structure-Property Relationships (QSPRs): Developed to correlate molecular structure with electrochemical properties (redox potential, Eº) and solubility.

The computational workflow involved optimizing thousands of conformers for each molecule, calculating their DFT total energies, and then using a calibrated theoretical model to predict Eº and ΔGsolv.

Datasets

The paper does not explicitly mention training on pre-existing experimental datasets for the entire screening library. However, it states that a "training set" of experimental data on aqueous redox Q/QH2 couples and their computed DFT energies was used to calibrate their theoretical model for predicting redox potentials (Eº). This calibration model showed a high linear correlation (R² = 0.974) with experimental data. The paper also references experimental data for AQDS (9,10-anthraquinone-2,7-disulphonic acid) as a reference point for both redox potential and solubility.

The primary "dataset" introduced by this work is the virtual library of 1710 quinone/hydroquinone redox couples, along with their computationally predicted redox potentials and solvation free energies.

Key Insights and Ideas

  • All-Quinone RFBs are Feasible: The study demonstrates that it's possible to find suitable quinone molecules for both the negative and positive electrodes of an RFB, enabling an "all-quinone" system. This is significant because it eliminates the need for expensive and potentially toxic metal-based electrolytes.
  • Structure-Property Relationships are Predictable: The QSPR analysis reveals clear trends:
  • Electron-Donating Groups (EDGs) like -OH and -NH2 generally decrease the redox potential, making them suitable for the negative electrode.
  • Electron-Withdrawing Groups (EWGs) like -SO3H, -PO3H2, and -NO2 generally increase the redox potential, making them suitable for the positive electrode.
  • Position Matters: The location of substituents significantly impacts both redox potential and solubility. Groups adjacent to the ketone functionalities have a more pronounced effect on redox potential, while groups further away can enhance solubility without drastically altering the redox potential.
  • Full Substitution is Powerful: Substituting all available hydrogens on the quinone core with R-groups often leads to wider windows of tunable redox potentials and improved solubility compared to single substitutions.
  • Solubility is Key: While redox potential is critical, aqueous solubility is equally important for practical flow batteries. The study highlights specific functional groups (-SO3H, -PO3H2, -OH, -NH2, -COOH) that are particularly effective at increasing solubility.

Key Results

  • Screened Library: A virtual library of 1710 quinone/hydroquinone redox couples was computationally screened.
  • Promising Candidates Identified:
    • 104 Q/QH2 couples with predicted redox potentials (Eº) lower than 0.2 V vs. SHE (suitable for the negative electrode).
  • 304 Q/QH2 couples with predicted Eº higher than 0.9 V vs. SHE (suitable for the positive electrode).
  • These candidates, when paired, can achieve a practical battery voltage greater than 0.7 V.
  • QSPR Insights: The study provides detailed correlations between functional groups, their positions, and their impact on Eº and ΔGsolv. For example, EDGs like -OH and -NH2 decrease Eº, while EWGs like -SO3H and -NO2 increase it. Hydrophilic groups like -SO3H and -PO3H2 are predicted to significantly increase aqueous solubility.
  • Reference Point: The 9,10-anthraquinone-2,7-disulphonic acid (AQDS) molecule, previously used in an AFB, is confirmed as a suitable negative electrode material with a low Eº (0.222 V calculated, 0.213 V measured) and high solubility (>1 M at pH 0).

Limitations and Future Work

  • Solubility Prediction: The authors acknowledge that their ΔGsolv calculation is a descriptor of solubility and that solid-state structures and crystal packing energies are not considered. Therefore, solubility predictions are qualitative and serve as a guide rather than definitive values.
  • Stability: The paper briefly discusses molecular stability, noting potential limitations like polymerization and oxidation. It suggests that replacing C-H bonds with groups like hydroxyl, sulfonyl, amino, or carboxyl can mitigate these issues. However, predicting decomposition pathways and reaction rates computationally is complex and remains an area for future research.
  • Experimental Validation: While the study provides a refined list of synthetically feasible quinones, experimental synthesis and testing are the crucial next steps to validate these computational predictions.
  • Electrode Interactions: The paper mentions that minimal membrane crossover is an advantage of quinones, but further investigation into interactions with electrodes and potential decomposition pathways due to solvent pH and electrical bias is suggested.

The future work clearly lies in synthesizing and experimentally testing the most promising candidates identified through this computational screening, and further refining stability predictions.

Summary updated

Abstract

Inspired by the electron transfer properties of quinones in biological systems, we recently showed that quinones are also very promising electroactive materials for stationary energy storage applications. Due to the practically infinite chemical space of organic molecules, the discovery of additional quinones or other redox-active organic molecules for energy storage applications is an open field of inquiry. Here, we introduce a high-throughput computational screening approach that we applied to an accelerated study of a total of 1710 quinone (Q) and hydroquinone (QH<sub>2</sub>) (<i>i.e.</i>, two-electron two-proton) redox couples. We identified the promising candidates for both the negative and positive sides of organic-based aqueous flow batteries, thus enabling an all-quinone battery. To further aid the development of additional interesting electroactive small molecules we also provide emerging quantitative structure-property relationships.
BibTeX
@inproceedings{er2014computational,
  author = {Süleyman Er and Changwon Suh and Michael P. Marshak and Alán Aspuru‐Guzik},
  title = {{Computational design of molecules for an all-quinone redox flow battery}},
  booktitle = {Chemical},
  year = {2014},
  doi = {10.1039/c4sc03030c},
}
Figure