[Paper Review] Towards more accurate First Principles prediction of redox potentials in transition-metal compounds with LDA+U

Fei Zhou, Matteo Cococcioni+3 more|arXiv (Cornell University)|Jun 16, 2004

Advancements in Battery Materials1 references3 citations

TL;DR

This paper proposes using self-consistently determined DFT+$U$ parameters to correct electron self-interaction errors in LDA/GGA, significantly improving the accuracy of first-principles predictions for redox potentials in transition-metal compounds. The method successfully reproduces experimental lithium intercalation voltages across diverse materials including olivine, layered, and spinel-type Li$_{x}$MPO$_{4}$, Li$_{x}$MO$_{2}$, and Li$_{x}$M$_{2}$O$_{4}$ compounds.

ABSTRACT

First-principles calculations within the Local Density Approximation (LDA) or Generalized Gradient Approximation (GGA), though very successful, are known to underestimate redox potentials, such as those at which lithium intercalates in transition metal compounds. We argue that this inaccuracy is related to the lack of cancellation of electron self-interaction errors in LDA/GGA and can be improved by using the DFT+$U$ method with a self-consistent evaluation of the $U$ parameter. We show that, using this approach, the experimental lithium intercalation voltages of a number of transition metal compounds, including the olivine Li$_{x}$MPO$_{4}$ (M=Mn, Fe Co, Ni), layered Li$_{x}$MO$_{2}$ ($x=$Co, Ni) and spinel-like Li$_{x}$M$_{2}$O$_{4}$ (M=Mn, Co), can be reproduced accurately.

Motivation & Objective

• To address the well-known underestimation of redox potentials in LDA/GGA calculations for transition-metal compounds.
• To identify the root cause of this inaccuracy as incomplete cancellation of electron self-interaction errors.
• To improve prediction accuracy by employing a self-consistent evaluation of the $U$ parameter in the DFT+$U$ method.
• To validate the method across a diverse set of lithium-ion battery materials, including olivine, layered, and spinel structures.

Proposed method

• Application of the DFT+$U$ method with a self-consistent determination of the Hubbard $U$ parameter to correct for self-interaction errors in LDA/GGA.
• Use of a linear response approach to compute the effective $U$ parameter within a self-consistent framework.
• Implementation of the method in first-principles calculations for transition-metal oxides with varying transition-metal cations.
• Systematic evaluation of redox potentials in Li$_{x}$MPO$_{4}$, Li$_{x}$MO$_{2}$, and Li$_{x}$M$_{2}$O$_{4}$ compounds using the self-consistent $U$ approach.
• Comparison of predicted voltages with experimental lithium intercalation potentials to assess accuracy.

Experimental results

Research questions

• RQ1 Can self-consistently determined $U$ parameters in DFT+$U$ correct the underestimation of redox potentials in LDA/GGA for transition-metal compounds?
• RQ2 To what extent does the self-consistent $U$ approach improve agreement with experimental redox potentials across diverse transition-metal oxide structures?
• RQ3 How does the cancellation of self-interaction errors influence the accuracy of redox potential predictions in lithium-ion battery materials?
• RQ4 Does the method maintain accuracy across different crystal structures, including olivine, layered, and spinel types?

Key findings

• The self-consistently determined $U$ parameter effectively reduces electron self-interaction errors, leading to more accurate redox potential predictions.
• The method reproduces experimental lithium intercalation voltages for olivine-type Li$_{x}$MPO$_{4}$ (M=Mn, Fe, Co, Ni) with high accuracy.
• Accurate prediction of redox potentials is achieved for layered Li$_{x}$MO$_{2}$ compounds (x=Co, Ni) using the self-consistent $U$ approach.
• The method also successfully predicts voltages for spinel-like Li$_{x}$M$_{2}$O$_{4}$ compounds (M=Mn, Co), demonstrating broad applicability across different transition-metal oxide structures.