Accurate prediction of inverted singlet-triplet excited states using self-consistent spin-opposite perturbation theory
Nhan Tri Tran, Hoang Thanh Nguyen, Lan Nguyen Tran · 2026
Read Accurate prediction of inverted singlet-triplet excited states using self-consistent spin-opposite perturbation theory by Nhan Tri Tran, Hoang Thanh Nguyen, Lan Nguyen Tran (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2603-11891.txt.
Introduction
Tran and co-authors test a cheaper way to predict inverted singlet-triplet gaps, where Hund's rule flips. Such inverted molecules could let OLEDs hit full internal quantum efficiency without heat. Standard predictors of these negative gaps cost too much to run widely. The team benchmarks their one-body perturbation methods, OBMP2 and O2BMP2, on 30 inverted-gap molecules. With the right spin-opposite scaling, O2BMP2 matches the accuracy of ADC(3) and EOM-CCSD. Its cost can fall to fourth-power scaling, suiting wide screening of new materials.
Catalog record
| CREATOR | Nhan Tri Tran, Hoang Thanh Nguyen, Lan Nguyen Tran |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 28146 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2603.11891
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Accurate prediction of inverted singlet-triplet excited states using self-consistent spin-opposite perturbation theory Nhan Tri Tran, Hoang Thanh Nguyen, Lan Nguyen Tran Abstract The violation of Hund’s rule, resulting in an inverted singlet-triplet (INVEST) gap, represents a paradigm shift in photophysics with major implications for OLED technology. INVEST molecules facilitate barrierless reverse intersystem crossing, theoretically permitting 100% internal quantum efficiency without thermal activation. However, accurately predicting negative singlet-triplet energy gaps typically demands prohibitive computational costs. In this study, we evaluate the efficacy of our recently developed one-body Møller–Plesset perturbation theory (OBMP2) and its spin-opposite variant (O2BMP2) as efficient alternatives. Benchmarking against 30 INVEST molecules reveals that O2BMP2, with appropriate spin-opposite scaling, achieves the accuracy of ADC(3) and EOM-CCSD. Furthermore, with the possibility of reducing computational complexity to N 4 N^{4} , O2BMP2 provides a robust balance of accuracy and efficiency, making it suitable for the high-throughput screening of next-generation INVEST materials. I Introduction Organic Light Emitting Diodes (OLEDs) have attracted significant attention in display and lighting technologies due to their outstanding energy efficiency, high contrast ratios, and design flexibility. However, the recombination of charge carriers results in a 1:3 ratio of singlet and triplet excited states, which limits the internal quantum efficiency of OLEDs and leads to energy losses. Numerous efforts have been dedicated to improving OLED efficiency. Among the most promising strategies is the development of organic molecules with inverted singlet-triplet energy gaps (INVEST), where the first excited singlet state (S1) is energetically lower than the lowest triplet state (T1), thus breaking Hund’s rule 10 ; 3 . This energetic inversion significantly enhances reverse intersystem crossing (RISC) rates and facilitates exothermic RISC, allowing for rapid and thermally barrier-free triplet-to-singlet upconversion. Consequently, the internal quantum efficiency could potentially reach 100%. However, experimentally quantifying the S1-T1 energy gap, \Delta E_{\text{ST}} , poses challenges due to the spin-forbidden nature of the S1 to T1 transitions observed in spectroscopy. As a result, quantum chemical calculations have been primarily employed to compute \Delta E_{\text{ST}} , thereby expanding the landscape of potential INVEST molecules 48 ; 47 . Recently, high-throughput virtual screening, coupled with symmetry-based design strategies, has facilitated a systematic exploration of vast chemical spaces, resulting in the identification of several new candidates that exhibit inverted singlet–triplet energy gaps 35 ; 34 . \Delta E_{\text{ST}} is conventionally defined as the difference between the Hamiltonians restricted to singlet and triplet subspaces, corresponding to twice the matrix of positive-semidefinite two-electron integrals 47 . Consequently, the inclusion of doubly excited configurations is essential for an accurate treatment of INVEST molecules 8 . As a result, widely used linear-response approaches for excited states, such as configuration interaction with single excitations (CIS) and time-dependent density functional theory (TD-DFT) under the Tamm–Dancoff approximation, fail to reliably predict \Delta E_{\text{ST}} . The inadequacy of TD-DFT for INVEST excited states and the significance of double excitation configurations have been examined in detail in several studies 37 ; 11 . Recently, it has been demonstrated that TD-DFT calculations employing frequency-dependent kernels, double-hybrid functionals, and other schemes that explicitly or implicitly incorporate double excitations can yield negative values of \Delta E_{\text{ST}} 5 . Nevertheless, the predictive performance of these approaches remains inconsistent. Delta self-consistent field methods ( \Delta SCF) 12 ; 6 ; 13 constitute a variational alternative to linear-response theory for the treatment of excited states. \Delta SCF has been extensively employed to describe X-ray absorption 32 ; 28 , valence excited states, and core-level excitations 4 ; 1 ; 1 ; 14 ; 15 , as well as nonadiabatic molecular dynamics simulations 46 ; 45 ; 25 . For INVEST molecules, several studies have computed vertical excitation energies with \Delta SCF 44 ; 18 . Numerical results indicate markedly improved agreement with experimental data for both S1 and T1 energies relative to TD-DFT, implying that orbital relaxation is crucial for an explicit description of double and higher-order excitations in these emissive states. Recently, quasiparticle energy DFT (QE-DFT), wherein STGs are obtained directly from differences in Kohn–Sham orbital energies, has been shown to provide a highly efficient and accurate alternative for predicting INVEST molecules 22 . Consequently, with favorable computational cost, \Delta DFT is anticipated to be suitable for high-throughput virtual screening. Nevertheless, one of the principal challenges of DFT remains system dependent 44 ; 18 . Correlated wavefunction methods, including second-order approximate coupled cluster (SCS-CC2) and its spin-component scaling variant (SCS-CC2), second- and third-order algebraic diagrammatic construction theory (ADC(2) and ADC(3)), equation-of-motion coupled-cluster singles and doubles (EOM-CCSD), N -electron valence state second-order perturbation theory (NEVPT2), and multireference second-order complete active space perturbation theory (CASPT2), have demonstrated negative \Delta E_{\text{ST}} for a broad class of systems 35 ; 34 ; 24 . Notably, Loos and co-workers showed that CC2 and ADC(2) can approach CC3 and EOM-CCSDT accuracies 24 . Despite their accuracy, correlated wavefunction methods are prohibitively expensive for high-throughput screening of thousands or millions of molecules 33 . Hence, there remains a strong demand for affordable yet reliable methods to predict INVEST excited states. We have developed a novel self-consistent perturbation theory, termed one-body Møller–Plesset perturbation theory (OBMP2) 39 ; 40 ; 41 ; 21 . In OBMP2, a canonical transformation 50 ; 51 ; 7 ; 29 ; 31 ; 30 followed by a cumulant approximation 19 ; 27 ; 26 ; 20 yields an effective one-body Hamiltonian. The resulting operator comprises the standard Fock term plus a one-body correlation potential that includes double-excitation MP2 amplitudes. Molecular orbitals and orbital energies are then optimized in the presence of correlation by diagonalizing a correlated Fock matrix. The self-consistency of OBMP2 mitigates issues associated with the non-iterative nature of standard MP2 for open-shell systems 40 ; 41 and for bond-stretching regimes 21 . In many cases, OBMP2 outperforms orbital optimized MP2 (OO-MP2) for open-shell systems 41 . We have further extended OBMP2 to a spin-opposite variant (O2BMP2) and demonstrated that OBMP2 and O2BMP2 can treat intermolecular charge-transfer excited states with accuracy comparable to high-level coupled-cluster methods 43 . Consequently, assessing the performance of these methods for INVEST excited states is particularly timely, given the need to account for double excitations, charge transfer, and stronger electron correlation effects relative to exchange. We have carried out an assessment on the performance of our methods for 30 INVEST molecules that are taken from Refs 24 and 34 . We find that O2BMP2, with an appropriate spin-opposite scaling ( c_{\text{os}} ), can describe INVEST excited states with accuracy comparable to ADC(3) and EOM-CCSD. Importantly, the formal scaling of O2BMP2 can be reduced from N^{5} to N^{4} , analogous to spin-opposite MP2 and OO-MP2 23 . Therefore, O2BMP2 offers high accuracy at affordable cost and is well suited for high-throughput screening aimed at identifying new molecular systems exhibiting INVEST excited states. II Theory and computational details A comprehensive derivation of the OBMP2 theoretical framework is provided in Refs. 40 ; 41 ; 21 ; 42 ; the method has been implemented within a locally modified version of the PySCF quantum chemistry package 38 . The OBMP2 effective Hamiltonian is constructed via the canonical transformation formalism 50 ; 51 ; 7 ; 29 ; 31 ; 30 : \displaystyle\hat{\bar{H}}=e^{\hat{A}^{\dagger}}\hat{H}e^{\hat{A}}, (1) where the molecular electronic Hamiltonian is expressed as \displaystyle\hat{H}=\sum_{pq}h^{p}_{q}\hat{a}_{p}^{q}+\tfrac{1}{2}\sum_{pqrs}g^{pr}_{qs}\hat{a}_{pr}^{qs}. (2)