betaFeedback

WorksPhysics and Astronomy

jahn-teller-dynamics: Python package to determine vibronic interaction demonstrated on molecules and trigonal defect qubits

Abstract

Trigonal solid-state defects are often subjects of spontaneous symmetry breaking driven by the E⊗eE\otimes e Jahn-Teller effect, reflecting strong electron-phonon coupling. These systems, particularly paramagnetic defect qubits in solids are central for quantum technology applications, where accurate knowledge of their fine-structure parameters - shaped by the complex interplay of spin-orbit and electron-phonon interactions - is essential. We introduce the jahn-teller-dynamics\mathtt{jahn {\text -} teller {\text -} dynamics} package, a Python code that implements the first-principles approach of [Phys. Rev. X 8, 021063 (2018)] to accurately compute the spin-orbit-phonon entanglement in trigonal defects utilizing the output from density functional theory calculations (DFT) to predict fine-structure parameters of zero-phonon lines (ZPLs), including Zeeman shifts under external magnetic fields. We demonstrate its capabilities on negatively charged Group-IV-vacancy (G4V) defects in diamond: SiV−^-, GeV−^-, SnV−^-, PbV−^- and the neutral N3_3V0^0 defect in diamond, and the CH3_3O0^0 methoxy radical. Additionally, we implement a generic electron-phonon code that is capable entangling an arbitrary amount of (i) vibration modes and (ii) electronic levels by (iii) arbitrarily high order of vibronic interaction terms. Exemplarily, we demonstrate the Jahn-Teller multimode problem on the CH3_3O0^0 methoxy radical and the Pseudo Jahn-Teller case on the C4H4+C_4H_4^+ butatrien cation.

Program Summary/New Version Program Summary

Program Title: jahn-teller-dynamics

CPC Library link to program files: (to be added by Technical Editor)

Developer’s repository link: https://github.com/tbalu98/Jahn-Teller-Dynamics https://pypi.org/project/jahn-teller-dynamics

Code Ocean capsule: (to be added by Technical Editor)

Licensing provisions(please choose one): GPLv3

Programming language: Python 3.11

Data availability: https://doi.org/21.15109/ARP/EXJKGL

Nature of problem(approx. 50-250 words): Electron states that have close or degenerate or degenerate energy levels are susceptible to atomic vibrations caused by atomic nuclei that causes the dynamic Jahn-Teller and pseudo Jahn-Teller interaction. Indeed, density functional theory (DFT) and multiconfigurational Hartree-Fock calculations within the Born-Oppenheimer approximation can be employed to determine the electronic structure in different geometrical configurations or determine the spin-orbit coupling strength. However, in order to solve Jahn-Teller problem one has to go beyond the Born-Oppenheimer approximation that is usually well beyond the capabilities of current state-of-art codes. Therefore, a first-principles approach is necessary to accurately determine the fine-structure parameters such as the spin-orbit splitting visible in optical measurements utilizing trigonal color centers and molecules.

Solution method(approx. 50-250 words): In jahn – teller – dynamics library we implemented a general formulation of dynamic Jahn-Teller theory, where arbitrary number of electron states and phonon modes interact with each other up to any order. Therefore, the user can freely customize the Hamilton operator of the electron-phonon interaction. We dedicated extra care to the E⊗eE \otimes e dynamic Jahn-Teller interaction, that is very common among point defects. We implemented methodology developed in Refs. [1, 2]. It utilizes the results directly obtained from the VASP[3] density functional theory (DFT) code to formulate the Hamiltonian of the E⊗eE \otimes e Jahn-Teller case. Our solution includes the non-perturbative effect of spin-orbit coupling and the dynamic Jahn-Teller interaction simultaneously, where we observe that these two interactions non-trivially entangle with each other. Our methodology determines the damping of experimentally visible spin-orbit splitting known as Ham [4, 5, 6] reduction factor pp that can be identified as electron-phonon renormalization of physical observables such as the renormalization of spin-orbit coupling strength [7, 8, 9, 10] that of trigonal defects. Our code provides a framework that automatically reads and post-processes the DFT results to compute fine-structure parameters and it includes the effect of external magnetic fields.

Introduction

Point defects, such as vacancies, dopants and defect complexes in semiconductors and insulators such as diamond and silicon carbide hold great promise to realize solid-state quantum bits [11, 12, 13, 14, 15, 16]. The optically active quantum bit candidates are fluorescent exhibiting coherent emission within their zero-phonon-line (ZPL) which play a crucial role in setting, manipulating and readout of defect-based quantum bits fully optically [17, 18, 19, 20, 21, 22, 23, 24, 25]. The underlying defect quantum bits often exhibit high symmetry, giving rise to degenerate orbitals that couple to spin, resulting in fine-structure splitting, and interact strongly with phonons also known as vibronic states. According to dynamic Jahn-Teller theory, which goes beyond the standard adiabatic approximation [26, 27, 28, 29], the strong electron-phonon interaction can quench orbital angular momentum and thereby reduce the effective spin-orbit coupling [1, 2].

In this work, we introduce Exe.py, an open-source Python code designed to simulate the fine structure parameters governing ZPL transitions of these defects under applied magnetic fields, using input from density functional theory (DFT) calculations. We note that our scheme can be applied to dynamic Jahn-Teller active trigonal molecules mostly exhibiting the XCH3_3 structure where X = Cd, Zn, Mg, Ca, O, Sr, etc. [5, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42]. The code is developed using widely adopted scientific libraries such as NumPy, Pandas, SciPy, and Matplotlib. It implements the formalism presented in Refs. [1, 2], allowing direct comparison with experimental observations.

Notable examples of such defects include nitrogen-vacancy center [17, 18] and group-IV–vacancy centers [2] in diamond or silicon-vacancy [43, 44, 45, 46] and divacancy centers [47, 48, 49, 50] in silicon carbide where other host materials with defects [13]

and molecular systems [51] have recently emerged as quantum bit candidates. Exemplarily, we compare the experimental ZPL spectrum that of SnV−^- [24] with our theoretically predicted spin-orbit coupling fine structure parameters (λexptgnd,ex/λtheorygnd,ex\lambda_{\mathrm{expt}}^{\mathrm{gnd,ex}}/\lambda_{\mathrm{theory}}^{\mathrm{gnd,ex}}). Figure 1 represents how they can be measured from the photoluminescence spectrum. The loss of degeneracy due to spin-orbit coupling manifests in ZPL splitting. Indeed there are four possible transition energies between the ground and excited state, however in low temperatures only the lowest orbital states are populated, therefore we anticipate two peaks. Higher orbital states are populated at higher temperatures (100 K), but they are hard to resolve because of the increasing intensity of the phonon bath.

Figure 1. Relationship between the experimentally observed photoluminescence spectrum of SnV−^- and the theoretical value of spin-orbit coupling λexptgnd,ex/λtheorygnd,ex\lambda_{\mathrm{expt}}^{\mathrm{gnd,ex}}/\lambda_{\mathrm{theory}}^{\mathrm{gnd,ex}} of the ground (gnd) and excited (ex) states. The data for the experimental spectrum is taken from Ref. [24].

Basic theory

Most generally, one may consider the following H^\hat{H} Hamiltonian

H^=H^e+H^ph+H^e-ph.(1)\hat{H} = \hat{H}_{e} + \hat{H}_{ph} + \hat{H}_{e\text{-}ph}. \tag*{(1)}

that depicts the interaction between electronic (H^e\hat{H}_{e}) and vibrational (H^vib\hat{H}_{\mathrm{vib}}) degrees of freedom spanned by a set of many-electron wavefunctions ∣n⟩\lvert n \rangle indexed by nn labels exhibiting εn\varepsilon_n eigenenergies and a set of quantum harmonic oscillators indexed by ii labels depicted by their creation ai†a_i^\dagger and annihilation aia_i ladder operators exhibiting ωi\omega_i energy quanta, respectively:

H^e=∑nεn∣n⟩⟨n∣,H^vib=∑iωi(ai†ai+12).(2)\hat{H}_{e} = \sum_n \varepsilon_n \lvert n \rangle\langle n \rvert,\qquad\hat{H}_{\mathrm{vib}} = \sum_i \omega_i \left(a_i^\dagger a_i + \frac{1}{2}\right). \tag*{(2)}

The occupation number representation of the N-mode quantum harmonic oscillator can be introduced by consecutive creation of phonon quanta on the vacuum state:

∣ν1,ν2,…,νN⟩=∏i(1νi!(ai†)νi)∣0,0,…,0⟩.(3)\lvert\nu_1,\nu_2,\ldots,\nu_N \rangle= \prod_i \left(\frac{1}{\sqrt{\nu_i!}}(a_i^\dagger)^{\nu_i}\right)\lvert0,0,\ldots,0\rangle. \tag*{(3)}

The resulting basis states of the Hilbert space H\mathcal{H} are

bj=∣n,ν1,ν2,…,νN⟩(4)b_j = \lvert n,\nu_1,\nu_2,\ldots,\nu_N\rangle \tag*{(4)}

The interaction Hamiltonian (H^e-vib=∑nm∣n⟩⟨m∣H^e-vibnm\hat{H}_{e\text{-}vib} = \sum_{nm} \lvert n\rangle\langle m\rvert\hat{H}_{e\text{-}vib}^{nm}) is a generic series expansion made of q^i=12(ai†+ai)\hat{q}_i = \frac{1}{\sqrt{2}}(a_i^\dagger+a_i) coordinate operators and ∣n⟩⟨m∣\lvert n\rangle\langle m\rvert many-electron excitation operators[37]:

H^e-vibnm=∑αFαnmq^α+∑αβFαβnmq^αq^β+∑αβγFαβγnmq^αq^βq^γ+∑αβγδFαβγδnmq^αq^βq^γq^δ+…(5)\begin{aligned} \hat{H}_{e\text{-}vib}^{nm} ={}& \sum_\alpha F_\alpha^{nm}\hat{q}_\alpha+ \sum_{\alpha\beta} F_{\alpha\beta}^{nm}\hat{q}_\alpha\hat{q}_\beta\\ &+ \sum_{\alpha\beta\gamma} F_{\alpha\beta\gamma}^{nm}\hat{q}_\alpha\hat{q}_\beta\hat{q}_\gamma+ \sum_{\alpha\beta\gamma\delta} F_{\alpha\beta\gamma\delta}^{nm}\hat{q}_\alpha\hat{q}_\beta\hat{q}_\gamma\hat{q}_\delta+ \ldots \tag*{(5)} \end{aligned}

The matrix elements describe the coupling between the ∣n⟩\lvert n\rangle and ∣m⟩\lvert m\rangle electron states as a Taylor series expansion up to an arbitrary order in terms of the normal modes q^α\hat{q}_\alpha. Fαnm=1k!fαnmF_\alpha^{nm} = \frac{1}{k!}f_\alpha^{nm}, where fαnmf_\alpha^{nm} is the kth order derivative of the adiabatic potential energy surface V(q^1,q^2,…)V(\hat{q}_1,\hat{q}_2,\ldots). Fαβ…nmF_{\alpha\beta\ldots}^{nm} for all permutations of indexes αβ…\alpha\beta\ldots coefficients are equal.

The position operators can be dimensionless x^i=12(a^i†+a^i)\hat{x}_i = \frac{1}{\sqrt{2}}(\hat{a}_i^\dagger+\hat{a}_i) or dimensional x^i=12ωi(a^i†+a^i)\hat{x}_i = \frac{1}{\sqrt{2\omega_i}}(\hat{a}_i^\dagger+\hat{a}_i). ωi\omega_i is the energy quantum of the iith vibrational mode, a^i†\hat{a}_i^\dagger is the creator and a^i\hat{a}_i is the annihilator operator.

This general Hamiltonian operator can be built by jtd_run program. Example files of user inputs are provided in the github repository along with a technical description. Here we aim to cover the Hamiltonian building principles of our code. The code requires the user to describe NeN_e electron states εn\varepsilon_n energy levels and the vibration energy quantum of the vibrational modes ϵi\epsilon_i. jtd_run code can build tensor product phonon states with respect to a maximum phonon quanta number on each modes NiN_i such that ∣ν1,ν2,…⟩\lvert\nu_1,\nu_2,\ldots\rangle such that for each vibrational quanta number νi≤Ni\nu_i \le N_i. The other option is that the user can define the maximum energy (EE) of the phonon basis, such that ∑iνiωi≤E\sum_i \nu_i\omega_i \le E, or the maximum number of phonon quanta NphN_{\mathrm{ph}}, such that ∑iνi≤Nph\sum_i \nu_i \le N_{\mathrm{ph}}.

Hamiltonian building principles of jtd_run

Due to symmetry considerations for any arbitrary order kk not all monomials have distinct coupling coefficients. For instance in case of E⊗eE \otimes e dynamic Jahn-Teller interaction, all second order monomials couples to the electron states with the same coefficients. Therefore, the coupling between two electron states ∣n⟩\lvert n\rangle and ∣m⟩\lvert m\rangle can be defined by arbitrary pnm,klp_{nm,kl} polynomial consists of kkth order monomials with a coupling coefficient cpklnmc_{p_{kl}}^{nm}. The kkth order polynomials are indexed by ll in order to take into account all the kkth order polynomial that couples ∣n⟩\lvert n\rangle and ∣m⟩\lvert m\rangle. jtd_run can be configured to create a permutation completion, which means that it automatically generate all distinct permutation of pnm,klp_{nm,kl}’s monomials. In this case of a tensor product phonon basis, position operators commute. Therefore, the monomials can simply be multiplied by the number of unique permutations. Otherwise all distinct monomials are generated, because the position operators do not commute. The resulting new polynomial noted by pnm,kl′p_{nm,kl}^{\prime}.

To concretely exemplify it, in case of a coupling polynomial x3+x2yx^3+x^2y for non commuting and commuting cases:

pkl′nm=x3+x2y+yxy+y2x=x3+3x2y(6)p_{kl}^{\prime nm}=x^3+x^2y+yxy+y^2x=x^3+3x^2y \tag*{(6)}

The code iterates over the user defined input and generates the Hamiltonian operator based on this sum:

H^built=∑nm,klcklnm∣n⟩⟨m∣⊗pkl′nm(7)\hat{H}_{\mathrm{built}}=\sum_{nm,kl}c_{kl}^{nm}\lvert n\rangle\langle m\rvert\otimes p_{kl}^{\prime nm} \tag*{(7)}

Here ∣n⟩⟨m∣\lvert n\rangle\langle m\rvert is an Ne×NeN_e\times N_e matrix. ∣n⟩⟨m∣κλ=1.0\lvert n\rangle\langle m\rvert_{\kappa\lambda}=1.0, if κ=n\kappa=n and λ=m\lambda=m, 0.00.0 otherwise.

For simplicity, we encourage users to define only the upper triangle of the Hamiltonian because Hermitian completion option can be set. Therefore the code automatically performs the summation (7), and includes the hermitian conjugate of each term.

In some cases it can take long time to generate all coupling polynomials, the user can choose to opt out Hermitian completion and set symmetrization such that at the end of the Hamiltonian building procedure H^e−ph=12(H^built+H^built†)\hat{H}_{\mathrm{e-ph}}=\frac{1}{2}(\hat{H}_{\mathrm{built}}+\hat{H}_{\mathrm{built}}^\dagger). Note that in that case the non diagonal coupling coefficients must be doubled.

Single mode E⊗eE\otimes e Jahn-Teller case for defects in solids

To sum up the component of the user defined coupling coefficients:

cklnm=2×1k!⏟Herm. comp.×fklnm(8)c_{kl}^{nm}=2\times\underbrace{\frac{1}{k!}}_{\text{Herm. comp.}}\times f_{kl}^{nm} \tag*{(8)}

Here fklnmf_{kl}^{nm} is the adiabatic potential energy surface derivative with respect to the coordinates of any monomial in pklp_{kl}.

Moreover, the code can optionally diagonally complete the Hamiltonian. Therefore, when the user specifies one diagonal element for an electron state, ii it generates that input for all other states as well. Figure 2 presents the operating principles of jtd_run.

Figure 2. jtd_run reads electron energies, phonon modes and coupling defined in a tabular form. It builds the general Hamiltonian H\mathcal{H} and returns the eigenvalues and eigenvectors, that are meant in terms of the basis states

jtd_run uses Lanczos algorithm as implemented in scipy and SPLPEc in order to find the first NeigenN_{\mathrm{eigen}} eigenvalues and eigenvectors in terms of the basis states required by the user and saves in .npz format, named eigenstates.npz. It contains the basis states of Hilbert space (H\mathcal{H}) in basis_labels.npy, the eigenvalues of the dynamic Jahn-Teller system EηE_\eta (eigenvalues.npy) and the eigenstates ψη\psi_\eta (eigenvectors.npy).

The general purpose of jahn-teller-dynamics is to evaluate the following interaction Hamiltonian [2, 28, 52]

H^=H^SOC+H^vib+H^DJT+H^ext,(9)\hat{H}=\hat{H}_{\mathrm{SOC}}+\hat{H}_{\mathrm{vib}}+\hat{H}_{\mathrm{DJT}}+\hat{H}_{\mathrm{ext}}, \tag*{(9)}

to theoretically predict the fine structure details for defects such as the λtheory≈λexpt\lambda_{\mathrm{theory}}\approx\lambda_{\mathrm{expt}} spin-orbit splitting parameter visible in experiments. In the following sections, we describe the terms of Eq. (9). First, we discuss the intrinsic spin-orbit coupling H^SOC\hat{H}_{\mathrm{SOC}} in § 3, that we determine by means of DFT calculations. Then, we introduce the dynamic Jahn-Teller effect in Sec. 2.4 where H^vib\hat{H}_{\mathrm{vib}} depicts the atomic vibrations that are represented by quantum harmonic oscillators while H^DJT\hat{H}_{\mathrm{DJT}} is the electron-phonon interaction term due to the dynamical Jahn-Teller effect. Finally, the code directly diagonalizes H^\hat{H} yielding the theoretical spin-orbit-phonon spectrum (see Sec. 2.5 for details).

The last term (H^ext=H^mag.+H^strain\hat{H}_{\mathrm{ext}} = \hat{H}_{\mathrm{mag.}} + \hat{H}_{\mathrm{strain}}) depicts the coupling of electronic orbital and spin with external magnetic field, electric field and strain that we discuss in Section 2.6.

We note that the fine structure parameters can be encompassed into an effective model Hamiltonian for the vibronic ground state where the vibrational degrees of freedom (XX, YY) can be traced out [2, 6] by averaging over the phonon bath. We discuss the properties of the effective Hamiltonian in Sec. 2.7 where the intrinsic λ0\lambda_0 spin-orbit parameter is renormalized into an effective λtheory=p×λ0≈λexpt\lambda_{\mathrm{theory}} = p \times\lambda_0 \approx\lambda_{\mathrm{expt}} energy gap visible in experiments through the introduction of Ham reduction factors. Whenever both ground and optical excited states are dynamic Jahn-Teller active multiplets, then (9) can be applied for both states to determine the fine structure features in the optical spectra. In such systems, jahn-teller-dynamics can simulate the fine levels in the optical spectrum as schematically depicted in Figure 3 and Figure 6.

Figure 3. This figure represents the physics behind the zero-phonon-line (ZPL) transitions in trigonal point defects with degenerate electronic ground and excited states. The spin-orbit electronic states of the ground and excited state are fourfold degenerate. Due to spin-orbit coupling they split into two E12E_{\frac{1}{2}} and E32E_{\frac{3}{2}} multiplets and the energy difference between them is λ0\lambda_0. If we take into consideration atomic vibrations by utilizing the E⊗eE \otimes e Jahn-Teller effect case, they will split to two distinct two-fold degenerate energy levels (∣ε1,2⟩\lvert\varepsilon_{1,2}\rangle and ∣ε3,4⟩\lvert\varepsilon_{3,4}\rangle). They are entangled states of phonons, orbitals and spin. Under a constant external magnetic field the system experiences total loss of degeneracy because of Zeeman-effect. A,B,C,D represents the transition types as defined in Ref. [23].

Figure 6. Magnetic field strength dependence of ZPL fine structure that of the SnV⁻ defect in diamond. Experimental data points are taken from Ref. [22] are compared to the result of our Exe.py code for the negatively charged SnV⁻ defect in diamond.

Finally, we exemplarily depict the workflow with our code on experimentally known defects such as G4V defects (SiV, GeV, SnV, PbV) in Sec. 3 including computational details (Sec. 3.1), where we highlight the N3_3V0^0 defect (Sec. 3.3) whose spin-orbit parameter has not been determined by ab-initio tools and compared to experimental data to our best knowledge.

Trigonal system and spin-orbit coupling

We investigate point defects in solids exhibiting trigonal symmetry and electronically degenerate states. According to Jahn-Teller theory, such degenerate many-body electronic states undergo spontaneous symmetry breaking due to geometrical distortions. In this context, adiabatic approximation generally breaks down for degenerate (or quasi-degenerate) many-body electronic defect states in solids, owing to strong electron-phonon interaction. In this work, we focus on defects exhibiting trigonal point-group symmetries such as D3dD_{3d} and C3vC_{3v} that can feature double degenerate orbitals or phonons labeled as ”E” or ”e” in Schönflies notation. For the D3dD_{3d} point group, parity labels are omitted, as they hold no relevance in the present context.

The doubly degenerate orbitals (∣exDFT⟩\lvert e_x^{\mathrm{DFT}}\rangle, ∣eyDFT⟩\lvert e_y^{\mathrm{DFT}}\rangle) of point defects in solids or molecules can be approximately well characterized by DFT methodology. When these degenerate Kohn-Sham orbitals are occupied by a single electron or equivalently left with a single hole results in a doubly degenerate many-body state. Furthermore, we note that taking into account the Kramers degeneracy of spin-12\frac{1}{2} systems yields a combined spin-orbital system exhibiting fourfold degeneracy. By performing non-collinear calculations with spin-orbit effects turned on [53], one can determine the spin-orbit splitting of these degenerate Kohn-Sham orbitals by setting half-half occupation of the degenerate Kohn-Sham levels [1, 2]. As a consequence, the Kohn-Sham states will acquire complex-valued character:

∣e±DFT⟩=∓(∣exDFT⟩±i∣eyDFT⟩),(10)\lvert e_{\pm}^{\mathrm{DFT}}\rangle= \mp\left(\lvert e_x^{\mathrm{DFT}}\rangle\pm i\lvert e_y^{\mathrm{DFT}}\rangle\right), \tag*{(10)}

where the phases are chosen in accordance with the Condon-Shortley convention. Their energy levels are separated by

λ0=E(e+DFT)−E(e−DFT),(11)\lambda_0 = E\left(e_{+}^{\mathrm{DFT}}\right) - E\left(e_{-}^{\mathrm{DFT}}\right), \tag*{(11)}

where E(e±DFT)E\left(e_{\pm}^{\mathrm{DFT}}\right) denote the eigenenergies of the corresponding orbital states. One may observe that the L^z=i(0−110)\hat{L}_z = i\begin{pmatrix}0 & -1 \\ 1 & 0\end{pmatrix} assumes a simpler diagonal form upon transformation to the complex basis of (10): L^z,complex=(100−1)\hat{L}_{z,\mathrm{complex}} = \begin{pmatrix}1 & 0 \\ 0 & -1\end{pmatrix} which explicitly reflects ∣e±DFT⟩\lvert e_{\pm}^{\mathrm{DFT}}\rangle’s the orbital angular momentum quantum number ml=±1m_l = \pm1. The spin states are ∣↑⟩\lvert\uparrow\rangle and ∣↓⟩\lvert\downarrow\rangle, and the spin operator is S^z=12(100−1)\hat{S}_z = \frac{1}{2}\begin{pmatrix}1 & 0 \\ 0 & -1\end{pmatrix}. The spin-orbit interaction operator is

H^SOC=λ0L^z⊗S^z.(12)\hat{H}_{\mathrm{SOC}} = \lambda_0 \hat{L}_z \otimes\hat{S}_z. \tag*{(12)}

As a result, we find that the fourfold degenerate electronic states split into two Kramers doublets:

∣E12⟩={∣e−DFT⟩⊗∣↓⟩, ∣e+DFT⟩⊗∣↑⟩},\lvert E_{\frac{1}{2}}\rangle= \{\lvert e_{-}^{\mathrm{DFT}}\rangle\otimes\lvert\downarrow\rangle,\ \lvert e_{+}^{\mathrm{DFT}}\rangle\otimes\lvert\uparrow\rangle\},
∣E32⟩={∣e−DFT⟩⊗∣↑⟩, ∣e+DFT⟩⊗∣↓⟩}.(13)\lvert E_{\frac{3}{2}}\rangle= \{\lvert e_{-}^{\mathrm{DFT}}\rangle\otimes\lvert\uparrow\rangle,\ \lvert e_{+}^{\mathrm{DFT}}\rangle\otimes\lvert\downarrow\rangle\}. \tag*{(13)}

where the spin-orbit interaction shifts ∣E12⟩\lvert E_{\frac{1}{2}}\rangle upwards while ∣E32⟩\lvert E_{\frac{3}{2}}\rangle downwards in energy by λ0/2\lambda_0/2. However, the computed λ0\lambda_0 substantially overestimates the experimentally measured λexpt\lambda_{\mathrm{expt}} for most defect-based quantum bits [1, 2]. This discrepancy arises from strong electron-phonon coupling: where λ0\lambda_0 represents only the spin-orbit splitting of the unrenormalized (bare) electronic states and is subsequently reduced by the Jahn-Teller effect, as discussed in the following section.

E⊗eE \otimes e Jahn-Teller effect

In this section, we briefly outline how strong electron-phonon coupling in trigonal systems can partially quench the spin-orbit interaction and thereby substantially modify the effective spin-orbit splitting. Within geometry optimization workflows during DFT calculations, one may observe that low symmetry configurations (e.g. C2hC_{2h} or C1hC_{1h}) are energetically favored over the high-symmetry configuration for trigonal defects. In such distorted cases ∣exDFT⟩|e_x^{\mathrm{DFT}}\rangle and ∣eyDFT⟩|e_y^{\mathrm{DFT}}\rangle orbitals become non-degenerate as the phonons perturb electronic structure. These trigonal systems can be described within the framework of the E⊗eE \otimes e Jahn-Teller case as explained in Ref. [28].

The localized vibration (EE) is also doubly degenerate, similarly to the orbital manifold, and thus it can be introduced as a two-dimensional harmonic oscillator:

H^vib=ω(a^X†a^X+a^Y†a^Y+1),(14)\hat{H}_{\mathrm{vib}}=\omega(\hat{a}_{X}^{\dagger}\hat{a}_{X}+\hat{a}_{Y}^{\dagger}\hat{a}_{Y}+1), \tag*{(14)}

where a^X,Y†\hat{a}_{X,Y}^{\dagger} and a^X,Y\hat{a}_{X,Y} are phonon creation and annihilation operators while ω\omega is the energy quantum of the oscillator. In an electron-phonon entangled manifold, the vibrational Hamiltonian is required to be extended to H^vib→H^vib⊗1orb\hat{H}_{\mathrm{vib}}\to\hat{H}_{\mathrm{vib}}\otimes\mathbb{1}_{\mathrm{orb}} where the additional identity operator 1orb=(1001)\mathbb{1}_{\mathrm{orb}}=\begin{pmatrix}1&0\\0&1\end{pmatrix} spans the orbital degrees of freedom to the coupled system. However, from now on, we will neglect the trivial ⊗\otimes and 1orb\mathbb{1}_{\mathrm{orb}} for simplicity. The two dimensional harmonic oscillation of the lattice couples to the doubly degenerate electron orbital states which can be described by the following Hamiltonian:

H^DJT=F(X^σ^z−Y^σ^x)+G[(X^2−Y^2)σ^z+2X^Y^σ^x],(15)\hat{H}_{\mathrm{DJT}}=F(\hat{X}\hat{\sigma}_{z}-\hat{Y}\hat{\sigma}_{x})+G\left[(\hat{X}^{2}-\hat{Y}^{2})\hat{\sigma}_{z}+2\hat{X}\hat{Y}\hat{\sigma}_{x}\right], \tag*{(15)}

where X^\hat{X} and Y^\hat{Y} are the position operators of the vibrational mode with energy quantum ω\omega while σ^z=∣ex⟩⟨ex∣−∣ey⟩⟨ey∣=(100−1)\hat{\sigma}_{z}=|e_x\rangle\langle e_x|-|e_y\rangle\langle e_y|=\begin{pmatrix}1&0\\0&-1\end{pmatrix} and σ^x=∣ex⟩⟨ey∣+∣ey⟩⟨ex∣=(0110)\hat{\sigma}_{x}=|e_x\rangle\langle e_y|+|e_y\rangle\langle e_x|=\begin{pmatrix}0&1\\1&0\end{pmatrix} Pauli matrices govern the orbital degrees of freedom. The quantum states of the system thus consist of doubly degenerate orbitals entangled with two-dimensional vibrational motion. We numerically diagonalize the H^vib+H^DJT\hat{H}_{\mathrm{vib}}+\hat{H}_{\mathrm{DJT}} Hamiltonian. The eigenstates of the coupled system can be expanded in the following series expansion:

∣κi⟩=∑k,l(ck,l(i)∣ex⟩+dk,l(i)∣ey⟩)⊗∣k,l⟩,(16)|\kappa_i\rangle=\sum_{k,l}\left(c_{k,l}^{(i)}|e_x\rangle+d_{k,l}^{(i)}|e_y\rangle\right)\otimes|k,l\rangle, \tag*{(16)}

where ∣ex,y⟩|e_{x,y}\rangle are electronic orbitals while ∣k,l⟩|k,l\rangle are vibrational wavefunctions. The latter represent the Fock states of the two-dimensional quantum harmonic oscillator along X,YX,Y directions, explicitly given by ∣k,l⟩=1k!l!(a^X†)k(a^Y†)l∣0,0⟩|k,l\rangle=\frac{1}{\sqrt{k!l!}}(\hat{a}_{X}^{\dagger})^{k}(\hat{a}_{Y}^{\dagger})^{l}|0,0\rangle. We truncate the phonons at an nn-phonon limit, where k+l≤nk+l\leq n. We note that the two lowest eigenstates are always degenerate, thus they can be grouped as a degenerate level exhibiting EE symmetry unless the quadratic coupling becomes extremely strong [28], which is not the case of normally behaved defect centers.

We now turn to the description of the DFT calculations, in which the coordinates XX and YY are treated as classical configurational variables entering the total energy of the full electron-nuclear system, rather than as quantized operators. The total energy can be evaluated as a function of these distortion coordinates, thereby defining the adiabatic potential energy surface (APES). The APES, which comprises two branches as illustrated in Figure 4, can be written as

Figure 4. (a) Adiabatic potential energy surface (APES) of the Jahn-Teller distortion. A one-dimensional cut through the APES surface is shown in (b). Panel (c) displays the atomic configuration of the SnV center in diamond, with the distortions indicated in the insets that of panel (b). By applying the same distortion in two opposite directions, the global energy minimum and saddle point configurations can be found.

E±=12ωR2±RF2+G2R2+2FGRcos⁡(3ϕ),(17)E_{\pm}=\frac{1}{2}\omega R^{2}\pm R\sqrt{F^{2}+G^{2}R^{2}+2FGR\cos(3\phi)}, \tag*{(17)}

where R=X2+Y2R=\sqrt{X^{2}+Y^{2}} is the distance from the conical intersection and ϕ=arctan⁡(Y/X)\phi=\arctan(Y/X) is the pseudo-rotation angle. The resulting APES exhibits three types of critical points: (i) the conical intersection at X=0X=0, Y=0Y=0 where ∣ex,yDFT⟩|e_{x,y}^{\mathrm{DFT}}\rangle is degenerate and exhibiting EhsE_{\mathrm{hs}} energy. It is unstable due to trigonal symmetry-breaking distortions. (ii) three equivalent minima, at which the system attains its lowest total energy Emin⁡E_{\min} due to trigonal symmetry-breaking distortion; (iii) three equivalent saddle points exhibiting EspE_{\mathrm{sp}} total energy. These special configurations can be identified through geometry optimizations performed with imposed symmetry constraints. The procedure begins by initializing the atomic structure in its high-symmetry configuration – e.g., D3dD_{3d} or C3vC_{3v} for group-IV—vacancy centers or the NV, N3_3V, (CH3_3O) centers (radical), respectively – and relaxing the atomic positions while preserving this high symmetry. Next, the lattice is distorted along +X+X, lowering the symmetry to C2hC_{2h} or C1hC_{1h}. Further optimizing the atomic positions by DFT yields either the saddle-point energy EspE_{\mathrm{sp}} or the minimum point Emin⁡E_{\min}; however, at this stage it is not known which of the two has been reached. To identify the remaining configuration, one has to distort the opposite −X-X additionally, which will identify the missing other special point. We note that retaining a mirror-symmetry constraint during the relaxation provides a practical advantage: it prevents the system from leaving the XX axis and relaxing to Emin⁡E_{\min} because the system is confined in Y=0Y = 0 thus ensuring both special points are being located unambiguously.

Our purpose of the DFT calculations is to parametrize the Jahn-Teller system. The full reconstruction of the APES makes possible to express the linear FF and quadratic GG Jahn-Teller coefficients as follows

F=2EJTω(1−δ2EJT−δ)andG=ωδ4EJT−2δ,(18)F = \sqrt{2E_{\mathrm{JT}}\omega\left(1-\frac{\delta}{2E_{\mathrm{JT}}-\delta}\right)} \qquad\text{and}\qquad G = \frac{\omega\delta}{4E_{\mathrm{JT}}-2\delta}, \tag*{(18)}

where EJT=Ehs−Emin⁡E_{\mathrm{JT}} = E_{\mathrm{hs}} - E_{\min} is the Jahn-Teller (JT) (stabilization) energy, δ=Esp−Emin⁡\delta= E_{\mathrm{sp}} - E_{\min} is the barrier height thus the quantities EJTE_{\mathrm{JT}} and δ\delta are readily obtained as total-energy differences directly from the DFT runs. To determine ω\omega, we proceed as follows. One may fit two one-dimensional harmonic oscillators along the XX coordinate (e.g. Y=0Y = 0) in the neighborhood of configurations Emin⁡E_{\min} and EspE_{\mathrm{sp}}. Specifically, the two oscillators will exhibit

ωmin⁡=2Ehs−Emin⁡dmin⁡2=ω+2G,ωsp=2Ehs−Espdsp2=ω−2G,(19)\omega_{\min} = \sqrt{2\frac{E_{\mathrm{hs}}-E_{\min}}{d_{\min}^{2}}} = \omega+ 2G,\qquad \omega_{\mathrm{sp}} = \sqrt{2\frac{E_{\mathrm{hs}}-E_{\mathrm{sp}}}{d_{\mathrm{sp}}^{2}}} = \omega- 2G, \tag*{(19)}

frequencies where dmin⁡d_{\min} and dspd_{\mathrm{sp}} denote the distances between the high-symmetry configuration and the Jahn–Teller minimum, and between the high-symmetry configuration and the saddle-point geometry, respectively. Both dmin⁡d_{\min} and dspd_{\mathrm{sp}} are generalized coordinates and can be expressed as follows:

dmin⁡/sp=∑i(dhs,i−dmin⁡/sp,i)2mi,(20)d_{\min/\mathrm{sp}} = \sqrt{\sum_i (d_{\mathrm{hs},i}-d_{\min/\mathrm{sp},i})^2m_i}, \tag*{(20)}

where dhs,id_{\mathrm{hs},i}, dmin⁡,id_{\min,i}, dsp,id_{\mathrm{sp},i} depicts the Cartesian coordinates of the ii-th atom in the high-symmetry, minimum, or saddle-point configuration, respectively, and mim_i is the mass of the ii-th nucleus. And thus finally, ω\omega of the two-dimensional oscillator can be expressed as the average of the two frequencies: ω=(ωsp+ωmin⁡)/2\omega= (\omega_{\mathrm{sp}}+\omega_{\min})/2.

We now quantify the energy splitting induced by the spin–orbit interaction, (H^SOC=λ0L^zS^z)(\hat{H}_{\mathrm{SOC}} = \lambda_0\hat{L}_z\hat{S}_z) acting within the subspace spanned by ∣κ1,κ2⟩\lvert\kappa_1,\kappa_2\rangle. For sufficiently small λ0\lambda_0, H^SOC\hat{H}_{\mathrm{SOC}} may be treated as a perturbation. By expanding ∣κ1,κ2⟩\lvert\kappa_1,\kappa_2\rangle in the real orbital basis {∣ex⟩,∣ey⟩}\{\lvert e_x\rangle,\lvert e_y\rangle\} and utilizing L^z=i(0−110)\hat{L}_z = i\begin{pmatrix}0&-1\\1&0\end{pmatrix} one finds that the diagonal matrix elements vanish: ⟨κ1∣L^z∣κ1⟩=⟨κ2∣L^z∣κ2⟩=0\langle\kappa_1|\hat{L}_z|\kappa_1\rangle= \langle\kappa_2|\hat{L}_z|\kappa_2\rangle= 0. Only purely imaginary off-diagonal matrix elements remain, with magnitude of ∣⟨κ1∣L^z∣κ2⟩∣=p×∣⟨ex∣L^z∣ey⟩∣\lvert\langle\kappa_1|\hat{L}_z|\kappa_2\rangle\rvert= p \times\lvert\langle e_x|\hat{L}_z|e_y\rangle\rvert where the pp Ham reduction factor [4] damps the LzL_z angular moment by vibronic coupling. Diagonalization of H^SOC\hat{H}_{\mathrm{SOC}} in the two-dimensional subspace spanned by ∣κ1,κ2⟩\lvert\kappa_1,\kappa_2\rangle shows that the bare DFT spin–orbit coupling λ0\lambda_0 is renormalized to an effective value:

λtheorypert.=ppert.×λ0≈λexpt..(21)\lambda_{\mathrm{theory}}^{\mathrm{pert.}} = p^{\mathrm{pert.}} \times\lambda_0 \approx\lambda_{\mathrm{expt.}}. \tag*{(21)}

We note that this perturbative approach is no longer valid for heavier impurities, for which λ0\lambda_0 is comparable or even exceeds EJTE_{\mathrm{JT}}. Therefore, we emphasize the perturbative approach by the pert. label for λtheorypert.\lambda_{\mathrm{theory}}^{\mathrm{pert.}} and ppert.p^{\mathrm{pert.}}. In such cases, the spin-orbit interaction can no longer be treated as a perturbation and must be included directly in the numerical diagonalization of the full Hamiltonian H^\hat{H}, as it can substantially affect and modify the vibronic expansion given in Eq. (16).

E⊗eE \otimes e Jahn-Teller interaction and spin-orbit coupling

In this section, we take into consideration an additional degrees of freedom: the electron spin of the unpaired electron (or hole) that can affect the electronic orbital through spin-orbit coupling from Eq. (12). Therefore, we extend the expansion of Eq. (16) with electronic spin ∣ξ⟩=∣↑⟩,∣↓⟩\lvert\xi\rangle= \lvert\uparrow\rangle,\lvert\downarrow\rangle. Therefore, we expand the eigenstates of the full H^=H^SOC+H^vib+H^DJT\hat{H} = \hat{H}_{\mathrm{SOC}} + \hat{H}_{\mathrm{vib}} + \hat{H}_{\mathrm{DJT}} Hamiltonian with the following expansion:

∣εi⟩=∑α∑ξ∑k,lck,l,α,ξ(i)∣eα⟩⊗∣ξ⟩⊗∣k,l⟩,(22)\lvert\varepsilon_i\rangle= \sum_{\alpha}\sum_{\xi}\sum_{k,l} c_{k,l,\alpha,\xi}^{(i)}\lvert e_\alpha\rangle\otimes\lvert\xi\rangle\otimes\lvert k,l\rangle, \tag*{(22)}

where ml={x,y}m_l = \{x,y\} labels depict the orbital quantum number while k,lk,l are phonon occupation numbers. We note that in this step, the Hamiltonian terms are promoted to the full vibrational–electronic–spin Hilbert space e.g.: H^DJT→H^DJT⊗1spin\hat{H}_{\mathrm{DJT}} \rightarrow\hat{H}_{\mathrm{DJT}}\otimes\mathbb{1}_{\mathrm{spin}} (Eq. (15)), H^SOC→H^SOC⊗1vib\hat{H}_{\mathrm{SOC}} \rightarrow\hat{H}_{\mathrm{SOC}}\otimes\mathbb{1}_{\mathrm{vib}} (Eq. (12)), H^vib→H^vib⊗1orb⊗1spin\hat{H}_{\mathrm{vib}} \rightarrow\hat{H}_{\mathrm{vib}}\otimes\mathbb{1}_{\mathrm{orb}}\otimes\mathbb{1}_{\mathrm{spin}} (Eq. (14)) where 1vib\mathbb{1}_{\mathrm{vib}}, 1orb\mathbb{1}_{\mathrm{orb}}, 1spin\mathbb{1}_{\mathrm{spin}} are identity operators for vibration, orbital, spin, respectively.

The two bare electronic spin-orbit Kramers doublets ∣E12⟩\lvert E_{\frac{1}{2}}\rangle, ∣E32⟩\lvert E_{\frac{3}{2}}\rangle introduced in Eq. (13) are transformed into ∣E~12⟩\lvert\tilde{E}_{\frac{1}{2}}\rangle, ∣E~32⟩\lvert\tilde{E}_{\frac{3}{2}}\rangle due to the combined action of dynamic Jahn-Teller effect and spin-orbit coupling. Therefore, the first four ∣ε1−4⟩\lvert\varepsilon_{1-4}\rangle eigenstates that of Eq. (22) will correspond to the two polaronic ∣E~12⟩,∣E~32⟩\lvert\tilde{E}_{\frac{1}{2}}\rangle,\lvert\tilde{E}_{\frac{3}{2}}\rangle doublets, whose energy separation is defined as

λtheory=⟨ε3,4∣H^∣ε3,4⟩−⟨ε1,2∣H^∣ε1,2⟩≈λexpt.,(23)\lambda_{\mathrm{theory}} = \langle\varepsilon_{3,4}|\hat{H}|\varepsilon_{3,4}\rangle- \langle\varepsilon_{1,2}|\hat{H}|\varepsilon_{1,2}\rangle\approx\lambda_{\mathrm{expt.}}, \tag*{(23)}

and thus it can be identified that λ0\lambda_0 becomes renormalized to λtheory\lambda_{\mathrm{theory}} in the combined vibrational-electronic-spin system. We note that λtheory=KJT+λSOC\lambda_{\mathrm{theory}} = K_{\mathrm{JT}} + \lambda_{\mathrm{SOC}} can be decomposed into two parts: (i) the Jahn-Teller- and vibronic-induced term KJT=⟨ε3,4∣H^DJT+H^vib∣ε3,4⟩−⟨ε1,2∣H^DJT+H^vib∣ε1,2⟩K_{\mathrm{JT}} = \langle\varepsilon_{3,4}|\hat{H}_{\mathrm{DJT}}+\hat{H}_{\mathrm{vib}}|\varepsilon_{3,4}\rangle- \langle\varepsilon_{1,2}|\hat{H}_{\mathrm{DJT}}+\hat{H}_{\mathrm{vib}}|\varepsilon_{1,2}\rangle and (ii) the spin-orbit contribution λSOC=⟨ε3,4∣H^SOC∣ε3,4⟩−⟨ε1,2∣H^SOC∣ε1,2⟩\lambda_{\mathrm{SOC}} = \langle\varepsilon_{3,4}|\hat{H}_{\mathrm{SOC}}|\varepsilon_{3,4}\rangle- \langle\varepsilon_{1,2}|\hat{H}_{\mathrm{SOC}}|\varepsilon_{1,2}\rangle. One can clearly recognize the non-perturbative character of the spin-orbit interaction: as λ0\lambda_0 increases, the expansion coefficients ck,l,α,ξ(i)c_{k,l,\alpha,\xi}^{(i)} from Eq. (22), associated with ∣ε1,2⟩\lvert\varepsilon_{1,2}\rangle and ∣ε3,4⟩\lvert\varepsilon_{3,4}\rangle will gradually diverge from one another. Consequently, KJTK_{\mathrm{JT}} becomes increasingly pronounced, and even the term λSOC\lambda_{\mathrm{SOC}} no longer the same as it was in Eq. (21) for example, in the case of the SnV−^{-} and PbV−^{-} vacancies.

The case of external magnetic, electric fields and strain

External magnetic field B⃗=(Bx,By,Bz)\vec{B} = (B_x, B_y, B_z) couples to both orbital (LL) and spin (SS) angular momentum of the electron via the Zeeman interaction:

H^mag.=μBgLBzL^z+μBgS(BxS^x+ByS^y+BzS^z),(24)\hat{H}_{\mathrm{mag.}} = \mu_{\mathrm{B}}g_{\mathrm{L}}B_z\hat{L}_z + \mu_{\mathrm{B}}g_{\mathrm{S}}(B_x\hat{S}_x + B_y\hat{S}_y + B_z\hat{S}_z), \tag*{(24)}

where μB=0.0579\mu_{\mathrm{B}} = 0.0579 meV/T is the Bohr magneton, and gS=2.0023g_{\mathrm{S}} = 2.0023 and gLg_{\mathrm{L}} are the orbital and electron spin gg-factor, respectively. We note that, here gLgnd,exg_{\mathrm{L}}^{\mathrm{gnd,ex}} factors are different values for optical ground (gnd) and excited (ex) states. Additionally, they differ from the gL=1g_{\mathrm{L}} = 1 value observed in full rotation group SO(3) because the defects we consider are exhibiting only trigonal 120° rotation symmetry thus gLgnd,exg_{\mathrm{L}}^{\mathrm{gnd,ex}}-s individual fractional values, see Appendix D in Ref. [2] or Refs. [54, 55, 56, 57] for details. Thus, the Zeeman Hamiltonian H^mag.\hat{H}_{\mathrm{mag.}} will lift the Kramers degeneracy for both ∣ε1,2⟩|\varepsilon_{1,2}\rangle, ∣ε3,4⟩|\varepsilon_{3,4}\rangle resulting in four states with distinct energies that can be seen in Figure 3 or Figure 6. Therefore, H^mag.\hat{H}_{\mathrm{mag.}} can be incorporated into the spin-orbit-phonon expansion that of Eq. (22) and then numerically determine the splitting due magnetic field. We note that the vibrational degrees of freedom can be traced out [4, 6, 58] whenever one is only interested in the lowest four ∣ε1−4⟩|\varepsilon_{1-4}\rangle states. We will discuss it in the upcoming Section 2.7.

The splitting induced by local strain or external electric field can be described by [17, 19, 59, 60]

H^strain=−Υxσ^z+Υyσ^x,(25)\hat{H}_{\mathrm{strain}} = -\Upsilon_x\hat{\sigma}_z + \Upsilon_y\hat{\sigma}_x, \tag*{(25)}

such that, in the presence of sufficiently strong strain or electric fields, the energy separation between the ∣E3⟩|E_3\rangle and ∣E1⟩|E_1\rangle Kramers doublets at zero magnetic field is enlarged to Δ>λtheory\Delta> \lambda_{\mathrm{theory}}. Since each specific defect center experiences a distinct local strain environment and a characteristic configuration of nearby charged defects (which can generate residual electric fields), the corresponding splitting, Δ\Delta is typically an emitter-specific parameter, measured individually for each single defect. For inversion-symmetric centers such as G4V, the coupling to external electric fields vanishes. In contrast, systems lacking inversion symmetry – such as the NV center – exhibit finite electric-field coupling. However, this coupling can be either an advantage or a disadvantage. (i) NV centers are prone to spectral diffusion [61, 62, 63], in which the position of the zero-phonon line (ZPL) fluctuates over time due to charge-transfer processes involving nearby defects, thereby complicating their use in practical qubit applications. (ii) On the other hand, the same coupling is advantageous enhancing qubit’s T2∗T_2^* coherence time [2, 59, 64, 65, 66, 67, 68, 69, 70, 71] by effectively quenching the decoherence induced by thermal phonons by enlarging Δ\Delta by means of strain engineering. Here, we note that tracing over vibrational degrees of freedom is also convenient for strain: we will further discuss the Δ\Delta splitting in the following Section 2.7.

Effective model Hamiltonian

The higher-lying vibronic states ∣εj>4⟩|\varepsilon_{j>4}\rangle appearing in Eq. (22) are irrelevant for standard optical measurements, as the system relaxes into one of the four lowest-lying states ∣ε1−4⟩|\varepsilon_{1-4}\rangle on a timescale in the ∼\simps to ∼\simfs regime. This situation corresponds to the Kasha rule [72, 73], according to which only the vibronic ground levels are typically populated under optical excitation, unless the system is probed by ultrafast spectroscopy [74, 75, 76, 77, 78, 79], where the dynamics of these short-lived resonances can be characterized. Thus, in typical state-of-the-art experiments, an effective four-level Hamiltonian (H^eff\hat{H}_{\mathrm{eff}}) can be employed [2, 6, 60, 80] to describe the ∣E~1⟩|\tilde{E}_1\rangle, ∣E~3⟩|\tilde{E}_3\rangle manifold by introducing effective orbital: L^zeff=σ^zeff=(0−ii0)\hat{L}_z^{\mathrm{eff}} = \hat{\sigma}_z^{\mathrm{eff}} = \begin{pmatrix} 0 & -i \\ i & 0 \end{pmatrix}, σ^zeff=(100−1)\hat{\sigma}_z^{\mathrm{eff}} = \begin{pmatrix} 1 & 0 \\ 0 & -1 \end{pmatrix}, σ^xeff=(0110)\hat{\sigma}_x^{\mathrm{eff}} = \begin{pmatrix} 0 & 1 \\ 1 & 0 \end{pmatrix} and spin: S^xeff=12(0110)\hat{S}_x^{\mathrm{eff}} = \frac{1}{2}\begin{pmatrix} 0 & 1 \\ 1 & 0 \end{pmatrix}, S^yeff=12(0−ii0)\hat{S}_y^{\mathrm{eff}} = \frac{1}{2}\begin{pmatrix} 0 & -i \\ i & 0 \end{pmatrix}, S^zeff=12(100−1)\hat{S}_z^{\mathrm{eff}} = \frac{1}{2}\begin{pmatrix} 1 & 0 \\ 0 & -1 \end{pmatrix} operators:

H^=H^SOC+H^vib+H^JT+H^mag.+H^strainH^eff=λtheoryL^zeffS^zeff+μB(fBzL^zeff+2δfBzS^zeff)+μBgS(BxS^xeff+ByS^yeff+BzS^zeff)−γxeffσ^zeff+γyeffσ^xeff(26)\begin{aligned} \hat{H} &= \hat{H}_{\mathrm{SOC}} + \hat{H}_{\mathrm{vib}} + \hat{H}_{\mathrm{JT}} + \hat{H}_{\mathrm{mag.}} + \hat{H}_{\mathrm{strain}} \\ \hat{H}_{\mathrm{eff}} &= \lambda_{\mathrm{theory}}\hat{L}_z^{\mathrm{eff}}\hat{S}_z^{\mathrm{eff}} + \mu_{\mathrm{B}}(fB_z\hat{L}_z^{\mathrm{eff}} + 2\delta_f B_z\hat{S}_z^{\mathrm{eff}}) \\ &\quad+ \mu_{\mathrm{B}}g_{\mathrm{S}}(B_x\hat{S}_x^{\mathrm{eff}} + B_y\hat{S}_y^{\mathrm{eff}} + B_z\hat{S}_z^{\mathrm{eff}}) \\ &\quad- \gamma_x^{\mathrm{eff}}\hat{\sigma}_z^{\mathrm{eff}} + \gamma_y^{\mathrm{eff}}\hat{\sigma}_x^{\mathrm{eff}} \tag*{(26)} \end{aligned}

The first term represents the effective spin-orbit coupling (λtheory\lambda_{\mathrm{theory}}). The subsequent terms describe the interaction of the magnetic field with the effective orbital and spin angular momenta of the electron. We note that f=pgLf = pg_{\mathrm{L}} indicates that the orbital gg-factor associated with LzL_z is reduced by the Ham reduction factor pp, since the orbital Zeeman term in Eq. (24) contains the operator L^z\hat{L}_z, which transforms according to the A2A_2 irreducible representation of the C3vC_{3v} point group. Therefore, projection onto the effective subspace reduces the orbital angular momentum such that: ⟨L^z⟩ph=pL^zeff\langle\hat{L}_z\rangle_{\mathrm{ph}} = p\hat{L}_z^{\mathrm{eff}}, where ⟨…⟩ph\langle\ldots\rangle_{\mathrm{ph}} denotes averaging over the phonon bath (see Appendices A and B of Ref. [6] for details). In contrast, the spin Zeeman interaction remains unchanged in the effective model, as it involves only spin operators {S^x,S^y,S^z}\{\hat{S}_x,\hat{S}_y,\hat{S}_z\} and no orbital contribution. However, an additional second-order term, 2δfBzS^zeff2\delta_fB_z\hat{S}_z^{\mathrm{eff}} may arise for heavier G4V centers such as SnV−^{-} and PbV−^{-}. This contribution originates from the orbital Zeeman interaction but manifests as an effective spin operator S^zeff\hat{S}_z^{\mathrm{eff}} in the effective system, see Appendix C in Refs. [2, 81] in the reduced Hamiltonian (see Appendix C of Refs. [2, 81]). The magnitude of this additional term can be determined utilizing the eigenstates ∣ε1−4⟩|\varepsilon_{1-4}\rangle defined in Eq. (22):

δf=gL×(⟨ε3,4∣L^z∣ε3,4⟩+⟨ε1,2∣L^z∣ε1,2⟩).(27)\delta_f = g_{\mathrm{L}} \times\left(\langle\varepsilon_{3,4}|\hat{L}_z|\varepsilon_{3,4}\rangle+ \langle\varepsilon_{1,2}|\hat{L}_z|\varepsilon_{1,2}\rangle\right). \tag*{(27)}

Additionally, we incorporate the effect of external strain or transverse electric fields through the effective parameters γxeff\gamma_x^{\mathrm{eff}} and γyeff\gamma_y^{\mathrm{eff}}. The bare strain Hamiltonian H^strain\hat{H}_{\mathrm{strain}} in Eq. (25) is renormalized by the Ham factor q=(1+p)/2q = (1 + p)/2, because orbital operators {−σ^z,σ^x}\{-\hat{\sigma}_z,\hat{\sigma}_x\} transform according to the {Ex,Ey}\{E_x,E_y\} irreducible representation of C3vC_{3v}, see the discussion below Eq. (11) in Ref. [6]. Consequently, in the effective description one obtains: {⟨σ^z⟩ph=qσ^zeff,⟨σ^x⟩ph=qσ^xeff}\{\langle\hat{\sigma}_z\rangle_{\mathrm{ph}} = q\hat{\sigma}_z^{\mathrm{eff}}, \langle\hat{\sigma}_x\rangle_{\mathrm{ph}} = q\hat{\sigma}_x^{\mathrm{eff}}\} which implies: {γxeff=qγx,\{\gamma_x^{\mathrm{eff}} = q\gamma_x, γyeff=gLγy\gamma_y^{\mathrm{eff}} = g_L\gamma_y. The resulting splitting between the two Kramers doublets at zero magnetic field is then enhanced by strain as we show below:

Δ=λtheory2+4(γxeff)2+4(γyeff)2.(28)\Delta= \sqrt{\lambda_{\mathrm{theory}}^2 + 4(\gamma_x^{\mathrm{eff}})^2 + 4(\gamma_y^{\mathrm{eff}})^2}. \tag*{(28)}

We remark that the 4×44 \times4 effective Hamiltonian H^eff\hat{H}_{\mathrm{eff}} reproduces the spectroscopic properties of the full Hamiltonian H^\hat{H} in Eq. (9), while avoiding the excessive computational cost associated with explicitly treating the large number of phononic degrees of freedom. Thus, for most practical purposes, the effective model provides a sufficient and efficient description.

Numerical results on solid state quantum bits in diamond

In this section, we present an illustrative example fine-structure calculation for G4V centers in diamond, a class of well-studied point defects shown in Figure 5. As a specific example, we also analyze the ZPL splitting of the SnV−^{-} center in an external magnetic field. The electronic structures of the high-symmetry (hs), minimum energy (min), and saddle-point (sp) geometries, as well as the spin–orbit coupling, can be evaluated by means of state-of-the-art DFT methodology. The gg-factor gLg_L, however, cannot be calculated by VASP faithfully. In certain special cases – specifically, when the relevant state is strongly localized on the dopant’s dd-orbitals such that the entire orbital moment is confined to a single atom and contained within its PAW sphere [82, 83, 84] – gLg_L can be extracted directly via the LORBMOM tag [7]. For the SiV center in diamond, however, the orbital moment is distributed over six carbon dangling bonds (see Fig. 8 in Ref. [2]), which renders this approach inapplicable. A more sophisticated treatment – such as those implemented in advanced quantum-chemistry packages [85, 86, 87] – would be required; however, these methods can model solids only through very small finite clusters, introducing severe finite-size artifacts. In that case, the ground- and excited-state values of gLg_L were obtained by fitting the effective Hamiltonian to experimental data, yielding gLgs=0.328g_L^{\mathrm{gs}} = 0.328 and gLexc=0.782g_L^{\mathrm{exc}} = 0.782, respectively. These values are commonly assumed to be transferable to the other G4V centers [2].

Illustration of group-IV vacancy defects embedded in diamond crystal, showing crystallographic and defect spin basis vectors

Figure 5. This figure illustrates a group-IV vacancy defect embedded in a diamond crystal. The crystallographic basis vectors (a⃗\vec{a}, b⃗\vec{b}, c⃗\vec{c}) are shown at the bottom of the figure in light red, green, and blue, respectively. The basis vectors for defect’s spin (S⃗x\vec{S}_x, S⃗y\vec{S}_y, S⃗z\vec{S}_z) are depicted in darker shades and need to be defined in the configuration file. For G4V centers, the spin zz-axis aligns with the [111] crystallographic direction, which corresponds to the C3C_3 rotation axis. The two xx- and yy-axes can be freely chosen freely. In panel (a), the [100] crystallographic direction points toward the viewer, while in panel (b), the defect’s C3C_3 rotational axis points toward the viewer.

It is important to emphasize that the natural basis of the defect spin system is, in general, not aligned with the crystallographic basis vectors. In the present example, we employ a simple-cubic supercell of diamond, where the supercell lattice vectors coincide with those of the Bravais cell. G4V centers exhibit a high-symmetry axis – corresponding to the C3C_3 rotation axis – aligned with the [111] crystallographic direction of diamond. It is therefore natural to adopt the [111] direction as the spin quantization axis, i.e., the zz-axis associated with the electronic spin operator S^z\hat{S}_z appearing in Eq. (21). The remaining basis vectors xx- and yy-axes may be chosen arbitrarily, provided they are orthogonal to the zz-axis and form a right-handed coordinate system for convenience. Exe.py requires a .cfg configuration file as input to specify the following attributes of the system in order to calculate the E⊗eE \otimes e Jahn-Teller interaction:

  • vasprun.xml outputs of VASP DFT calculations with relaxed atomic positions at EhsE_{\mathrm{hs}}, EminE_{\mathrm{min}}, EspE_{\mathrm{sp}} special points, see Figure 4

  • maximum number of phonon quanta (k+l≤n)(k + l \le n) in Eqs. (22), (16). Figure 9 shows how the pp factor converges with respect to this parameter. In practice we set it to 12.

  • spin-orbit coupling (λ0)(\lambda_0) as described in § 3

Figure 9. This figure shows the convergence of the SnV⁻ center’s pp factor with respect to the maximum number of vibrational quanta in each direction X^,Y^\hat{X}, \hat{Y}. It goes to p=0.4706p = 0.4706 after having at least 6 maximum phonon quanta in the system.

For calculating the ZPL fine structure in the presence of magnetic field, it is further necessary to specify the following parameters:

  • orbital reduction factor gLg_L

  • the basis vectors for electronic spin quantization

  • range and directions of the applied magnetic field in terms of the basis vectors of geometrical configurations

The code automatically normalizes the user-specified basis vectors. The external magnetic field is then transformed into the coordinate system defined by the defect’s spin basis. The implementation is explicitly interfaced with the VASP package [3]. Nevertheless, the code is also compatible with results from other electronic-structure calculations, since the relevant energies can be provided directly via the configuration file and the geometries can be supplied in .csv format. In this latter case, the user must additionally specify the following attributes in the .cfg file:

  • energies of the geometries (Ehs,Emin,EspE_{\mathrm{hs}}, E_{\mathrm{min}}, E_{\mathrm{sp}})

  • basis vectors of the geometry

  • atomic masses for ions

Moreover, the input may be specified either in terms of the Jahn-Teller parameters (EJT,δ,dmin,dsp)(E_{\mathrm{JT}}, \delta, d_{\mathrm{min}}, d_{\mathrm{sp}}) or via Taylor-expansion coefficients (F,G)(F,G) of H^DIT\hat{H}_{\mathrm{DIT}} as defined in (15). All supported use cases are documented in the point_defect_config_files directory of the GitHub repository. The raw parameters extracted from the input files are summarized in Table 1. The code generates the following output files:

defect typeEhsE_{\mathrm{hs}} (eV)Emin⁡E_{\min} (eV)EspE_{\mathrm{sp}} (meV)dmin⁡d_{\min} (Åamu\sqrt{\mathrm{amu}})dspd_{\mathrm{sp}} (Åamu\sqrt{\mathrm{amu}})gLg_Lλ0\lambda_0 (meV)
SiV gnd-5375.9368-5375.9791-5375.97630.20850.21090.3280.8
SiV ex-5374.1818-5374.2603-5374.25770.34420.34500.7826.9
GeV gnd-5372.7556-5372.7858-5372.78370.18540.18840.3282.2
GeV ex-5370.5590-5370.6440-5370.63900.35740.35300.78236.1
SnV gnd-5368.3068-5368.3284-5368.32680.16440.16760.3288.6
SnV ex-5366.1366-5366.2197-5366.21290.34070.34210.78295.9
PbV gnd-5364.4629-5364.4785-5364.47790.15380.14830.32834.6
PbV ex-5361.9405-5362.0324-5362.01980.34360.33620.782245.2

Table 1.

  • eigenvalues and eigenstates in separate .csv files

  • all raw parameters and theoretical values that are calculated, describes in § 2.1

  • energies of A,B,C,D transitions

  • expectation values of Lz⊗SzL_z \otimes S_z.

Computational details

The accuracy of the results obtained with jahn-teller-dynamics is profoundly impacted by the precision of input parameters. In this work, we employ density functional theory (DFT) within the Born-Oppenheimer approximation, as implemented in the VASP 5.4.1 code [3] within plane-wave supercell framework and projector-augmented-wave (PAW) formalism [82, 88]. We note that our code is also compatible with newer 6.x.x versions. However, we observed that numerical stability of constrained occupation Δ\DeltaSCF calculations by setting FERWE, FERDO tags often vary from version to version see Refs. [89, 90] for additional details reported by other research groups. Therefore, we opted to VASP 5.4.1 where we found the best convergence over the years. Calculations are performed utilizing the Perdew-Burke-Ernzerhof (PBE) [91] generalized-gradient functional or the Heyd-Scuseria-Ernzerhof [92, 93] (HSE06) hybrid functional. Point defects are modeled with an plane-wave energy cutoff of 370 eV, and strict convergence criteria on forces action on ions: 10−410^{-4} eV/Å during the geometry optimization. We used 512-atom supercells to mitigate finite size effects and obtain results suitable for comparison with experiment except the spin-orbit parameters that we discuss next.

We employed the noncollinear formalism implemented in VASP 5.4.1 [84] [53] to determine the spin-orbit (λ0\lambda_0) parameters. We note that fully converged spin-orbit splittings generally require supercells larger than the 512-atom cell mentioned before. The spin quantization axis was chosen along the [111] direction (C3C_3 rotation axis). We determine the spin-orbit parameters in supercells optimized in optimized geometries previously by means of conventional collinear calculations; thus, the influence of spin-orbit coupling on the ionic positions was neglected. The occupations of the relevant degenerate Kohn-Sham orbitals in the band gap were constrained to an equal (half-half) distribution, corresponding to a single-electron configuration (e+1e−1e0.50.5e0.50.5e_{+}^{1}e_{-}^{1}e_{0.5}^{0.5}e_{0.5}^{0.5}).

Optical transition energies under magnetic fields

In this section we provide exemplary calculations with the negatively charged group-IV-vacancy (G4V) defects in diamond. In these defects, the impurity atom (Si, Ge, Sn, or Pb) adopts the split-interstitial configuration (see Figure 4(c)) exhibiting D3dD_{3d} point group symmetry. G4V centers exhibits an (eu4eg3)(e_u^4e_g^3) occupation, leaving a single hole in the degenerate ege_g gerade orbital predominantly localized on carbon dangling bonds surrounding the two vacancies [81]. Their optically excited state can be characterized by (eu3eg4)(e_u^3e_g^4), thus, the hole is now present in the degenerate eue_u ungerade orbital. Consequently, both the ground and excited states are subjects of the E⊗eE \otimes e Jahn-Teller problem [2], and the corresponding spin-orbit coupling enters with negative sign, −λ0-\lambda_0 according to Hund’s third law, that fact also visible in VASP calculations.

In order to simulate the ZPL fine structure, parameters for both ground and excited states must be specified along with the applied magnetic field. In the examples considered here, the magnetic field is aligned with the defect spin, i.e., along the [111] crystallographic direction. The script Exe.py extracts the total energies of the geometries (EhsE_{\mathrm{hs}}, Emin⁡E_{\min}, EspE_{\mathrm{sp}}), the distances between them (dhsd_{\mathrm{hs}}, dmin⁡d_{\min}, dspd_{\mathrm{sp}}), orbital gg-factors, gLg_L, and spin-orbit λ0\lambda_0 parameters, are summarized in Table 1 as obtained from DFT. The code iterates over the user-specified magnetic-field range and constructs the corresponding Hamiltonian for each field value. It then computes the eigenenergies (∣εi⟩\lvert\varepsilon_i\rangle-s) for both the ground and excited states and subsequently determines the resulting ZPL fine structure, which can be directly compared with experimental observations, see Figure 6.

At zero magnetic field, four optically allowed transitions (A, B, C, D) connect the fourfold-degenerate (ege_g, eue_u) manifolds (Figure 3). Under nonzero magnetic field, each of these transitions acquires a distinct energy, giving rise to a characteristic fine structure in the spectrum with a total 16 individual transitions. In addition, the code determines the Jahn-Teller parameters and the theoretical spin-orbit splitting λtheory\lambda_{\mathrm{theory}} that is comparable with experimental data λexpt\lambda_{\mathrm{expt}} as shown in Table 2.

defect typeEJTE_{\mathrm{JT}} (meV)δ\delta (meV)ω\omega (meV)ppλ0\lambda_0 (meV)λtheory\lambda_{\mathrm{theory}} (meV)λexpt\lambda_{\mathrm{expt}} (meV)ffδf\delta_f
SiV gnd42.32.988.20.3160.80.250.21[19]0.1040.001
SiV ex78.52.773.70.1286.90.901.08[19]0.1000.018
GeV gnd30.12.183.40.3942.20.870.75[20]0.1290.004
GeV ex85.05.073.90.11736.14.214.63[20]0.0910.089
SnV gnd21.61.679.50.4718.33.923.52[22]0.1540.015
SnV ex83.26.875.60.12595.912.1012.41[22]0.0980.232
PbV gnd15.60.674.90.49434.618.1417.5[25]0.1620.069
PbV ex91.912.678.60.105245.228.570.0820.472

Table 2.

N3_3V0^0 defect in diamond

In this section, we demonstrate the predictive capability of our method by applying Exe.py on the neutral N3_3V0^0 defect in diamond. N3_3V$^0 center is made of three substitutional nitrogen atoms surrounding a vacancy embedded in the diamond host exhibiting C3vC_{3v} symmetry [94]. This defect gives rise to a prominent optical feature in most type-Ia natural diamonds containing B-type nitrogen aggregates and can also be generated artificially via nitrogen ion implantation followed by annealing at temperatures above 1200 °C [95]. Figure 7 schematically illustrates the electronic structure of the defect: a1a_1 and ee levels appear within the boundaries of the band gap, with the corresponding orbitals predominantly localized on carbon and nitrogen dangling bonds, respectively. In the ground state configuration (e4a11e^4a_1^1), the ee level is fully occupied, therefore, it is not Jahn-Teller active $3multiplet.Uponopticalexcitation,anelectronispromotedfromthemultiplet. Upon optical excitation, an electron is promoted from theeleveltothelevel to thea_1level,leavingaholeinthelevel, leaving a hole in thee level [96] and thus, the excited state is Jahn-Teller active 4multiplet.Experimentally,theZPLofNmultiplet. Experimentally, the ZPL of N_3VV^0$ has been reported at 2.985 eV, with a fine-structure splitting of 0.59 meV in photoluminescence (PL) spectra [95]. Utilizing the HSE06 functional, we obtain a ZPL energy as 3.059 eV, representing an improvement over an earlier (semi)local DFT Table 1: Raw parameters extracted from the vasprun.xml VASP calculations supplemented with additional data for gLg_L and λ0\lambda_0 in an accompanying configuration file. Those are the energies of the high symmetric, global energy minimum and saddle point geometries (Ehs,Emin⁡,EspE_{\mathrm{hs}}, E_{\min}, E_{\mathrm{sp}}) and the distances between them (dmin⁡,dspd_{\min}, d_{\mathrm{sp}}). From these values the parameters of the Jahn-Teller interaction are calculated. λ0\lambda_0 is the energy splitting between degenerate orbitals (Kohn-Sham levels) caused by spin-orbit coupling calculated using DFT. The orbital reduction factor (gLg_L) appears in the magnetic field - angular momentum interaction in Eq. (24).

Figure 7. In the ground state configuration, ee electron state is fully occupied. As we excite the system, the electron is promoted to the a1a_1 state and leaves a hole behind in the ee state. This leads to a dynamic Jahn-Teller system in the optical excited state.

Table 2: Jahn-Teller parameters and reduction factors that describe the E⊗eE \otimes e system, λ0\lambda_0 is the spin-orbit splitting calculated by DFT, EJTE_{\mathrm{JT}} is the Jahn-Teller energy, δ\delta is the barrier energy, ω\omega is the vibration energy quantum, pp is the Ham reduction factor, gLg_L is the orbital reduction factor, λtheory\lambda_{\mathrm{theory}} is the theoretical spin-orbit splitting energy, λexpt\lambda_{\mathrm{expt}} is the experimental value taken from [19, 20, 22, 25]. The table also contains ff, δf\delta_f factors that are necessary in order to configure the four state model described by Eq. (26). We note that pp and λtheory\lambda_{\mathrm{theory}} that of Eq. (26) slightly differs than the perturbative approach of λtheorypert.\lambda_{\mathrm{theory}}^{\mathrm{pert.}} and ppert.p^{\mathrm{pert.}} that was in Eq. (21). The difference negligible for SiV, GeV, but becomes non-negligible for SnV, PbV please see the last four columns of Table. III. in Ref. [81], where the last four {p,λHam,λ,λexpt}\{p, \lambda_{\mathrm{Ham}}, \lambda, \lambda_{\mathrm{expt}}\} columns depicts the {ppert.,λtheorypert.,λtheory,λexpt}\{p^{\mathrm{pert.}}, \lambda_{\mathrm{theory}}^{\mathrm{pert.}}, \lambda_{\mathrm{theory}}, \lambda_{\mathrm{expt}}\} parameters in the present article.

result of 2.8 eV [97]. We attribute the observed ZPL splitting to (e+1e+1e−1e−0)(e_{+}^{1}e_{+}^{1}e_{-}^{1}e_{-}^{0}) and (e+1e+1e−0e−1)(e_{+}^{1}e_{+}^{1}e_{-}^{0}e_{-}^{1}) with integer occupations and spin-orbit splitting within the 2E^{2}E excited state; thus, we employ our Exe.py code to calculate λtheory\lambda_{\mathrm{theory}} for the PL spectrum. To generate all input parameters for Exe.py, we performed constrained DFT calculations by means of the HSE06 functional on an 1728-atom supercell that holds 6911 valence electrons. In order to simulate the excited state (Figure 7) we employ constrained occupation DFT calculations with FERWE = 3456 * 1.0 0.0 1.0 tags. and FERDO = 3454 * 1.0 0.0 1.0 tags. These points perfectly fit two quadratic polynomials that intersect exactly at the C3vC_{3v} high-symmetry configuration, see Figure 10. Finally, the derived Jahn-Teller energy of the system EJT=32.12E_{\mathrm{JT}} = 32.12 meV, the barrier energy δ=2.68\delta= 2.68 meV, and the vibrational energy quantum ω=89.54\omega= 89.54 meV as yielded by Exe.py code which results in a Ham reduction factor p=0.40p = 0.40 for this system.

Figure 10. This figure shows the calculated APES of the excited state of the neutral N₃V⁰ defect. The a) and b) part of the figure represents the N₃V⁰ point defect.

Next, we determine the bare spin-orbit λ0\lambda_0 parameter for the 2E^{2}E excited state, where we force the (e+1e+1e−0.5e−0.5)(e_{+}^{1}e_{+}^{1}e_{-}^{0.5}e_{-}^{0.5}) occupation – for example on the 1727-atom supercell – by

FERWE=3455∗1.0 3453∗1.0 0.5 0.5 2∗1.0 0.0 …(29)\mathrm{FERWE}=3455 * 1.0\ 3453 * 1.0\ 0.5\ 0.5\ 2 * 1.0\ 0.0\ \ldots \tag*{(29)}

where Kohn-Sham energy difference between e+0.5e_{+}^{0.5} and e−0.5e_{-}^{0.5} single particle orbitals yields λ0\lambda_0. The straightforward approach would be calculating the system in both orbital configurations (e+1e+1e−1e−0)(e_{+}^{1}e_{+}^{1}e_{-}^{1}e_{-}^{0}) and (e+1e+1e−0e−1)(e_{+}^{1}e_{+}^{1}e_{-}^{0}e_{-}^{1}) with integer occupations and identify their total energy difference as the λ0\lambda_0. However, in our experience this approach is not only computationally more demanding. Unlike molecular systems with limited degrees of freedom, defects in solids experience more severe More importantly, the latter case (e+1e+1e−0e−1)(e_{+}^{1}e_{+}^{1}e_{-}^{0}e_{-}^{1}) can often be numerically unstable because it is not a ground state configuration, and thus it may relax to (e+1e+1e−1e−0)(e_{+}^{1}e_{+}^{1}e_{-}^{1}e_{-}^{0}) or even quench spin-orbit coupling by mixing into a real valued solution (ex1ey1ex1ey0)(e_{x}^{1}e_{y}^{1}e_{x}^{1}e_{y}^{0}). Therefore, it is more convenient and reliable to depend on the (e+1e+1e−0.5e−0.5)(e_{+}^{1}e_{+}^{1}e_{-}^{0.5}e_{-}^{0.5}) occupation even when the two electron halves effectively repel each other, thus one would obtain a significantly larger total energy (∼0.2\sim0.2 eV) in the fractional case by means of hybrid functionals such as HSE06. We note that semilocal GGA (generalized-gradient approximation) functionals such as PBE this on-site interaction seemingly does not exist thus the fractional occupation case for PBE exhibits exactly the same total energy as the two integer occupation cases, of course, except for the small difference due to λ0\lambda_0. Nevertheless, we have shown in our previous [1] work that the Kohn-Sham energy difference can be interpreted as λ0\lambda_0 because it is numerically exactly the same as the aforementioned total energy difference between the (e+1e−1e+1e−0)(e_{+}^{1}e_{-}^{1}e_{+}^{1}e_{-}^{0}) and (e+1e−1e+0e−1)(e_{+}^{1}e_{-}^{1}e_{+}^{0}e_{-}^{1}) occupations.

We used a sequence of diamond supercells of increasing size 63, 215, 511, 999, 1727, 2743, 4095 atoms to fully avoid the spurious interaction between periodic images. Figure 8 shows the calculated spin-orbit energies λ0\lambda_0 as a function of supercell size; the values decrease exponentially with increasing superlattice length and converge to 1.81 meV (1.74 meV) within PBE (HSE06).

Figure 8. Spin-orbit λ0\lambda_0 parameters using PBE (HSE06) functional in diamond supercells with superlattices ranging 63-4095 (63-1727) atoms. Both functionals show an exponential decay with respect to supercell length converging to 1.81 meV (1.74 meV)

Combining this with the Ham reduction factor p=0.40p = 0.40, we obtain an effective spin-orbit splitting λtheory=0.69\lambda_{\mathrm{theory}} = 0.69 meV which is a very good agreement against the experimental data at λexpt=0.59\lambda_{\mathrm{expt}} = 0.59 meV, and thus, reaffirms N3_3V0^0 assignment for the 3.059 eV color center in diamond.

CH3_3O$ radical in the single mode case

In this section we present how our code can be used to calculate the spin-orbit coupling in the CH3_3O radical. This radical possess C3v_{3v} symmetry. We performed density functional theory calculations using the PBE0 functional [98] which combines the generalized gradient functional with an amount of predefined full-range exact exchange interaction of 0.25 and uses no cutoff in the calculation of Coulomb interaction. In the ground state it has an ee orbital partially filled by three electrons. The hole that represents the three electrons that occupy a fourfold degenerate spin-orbit state is dynamic Jahn-Teller active. We found the local minimum and saddle point geometrical configurations; however, in the optimization process the molecules experienced spurious translations and rotations. To eliminate these numerical artifacts, we calculated the vibrational modes by means of the PBE functional and projected out the last six degrees modes associated with translations and rotations from atomic displacements. As a result, we found that the doubly degenerate orbital states of the CH₃O interacts with an effective vibrational mode whose frequency is 130.50 meV. We report the Ham reduction factor as p=0.47p = 0.47. We note that a similar quenching factor (d=0.478d = 0.478) was reported in Ref. [37], where EOM-CCSDT/ANO1 level of theory [99] was applied to approximate the potential energy surface until the quartic order. Moreover, we performed non-collinear calculation of the energy splitting due to spin-orbit coupling as described in § 4.3. However, we had to smear the three electrons that resides on the four spin-orbital states of the ee orbital by forced 0.75 occupations by (e+↑0.75e−↑0.75−e+↓0.75e−↓0.75)(e_{+\uparrow}^{0.75}e_{-\uparrow}^{0.75}-e_{+\downarrow}^{0.75}e_{-\downarrow}^{0.75}) occupation in order to avoid the disastrous effect of spin contamination. Otherwise, we experienced 3× underestimation upon calculating λ0\lambda_0 within the (e+↑1e+↓1e−↑0.5e−↓0.5)(e_{+\uparrow}^{1}e_{+\downarrow}^{1}e_{-\uparrow}^{0.5}e_{-\downarrow}^{0.5}) constraint. Our intrinsic value of λ0=17.99\lambda_0 = 17.99 meV is further reduced by the Ham factor that results in λtheory=8.47\lambda_{\mathrm{theory}} = 8.47 meV which is in good agreement with the experimental value in Ref. [100] see Table 4 where we additionally compare the higher level multimode case that we will discuss in the following section.

ᵃp.w.ᵇcᵈ
λ0\lambda_018.016.516.516.5
λtheory\lambda_{\mathrm{theory}}8.477.537.87 (7.80)7.69

Table 4. Comparison of our single mode Jahn-Teller treatment on top of DFT calculations as described in Section 3.4 to the multimode treatments that of Refs. [37, 40, 106] and the experimentally reported value λexp=7.94\lambda_{\mathrm{exp}} = 7.94 meV from Ref. [100]. All values are in meV units while "p.w." refers to the present work. λtheory\lambda_{\mathrm{theory}} values under () parentheses refers recalculated values by our jahn-teller-dynamics code.

The general multimode treatment for CH₃O⁰

We determine the (A1⊕A1⊕…⊕E⊕E⊕…)⊗e(A_1 \oplus A_1 \oplus\ldots\oplus E \oplus E \oplus\ldots) \otimes e multimode Jahn-Teller case where an arbitrary amount of vibration modes transforming as EE and A1A_1 irreducible representations that of C3vC_{3v} are being coupled to a Jahn-Teller unstable degenerate electronic multiplet spanned by ee. We apply the Condon-Shortley convention that takes advantage of the orbital states (∣ex⟩,∣ey⟩)(|e_x\rangle, |e_y\rangle) and phonon vibration (q^ix,q^iy)(\hat{q}_{ix},\hat{q}_{iy}) double degeneracy both on electronic (∣e±⟩=∓(∣ex⟩±i∣ey⟩)/2)(|e_{\pm}\rangle= \mp(|e_x\rangle\pm i|e_y\rangle)/\sqrt{2}) and nuclear q^i±=∓(q^ix±iq^iy)\hat{q}_{i\pm} = \mp(\hat{q}_{ix} \pm i\hat{q}_{iy}) degrees of freedom. This convention allows us to express the polynomial basis of the phonon normal modes representation of nnth order by simple monomials. Methoxy molecule has three modes transforming as A1A_1 (q^1,q^2,q^3)(\hat{q}_1,\hat{q}_2,\hat{q}_3) and three doubly degenerate modes (q^4x/y,q^5x/y,q^6x/y)(\hat{q}_{4x/y},\hat{q}_{5x/y},\hat{q}_{6x/y}). We designed our code so when the user describe phonon modes, using these convention, the code can parse polinomial expressions that contain the corresponding complex basis polynomials: (q^i+/−)(\hat{q}_{i+/-}). It is important to note that the user can easily include the spin-orbit coupling Hamiltonian by defining a zeroth order polynomial between the electron states.

The fourth order dynamic Jahn-Teller Hamiltonian of the methoxy radical contains hundreds of terms. In order to simplify the inputs, and lower running time we set jtd_run to symmetrize Hamiltonian and complete the diagonal terms. We calculated the 10 lowest vibronic states of methoxy radical using various phonon energy cuts (12000 – 24000 cm−1^{-1}) with Hilbert space dimensions 22882-1767930. We set the number of Lanczos iterations to 2000. We tested the convergence and found that by changing the phonon energy cut from 20000 cm−1^{-1} (520484 dimensions) to 24000 cm−1^{-1} (1767930 dimensions), the eigenvalues improve by less then 0.1 cm−1^{-1}. Our vibronic energy level values are presented in Table 3 along with experimental values [37, 102, 103, 104, 105] [101]. Table 3 shows the splitting of the first two energy levels using different methods. Figure 11 demonstrates how the methoxy’s ground state is converging with respect to the Hilbert space dimension.

eigenvalue (cm−1^{-1})energy level (cm−1^{-1})experiment (cm−1^{-1})
7721.630.000
7785.9564.3361.5
8417.04695.41689.3
8672.51950.88945.9
8771.571049.941041.8
8837.331115.711108
8961.151239.521224.6
8971.021249.391233.7
9080.511358.891346.8
9101.661380.041365.7
9144.701423.071411.3

Table 3. This table presents the eigenvalues of the 1767930 dimensional Hamiltonian that describe the methoxy radical, compared with the experimental data taken from [37, 102, 103, 104, 105] [101].

Figure 11. This figure shows the calculated APES of the excited state of the neutral N3_3V0^0 defect. The a) and b) part of the figure represents the N3_3V0^0 point defect.

ᵃ single mode E⊗EE \otimes E JT case, up to 2nd order terms

ᵇ multimode JT, up to 2nd order terms without cross-terms, see Supp. Mat. of Ref. [40]

ᶜ multimode JT, up to 4th order terms, see Supp. Mat. of Ref. [37]

ᵈ multimode JT, up to 4th order terms, see Ref. [106]

The Taylor expansion of the coupling of electron orbitals and phonons is the same as in equation (5). In order to express the electron-vibration interaction from Eq. (5) in the circular basis for the first order:

H^1st=∣e+⟩⟨e+∣∑iFi(++)q^i+∣e−⟩⟨e−∣∑iFi(++)q^i†+∣e+⟩⟨e−∣∑iFi(+−)q^i+∣e−⟩⟨e+∣∑iFi(+−)∗q^i†(30)\begin{aligned} \hat{H}_{1\mathrm{st}} ={}& |e_{+}\rangle\langle e_{+}| \sum_i F_i^{(++)}\hat{q}_i + |e_{-}\rangle\langle e_{-}| \sum_i F_i^{(++)}\hat{q}_i^\dagger\\ &+ |e_{+}\rangle\langle e_{-}| \sum_i F_i^{(+-)}\hat{q}_i + |e_{-}\rangle\langle e_{+}| \sum_i F_i^{(+-)*}\hat{q}_i^\dagger \tag*{(30)} \end{aligned}

generally in the arbitrary order of methoxy’s dynamical Jahn-Teller interaction ii runs over the real coordinates of three A1A_1 modes (1, 2, 3) and as well as three double degenerate EE modes (4+,4−,5+,5−,6+,6−), however in the first order, only 4±, 5±, 6± coordinates are involved in the upper triangle of the Hamiltonian leading to a (E⊕E⊕E)⊗e(E \oplus E \oplus E)\otimes e three-mode JT case. Eq. (30) shows how Hermitian completion diagonal completion works. The higher order terms are generated based of Eq. (9). The other option, what we used is to define the upper triangle of the system such that the offdiagonal coupling coefficients are multiplied by a factor of two, and then the code symmetrize the Hamiltonian.

H^1stupper=∣e+⟩⟨e+∣∑iFi+−q^i+∣e−⟩⟨e−∣∑iFi−+q^i+∣e+⟩⟨e−∣∑i2×Fi+−q^i(31)\hat{H}_{1\mathrm{st}}^{\mathrm{upper}} = |e_{+}\rangle\langle e_{+}| \sum_i F_i^{+-}\hat{q}_i + |e_{-}\rangle\langle e_{-}| \sum_i F_i^{-+}\hat{q}_i + |e_{+}\rangle\langle e_{-}| \sum_i 2 \times F_i^{+-}\hat{q}_i \tag*{(31)}
H^1st=12(H^1st+H^1st†)\hat{H}_{1\mathrm{st}} = \frac{1}{2}\left(\hat{H}_{1\mathrm{st}}+\hat{H}_{1\mathrm{st}}^\dagger\right)

Our jahn-teller-dynamics package can parse and interpret polynomials in any order in the circularly polarized basis (q^i+,∣e±⟩)(\hat{q}_{i+}, |e_{\pm}\rangle). The user can set Hermitian completion and diagonal completion which follows the symmetry arguments explained by the Supp. Mat. of Ref. [37]. We demonstrate the capabilities of our code by loading up all polynomial expressions and dynamic Jahn-Teller coefficients up to quartic order directly obtained from the data presented in the Supp. Mat. that of Ref. [37] where six vibration modes (three non degenerate and three degenerate) are coupled to the electronic structure.

Pseudo Jahn-Teller case of the butatrien molecule

The pseudo Jahn-Teller effect occurs between electron orbital states that lies so close to each other that the nuclear vibrations are strong enough to create coupling between them. One remarkable example of this interaction is the butatrien cation (C4H4)+(C_4H_4)^+ that’s absorption spectrum is a well studied phenomena. It’s two lowest energy states 2B3g^2B_{3g} and 2B3u^2B_{3u} are coupled together[107, 108]. In this study we use vibrational modes and coupling coefficients of the two-mode linear model provided by [109]’s Supplementary Material and set the maximum number of vibrational quanta N=40N = 40. They do not consider the null point vibration in the vibrational energy Hamiltonian, and they use dimensional coordinate, therefore we implemented position operators in our codebase such that (q^i=12ωi(a^+a^†))(\hat{q}_i = \frac{1}{\sqrt{2\omega_i}}(\hat{a}+\hat{a}^\dagger)), where ωi\omega_i is the energy quantum of the iith harmonic oscillator, that in this case represents nuclear motion in the butatrien cation. The pseudo Jahn-Teller interaction between ∣2B3g⟩|^2B_{3g}\rangle and ∣2B3u⟩|^2B_{3u}\rangle states can be written as

Hpseudo=∣2B3g⟩⟨2B3u∣⊗F1guq^1+∣2B3g⟩⟨2B3g∣⊗F2ggq^2+∣2B3u⟩⟨2B3g∣⊗F1guq^1+∣2B3u⟩⟨2B3u∣⊗F2uuq^2(32)\begin{aligned} H_{\mathrm{pseudo}} ={}& |^2B_{3g}\rangle\langle^2B_{3u}| \otimes F_1^{gu}\hat{q}_1 + |^2B_{3g}\rangle\langle^2B_{3g}| \otimes F_2^{gg}\hat{q}_2 \\ &+ |^2B_{3u}\rangle\langle^2B_{3g}| \otimes F_1^{gu}\hat{q}_1 + |^2B_{3u}\rangle\langle^2B_{3u}| \otimes F_2^{uu}\hat{q}_2 \tag*{(32)} \end{aligned}

unbraced terms are defined by the user.

The vertical excitation spectrum can be seen in Figure 12 experimental values are taken from [110].

Vertical excitation spectrum of C4H4+

Figure 12. The vertical spectrum of the butatrien cation (C4H4+\mathrm{C}_4\mathrm{H}_4^+), Gaussians with fwhm = 0.05 is fitted to each points.

I(Eη)=∣⟨ψ0∣ψη⟩∣2 where ψ0=12(∣2B3g,0,0⟩+∣2B3u,0,0⟩)(33)I(E_\eta)=|\langle\psi_0|\psi_\eta\rangle|^2\ \text{where}\ \psi_0=\frac{1}{\sqrt{2}}\left(|^2B_{3g},0,0\rangle+|^2B_{3u},0,0\rangle\right) \tag*{(33)}

and ψη\psi_\eta is the iith eigenstate of the system with energy EηE_\eta. jtd_run provides the results in eigenstates.npz. It contains the basis states (basis_labels.npy), the eigenvalues (eigenvalues.npy) and the eigenstates (eigenvectors.npy), therefore the user can easily post-process the results to produce a vertical excitation or absorption spectrum.

Summary

In this work, we introduced jahn-teller-dynamics, a Python-based library designed to model any dynamic Jahn-Teller systems. On top of this library we created Exe.py, a

Python script specifically designed to perform E⊗eE \otimes e Jahn-Teller active systems that experience spin-orbit coupling using outputs of density functional theory calculation. The code enables accurate calculation of energy splittings induced by vibronic and spin-orbit interactions and then simulates optical transition energies in the presence of an external magnetic field – parameters of direct relevance to current experimental and theoretical studies in the recent years. Its capabilities were demonstrated by extracting key theoretical parameters for G4V centers in diamond, simulating the ZPL fine structure of the SnV−^- center, and we applied the code to predict the ZPL fine structure splitting energy of the neutral N3_3V0^0 defect in diamond and the reduced spin-orbit coupling in the methoxy (CH3_3O$∙{}^{\bullet}) radical.

The general dynamic Jahn-Teller modelling infrastructure is implemented in jtd_run. This script enables the user do define any type of dynamic Jahn-Teller system with various degenerate and non-degenerate electron states and vibrational modes. In order to present it’s capabilities we calculated the energy levels of the methoxy (CH3_3O$∙{}^{\bullet}) radical using the fourth order multimode dynamic Jahn-Teller model that was introduced by [37]. Moreover, we calculated the vertical excitation spectrum of the butatrien cation (C4_4H4+_4^+) that is a pseudo Jahn-Teller system. In all cases, the results show good agreement with experimental data. Due to it’s user friendly design, it can be readily extended to treat more complex scenarios, such as the T⊗(t+e)T \otimes(t + e) Jahn-Teller system [111, 112, 113, 114], pseudo Jahn-Teller effect [115, 116, 117, 118], and product Jahn-Teller systems [118, 119, 120, 121].

Therefore we can conclude that jahn-teller-dynamics is a robust codebase that can solve any dynamical Jahn-Teller problem and can serve as a useful asset for molecular and solid state physicists.

CRediT authorship contribution statement

Balázs Tóth: Writing – original draft, Software, Validation, Methodology, Investigation, Conceptualization, Data curation; Ádám Gali: Writing – review & editing, Supervision, Funding acquisition Gergő Thiering: Writing – review & editing, Supervision, Project administration, Formal analysis, Conceptualization, Validation, Funding acquisition

Data availability

The data that support the findings of this article will be openly available in the HUN-REN ARP database under the entry [122]. Additional exemplary data (NV center in diamond) can be found in the ARP database entry [123] of Ref. [10].

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Declaration of generative AI and AI-assisted technologies in the writing process

The authors used the grammar assisting tools provided by Writefull (within Overleaf) and ChatGPT to improve the overall readability and the quality of the text and the styling of figures. The authors reviewed and edited the content from sentence to sentence and acted appropriately for every single suggestion made by these tools, and thus the authors take full responsibility for the content of the manuscript.

Acknowledgements

The project supported by the Doctoral Excellence Fellowship Programme (DCEP) is funded by the National Research Development and Innovation Fund of the Ministry of Culture and Innovation and the Budapest University of Technology and Economics. Support by the Ministry of Culture and Innovation and the National Research, Development and Innovation Office (NKFIH) within the Quantum Information National Laboratory of Hungary (Grant No. 2022-2.1.1-NL-2022-00004) is much appreciated. A.G. acknowledges the high-performance computational resources provided by KIFÜ (Governmental Agency for IT Development), the European Commission for the projects SPINUS (Grant No. 101135699) and QuSPARC (Grant No. 101186889) as well as the QuantERA II projects Maestro (Grant No. 2019-2.1.7-ERA-NET-2022-00045) and Sensextreme (NKFIH Grant No. 2019-2.1.7-ERA-NET-2022-00040). G.T. was supported by the János Bolyai Research Scholarship of the Hungarian Academy of Sciences and by NKFIH under Grant No. STARTING 150113.

References

  1. [1]G. Thiering, A. Gali, Ab initio calculation of spin-orbit coupling for an nv center in diamond exhibiting dynamic jah n-teller effect, Phys. Rev. B 96 (2017) 081115. doi:10.1103/PhysRevB.96.081115. URL https://link.aps.org/doi/10.1103/PhysRevB.96.081115DOI
  2. [2]G. Thiering, A. Gali, Ab initio magneto-optical spectrum of group-iv vacancy color centers in diamond, Phys. Rev. X 8 (2018) 021063. doi:10.1103/PhysRevX.8.021063. URL https://link.aps.org/doi/10.1103/PhysRevX.8.021063DOI
  3. [3]G. Kresse, J. Furthmüller, Efficient iterative schemes for ab initio total-energy calculations using a plane-wave basis set, Phys. Rev. B 54 (1996) 11169–11186. doi:10.1103/PhysRevB.54.11169. URL https://link.aps.org/doi/10.1103/PhysRevB.54.11169DOI
  4. [4]F. S. Ham, Dynamical jahn-teller effect in paramagnetic resonance spectra: Orbital reduction factors and partial quenching of spin-orbit interaction, Phys. Rev. 138 (1965) A1727–A1740. doi:10.1103/PhysRev.138.A1727. URL https://link.aps.org/doi/10.1103/PhysRev.138.A1727DOI
  5. [5]G. D. Bent, Method of measuring the ham reduction factor q in c3v molecules, The Journal of Chemical Physics 92 (3) (1990) 1547–1552. doi:10.1063/1.458085. URL http://dx.doi.org/10.1063/1.458085
  6. [6]A. Norambuena, A. Jimenez, C. Becher, J. R. Maze, Effect of phonons on the electron spin resonance absorption spectrum, New Journal of Physics 22 (7) (2020) 073068. doi:10.1088/1367-2630/ab9da0. URL https://doi.org/10.1088/1367-2630/ab9da0
  7. [7]A. Csoré, A. Gali, Ab initio determination of pseudospin for paramagnetic defects in sic, Phys. Rev. B 102 (2020) 241201. doi:10.1103/PhysRevB.102.241201. URL https://link.aps.org/doi/10.1103/PhysRevB.102.241201DOI
  8. [8]G. Thiering, A. Gali, Photoexcitation and recombination processes of the neutral nitrogen-vacancy center in diamond from first principles, Journal of Applied Physics 136 (8) (Aug. 2024). doi:10.1063/5.0221228. URL https://dx.doi.org/10.1063/5.0221228
  9. [9]K. Li, V. D. Dergachev, I. D. Dergachev, S. Zhang, S. A. Varganov, Y. Ping, Excited-state dynamics and optically detected magnetic resonance of solid-state spin defects from first principles, Phys. Rev. B 110 (2024) 184302. doi:10.1103/PhysRevB.110.184302. URL https://link.aps.org/doi/10.1103/PhysRevB.110.184302DOI
  10. [10]G. Thiering, A. Gali, Nuclear-spin relaxation in solid-state-defect quantum bits via electron-phonon coupling in the optically excited state, Phys. Rev. Appl. 24 (2025) 044027. doi:10.1103/1q3f-7zvl. URL https://link.aps.org/doi/10.1103/1q3f-7zvlDOI
  11. [11]J. R. Weber, W. F. Koehl, J. B. Varley, A. Janotti, B. B. Buckley, C. G. Van de Walle, D. D. Awschalom, Quantum computing with defects, Proceedings of the National Academy of Sciences 107 (19) (2010) 8513–8518. doi:10.1073/pnas.1003052107. URL http://dx.doi.org/10.1073/pnas.1003052107
  12. [12]M. Atatüre, D. Englund, N. Vamivakas, S.-Y. Lee, J. Wrachtrup, Material platforms for spin-based photonic quantum technologies, Nature Reviews Materials 3 (5) (2018) 38–51. doi:10.1038/s41578-018-0008-9. URL http://dx.doi.org/10.1038/s41578-018-0008-9
  13. [13]G. Zhang, Y. Cheng, J.-P. Chou, A. Gali, Material platforms for defect qubits and single-photon emitters, Applied Physics Reviews 7 (3) (2020) 031308. doi:10.1063/5.0006075. URL https://doi.org/10.1063/5.0006075
  14. [14]L. C. Bassett, A. Alkauskas, A. L. Exarhos, K.-M. C. Fu, Quantum defects by design, Nanophotonics 8 (11) (2019) 1867–1888. doi:10.1515/nanoph-2019-0211. URL http://dx.doi.org/10.1515/nanoph-2019-0211
  15. [15]A. Chatterjee, P. Stevenson, S. De Franceschi, A. Morello, N. P. de Leon, F. Kuemmeth, Semiconductor qubits in practice, Nature Reviews Physics 3 (3) (2021) 157–177. doi:10.1038/s42254-021-00283-9. URL http://dx.doi.org/10.1038/s42254-021-00283-9
  16. [16]G. Wolfowicz, F. J. Heremans, C. P. Anderson, S. Kanai, H. Seo, A. Gali, G. Galli, D. D. Awschalom, Quantum guidelines for solid-state spin defects, Nature Reviews Materials 6 (10) (2021) 906–925. doi:10.1038/s41578-021-00306-y. URL http://dx.doi.org/10.1038/s41578-021-00306-y
  17. [17]M. W. Doherty, N. B. Manson, P. Delaney, F. Jelezko, J. Wrachtrup, L. C. Hollenberg, The nitrogen-vacancy colour centre in diamond, Physics Reports 528 (1) (2013) 1–45, the nitrogen-vacancy colour centre in diamond. doi:https://doi.org/10.1016/j.physrep.2013.02.001. URL https://www.sciencedirect.com/science/article/pii/S0370157313000562
  18. [18]A. Gali, Ab initio theory of the nitrogen-vacancy center in diamond, Nanophotonics 8 (11) (2019) 1907–1943 [cited 2025-07-28]. doi:10.1515/nanoph-2019-0154. URL https://doi.org/10.1515/nanoph-2019-0154
  19. [19]C. Hepp, T. Müller, V. Waselowski, J. N. Becker, B. Pingault, H. Sternschulte, D. Steinmüller-Nethl, A. Gali, J. R. Maze, M. Atatüre, C. Becher, Electronic structure of the silicon vacancy color center in diamond, Phys. Rev. Lett. 112 (2014) 036405. doi:10.1103/PhysRevLett.112.036405. URL https://link.aps.org/doi/10.1103/PhysRevLett.112.036405DOI
  20. [20]E. Ekimov, S. Lyapin, K. Boldyrev, M. Kondrin, R. Khmelnitskii, V. Gavva, T. Kotereva, M. Popova, Germanium–vacancy color center in isotopically enriched diamonds synthesized at high pressures, JETP Letters 102 (2015) 701–706. doi:10.1134/S00213640152030034.DOI
  21. [21]P. Siyushev, M. H. Metsch, A. Ijaz, J. M. Binder, M. K. Bashkar, D. D. Suckachev, A. Sipahigil, R. E. Evans, C. T. Nguyen, M. D. Lukin, P. R. Hemmer, Y. N. Palyanov, I. N. Kupriyanov, Y. M. Borzdov, L. J. Rogers, F. Jelezko, Optical and microwave control of germanium-vacancy center spins in diamond, Phys. Rev. B 96 (2017) 081201. doi:10.1103/PhysRevB.96.081201. URL https://link.aps.org/doi/10.1103/PhysRevB.96.081201DOI
  22. [22]T. Iwasaki, Y. Miyamoto, T. Taniguchi, P. Siyushev, M. H. Metsch, F. Jelezko, M. Hatano, Tin-vacancy quantum emitters in diamond, Phys. Rev. Lett. 119 (2017) 253601. doi:10.1103/PhysRevLett.119.253601. URL https://link.aps.org/doi/10.1103/PhysRevLett.119.253601DOI
  23. [23]A. E. Rugar, C. Dory, S. Sun, J. Vučković, Characterization of optical and spin properties of single tin-vacancy centers in diamond nanopillars, Phys. Rev. B 99 (2019) 205417. doi:10.1103/PhysRevB.99.205417. URL https://link.aps.org/doi/10.1103/PhysRevB.99.205417DOI
  24. [24]J. Görlitz, D. Herrmann, G. Thiering, P. Fuchs, M. Gandl, T. Iwasaki, T. Taniguchi, M. Kieschnick, J. Meijer, M. Hatano, A. Gali, C. Becher, Spectroscopic investigations of negatively charged tin-vacancy centres in diamond, New Journal of Physics 22 (1) (2020) 013048. doi:10.1088/1367-2630/ab6631 URL https://dx.doi.org/10.1088/1367-2630/ab6631
  25. [25]S. Ditalia Tchernij, T. Lühmann, T. Herzig, J. Küpper, A. Damin, S. Santonocito, M. Signorile, P. Traina, E. Moreva, F. Celegato, S. Pezzagna, I. P. Degiovanni, P. Olivero, M. Jakišić, J. Meijer, P. M. Genovese, J. Forneris, Single-photon emitters in lead-implanted single-crystal diamond, ACS Photonics 5 (12) (2018) 4864–4871. arXiv:https://doi.org/10.1021/acsphotonics.8b01013, doi:10.1021/acsphotonics.8b01013. URL https://doi.org/10.1021/acsphotonics.8b01013
  26. [26]H. A. Jahn, E. Teller, Stability of Polyatomic Molecules in Degenerate Electronic States. I. Orbital Degeneracy, Proceedings of the Royal Society of London Series A 161 (905) (1937) 220–235. doi:10.1098/rspa.1937.0142.DOI
  27. [27]H. C. Longuet-Higgins, Some recent developments in the theory of molecular energy levels, Advances in Spectroscopy, Volume II. Edited by H. W. Thompson. II. (1961) 429–472.
  28. [28]I. Bersuker, The Jahn-Teller Effect, Cambridge University Press, 2006.DOI
  29. [29]I. Bersuker, V. Polinger, Vibronic interactions in molecules and crystals, Vol. 49, Springer Berlin, 2011.
  30. [30]T. Momose, Y. Endo, E. Hirota, T. Shida, The submillimeter-wave spectrum of the 13ch3o radical, The Journal of Chemical Physics 88 (9) (1988) 5338–5343. doi:10.1063/1.454593. URL http://dx.doi.org/10.1063/1.454593
  31. [31]T. Momose, Y. Endo, E. Hirota, T. Shida, Erratum: The submillimeter-wave spectrum of the 13ch3o radical [j. chem. phys. 8 8, 5338 (1988)], The Journal of Chemical Physics 90 (8) (1989) 4636–4637. doi:10.1063/1.456703. URL http://dx.doi.org/10.1063/1.456703
  32. [32]C. R. Brazier, P. F. Bernath, The òE − òA1 transition of monomethyl calcium: A rotational analysis, The Journal of Chemical Physics 91 (8) (1989) 4548–4554. doi:10.1063/1.456742. URL http://dx.doi.org/10.1063/1.456742
  33. [33]T. A. Barckholtz, T. A. Miller, The calculation of spectroscopic jahn-teller parameters by ab initio methods, The Journal of Physical Chemistry A 103 (14) (1999) 2321–2336. doi:10.1021/jp983829x. URL http://dx.doi.org/10.1021/jp983829x
  34. [34]D. G. Melnik, J. Liu, M.-W. Chen, T. A. Miller, R. F. Curl, The spectroscopic characterization of the methoxy radical. III. rotationally resolved òA1−−òE electronic and òE submillimeter wave spectra of partially deuterated ch3do and ch2do radicals, The Journal of Chemical Physics 135 (9) (Sep. 2011). doi:10.1063/1.3615724. URL http://dx.doi.org/10.1063/1.3615724
  35. [35]J. Liu, M.-W. Chen, D. Melnik, J. T. Yi, T. A. Miller, The spectroscopic characterization of the methoxy radical. i. rotationally resolved òE −−òA1 electronic spectra of CH3O, The Journal of Chemical Physics 130 (7) (Feb. 2009). doi:10.1063/1.3072104. URL http://dx.doi.org/10.1063/1.3072104
  36. [36]M.-A. Martin-Drumel, J.-T. Spaniol, O. Chitarra, O. Pirali, H. S. P. Müller, Submillimeter-wave spectroscopy of the ch3o radical, The Journal of Physical Chemistry A 129 (3) (2025) 754–761. doi:10.1021/acs.jpca.4c07638. URL http://dx.doi.org/10.1021/acs.jpca. 4c07638
  37. [37]K. Sharma, O. A. Vasilyev, J. F. Stanton, T. A. Miller, Ab initio simulation of spin-vibronic spectra of methoxy radical, The Journal of Chemical Physics 162 (18) (May 2025). doi:10.1063/5.0266367. URL http://dx.doi.org/10.1063/5.0266367
  38. [38]J. Nagesh, E. L. Sibert, J. F. Stanton, Simulation of laser excitation spectrum of ch3o and cd3o, Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 119 (2014) 90–99. doi:10.1016/j.saa.2013.02.037. URL http://dx.doi.org/10.1016/j.saa.2013.02.037
  39. [39]B. A. Johnson, E. L. Sibert, Assigning the low lying vibronic states of ch3o and cd3o, The Journal of Chemical Physics 146 (17) (May 2017). doi:10.1063/1.4981795. URL http://dx.doi.org/10.1063/1.4981795
  40. [40]K. Sharma, S. Garner, T. A. Miller, J. F. Stanton, First-principles calculation of jahn–teller rotational distortion parameters, The Journal of Physical Chemistry A 123 (23) (2019) 4990–5004. doi:10.1021/acs.jpca.9b03360. URL http://dx.doi.org/10.1021/acs.jpca.9b03360
  41. [41]L. B. Knight, B. W. Gregory, D. W. Hill, C. A. Arrington, T. Momose, T. Shida, Electron-spin-resonance studies on 12ch3f+, 13ch3f+, and 12ch2df+ in neon matrices at 4 k: Comparison with theoretical calculations, The Journal of Chemical Physics 94 (1) (1991) 67–79. doi:10.1063/1.460335. URL http://dx.doi.org/10.1063/1.460335
  42. [42]M. J. Dick, P. M. Sheridan, J.-G. Wang, P. F. Bernath, High-resolution laser excitation spectroscopy of the Ã2E − X̃2A1 transition of SrCH3, The Journal of Chemical Physics 124 (17) (May 2006). doi:10.1063/1.2189236. URL http://dx.doi.org/10.1063/1.2189236
  43. [43]E. Janzén, A. Gali, P. Carlsson, A. Gällström, B. Magnusson, N. Son, The silicon vacancy in sic, Physica B: Condensed Matter 404 (22) (2009) 4354–4358. doi:https://doi.org/10.1016/j.physb.2009.09.023. URL https://www.sciencedirect.com/science/article/pii/S0921452609010862
  44. [44]P. G. Baranov, A. P. Bundakova, A. A. Soltamova, S. B. Orlinskii, I. V. Borovykh, R. Zondervan, R. Verberk, J. Schmidt, Silicon vacancy in sic as a promising quantum system for single-defect and single-photon spectroscopy, Phys. Rev. B 83 (2011) 125203. doi:10.1103/PhysRevB.83.125203. URL https://link.aps.org/doi/10.1103/PhysRevB.83.125203DOI
  45. [45]F. Fuchs, V. A. Soltamov, S. Väth, P. G. Baranov, E. N. Mokhov, G. V. Astakhov, V. Dyakonov, Silicon carbide light-emitting diode as a prospective room temperature source for single photons, Scientific Reports 3 (1) (2013) 1637. doi:10.1038/srep01637. URL https://doi.org/10.1038/srep01637
  46. [46]M. Widmann, S.-Y. Lee, T. Rendler, N. T. Son, H. Fedder, S. Paik, L.-P. Yang, N. Zhao, S. Yang, J. Booker, A. Denisenko, M. Jamali, S. A. Momenzadeh, I. Gerhardt, T. Ohshima, A. Gali, E. Janzén, J. Wachtrup, Coherent control of single spins in silicon carbide at room temperature, Nature Materials 14 (2) (2015) 164–168. doi:10.1038/nmat4145. URL https://doi.org/10.1038/nmat4145
  47. [47]N. T. Son, P. Carlsson, J. ul Hassan, E. Janzén, T. Umeda, J. Isoya, A. Gali, M. Bockstedte, N. Morishita, T. Ohshima, H. Itoh, Divacancy in 4h-sic, Phys. Rev. Lett. 96 (2006) 055501. doi:10.1103/PhysRevLett.96.055501. URL https://link.aps.org/doi/10.1103/PhysRevLett.96.055501DOI
  48. [48]A. Gali, Time-dependent density functional study on the excitation spectrum of point defects in semiconductors, physica status solidi (b) 248 (6) (2011) 1337–1346. arXiv:https://onlinelibrary.wiley.com/doi/pdf/10.1002/pssb.201046254. doi:https://doi.org/10.1002/pssb.201046254. URL https://onlinelibrary.wiley.com/doi/abs/10.1002/pssb.201046254
  49. [49]W. F. Koehl, B. B. Buckley, F. J. Heremans, G. Calusine, D. D. Awschalom, Room temperature coherent control of defect spin qubits in silicon carbide, Nature 479 (7371) (2011) 84–87. doi:10.1038/nature10562. URL http://dx.doi.org/10.1038/nature10562
  50. [50]D. J. Christle, P. V. Klimov, C. F. de las Casas, K. Szász, V. Ivády, V. Jokubavicius, J. Ul Hassan, M. Syväjärvi, W. F. Koehl, T. Ohshima, N. T. Son, E. Janzén, A. Gali, D. D. Awschalom, Isolated spin qubits in sic with a high-fidelity infrared spin-to-photon interface, Phys. Rev. X 7 (2017) 021046. doi:10.1103/PhysRevX.7.021046. URL https://link.aps.org/doi/10.1103/PhysRevX.7.021046DOI
  51. [51]S. L. Bayliss, D. W. Laorenza, P. J. Mintun, B. D. Kovos, D. E. Freedman, D. D. Awschalom, Optically addressable molecular spins for quantum information processing, Science 370 (6552) (2020) 1309–1312. arXiv:https://www.science.org/doi/pdf/10.1126/science.abb9352, doi:10.1126/science.abb9352. URL https://www.science.org/doi/abs/10.1126/science.abb9352DOI
  52. [52]K. Wang, T. Zeng, Hamiltonian formalism of spin–orbit jahn–teller and pseudo-jahn–teller problems in trigonal and tetragonal symmetries, Physical Chemistry Chemical Physics 21 (35) (2019) 18939–18957. [53] S. Steiner, S. Khmelevskyi, M. Marsmann, G. Kresse, Calculation of the magnetic anisotropy with projected-augmented-wave methodology and the case study of disordered fe₁₋ₓcoₓ alloys, Phys. Rev. B 93 (2016) 224425. doi:10.1103/PhysRevB.93.224425. URL https://link.aps.org/doi/10.1103/PhysRevB.93.224425
  53. [54]K. W. H. Stevens, On the Magnetic Properties of Co-valent XY₆ Complexes, Proceedings of the Royal Society of London Series A 219 (1139) (1953) 542–555. doi:10.1098/rspa.1953.0166.
  54. [55]M. Gerloch, J. R. Miller, Covalence and the Orbital Reduction Factor, k, in Magnetochemistry, John Wiley & Sons, Ltd, 1968, Ch. 1, pp. 1–47. arXiv:https://onlinelibrary.wiley.com/doi/pdf/10.1002/9780470166116.ch1 doi:https://doi.org/10.1002/9780470166116.ch1. URL https://onlinelibrary.wiley.com/doi/abs/10.1002/9780470166116.ch1
  55. [56]Z. Gajek, J. Mulak, Steven’s orbital reduction factor in ionic clusters, Journal of Magnetism and Magnetic Materials 53 (1–2) (1985) 63–70. doi:10.1016/0304-8853(85)90130-1. URL http://dx.doi.org/10.1016/0304-8853(85)90130-1
  56. [57]Q. Wei, M. G. Brik, L.-X. Guo, B. Wei, Investigations of g factors for the 2e state of v²+ ions in al₂o₃ crystals, Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 97 (2012) 50–53. doi:10.1016/j.saa.2012.05.074. URL http://dx.doi.org/10.1016/j.saa.2012.05.074
  57. [58]F. S. Ham, Effect of linear jahn-teller coupling on paramagnetic resonance in a ²e state, Phys. Rev. 166 (1968) 307–321. doi:10.1103/PhysRev.166.307. URL https://link.aps.org/doi/10.1103/PhysRev.166.307DOI
  58. [59]S. Meesala, Y.-I. Sohn, B. Pingault, L. Shao, H. A. Atikian, J. Holzgrafe, M. Gündoğan, C. Stavrakas, A. Sipahigil, C. Chia, R. Evans, M. J. Burek, M. Zhang, L. Wu, J. L. Pacheco, J. Abraham, E. Bielejec, M. D. Lukin, M. Atatüre, M. Lončar, Strain engineering of the silicon-vacancy center in diamond, Physical Review B 97 (20) (May 2018). doi:10.1103/physrevb.97.205444. URL http://dx.doi.org/10.1103/PhysRevB.97.205444
  59. [60]E. I. Rosenthal, C. P. Anderson, H. C. Kleidermacher, A. J. Stein, H. Lee, J. Grzesik, G. Scuri, A. E. Rugar, D. Riedel, S. Aghaeimeibodi, G. H. Ahn, K. Van Gasse, J. Vučković, Microwave spin control of a tin-vacancy qubit in diamond, Phys. Rev. X 13 (2023) 031022. doi:10.1103/PhysRevX.13.031022. URL https://link.aps.org/doi/10.1103/PhysRevX.13.031022DOI
  60. [61]B. McCullian, H. Cheung, H. Chen, G. Fuchs, Quantifying the spectral diffusion of n-v centers by symmetry, Phys. Rev. Appl. 18 (2022) 064011. doi:10.1103/PhysRevApplied.18.064011. URL https://link.aps.org/doi/10.1103/PhysRevApplied.18.064011DOI
  61. [62]J. Wolters, N. Sadzak, A. W. Schell, T. Schröder, O. Benson, Measurement of the ultrafast spectral diffusion of the optical transition of nitrogen vacancy centers in nano-size diamond using correlation interferometry, Phys. Rev. Lett. 110 (2013) 027401. doi:10.1103/PhysRevLett.110.027401. URL https://link.aps.org/doi/10.1103/PhysRevLett.110.027401DOI
  62. [63]H. F. Fotso, A. E. Feiguin, D. D. Awschalom, V. V. Dobrovitski, Suppressing spectral diffusion of emitted photons with optical pulses, Phys. Rev. Lett. 116 (2016) 033603. doi:10.1103/PhysRevLett.116.033603. URL https://link.aps.org/doi/10.1103/PhysRevLett.116.033603DOI
  63. [64]K. D. Jahnke, A. Sipahigil, J. M. Binder, M. W. Doherty, M. Metsch, L. J. Rogers, N. B. Manson, M. D. Lukin, F. Jelezko, Electron–phonon processes of the silicon-vacancy centre in diamond, New Journal of Physics 17 (4) (2015) 043011. doi:10.1088/1367-2630/17/4/043011. URL http://dx.doi.org/10.1088/1367-2630/17/4/043011
  64. [65]I. B. W. Harris, D. Englund, Coherence of group-iv color centers, Phys. Rev. B 109 (2024) 085414. doi:10.1103/PhysRevB.109.085414. URL https://link.aps.org/doi/10.1103/PhysRevB.109.085414DOI
  65. [66]M. Mohseni, I. Razinkovas, V. Žalandauskas, G. Thiering, A. Gali, Magneto-optical properties of group-iv vacancy centers in diamond upon hydrostatic pressure, Phys. Rev. B 112 (2025) 155201. doi:10.1103/fq19-lfmv. URL https://link.aps.org/doi/10.1103/fq19-lfmvDOI
  66. [67]J. M. Brevoord, L. G. C. Wienhoven, N. Codreanu, T. Ishiguro, E. Van Leeuwen, M. Iuliano, L. De Santis, C. Waas, H. K. C. Beukers, T. Turan, C. Errando-Herranz, K. Kawaguchi, R. Hanson, Large-range tuning and stabilization of the optical transition of diamond tin-vacancy centers by in situ strain control, Applied Physics Letters 126 (17) (Apr. 2025). doi:10.1063/5.0251211. URL http://dx.doi.org/10.1063/5.0251211
  67. [68]X. Guo, A. M. Stramma, Z. Li, W. G. Roth, B. Huang, Y. Jin, R. A. Parker, J. Arjona Martínez, N. Shofer, C. P. Michaels, C. P. Purser, M. H. Appel, E. M. Alexeev, T. Liu, A. C. Ferrari, D. D. Awschalom, N. Delegan, B. Pingault, G. Galli, F. J. Heremans, M. Atatüre, A. A. High, Microwave-based quantum control and coherence protection of tin-vacancy spin qubits in a strain-tuned diamond-membrane heterostructure, Phys. Rev. X 13 (2023) 041037. doi:10.1103/PhysRevX.13.041037. URL https://link.aps.org/doi/10.1103/PhysRevX.13.041037DOI
  68. [69]Y.-I. Sohn, S. Meesala, B. Pingault, H. A. Atikian, J. Holzgrafe, M. Gündoğan, C. Stramma, M. J. Stanley, A. Sipahigil, J. Choi, M. Zhang, J. L. Pacheco, J. Abraham, E. Bielejec, M. D. Lukin, M. Atatüre, M. Lončar, Controlling the coherence of a diamond spin qubit through its strain environment, Nature Communications 9 (1) (May 2018). doi:10.1038/s41467-018-04340-3. URL http://dx.doi.org/10.1038/s41467-018-04340-3
  69. [70]M. Klotz, K. G. Fehler, R. Waltrich, E. S. Steiger, S. Häußler, P. Reddy, L. F. Kulikova, V. A. Davydov, V. N. Agafonov, M. W. Doherty, A. Kubanek, Prolonged orbital relaxation by locally modified phonon density of states for the siv− center in nanodiamonds, Phys. Rev. Lett. 128 (2022) 153602. doi:10.1103/PhysRevLett.128.153602. URL https://link.aps.org/doi/10.1103/PhysRevLett.128.153602DOI
  70. [71]K. Kuruma, B. Pingault, C. Chia, M. Haas, G. D. Joe, D. R. Assumpcao, S. W. Ding, C. Jin, C. J. Xin, M. Yeh, N. Sinclair, M. Lončar, Controlling interactions between high-frequency phonons and single quantum systems using phononic crystals, Nature Physics 21 (1) (2024) 77–82. doi:10.1038/s41567-024-02697-5. URL http://dx.doi.org/10.1038/s41567-024-02697-5
  71. [72]M. Kasha, Characterization of electronic transitions in complex molecules, Discussions of the Faraday Society 9 (1950) 14. doi:10.1039/df9500900014. URL http://dx.doi.org/10.1039/DF950900014
  72. [73]M. Kasha, S. P. McGlynn, Molecular electronic spectroscopy, Annual Review of Physical Chemistry 7 (1) (1956) 403–424. doi:10.1146/annurev.pc.07.100156.002155. URL http://dx.doi.org/10.1146/annurev.pc.07.100156.002155
  73. [74]A. Liu, S. T. Cundiff, D. B. Almeida, R. Ulbricht, Spectral broadening and ultrafast dynamics of a nitrogen-vacancy center ensemble in diamond, Materials for Quantum Technology 1 (2) (2021) 025002. doi:10.1088/2633-4356/abf330. URL http://dx.doi.org/10.1088/2633-4356/abf330
  74. [75]R. Ulbricht, S. Dong, A. Gali, S. Meng, Z.-H. Loh, Vibrational relaxation dynamics of the nitrogen-vacancy center in diamond, Phys. Rev. B 97 (2018) 220302. doi:10.1103/PhysRevB.97.220302. URL https://link.aps.org/doi/10.1103/PhysRevB.97.220302DOI
  75. [76]V. M. Huxter, T. A. A. Oliver, D. Budker, G. R. Fleming, Vibrational and electronic dynamics of nitrogen–vacancy centres in diamond revealed by two-dimensional ultrafast spectroscopy, Nature Physics 9 (11) (2013) 744–749. doi:10.1038/nphys2753. URL http://dx.doi.org/10.1038/nphys2753
  76. [77]R. Ulbricht, S. Dong, I.-Y. Chang, B. M. K. Mariserla, K. M. Dani, K. Hyeon-Deuk, Z.-H. Loh, Jahn-teller-induced femtosecond electronic depolarization dynamics of the nitrogen-vacancy defect in diamond, Nature Communications 7 (1) (Nov. 2016). doi:10.1038/ncomms13510. URL http://dx.doi.org/10.1038/ncomms13510
  77. [78]W. P. Carbery, C. A. Farfan, R. Ulbricht, D. B. Turner, The phonon-modulated jahn–teller distortion of the nitrogen vacancy center in diamond, Nature Communications 15 (1) (Oct. 2024). doi:10.1038/s41467-024-52712-9. URL http://dx.doi.org/10.1038/s41467-024-52712-9
  78. [79]J. Eng, C. Gourlaouen, E. Gindensperger, C. Daniel, Spin-vibronic quantum dynamics for ultrafast excited-state processes, Accounts of Chemical Research 48 (3) (2015) 809–817. doi:10.1021/ar500369r. URL http://dx.doi.org/10.1021/ar500369r
  79. [80]A. E. Rugar, S. Aghaimeibodi, D. Riedel, C. Dory, H. Lu, P. J. McQuade, Z.-X. Shen, N. A. Melosh, J. Vučković, Quantum photonic interface for tin-vacancy centers in diamond, Phys. Rev. X 11 (2021) 031021. doi:10.1103/PhysRevX.11.031021. URL https://link.aps.org/doi/10.1103/PhysRevX.11.031021DOI
  80. [81]G. Thiering, A. Gali, Erratum: Ab initio magneto-optical spectrum of group-iv vacancy color centers in diamond [phys. rev. x 8, 021063 (2018)], Phys. Rev. X 10 (2020) 039901. doi:10.1103/PhysRevX.10.039901. URL https://link.aps.org/doi/10.1103/PhysRevX.10.039901
  81. [82]P. E. Blöchl, Projector augmented-wave method, Phys. Rev. B 50 (1994) 17953–17979. doi:10.1103/PhysRevB.50.17953. URL https://link.aps.org/doi/10.1103/PhysRevB.50.17953DOI
  82. [83]G. Kresse, D. Joubert, From ultrasoft pseudopotentials to the projector augmented-wave method, Phys. Rev. B 59 (1999) 1758–1775. doi:10.1103/PhysRevB.59.1758. URL https://link.aps.org/doi/10.1103/PhysRevB.59.1758DOI
  83. [84]D. Hobbs, G. Kresse, J. Hafner, Fully unconstrained noncollinear magnetism within the projector augmented-wave method, Phys. Rev. B 62 (2000) 11556–11570. doi:10.1103/PhysRevB.62.11556. URL https://link.aps.org/doi/10.1103/PhysRevB.62.11556DOI
  84. [85]F. Neese, Efficient and accurate approximations to the molecular spin–orbit coupling operator and their use in molecular g-tensor calculations, The Journal of Chemical Physics 122 (3) (Jan. 2005). doi:10.1063/1.1829047. URL http://dx.doi.org/10.1063/1.1829047
  85. [86]F. Bruder, Y. J. Franzke, F. Weigend, Paramagnetic nmr shielding tensors based on scalar exact two-component and spin–orbit perturbation theory, The Journal of Physical Chemistry A 126 (30) (2022) 5050–5069. doi:10.1021/acs.jpca.2c03579. URL http://dx.doi.org/10.1021/acs.jpca.2c03579
  86. [87]Y. J. Franzke, J. M. Yu, Quasi-relativistic calculation of epr g tensors with derivatives of the decoupling transformation, gauge-including atomic orbitals, and magnetic balance, Journal of Chemical Theory and Computation 18 (4) (2022) 2246–2266. doi:10.1021/acs.jctc.1c01175. URL http://dx.doi.org/10.1021/acs.jctc.1c01175
  87. [88]O. Bengone, M. Alouani, P. Blöchl, J. Hugel, Implementation of the projector augmented-wave lda+u method: Application to the electronic structure of nio, Phys. Rev. B 62 (2000) 16392–16401. doi:10.1103/PhysRevB.62.16392. URL https://link.aps.org/doi/10.1103/PhysRevB.62.16392DOI
  88. [89]O. Bulancea-Lindvall, J. Davidsson, I. G. Ivanov, A. Gali, V. Ivády, R. Armiento, I. A. Abrikosov, Temperature dependence of the ab lines and optical properties of the carbon–antisite–vacancy pair in 4h–SiC, Phys. Rev. Appl. 22 (2024) 034056. doi:10.1103/PhysRevApplied.22.034056. URL https://link.aps.org/doi/10.1103/PhysRevApplied.22.034056
  89. [90]Y. Xiong, G. Hautier, dscf in VASP for excited-state defect computations: tips and pitfalls (2025). arXiv:2505.04748. URL https://arxiv.org/abs/2505.04748
  90. [91]J. P. Perdew, K. Burke, M. Ernzerhof, Generalized gradient approximation made simple, Phys. Rev. Lett. 77 (1996) 3865–3868. doi:10.1103/PhysRevLett.77.3865. URL https://link.aps.org/doi/10.1103/PhysRevLett.77.3865DOI
  91. [92]J. Heyd, G. E. Scuseria, M. Ernzerhof, Hybrid functionals based on a screened coulomb potential, The Journal of Chemical Physics 118 (18) (2003) 8207–8215. arXiv:https://pubs.aip.org/aip/jcp/article-pdf/118/18/8207/19093575/8207_1_online.pdf, doi:10.1063/1.1564060. URL https://doi.org/10.1063/1.1564060
  92. [93]A. V. Krukau, O. A. Vydrov, A. F. Izmaylov, G. E. Scuseria, Influence of the exchange screening parameter on the performance of screened hybrid functionals, The Journal of Chemical Physics 125 (22) (2006) 224106. arXiv:https://pubs.aip.org/aip/jcp/article-pdf/doi/10.1063/1.2404663/13263224/224106_1_online.pdf, doi:10.1063/1.2404663. URL https://doi.org/10.1063/1.2404663
  93. [94]M. N. R. Ashfold, J. P. Goss, B. L. Green, P. W. May, M. E. Newton, C. V. Peaker, Nitrogen in diamond, Chemical Reviews 120 (12) (2020) 5745–5794. doi:10.1021/acs.chemrev.9b00518. URL https://doi.org/10.1021/acs.chemrev.9b00518
  94. [95]A. M. Zaitsev, Optical Properties of Diamond, Springer Berlin, Heidelberg, 2001.DOI
  95. [96]M. Babamoradi, S. Asgari, A. Ranjbar, R. V. Belosludov, S. Yunoki, Many-electron states of the n2 and n3 color centers in diamond: A first-principles and many-body study, Physica B: Condensed Matter 505 (2017) 17–21. doi:https://doi.org/10.1016/j.physb.2016.10.021. URL https://www.sciencedirect.com/science/article/pii/S0921452616304835
  96. [97]J. P. Goss, R. Jones, S. J. Breuer, P. R. Briddon, S. Öberg, The twelve-line 1.682 ev luminescence center in diamond and the vacancy-silicon complex, Phys. Rev. Lett. 77 (1996) 3041–3044. doi:10.1103/PhysRevLett.77.3041. URL https://link.aps.org/doi/10.1103/PhysRevLett.77.3041DOI
  97. [98]C. Adamo, V. Barone, Toward reliable density functional methods without adjustable parameters: The pbe0 model, The Journal of Chemical Physics 110 (13) (1999) 6158–6170. doi:10.1063/1.478522. URL https://doi.org/10.1063/1.478522
  98. [99]D. A. Matthews, J. F. Stanton, A new approach to approximate equation-of-motion coupled cluster with triple excitations, The Journal of Chemical Physics 145 (12) (2016) 124102. doi:10.1063/1.4962910. URL https://doi.org/10.1063/1.4962910
  99. [100]Y. Lee, G. Wann, Y. Lee, Vibronic analysis of the à → X̃ laser-induced fluorescence of jet-cooled methoxy (ch3o) A. Geers, J. Kappert, F. Temps, J. W. Wiebrecht, Stimulated emission pumping spectroscopy of ch₃o (X̃²e) at the dissociation limit, The Journal of Chemical Physics 93 (2) (1990) 1472–1473. doi:10.1063/1.459157. URL https://doi.org/10.1063/1.459157
  100. [102]A. Geers, J. Kappert, F. Temps, J. W. Wiebrecht, Stimulated-emission pumping spectroscopy of highly excited states of ch₃o (X̃²e): zero-order vibrational states at 3000 cm⁻¹ ≤ 9000 cm⁻¹, J. Opt. Soc. Am. B 7 (9) (1990) 1935–1943. doi:10.1364/JOSAB.7.001935. URL https://opg.optica.org/josab/abstract.cfm?URI=josab-7-9-1935DOI
  101. [103]A. Geers, J. Kappert, F. Temps, T. J. Sears, Stimulated emission pumping spectroscopy of ch₃o (X̃²e, ν6): New observations on the jahn–teller effect, The Journal of Chemical Physics 98 (5) (1993) 4297–4300.DOI
  102. [104]A. Geers, J. Kappert, F. Temps, J. W. Wiebrecht, Rotation–vibration state resolved unimolecular dynamics of highly vibrationally excited ch₃o (X̃²e). i. observed stimulated emission pumping spectra, The Journal of Chemical Physics 101 (5) (1994) 3618–3633. doi:10.1063/1.467547. URL https://doi.org/10.1063/1.467547
  103. [105]M. L. Weichman, L. Cheng, J. B. Kim, J. F. Stanton, D. M. Neumark, Low-lying vibronic level structure of the ground state of the methoxy radical: Slow electron velocity-map imaging (sevi) spectra and köppel-domcke-cederbaum (kdc) vibronic hamiltonian calculations, The Journal of Chemical Physics 146 (22) (2017) 224309. doi:10.1063/1.4984963. URL https://doi.org/10.1063/1.4984963
  104. [106]M. L. Weichman, L. Cheng, J. B. Kim, J. F. Stanton, D. M. Neumark, Low-lying vibronic level structure of the ground state of the methoxy radical: Slow electron velocity-map imaging (sevi) spectra and köppel-domcke-cederbaum (kdc) vibronic hamiltonian calculations, The Journal of Chemical Physics 146 (22) (Jun 2017). doi:10.1063/1.4984963. URL http://dx.doi.org/10.1063/1.4984963
  105. [107]L. Cederbaum, W. Domcke, H. Köppel, W. V. Niessen, Strong vibronic coupling effects in ionization spectra: The “mystery band” of butatriene, Chemical Physics 26 (2) (1977) 169–177. doi:https://doi.org/10.1016/0301-0104(77)87041-9. URL https://www.sciencedirect.com/science/article/pii/0301010477870419
  106. [108]C. Cattarius, G. A. Worth, H.-D. Meyer, L. S. Cederbaum, All mode dynamics at the conical intersection of an octa-atomic molecule: Multi-configuration time-dependent hartree (mctdh) investigation on the butatriene cation, The Journal of Chemical Physics 115 (5) (2001) 2088–2100. arXiv:https://pubs.aip.org/aip/jcp/article-pdf/115/5/2088/19066005/2088_1_online.pdf, doi:10.1063/1.1384872. URL https://doi.org/10.1063/1.1384872
  107. [109]C. Fábri, G. J. Halász, L. S. Cederbaum, Á. Vibók, Impact of cavity on molecular ionization spectra, The Journal of Physical Chemistry Letters 15 (17) (2024) 4655–4661. doi:10.1021/acs.jpclett.4c00247. URL https://doi.org/10.1021/acs.jpclett.4c00247
  108. [110]F. Brogli, E. Heilbronner, E. Kloster-Jensen, A. Schmeltzer, A. Manocha, J. Pople, L. Radom, The photoelectron spectrum of butatriene, Chemical Physics 4 (1) (1974) 107–119. doi:https://doi.org/10.1016/0301-0104(74)80051-0. URL https://www.sciencedirect.com/science/article/pii/0301010474800510
  109. [111]S. Bhattacharyya, D. Opalka, L. V. Poluyanov, W. Domcke, Jahn-teller theory beyond the standard model, Journal of Physics: Conference Series 428 (2013) 012015. doi:10.1088/1742-6596/428/1/012015. URL http://dx.doi.org/10.1088/1742-6596/428/1/012015
  110. [112]G. Thiering, A. Gali, Spin–orbit coupling and jahn-teller effect in i_d symmetry: an ab initio study on the substitutional nickel defect in diamond, Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences 382 (2265) (Dec. 2023). doi:10.1098/rsta.2022.0310. URL http://dx.doi.org/10.1098/rsta.2022.0310
  111. [113]G. Herrero-Saboya, L. Martin-Samos, N. Richard, A. Hemeryck, Common defects in diamond lattices as instances of the general t₁ ⊗ (e + t₂) jahn-teller effect, Phys. Rev. Mater. 6 (2022) 034601. doi:10.1103/PhysRevMaterials.6.034601. URL https://link.aps.org/doi/10.1103/PhysRevMaterials.6.034601
  112. [114]J. C. A. Prentice, B. Monserrat, R. J. Needs, First-principles study of the dynamic jahn-teller distortion of the neutral vacancy in diamond, Phys. Rev. B 95 (2017) 014108. doi:10.1103/PhysRevB.95.014108. URL https://link.aps.org/doi/10.1103/PhysRevB.95.014108DOI
  113. [115]I. B. Bersuker, Pseudo-jahn-teller effect–a two-state paradigm in formation, deformation, and transformation of molecular systems and solids, Chemical Reviews 113 (3) (2013) 1351–1390. doi:10.1021/cr300279n. URL http://dx.doi.org/10.1021/cr300279n
  114. [116]J. Coutinho, V. J. B. Torres, K. Demmouche, S. Öberg, Theory of the carbon vacancy in 4h-sic: Crystal field and pseudo-jahn-teller effects, Phys. Rev. B 96 (2017) 174105. doi:10.1103/PhysRevB.96.174105. URL https://link.aps.org/doi/10.1103/PhysRevB.96.174105DOI
  115. [117]G. Thiering, A. Gali, Theory of the optical spin-polarization loop of the nitrogen-vacancy center in diamond, Phys. Rev. B 98 (2018) 085207. doi:10.1103/PhysRevB.98.085207. URL https://link.aps.org/doi/10.1103/PhysRevB.98.085207DOI
  116. [118]I. B. Bersuker, Jahn–teller and pseudo-jahn–teller effects: From particular features to general tools in exploring molecular and solid state properties, Chemical Reviews 121 (3) (2020) 1463–1512. doi:10.1021/acs.chemrev.0c00718. URL http://dx.doi.org/10.1021/acs.chemrev.0c00718
  117. [119]Q. C. Qiu, L. F. Chibotaru, A. Ceulemans, Product jahn–teller systems: The T₁ ⊗ H ⊗ (g + 2h) icosahedral exciton, Phys. Rev. B 65 (2001) 035104. doi:10.1103/PhysRevB.65.035104. URL https://link.aps.org/doi/10.1103/PhysRevB.65.035104DOI
  118. [120]G. Thiering, A. Gali, The (e_g ⊕ e_u) ⊗ e_g product jahn–teller effect in the neutral group-iv vacancy quantum bits in diamond, npj Computational Materials 5 (1) (Feb. 2019). doi:10.1038/s41524-019-0158-3. URL http://dx.doi.org/10.1038/s41524-019-0158-3
  119. [121]C. J. Ciccarino, J. Flick, I. B. Harris, M. E. Trusheim, D. R. Englund, P. Narang, Strong spin–orbit quenching via the product jahn–teller effect in neutral group iv qubits in diamond, npj Quantum Materials 5 (1) (Oct. 2020). doi:10.1038/s41535-020-00281-7. URL http://dx.doi.org/10.1038/s41535-020-00281-7
  120. [122]B. Tóth, A. Gali, G. Thiering, Exe.py: Ab initio fine structure parameters for trigonal defect qubits within the E ⊗ e Jahn-Teller case, HUN-REN Data Repository Platform (2025). doi:21.15109/ARP/EXJKGL. URL https://hdl.handle.net/21.15109/ARP/EXJKGLDOI
  121. [123]G. Thiering, A. Gali, Nuclear spin relaxation in solid state defect quantum bits via electron-phonon coupling in their optical excited state, HUN-REN Data Repository Platform (2025). doi:21.15109/ARP/04CEI6. URL https://hdl.handle.net/21.15109/ARP/04CEI6DOI

Paper details

vlnc:W_q8xyb3n3
Version
v2 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Writefull, ChatGPT

View statement
Looking for related papers…
Contents