Position-Dependent Epoxidation as a Strategy to Tune Optical Transitions in Nanographenes: Coronene Model Insights ^†

Abstract

This study investigated the influence of position-dependent edge epoxidation on the electronic and optical properties of nanographenes, using coronene as a model system. Quantum-chemical calculations were employed to examine how local functionalization alters the carbon framework and the nature of π-conjugation, thereby influencing the distribution of frontier molecular orbitals. It was established that the position of the epoxy group determines the nature of the changes within the π-system and the resulting optical response of the functionalized coronene. These structural and electronic modifications are accompanied by variations in the spectral characteristics and the transition probabilities of low-energy electronic transitions. The obtained results demonstrate the interplay between the functional group position, electronic structure, and optical properties of nanographenes, providing a foundation for the rational tuning of their optical response.

1. Introduction

The ability to precisely tune the electronic and optical properties makes nanographenes (NGs) highly promising for applications in optoelectronic and photosensitive materials [1,2]. However, despite their high thermodynamic stability, symmetric NGs exhibit limited optical and electronic properties. The high structural symmetry leads to the degeneracy of their frontier molecular orbitals (FMOs) [3], thereby rendering the low-lying electronic transitions symmetry-forbidden [4].

Controlled functionalization represents an effective approach for lowering the point group symmetry of NGs and tuning their optical properties [5,6]. Specifically, edge epoxidation changes the hybridization state of carbon atoms from sp² to sp³, disrupts the original π-conjugation of the nanographenes, and can significantly alter the electronic structure and spectral characteristics of the system [7,8].

At the same time, our recent study [9] demonstrated that symmetry breaking induced by epoxidation can either render the S₀ → S₁ transition allowed or leave it forbidden. However, the mechanism by which the position of the epoxy group influences the distribution of frontier molecular orbitals (FMOs) and the nature of quantum transitions remains poorly understood. Understanding this mechanism is a critical prerequisite for the controlled tuning of the electronic and optical properties of nanographenes via epoxidation.

The aim of the present study is to determine how the epoxidation of various edge configurations of nanographenes (zigzag and K-region) influences their electronic and optical properties. To achieve this goal, we compared two non-equivalent edge sites of coronene, the epoxidation of one of which lifts the symmetry restriction on the S₀ → S₁ transition, rendering it allowed. This comparison allows us to trace how the position of the epoxy group correlates with the redistribution of FMOs and the resulting changes in the optical response. Clar’s rule [10,11,12], which has proven highly effective in analyzing the electronic properties of nanographenes [9], was employed to analyze and interpret the obtained results.

2. Models and Methods

In the present study, coronene, a polycyclic aromatic hydrocarbon (PAH) with the chemical formula C₂₄H₁₂ and D_6h point group symmetry, was selected as the model structure. Owing to its well-characterized nature [13,14] and high symmetry, coronene serves as a convenient model system for investigating the impact of edge epoxidation on the electronic and optical properties of NGs. Furthermore, the high symmetry of coronene restricts the number of non-equivalent edge functionalization sites to two, thereby simplifying the comparative analysis. Unmodified coronene, along with its epoxidized forms, is shown in Figure 1.

Geometry optimizations of pristine coronene and its epoxidized isomers for the closed-shell singlet ground state (spin multiplicity = 1) and a total charge of zero, along with calculations of their FMOs, were performed using density functional theory (DFT) [15]. The hybrid B3LYP functional [16] and the def2-SVP basis set [17] were employed in the calculations. The choice of this functional is justified by its ability to accurately describe the electronic structure of π-conjugated systems while providing an optimal balance between accuracy and computational efficiency.

Dispersion interactions were accounted for using the DFT-D3(BJ) empirical correction [18]. Vibrational frequency calculations were performed to verify the absence of imaginary frequencies for both pristine and epoxidized coronene following geometry optimization. Transition states (TS) for the transformation of the epoxy group into an ether bridge were also calculated to assess the kinetic stability of the epoxidized coronene isomers. All calculations were performed using the ORCA 6.1 program system. An SCF convergence threshold of 1 × 10⁻⁸ Ha was used for all calculations. Geometry optimizations were performed with the following convergence criteria: energy change 5 × 10⁻⁶, maximum gradient 3 × 10⁻⁴.

The excited-state geometries of pristine and epoxidized coronene were obtained using single-reference time-dependent DFT (TD-DFT) at the B3LYP-D3(BJ)/def2-SVP level of theory. Absorption and emission wavelengths and oscillator strengths were computed using the density functional theory/multireference configuration interaction (DFT/MRCI) method [19]. This approach ensures high accuracy in calculating excited-state energies and transition dipole moments. In the DFT/MRCI calculations, we used a restricted active space (RAS)-like reference space of single and double excitations from the highest-lying occupied to the lowest-lying empty orbitals. In the active space 12 highest-lying π-electrons were distributed among 12 orbitals (6 occupied and 6 virtual). The configuration selection threshold of 10.9 eV was used in all calculations. For each structure 10 roots were calculated. Absorption wavelengths were calculated at the ground state S₀ geometry (vertical excitations), while fluorescence wavelengths were obtained using S₁ geometries.

3. Results and Discussion

The high symmetry of coronene imposes strict selection rules, rendering the S₀ → S₁ and S₀ → S₂ transitions forbidden, as evidenced by their zero oscillator strength. Consequently, the first allowed transition is S₀ → S₃, whose absorption band is located in the UV region (~305 nm) [20]. Symmetry constraints are also responsible for the zero oscillator strength of the reverse S₁ → S₀ transition, rendering it forbidden. These findings are consistent with experimental data demonstrating the low optical activity of pristine coronene in the visible spectral region [20,21,22].

Edge epoxidation of coronene enables symmetry breaking and lifts the forbiddenness of low-energy optical transitions. However, the change in hybridization state of the epoxidized carbon atoms can induce π-electron localization and disrupt π-conjugation within specific benzene rings. The extent to which the epoxy group influences the delocalized π-system largely depends on the position of functional group attachment. As demonstrated in our recent study [9], Clar’s rule is an effective tool for assessing the impact of epoxidation on the structural and electronic properties of NGs. Unmodified coronene and its epoxidized forms, with visualization of aromatic sextets, are presented in Figure 2. The positions of aromatic sextets are indicated by circles, and their possible migration pathways are shown by arrows.

As evident from Figure 2, epoxidation at the β-position induces a profound alteration of the electronic properties of coronene, characterized by extensive π-electron localization and the disruption of the aromatic sextet. Consequently, the epoxy group introduces severe local structural strain, rendering it a kinetically unstable intermediate in the functionalization process [23], as evidenced by the low activation barrier for its transformation into an ether bridge (≈0.1 kcal/mol).

The metastable nature of the epoxy group at the β-position renders its consideration of limited value. Instead, a more rational approach for the β-position is to investigate the impact of the ether bridge on the electronic and optical properties of coronene. Coronene modified at the β-position, both before and after the transformation of the epoxy group into an ether bridge, along with the visualization of aromatic sextets, is depicted in Figure 3. The positions of aromatic sextets are indicated by circles, and their possible migration pathways are shown by arrows. The vertical absorption and emission wavelengths, the configurational composition of the excited states, and the oscillator strengths for pristine coronene and its modified forms are summarized in

Table 1. The vertical absorption and emission wavelengths, the configurational composition of the excited states, and the oscillator strengths for pristine coronene and its modified forms.

Structure	Singlet Excited State	Vertical Excitation (nm)	Oscillator Strength, fabs	Transition Configurations with Dominant Orbital Contributions	Vertical Emission (nm)	Oscillator Strength, fem
Pristine coronene	S1	404 (375 *)	0.00	H → L (22%); H-1 → L + 1 (21%); H-1 → L (21%); H → L + 1 (21%)	431 (420 *)	0.00
	S2	344 (340 *)	0.00	H-1 → L (23%); H → L + 1 (23%); H → L (22%); H-1 → L + 1 (22%);	—	—
Epoxy-α	S1	390	0.00	H-1 → L (44%); H → L + 1 (39%)	414	0.00
	S2	373	0.35	H → L (85%)	—	—
Ether-β	S1	409	0.21	H → L (80%)	688	0.04
	S2	397	0.02	H-1 → L (39%); H → L + 1 (39%)	—	—

* Experimental values taken from Ref. [20].

.

The epoxy group at the α-position and the ether bridge at the β-position disrupt the planarity of coronene and lower its point group symmetry. As a result, the energies of the frontier and adjacent molecular orbitals (HOMO, LUMO, HOMO−1, LUMO+1) are altered, their degeneracy is lifted, and their spatial distributions are modified. Consequently, this induces a shift in the absorption and emission bands and lifts the symmetry restriction on the dipole transition corresponding to the p-band, to which the HOMO → LUMO configuration provides the main contribution. The molecular orbital energy levels of pristine coronene and its modified derivatives are summarized in

Table 2. Molecular orbital energy levels (in eV) of pristine coronene and its modified derivatives.

Structure	HOMO	LUMO	H/L Gap	HOMO−1	LUMO+1	H-1/L Gap	H/L + 1 Gap
Pristine coronene	−5.55	−1.55	4.00	−5.55	−1.55	4.00	4.00
Epoxy α	−5.44	−1.96	3.48	−6.01	−1.28	4.05	4.16
Ether β	−5.22	−1.90	3.32	−5.90	−1.20	4.00	4.02

. The spatial distributions of the molecular orbitals for pristine coronene and its α- and β-functionalized derivatives are depicted in Figure 4.

As evident from Table 2, functionalization at the α-position leads to a narrowing of the energy gap due to the symmetry lowering of coronene to C_S. Meanwhile, the epoxy group does not induce π-electron localization, and its effect is predominantly local: the epoxidized carbon atoms, owing to the change in their hybridization state (sp² → sp³), cease to participate in the global resonance system.

Meanwhile, the ether bridge at the β-position induces a moderate redistribution of the coronene molecular orbitals, along with a significant localization of the HOMO and HOMO−1 near the functionalization site. As evident from Figure 3, the transformation of the epoxy group into an ether bridge at the β-position alters the local structure of coronene. Specifically, C–C bond cleavage and the restoration of sp² hybridization at the carbon atoms lead to a reorganization of the local π-conjugation near the functionalization site. Furthermore, Figure 4 shows that the amplitudes of the HOMO and HOMO−1 are localized precisely within the newly formed π-conjugated fragment. Consequently, this results not only in the narrowing of the structure’s energy gap but also in the reordering of its α- and p-bands: the p-band now corresponds to the S₁ state and acquires a non-zero oscillator strength, rendering the S₀ → S₁ transition dipole-allowed.

This observation is particularly important because the non-zero oscillator strength of the S₀ → S₁ transition also renders the reverse S₁ → S₀ transition dipole-allowed (f = 0.04). It is important to note, however, that while a non-zero oscillator strength is a necessary condition for the structure to exhibit photoluminescence, it does not guarantee its occurrence. In particular, for efficient luminescence, one must also account for singlet–triplet interactions and the associated non-radiative relaxation channels, which, however, fall beyond the scope of the present study.

4. Conclusions

This study has demonstrated that the epoxidation of coronene affects its optical properties by modifying its delocalized π-system. Specifically, a redistribution of the molecular orbitals occurs, leading to a more pronounced spatial overlap of the FMO wavefunctions for the dipole transition. As a result, the p-band of coronene becomes associated with the S₁ state and acquires a non-zero oscillator strength, rendering the S₀ → S₁ transition dipole-allowed.

Furthermore, it was demonstrated that the redistribution of coronene’s molecular orbitals depends on the degree of structural perturbation: a greater disruption of the delocalized π-system results in a more pronounced overlap of the HOMO and LUMO wavefunctions, whereas a minor perturbation is accompanied by a less pronounced overlap.

The results obtained reveal underlying trends, the further investigation of which paves the way for the controlled tuning of the position and intensity of low-energy absorption and emission bands in symmetric NGs via epoxidation. Importantly, the conclusions drawn herein require further verification and validation on an extended set of model nanographenes of various sizes and point group symmetries.