Aromaticity recovery as a driving force for electron transfer and radical addition in 5-hydroxy-1,4-naphthoquinone derivatives

Abstract

This study introduces the concept of aromaticity recovery as a unified framework to rationalize electron transfer and radical reactivity in 5-hydroxy-1,4-naphthoquinone (HXNQ) derivatives. Density functional theory (M06-2X/6-311++G(d,p), SMD) was employed to investigate hydroxyl radical (HO˙) scavenging via electron transfer (ET), radical adduct formation (RAF), and hydrogen atom transfer (HAT) pathways. The results show that electron addition or radical attack promotes a more delocalized and partially aromatized electronic structure, providing an electronic driving force for the observed reactivity. Environmental polarity significantly stabilizes ionic intermediates, thereby favoring ET processes. Among the pathways examined, RAF at the C2 C3 bond is both kinetically and thermodynamically preferred, exhibiting near-barrierless activation and strong exothermicity. In contrast, HAT is hindered by a stabilizing intramolecular hydrogen bond that imposes conformational constraints. Aromaticity analyses based on HOMA and NICS descriptors, supported by MEP, HOMO–LUMO, and ELF calculations, reveal that the radical anion (HHNQ˙⁻) represents a unique electronic state in which partial aromaticity recovery occurs within the quinoid ring while the aromatic character of the benzenoid ring is largely preserved. This dual stabilization provides a structural basis for the favorable ET and RAF pathways. These findings establish aromaticity recovery as a useful framework for understanding and predicting radical-scavenging behavior in quinone systems.

Aromaticity recovery (HOMA and NICS) links electron transfer, radical addition, and antioxidant activity in HXNQ derivatives, providing a predictive framework for quinone radical-scavenging reactivity.

1. Introduction

Quinone derivatives constitute an important class of redox-active organic compounds widely used in medicinal chemistry and materials science. Among them, 5-hydroxy-1,4-naphthoquinone (juglone) and its derivatives exhibit diverse biological activities, primarily arising from the unique electronic properties of the 1,4-naphthoquinone core. Their reactivity is governed by electrophilic character and redox cycling, which are intrinsically linked to π-electron delocalization within the fused ring system.^1–5

At the molecular level, the biological activity of naphthoquinones is governed by two key mechanisms: (i) electrophilic reactivity toward nucleophiles (nucleophilic arylation), where the electron-deficient quinoid moiety acts as a reactive center, and (ii) intracellular redox cycling, leading to the generation of reactive oxygen species (ROS) and oxidative stress. These processes are intrinsically linked to the electronic distribution and conjugation within the molecular framework.^6–8

Structurally, 5-hydroxy-1,4-naphthoquinones (HXNQ) possess a cross-conjugated framework comprising an electron-deficient quinoid ring and an electron-rich benzenoid ring, giving rise to localized electronic strain and tunable reactivity. Substituents further modulate charge distribution and influence both electrophilic behavior and redox properties (Fig. 1).

Fig. 1. Chemical structures of 2-substituted 5-hydroxy-1,4-naphthoquinone (HXNQ) derivatives. A: quinoid ring and B: benzenoid ring.

Traditionally, understanding such reactivity requires a detailed exploration of potential energy surfaces (PES), which is computationally demanding. As an alternative, aromaticity-based descriptors have emerged as powerful and cost-efficient tools for probing electronic structure and reactivity. Aromaticity reflects the degree of π-electron delocalization and is associated with structural, energetic, and magnetic stabilization. Although widely applied to classical conjugated systems, this concept has also been extended to non-traditional cases such as quasi-aromatic systems, radicals, and charged species.^9–12

Recent experimental and theoretical studies have provided deeper insight into the structure–activity relationships of naphthoquinone derivatives.^13–16 Computational approaches, including density functional theory (DFT), have been widely employed to elucidate how electronic distribution, charge localization, and substituent effects influence reactivity and stability within the quinone framework. These studies consistently highlight the pivotal role of the quinoid π-system in controlling electrophilic behavior and redox properties, while also emphasizing how substituents modulate electron density across the fused ring system.^15,17

From a biological perspective, numerous investigations have demonstrated that the activity of HXNQ derivatives is closely associated with their ability to undergo redox cycling and interact with biomolecular nucleophiles.^15,18 The balance between electrophilicity and redox potential determines their effectiveness as antimicrobial and anticancer agents, as well as their selectivity toward biological targets. In addition, the presence of hydroxyl substituents significantly enhances biological performance by facilitating hydrogen bonding and modulating electron-donating effects within the aromatic framework.¹⁹

Particularly relevant is the antioxidant behavior of hydroxy-substituted naphthoquinones, which has been extensively studied through both experimental assays and theoretical modeling. These compounds can act as radical scavengers or pro-oxidant agents depending on their redox environment, with activity governed by the ease of electron or hydrogen atom transfer. The interplay between conjugation, aromaticity, and charge delocalization is therefore central to understanding their dual antioxidant/pro-oxidant nature.

Spectroscopic investigations have further contributed to the characterization of these systems. Techniques such as UV-vis, IR, NMR, and electron paramagnetic resonance (EPR) spectroscopy have been employed to probe electronic transitions, functional group interactions, and radical intermediates. In particular, UV-vis spectra provide insight into π–π* and n–π* transitions associated with the quinone core, while IR and NMR analyses reveal structural changes linked to conjugation and hydrogen bonding. Combined with theoretical calculations, these spectroscopic methods enable a comprehensive understanding of the electronic structure and charge-dependent behavior of HXNQ derivatives.²⁰

Within this context, quantitative descriptors of electron delocalization are essential for linking electronic structure to reactivity. Among various aromaticity indices, the Harmonic Oscillator Model of Aromaticity (HOMA)^10,11,21,22 provides a reliable, geometry-based measure of electron delocalization derived from bond length equalization. However, because aromaticity is a multidimensional concept and no single index can fully characterize aromatic behavior, aromaticity was assessed using complementary structural (HOMA) and magnetic (NICS) criteria. While HOMA is typically used for static analysis, its application alongside NICS in the present study enables a more comprehensive evaluation of aromatic character and its relationship to reactivity.^23–25

In this work, we apply the concept of “aromaticity recovery” as a fundamental driving force underlying electron transfer and radical reactions in substituted naphthoquinones. We propose that upon electron addition or radical attack, the quinoid system undergoes a transition from a localized, strained configuration toward a more delocalized and stabilized aromatic-like structure. This process provides a thermodynamic incentive for chemical transformation.

To investigate this hypothesis, we perform a systematic density functional theory (DFT) study on a series of 2-substituted 5-hydroxy-1,4-naphthoquinone derivatives (HXNQ), incorporating substituents spanning a wide electronic range from electron-donating to electron-withdrawing groups. Calculations are conducted at the M06-2X functional²⁶ in combination with the basis set of 6-311++G(d,p) along with the SMD solvation model²⁷ to capture both structural and environmental effects. Multiple electronic states, including neutral, radical, ionic, and deprotonated species, are considered to track the evolution of electronic structure.

By analyzing ring-specific HOMA indices alongside electronic structure descriptors,²⁸ we elucidate how substituent effects modulate aromaticity and influence preferred reaction pathways. Furthermore, solvent effects are examined to evaluate the role of intermolecular interactions in modulating intramolecular hydrogen bonding and reactivity.

Overall, this study establishes a quantitative relationship between aromaticity, electronic structure, and chemical reactivity, offering a predictive and mechanistic framework for the rational design of functionalized naphthoquinone systems in redox-active and bioactive applications.

2. Computational methods

All electronic structure calculations were performed using the Gaussian software package.²⁹ Density Functional Theory (DFT) calculations were carried out employing the M06-2X global hybrid functional,²⁶ well suited for non-covalent interactions, thermochemistry, and kinetics in organic systems. M06-2X was chosen because it has been extensively validated for thermochemistry, kinetics, electron-transfer processes, and radical reactions in main-group organic systems.³⁰ Since the present study focuses on aromaticity changes, electron-transfer mechanisms, and radical addition reactions of hydroxynaphthoquinone derivatives, M06-2X provides an appropriate balance between computational efficiency and predictive accuracy. The 6-311++G(d,p) triple-zeta basis set, augmented with both diffuse and polarization functions on all atoms, was utilized to ensure an accurate description of the electronic structures and hydrogen-bonding networks across all neutral, ionic, and radical forms of the HXNQ derivatives.

Optimized Cartesian coordinates are provided in the SI (Table S1). Harmonic frequency calculations were performed at the same level of theory to characterize all stationary points, with minima exhibiting no imaginary frequencies and transition states exactly one imaginary frequency. All energies include zero-point energy (ZPE) corrections.

Solvent effects were modeled using the SMD implicit solvation model. Two representative media were considered: non-polar pentyl ethanoate (to mimic hydrophobic/lipophilic environments) and water (to represent highly polar aqueous conditions).

To gain deeper insight into the electronic structure, topological analyses based on frontier orbitals³¹ and the electron localization function (ELF)³² were performed.

HOMA values were calculated using the standard Krygowski formula¹¹ for carbon–carbon bonds as given in the eqn (1).

1

where n = 6 (the number of bonds in each individual ring). α = 257.7 Å⁻² (the empirical normalization constant defined in the original HOMA formalism).¹¹R_opt = 1.388 Å (the optimized reference bond length for a perfectly aromatic, fully delocalized benzene system).¹¹R_i represents the individual calculated bond lengths of the corresponding ring. HOMA = 1.00 for ideal benzene and HOMA = 0.00 for a localized, non-aromatic Kekulé structure.

In parallel with HOMA, NICS(1) was employed as a complementary magnetic descriptor of aromaticity. Nucleus-independent chemical shift (NICS) values were calculated at a ghost atom position located 1 Å above the ring center, where negative values indicate aromatic character and positive values indicate antiaromatic character. In addition, NICS(1)_zz values were calculated and used as a complementary magnetic aromaticity descriptor to assess out-of-plane ring-current effects. By combining geometric (HOMA) and magnetic (NICS) criteria, a more comprehensive evaluation of substituent-induced changes in local aromaticity within the bicyclic framework was achieved.

The reactivity of HHNQ toward hydroxyl radicals (HO˙) was evaluated by analyzing the activation Gibbs free energies of the competing reaction pathways. The calculated barriers indicate that HO˙ preferentially reacts via radical addition to electron-rich sites rather than hydrogen abstraction. These reactivity trends are further supported by HOMA analysis, which shows that regions with lower aromatic character are more susceptible to HO˙ attack. Additionally, the relatively small changes in HOMA values along the preferred reaction pathway highlight the role of aromatic stabilization in governing the reaction kinetics.

Rate constants were calculated using conventional transition state theory (TST) at 298.15 K and 1 atm, in accordance with the standard Eyring–Polanyi equation³³ (Eqn (2)). The activation Gibbs free energy (ΔG^‡) includes the corresponding zero-point energy corrections:

2

where σ is the reaction symmetry number, κ represents the tunneling correction, k_B is the Boltzmann constant, h is the Planck constant, ΔG^‡ is the Gibbs free energy of activation. All rate constant calculations were performed using the Eyringpy package.^34–36

3. Results and discussions

3.1. Geometric structural response to C2-substituents

To elucidate the localized electronic perturbations induced by the functional groups introduced at the C2 position, the calculated bond lengths (Å) of both the quinoid framework (Ring A) and the benzenoid core (Ring B) of the HXNQ derivatives were systematically evaluated (X= H, F, Cl, CH₃, OCH₃, NH₂, CN, CF₃ and NO₂). The optimized geometric bond lengths (see Cartesian Coordinates in Table S1 of the SI) collected in Table 1 exhibit a highly significant, position-specific sensitivity to the electronic nature, namely the inductive (±I) and resonance (±R) behaviors, of the C2-substituents.

Table 1. Optimized bond lengths^a (in Å) of Ring A and B in HXNQ derivatives at M06-2X/6-311++G(d,p) level of theory.

Bond distance	H	F	Cl	CH3	OCH3	NH2	CN	CF3	NO2
Ring A
C1–C2	1.489	1.497	1.505	1.500	1.508	1.512	1.504	1.497	1.497
C2–C3	1.334	1.331	1.335	1.339	1.345	1.356	1.339	1.332	1.327
C3–C4	1.483	1.475	1.479	1.477	1.463	1.449	1.485	1.487	1.487
C4–C10	1.470	1.474	1.469	1.471	1.476	1.483	1.466	1.466	1.467
C10–C9	1.410	1.410	1.409	1.408	1.406	1.406	1.411	1.410	1.412
C9–C1	1.495	1.491	1.494	1.494	1.492	1.483	1.491	1.493	1.492
Ring B
C5–C6	1.403	1.404	1.403	1.402	1.403	1.404	1.403	1.403	1.404
C6–C7	1.380	1.380	1.380	1.381	1.381	1.381	1.379	1.380	1.379
C7–C8	1.399	1.399	1.399	1.398	1.398	1.397	1.400	1.400	1.400
C8–C9	1.382	1.381	1.382	1.382	1.383	1.384	1.381	1.381	1.380
C9–C10	1.410	1.410	1.409	1.408	1.406	1.406	1.411	1.410	1.412
C10–C5	1.407	1.406	1.406	1.406	1.404	1.404	1.407	1.407	1.407

^aNote: the values are rounded off to three digits. The full numerical values are provided in Table S2.

3.1.1. Structural tuning of the quinoid domain (Ring A)

Ring A displays substantial bond length alterations when transitioning from strong electron-donating groups (EDGs) to powerful electron-withdrawing groups (EWGs). The localized endocyclic double bond C2 C3, which possesses a baseline length of 1.334 Å in the unsubstituted reference (X = H), undergoes significant modifications, as follows:

Under the influence of EDGs (NH₂, OCH₃, CH₃), a substantial elongation of the C2 C3 bond is observed, reaching its maximum at 1.356 Å for the –NH₂ derivative. Concurrently, the adjacent nominal C3–C4 single bond contracts sharply from 1.483 Å (H) down to 1.449 Å (–NH₂). This systematic bond equalization, where the double bond lengthens and the single bond shortens, strongly indicates an efficient π-electron delocalization via the positive resonance effect (+R) of the lone pairs on the substituent. This electronic push diminishes the rigid quinoid character of Ring A, shifting its geometry closer to a conjugated, semi-aromatic state.

Conversely, under the influence of EWGs (–CN, –CF₃, –NO₂): strong EWGs enforce the opposite geometric trend. For instance, the highly electronegative NO₂ group induces a severe contraction of the C2 C3 double bond to 1.327 Å while keeping the C3–C4 bond relatively stretched (1.487 Å). This outcome reinforces the localized, classical α,β-unsaturated carbonyl nature of the quinoid ring, locking the π-electrons into distinct single and double bonds.

Interestingly, the halogen substituents (–F and –Cl) showcase an expected ambivalent structural response; the C2 C3 bond in the F derivative contracts slightly (1.331 Å) due to dominant strong inductive pulling (−I), whereas the Cl derivative expands slightly (1.335 Å), indicating a minor local lone-pair resonance intervention (+R) on the carbon framework.

3.1.2. Structural rigidity of the benzenoid domain (Ring B)

In stark contrast to the highly responsive quinoid ring, the bond distances within the phenolic benzenoid core (Ring B, positions C5 to C10) exhibit notable structural resilience. Across the entire substituent spectrum, ranging from the strongly donating –NH₂ to the intensely withdrawing –NO₂, the carbon–carbon bond lengths fluctuate within an incredibly narrow margin (typically < 0.005 Å). For example, the C6–C7 bond remains nearly locked between 1.3789 Å and 1.38086 Å. The bridging C9–C10 bond, shared between both domains, settles consistently around 1.405 Å–1.412 Å. This exceptional rigidity demonstrates that the intrinsic aromaticity of Ring B acts as a thermodynamic sink, effectively absorbing electronic stress and isolating its benzenoid π-system from direct structural disruption caused by modifications at the distant C2 site.

Under the influence of strong EDGs, the baseline single-double Bond Length Alternation (BLA) of the diene–dione system undergoes a substantial compression, pulling the localized carbon matrix into a more uniform, pre-relaxed configuration. Conversely, EWGs induce a severe pinching effect, sharpening the structural disparity between the single and double bonds to firmly lock the quinoid framework into its classical, highly strained state. This intricate coordination between inductive/resonance fields and skeletal topography establishes the physical foundation required to transition from raw structural coordinates to standardized aromaticity indices.

3.2. Local aromaticity analysis via the HOMA index and NICS(1) values

3.2.1. Aromaticity of HHNQ across different electronic states

The HOMA index was employed as one of a quantitative descriptor of aromaticity based on bond-length equalization within the ring. Higher HOMA values correspond to a greater degree of aromatic character, whereas lower values indicate diminished aromaticity and increased structural distortion. In the present study, HOMA is particularly valuable because it enables direct monitoring of aromaticity changes during electron-transfer and radical-addition processes. The increase in HOMA values observed upon reaction provides quantitative evidence for the recovery of aromatic character, supporting the hypothesis that restoration of aromatic stabilization contributes to the thermodynamic driving force of these transformations. Nevertheless, aromaticity is widely recognized as a multidimensional phenomenon that cannot be fully described by a single index. Therefore, in accordance with current recommendations, aromaticity was evaluated using complementary structural (HOMA) and magnetic (NICS, nucleus-independent chemical shift) criteria, providing a more comprehensive assessment of electron delocalization and aromatic behavior.²⁵

The HOMA index was calculated according to eqn (1) using optimized geometric bond lengths obtained at the M06-2X/6-311++G(d,p) level of theory. Based on the bond-length data summarized in Table 1 (as well Table S1), the local aromaticity of the unsubstituted naphthoquinone core (X = H) was evaluated across five electronic configurations and charge states (Table 2) using complementary HOMA and NICS descriptors. Both analyses reveal a substantial electronic-state-dependent modulation of aromaticity, characterized by a clear differentiation between the quinoid Ring A and the benzenoid Ring B. In the ground-state neutral species (HHNQ), a distinct split-domain structure is observed, with Ring B exhibiting strong aromatic delocalization (HOMA = +0.853; NICS(1) = −9.787 ppm) while Ring A remains distinctly quinoidal and nonaromatic (HOMA = −0.450; NICS(1) = +1.314 ppm). Upon changes in charge state, substantial variations in aromaticity occur. Notably, the radical anion HHNQ˙⁻ displays a partial recovery of aromatic character in Ring A, evidenced by bond equalization within the quinoid framework, including elongation of the C2 C3 bond from 1.334 Å to 1.374 Å (Table 1), resulting in a positive HOMA value (+0.125) and a more aromatic NICS value (−4.672 ppm). Simultaneously, Ring B retains substantial aromaticity (HOMA = +0.781), indicating that electron capture effectively relieves quinoid strain while preserving benzenoid delocalization.

Table 2. Local HOMA indices^a and NICS(1) values^b for Ring A and Ring B across different electronic states.

Domain	Electronic states
	HHNQ	HHNQ˙+	HHNQ˙−	HNQ−	HNQ˙
Ring A (quinoid)	−0.450 (1.314 ppm)	−0.413 (0.917 ppm)	+0.125 (−4.672 ppm)	−0.476 (1.021 ppm)	−0.490 (1.995 ppm)
Ring B (benzenoid)	+0.853 (−9.787 ppm)	+0.362 (1.827 ppm)	+0.781 (−9.989 ppm)	−0.048 (−5.115 ppm)	+0.140 (−0.253 ppm)
Aromaticity state	Split domain	Disrupted split	Aromatic co-recovery	Severe dearomatization	Severely weakened

^aUpper values are HOMA indices, and.

^bLower values in parentheses are NICS(1) values. Note: the full numerical values for isotropic (σ_iso), anisotropic (Δσ_iso), and out-of-plane tensor (σ_zz) parameters are provided in Table S3.

In contrast, ionization (HHNQ˙⁺) and neutral radical formation (HNQ˙) do not restore the aromaticity of Ring A, which retains negative HOMA values (−0.413 and −0.490, respectively) and positive NICS(1) values (+0.917 and +1.995 ppm). Instead, both processes substantially diminish the aromatic character of Ring B, with HOMA values decreasing from +0.853 in HHNQ to +0.362 and +0.140, accompanied by NICS(1) values shifting from −9.787 ppm to +1.827 ppm and −0.253 ppm, respectively, indicating severe disruption of benzenoid electron delocalization. The deprotonated species (HNQ⁻) exhibits the most significant structural dearomatization, particularly in Ring B, where the HOMA value decreases to −0.048, while Ring A also remains nonaromatic (HOMA = −0.476). Although the corresponding NICS(1) value for Ring B (−5.115 ppm) still indicates some residual diatropic character, the combined HOMA and NICS data reveal substantial disruption of electron delocalization throughout the bicyclic framework. Notably, the corresponding NICS(1)_zz values (Table S3) follow the same qualitative trends as the isotropic NICS(1) values, providing additional magnetic evidence for the aromaticity changes discussed above.

Overall, the HOMA and NICS results consistently demonstrate an evolution from a split aromatic domain in HHNQ, through partial aromaticity recovery in the radical anion HHNQ˙⁻, to progressively dearomatized electronic states. These findings suggest that chemical addition reactions at the C2 C3 double bond are most likely to be kinetically and thermodynamically favored through pathways involving the radical anion intermediate, which uniquely benefits from the simultaneous partial restoration of aromaticity in Ring A and preservation of aromaticity in Ring B.

3.2.2. Electronic structure analysis via MEP, HOMO–LUMO, and ELF

The electronic structure of HHNQ is strongly influenced by its charge state, as supported by HOMA, MEP, HOMO–LUMO, and ELF analyses (Fig. 2–4). In the neutral form, HHNQ displays a split-domain electronic structure, with an aromatic benzenoid Ring B (HOMA = +0.853) and a localized quinoid Ring A (HOMA = −0.450). The MEP, frontier orbital, and ELF distributions consistently reflect this differentiation, revealing electrophilic character around the carbonyl/C2 C3 region and pronounced bond localization within the quinoid framework.

Fig. 2. Molecular electrostatic potential (MEP) maps of HHNQ across five electronic states: (a) neutral, (b) radical cation, (c) radical anion, (d) deprotonated anion, and (e) neutral radical.

Fig. 3. Frontier molecular orbital (HOMO–LUMO) distributions of HHNQ in different electronic states, illustrating charge-dependent orbital localization and reactivity. (a) HHNQ, (b) HHNQ˙⁺, (c) HHNQ˙⁻, (d) HNQ⁻, and (e) HNQ˙ radical.

Fig. 4. Electron localization function (ELF) maps of HHNQ across different charge states, highlighting changes in π-electron delocalization and aromaticity. (a) HHNQ, (b) HHNQ˙⁺, (c) HHNQ˙⁻, (d) HNQ⁻, and (e) HNQ˙ radical.

Ionization to HHNQ˙⁺ reduces aromaticity (HOMA = +0.362) and leads to increased electrostatic polarization and fragmentation of orbital (Fig. 2b) and ELF distributions (Fig. 4b), indicating weakened π-conjugation (Fig. 3b). In contrast, electron addition produces a distinct stabilization pathway: in HHNQ˙⁻, partial delocalization of the quinoid ring (HOMA = +0.125) occurs while aromaticity is largely preserved in the benzenoid ring (HOMA = +0.781). This state is characterized by a more homogeneous MEP surface, HOMO delocalization over the quinoid framework (Fig. 2c and 3c), and ELF basin merging (Fig. 4c), evidencing suppression of bond alternation and enhanced π-electron continuity. By comparison, deprotonation (HNQ⁻) leads to severe electronic localization (Fig. 2d), near loss of aromaticity (HOMA = −0.048), and confined orbital and ELF features (Fig. 3d and 4d), while the neutral radical (HNQ˙) displays incomplete delocalization and weakened aromaticity (see Fig. 2e, 3e and 4e).

Collectively, these results demonstrate that reactivity is governed by the ability to achieve quinoid delocalization while preserving benzenoid aromaticity, a condition uniquely satisfied by the radical anion through an aromaticity co-recovery mechanism.

The electron localization function (ELF) provides a real-space description of electron pairing and delocalization, enabling direct visualization of bonding patterns and π-electron distribution across the HHNQ framework. As shown in Fig. 4, ELF values range from 0 to 1, where values close to 1 indicate highly localized electrons (e.g., lone pairs and σ bonds), while values near 0.5 represent delocalized electron density typical of aromatic systems. In the neutral HHNQ state (Fig. 4a), ELF analysis reveals strong localization at the carbonyl oxygen atoms and pronounced bond alternation within the quinoid Ring A, evidenced by alternating high and low ELF regions, while Ring B displays a continuous ELF domain consistent with aromatic π-delocalization. Upon formation of the radical cation (HHNQ˙⁺) (Fig. 4b), ELF basins become fragmented and less continuous, particularly over the benzenoid ring, indicating disruption of π-conjugation and reduced aromatic stabilization. In contrast, the radical anion (HHNQ˙⁻) in Fig. 4c shows a marked redistribution of electron density, where ELF basins in Ring A become more merged and homogeneous, reflecting suppression of bond alternation and increased π-delocalization, while Ring B largely retains its continuous aromatic basin. This behavior provides clear real-space evidence of aromaticity co-recovery, where both rings approach a more delocalized electronic structure. For the deprotonated anion (HHNQ⁻), ELF indicates strong charge localization around the oxygen center, forming dominant lone-pair basins and disrupting π-electron continuity across the rings, consistent with severe loss of aromaticity (Fig. 4d). Similarly, in Fig. 4e the neutral radical (HHNQ˙) exhibits incomplete basin connectivity and mixed localized–delocalized features, reflecting partial electron delocalization and reduced electronic stability. Overall, ELF analysis demonstrates that the key factor governing stability and reactivity in HHNQ systems is the ability to transform localized quinoid electron density into a continuous delocalized π-network while preserving benzenoid aromaticity, a condition uniquely achieved in the radical anion state.

3.2.3. Influence of C2-substituents on ground-state local aromaticity

To analyze the effect of C2-substituents on the local aromaticity of 5-hydroxy-1,4-naphthoquinone derivatives, geometric bond lengths (Table 1) were evaluated using the standard HOMA formalism. The substituents were systematically arranged along an electronic gradient from strong electron-donating groups (EDGs) to halogens and strong electron-withdrawing groups (EWGs), enabling a clear interpretation of structural trends.

Building upon these baseline trajectories, Table 3 and Fig. 5 illustrate the broader derivative series, explicitly capturing how the structural configurations of these electronic states are modulated by the inductive and resonance fields of electron-donating (EDG) and electron-withdrawing (EWG) groups as a function of C2-functionalization.

Table 3. Local HOMA index and NICS(1) profiles of neutral derivatives along substituents.

Electronic nature	Substituent	HOMA index		HOMA gap (ΔHOMA)a	NICS(1) (ppm)
		Ring A (quinoid)	Ring B (benzenoid)		Ring A (quinoid)	Ring B (benzenoid)
Strong EDG	–NH2	−0.138	+0.852	0.990	1.451	−9.189
Moderate EDG	–OCH3	−0.264	+0.851	1.115	1.063	−9.300
Weak EDG	–CH3	−0.383	+0.854	1.237	1.476	−9.283
Reference	–H	−0.45	+0.853	1.303	1.314	−9.787
Halogen	–F	−0.449	+0.854	1.303	1.013	−9.292
	–Cl	−0.463	+0.850	1.313	1.231	−9.289
Strong EWG	–CF3	−0.505	+0.853	1.358	1.152	−9.294
	–CN	−0.528	+0.856	1.384	1.076	−9.312
Very strong EWG	–NO2	−0.579	+0.855	1.434	0.825	−9.215

^aΔHOMA = HOMA(B) − HOMA(A). Note: the full numerical values are provided in Table S4.

Fig. 5. Linear regression mapping of static HOMA values for Ring A and ΔHOMA across the nine C2-substituents.

The calculated HOMA values (Table 3) reveal a significant asymmetric response between the two rings of the bicyclic system. The phenolic Ring B remains remarkably stable, exhibiting consistently high aromaticity (HOMA ≈ +0.853 ± 0.003) across all substituents. This invariance indicates that the C5–OH group and the structural separation from the C2 position effectively preserve its π-electron delocalization.

In contrast, the quinoid Ring A displays strong sensitivity to substituent effects, with its HOMA value varying significantly by 0.441 units (from −0.138 to −0.579). Electron-donating groups promote bond equalization within the quinoid framework through positive resonance (+R) interactions. For example, the –NH₂ substituent induces elongation of the C2 C3 bond (from 1.334 Å to 1.356 Å) and contraction of the adjacent C3–C4 bond (from 1.483 Å to 1.449 Å), leading to an increase in the HOMA value from −0.450 to −0.138. This behavior reflects partial relief of the rigid diene–dione structure and enhanced π-delocalization.

Conversely, electron-withdrawing groups reinforce bond localization by depleting electron density from the conjugated system. In the case of –NO₂, the C2 C3 bond shortens to 1.327 Å, stabilizing the classical quinoid structure and driving the HOMA value to a minimum of −0.579. Halogen substituents exhibit intermediate behavior due to competing inductive (−I) and resonance (+R) effects. Fluorine shows a nearly neutral influence (HOMA ≈ −0.449), whereas chlorine slightly enhances localization (HOMA ≈ −0.463), reflecting a modest dominance of its inductive effect.

Overall, the HOMA results are fully consistent with the bond-length variations discussed in Section 3.1, indicating that the substituent-dependent structural changes are directly reflected in the aromatic properties of the two rings. To complement this geometric analysis, NICS(1) calculations were performed as an independent magnetic measure of aromaticity.

Nucleus-independent chemical shift (NICS) calculations were further performed as a magnetic criterion of aromaticity to validate the geometric trends derived from the HOMA indices. As shown in Table 3, the quinoid Ring A exhibits a pronounced substituent dependence. Electron-donating groups (EDGs), particularly –NH₂ and –CH₃, enhance π-electron delocalization within the quinoid framework, giving rise to NICS(1) values of 1.451 and 1.476 ppm, respectively. In contrast, strong electron-withdrawing groups (EWGs), such as –CN and –NO₂, produce lower NICS(1) values of 1.076 and 0.825 ppm, respectively, consistent with the corresponding HOMA trends. Halogen substituents show only minor effects, with the F- and Cl-substituted systems displaying nearly identical HOMA and NICS(1) values, indicating that Ring A largely retains its quinoidal character.

By contrast, the benzenoid Ring B remains remarkably insensitive to substituent effects. Its NICS(1) values vary only slightly, from −9.189 to −9.312 ppm (excluding the parent compound, −9.787 ppm), consistent with the nearly invariant HOMA values (≈+0.85). The halogen derivatives further illustrate this stability, exhibiting almost identical aromaticity descriptors for both HOMA and NICS(1). These results confirm that Ring B serves as a robust aromatic domain that is largely insulated from electronic perturbations transmitted from the C2 substituent.

The aromaticity assignments based on isotropic NICS(1) values are further supported by the corresponding NICS(1)_zz data (Table S4), which exhibit the same qualitative behavior across the entire substituent series. Ring A consistently shows positive NICS(1)_zz values (+5.997 to +8.657 ppm), confirming its predominantly quinoidal nature, whereas Ring B displays strongly negative NICS(1)_zz values (−22.414 to −23.534 ppm), indicative of persistent benzenoid aromaticity. The excellent agreement among HOMA, NICS(1), and NICS(1)_zz demonstrates that substituent effects primarily modulate the degree of quinoid localization in Ring A while leaving the aromatic character of Ring B essentially unchanged.

Although the substituent (X) is introduced at a position adjacent to the C1 O group to maximize electronic interaction with the quinoid system, Hammett σ_p constants are employed as a standardized and transferable measure of intrinsic substituent electronic effects. Unlike ortho parameters, σ_p values are largely free from steric and proximity influences, enabling a more reliable quantitative correlation between electronic effects and HOMA-derived aromaticity.

To elucidate this relationship, HOMA indices were correlated with σ_p values (detailed in Table S5 of the SI) as shown in Fig. 5. The regression analysis reveals that substituent effects are highly localized and unevenly distributed between the two rings, and are most effectively interpreted through Ring A and the HOMA gap (ΔHOMA), rather than Ring B. The benzenoid Ring B remains essentially invariant, with HOMA values narrowly clustered between +0.850 and +0.856, showing no meaningful dependence on σ_p. This invariance confirms that Ring B is electronically insulated from substituent perturbations and serves as a stable aromatic reference.

In contrast, the quinoid Ring A exhibits a strong linear dependence on substituent electronic properties, following the relationship HOMA(A) = −0.276σ_p − 0.382 (R² = 0.884). Electron-withdrawing groups (σ_p > 0) drive HOMA values toward more negative values (≈−0.6), indicating enhanced bond localization and stabilization of the quinoid framework, whereas electron-donating groups (σ_p < 0) promote bond equalization and increase HOMA values toward −0.1, reflecting partial recovery of π-delocalization.

The most informative descriptor is therefore the HOMA gap (ΔHOMA = Ring B − Ring A), which shows a strong positive correlation with σ_p (ΔHOMA = 0.278σ_p + 1.235, R² = 0.885). This trend reflects progressive electronic decoupling between the invariant aromatic Ring B and the increasingly localized quinoid Ring A as substituent withdrawing strength increases. Mechanistically, electron-donating substituents reduce ΔHOMA by enhancing delocalization within Ring A, thereby lowering structural strain and pre-activating the system for radical addition. In contrast, electron-withdrawing substituents increase ΔHOMA, reinforcing quinoid localization and stabilizing the ground state.

3.2.4. Environmental perturbations: solvation effects on local aromaticity

To evaluate the influence of external condensed environments on the electronic structure of the unsubstituted 5-hydroxy-1,4-naphthoquinone (HHNQ) system, HOMA indices were calculated using geometries optimized in the gas phase (HHNQ-G), pentyl ethanoate (HHNQ-P), and water (HHNQ-W) (Table 4).

Table 4. Local HOMA indices across solvation media of increasing polarity.

Domain	Gas phase (ε = 1.0)	Pentyl ethanoate (ε = 4.8)	Water (ε = 78.4)
	HHNQ-G	HHNQ-P	HHNQ-W
Ring A (quinoid)	−0.450	−0.467	−0.354
Ring B (benzenoid)	+0.853	+0.843	+0.858
Total system aromaticity gap (Δ)	1.303	1.310	1.212

Water and pentyl ethanoate were selected to represent highly polar and weakly polar environments, respectively. These solvents provide chemically distinct dielectric media that allow evaluation of the sensitivity of electron-transfer and radical-addition processes to solvent polarity. While the SMD continuum model captures bulk electrostatic solvation effects, it does not explicitly account for specific solvent–solute interactions such as hydrogen bonding. Therefore, the observed trends should be interpreted primarily in terms of dielectric stabilization rather than detailed solvation structures.

The results show that solvent effects do not produce a simple linear trend in aromaticity but instead reflect a balance between dielectric stabilization and disruption of the intramolecular C5O–H⋯OC4 hydrogen bond (hereafter denoted as O5–H⋯O4). In a moderately polar medium such as pentyl ethanoate (ε = 4.8), the molecular dipole is stabilized while the internal hydrogen bond remains largely intact. As a result, the quinoid Ring A becomes slightly more localized (HOMA = −0.467), while Ring B shows a small decrease in aromaticity (HOMA = +0.843). This leads to a marginal increase in the aromaticity gap (Δ = 1.310), indicating a more structurally confined quinoid framework.

In contrast, in a highly polar protic solvent such as water (ε = 78.4), intermolecular hydrogen bonding competes with and disrupts the internal O5–H⋯O4 interaction. This effect reduces the rigidity of the quinoid system and promotes bond equalization in Ring A, as reflected by an increase in its HOMA value (−0.354). At the same time, Ring B exhibits a slight enhancement of aromaticity (HOMA = +0.858), while the overall aromaticity gap decreases to 1.212.

These results suggest that polar protic environments facilitate structural relaxation and increased π-delocalization within the quinoid domain. Consequently, such environments are expected to lower activation barriers for subsequent reactions, including electrophilic protonation and radical processes, by pre-activating the molecular framework.

3.3. Divergence: the interplay of ET, RAF, and HAT competitions

The capacity of 5-hydroxy-1,4-naphthoquinone derivatives to act as biological redox switches or antioxidant shields relies on a fierce kinetic and thermodynamic competition among three primary chemical pathways: Electron Transfer (ET), Radical Adduct Formation (RAF), and Hydrogen Atom Transfer (HAT).^37–39 In this asymmetric bicyclic framework, ET and RAF share the mechanistic crown, as both pathways directly exploit the structural vulnerabilities and latent energetic drives of the quinoid domain, while the traditionally dominant phenolic HAT pathway is heavily penalized by an intramolecular hydrogen lock of the molecule. To clarify these competitive behaviors, the three distinct mechanistic channels between HXNQ and the HO˙ radical are illustrated in Fig. 6.

Fig. 6. Schematic representation of the competing radical-trapping mechanisms between HXNQ and the HO˙, highlighting Single Electron Transfer (SET), Hydrogen Atom Transfer (HAT), and Radical Adduct Formation (RAF) at the C2 and C3 active sites.

In this Section, we systematically evaluate the thermodynamic parameters of the unsubstituted core (X = H) in the three distinct media, followed by the construction of explicit Potential Energy Surfaces (PES) to map the corresponding kinetic barriers. Our energetic landscapes, compiled in Table 5, confirm that covalent radical addition (RAF) at the C2 C3 bond is highly favored. In parallel, local HOMA geometric indices were tracked across all reaction channels to establish a quantitative bridge between static ground-state aromaticity and dynamic chemical reactivity.

Table 5. Computed adiabatic energetics (kcal mol⁻¹) of the HXNQ (X = H) across different media at M06-2X/6-311++G(d,p).

Phase	IE	EA	DPE	BDE
Gas	203.8	52.8	338.0	102.5
Pentyl ethanoate	163.5	82.9	304.9	99.5
Water	152.4	80.4	285.2	98.7

3.3.1. Intrinsic energetic profiles of the unsubstituted scaffold (X = H)

To comprehensively map the competitive landscape between the SET, HAT, and RAF pathways, the intrinsic energetic properties of the unsubstituted scaffold (X = H) were first evaluated across three representative environments: gas phase, pentyl ethanoate, and water. The corresponding mechanistic pathways are illustrated in Fig. 6, while the computed energetic parameters are summarized in Table 5.

In the gas phase, the molecule exhibits strong resistance to both homolytic and heterolytic processes, as reflected by high ionization energy (IE = 203.8 kcal mol⁻¹) and deprotonation enthalpy (DPE = 338.0 kcal mol⁻¹). This behavior arises from the rigid O5–H⋯O4 intramolecular hydrogen bond, which stabilizes the localized quinoid framework and imposes a significant energetic penalty on bond cleavage.

Transitioning to condensed media substantially alters this energetic landscape. In both pentyl ethanoate and water, a significant decrease in IE (down to 152.4 kcal mol⁻¹ in water) and increase in electron affinity (EA > 80 kcal mol⁻¹) are observed, indicating strong solvent stabilization of ionic species. This effect is especially significant in water, where competitive intermolecular hydrogen bonding disrupts the internal O5–H⋯O4 interaction, leading to structural relaxation and reduced rigidity of the quinoid system.

As a result, the deprotonation enthalpy (DPE) decreases significantly from 338.0 kcal mol⁻¹ in the gas phase to 285.2 kcal mol⁻¹ in water, highlighting a substantial lowering of the energetic barrier for heterolytic processes. In contrast, the bond dissociation enthalpy (BDE) remains relatively constant across environments (∼98–102 kcal mol⁻¹), indicating that the hydrogen atom transfer (HAT) pathway is largely insensitive to solvent polarity.

Overall, these results demonstrate that increasing environmental polarity selectively stabilizes ionic pathways and promotes single electron transfer (SET) and deprotonation processes, while leaving the homolytic HAT pathway largely unaffected. This solvent-induced pre-activation establishes a thermodynamic bias toward ionic and radical-driven mechanisms even before substituent effects are introduced.

3.3.2. The electron transfer (ET) pathway: charge-driven feasibility

The outer-sphere ET pathway initiates via a single-electron reduction, transforming the neutral naphthoquinone into a radical anion intermediate (HXNQ˙⁻). This transition represents the cleanest electronic channel available to the system. The incoming electron populates the Lowest Unoccupied Molecular Orbital (LUMO), which is localized cleanly over the quinoid domain (Ring A). This injection of negative charge breaks the rigid carbonyl localization (C1 O and C4 O) and equalizes the highly localized C2 C3 double bond. According to Marcus Theory, the rate of this transfer (k_ET) is governed by the inner-sphere reorganization energy (λ_int), or the energetic cost of distorting the ground-state nuclear geometry to accommodate the new charge configuration. Because LUMO occupation suppresses bond localization within the quinoid framework, Ring A undergoes partial recovery of aromatic character, with its HOMA value increasing from approximately −0.450 to +0.125. At the same time, the benzenoid Ring B remains largely aromatic (HOMA ≈ +0.781), indicating that the added electron relieves quinoid strain while preserving benzenoid delocalization.

3.3.3. The radical adduct formation (RAF) pathway: kinetic regioselectivity at C2 C3

In parallel, the RAF pathway represents the cleanest covalent channel for radical mitigation, involving a homolytic attack by an external radical species (such as HO˙, for example) directly onto the active carbon scaffold of Ring A: HHNQ + HO˙ → (HHNQ⋯OH)˙ (adduct radical intermediate). This mechanism forces a profound hybridization shift, converting either the C2 or C3 carbon atom from a planar, conjugated sp² state to a tetrahedral sp³ configuration. While breaking conjugation normally incurs a steep thermodynamic penalty in linearly conjugated molecules, in the HNQ system, it provides the exact mechanical relief Ring A needs to snap into an aromatic conformation. The localized diene strain is broken, and the remaining unhybridized carbon atoms in the ring spontaneously pool their electrons to form a stable, benzenoid-like arrangement. The rate of reaction and regioselectivity of this addition are highly customizable via the C2-substituents; for example, electron-withdrawing groups (EWGs) strongly polarize the C2 C3 π-cloud, lowering the activation barrier (ΔE^‡) specifically for a C3 attack to yield clean, predictable intermediate stability.

To unravel the exact kinetic profile of the covalent trapping channel, explicit Potential Energy Surfaces (PES) were mapped for the regioselective addition of the HO˙ radical at both the C2 and C3 positions of the quinoid ring as well as well as for the subsequent hydrogen abstraction following the cis–trans rotameric isomerization of HHNQ (Fig. 7).

Fig. 7. Potential energy surfaces (PES) for the competitive pathways of the unsubstituted core (HHNQ) against the HO˙ radical at the M06-2X/6-311++G(d,p) level of theory: (left) low-barrier regioselective radical adduct formation (RAF) at C2 and C3 sites; (right) sequential hydrogen atom transfer (HAT) hindered by the structural cis–trans Srotamer lock. Relative energies are reported in kcal mol⁻¹.

The explicit potential energy surface (PES) for the covalent trapping of the HO˙ radical within the unsubstituted core (HHNQ) reveals an exceptionally fluid kinetic profile for the RAF channel. The reaction initiates with the formation of a pre-reactive van der Waals complex (RC_RAF), which sits in a shallow thermodynamic well below the reactant baseline. From this state, the addition of HO at either the C2 or C3 position of the quinoid domain (Ring A) proceeds through very low, near-barrierless activation barriers. Specifically, the activation energies (ΔE^‡) for TS_RAF-C2 range between ∼0.1 and 1.0 kcal mol⁻¹, while TS_RAF-C3 remains tightly constrained between 0.2 and 1.1 kcal mol⁻¹ across gas (G), pentyl ethanoate (P), and water (W) phases (Table 6).

Table 6. Activation energies (ΔE^‡) and reaction free energies (ΔG_rxn) in kcal mol⁻¹ for the competing pathways of HXNQ against the HO˙ radical.

Media	RAF				HAT following cis–trans rotamer
	ΔE‡RAF-C2	ΔGRAF-C2	ΔE‡RAF-C3	ΔGRAF-C3	ΔE‡cis–trans	ΔE‡HAT	ΔGHAT
Gas	0.05	−26.15	0.15	−25.01	13.12	9.58	−15.75
Pentyl ethanoate	1.00	−24.36	1.14	−23.90	11.5	6.96	−22.70
Water	0.19	−24.75	0.54	−23.39	8.29	6.10	−23.68

This low activation barrier can be understood from the perspective of ground-state local aromaticity. Structural pre-activation arises from the substantial quinoid character of Ring A, reflected in its strongly negative HOMA value and significant bond-length alternation. As a result, the localized C2 C3 double bond is predisposed toward radical addition. In addition, the required structural reorganization is relatively small, as the conversion of the reacting carbon atom from a planar sp² configuration to the tetrahedral sp³ geometry of the radical adduct (P_RAF-C2 or P_RAF-C3) involves only a modest distortion of the quinoid framework.

Thermodynamic driving force: this kinetic fluidity is reinforced by a massive thermodynamic driving force, with exothermic drops exceeding −33.0 kcal mol⁻¹ in all media. The addition of the radical breaks the anti-aromatic/quinoid strain of Ring A, prompting a significant structural relaxation that can be tracked by a dramatic upward shift in local aromaticity (ΔHOMA ≫ 0) as the ring moves toward a more stable radical state.

3.3.4. The hydrogen atom transfer (HAT) pathway: the structural intramolecular lock

In sharp contrast to traditional phenolic antioxidants where O–H homolytic cleavage serves as the frontline defense, the kinetic evaluation of the HAT pathway within the HHNQ framework reveals that this channel is severely bottlenecked. As illustrated by the comparative PES in Fig. 7, the native molecule resides exclusively in the cis-HHNQ conformation, stabilized by a formidable O5–H⋯O4 intramolecular hydrogen bond ring lock. To quantify the structural stabilization afforded by this chelate ring, the trans-HHNQ rotamer serves as an ideal, structurally decoupled reference model within the framework of the conformational/isodesmic method. By rotating the phenolic O5–H bond by 180° away from the neighboring carbonyl group, the intense non-covalent interaction is cleanly severed without altering the fundamental atomic connectivity or introducing confounding global electronic rearrangements to the bicyclic scaffold. Consequently, the Intramolecular Hydrogen Bonding Energy (IHBE) can be rigorously isolated and calculated as the pure energetic discrepancy between the two relaxed ground-state conformers using the following thermodynamic expression: IHBE = E_trans − E_cis. Where E_trans and E_cis represent the fully optimized total electronic energies (including zero-point vibrational energy corrections) of the trans and cis rotamers, respectively. The calculated IHBE for HHNQ in the gas phase, pentyl ethanoate, and water are 10.5, 8.3, and 3.9 kcal mol⁻¹, respectively (Table 6).

Crucially, this energetic rupture is mirrored by a distinct topological relaxation in the local aromaticity profiles. The transition from the locked cis state to the unconstrained trans state induces a measurable redistribution of the bond alternation within Ring B, demonstrating that the dynamic behavior of the structural index (HOMA) is inherently coupled to the strength and presence of this internal hydrogen bridge. Because of this rotamer lock, the mechanism of abstraction must unfold across a two-step kinetic crossroads. First, the radical must overcome a substantial conformational barrier (TS_cis–trans = 11.5 kcal mol⁻¹ in gas; 11.5 kcal mol⁻¹ in pentyl ethanoate, and 8.3 kcal mol⁻¹ in water) simply to break the internal hydrogen bond and swing the phenolic hydrogen outward. Once the trans-rotamer is populated, the actual abstraction step introduces a secondary kinetic barrier (TS_HAT = 9.6 kcal mol⁻¹ in gas; 6.1 kcal mol⁻¹ in water). While the net reaction is ultimately exothermic (ΔG_rxn≈ −14.2 to −22.1 kcal mol⁻¹), the high cumulative kinetic barrier (TS_cis–trans + TS_HAT) forces a profound mechanistic divergence.

3.3.5. Kinetics study

In this subsection, the rate constants (k_eck) at 298.15 K for two competitive reaction mechanisms—HAT following the cis–trans isomerization of HHNQ proceeds via OH-bond rotation and RAF at the C2 and C3 sites, were calculated in the gas phase (Table 7 and details in Table S6). While the cis–trans isomerization is a rapid unimolecular process facilitating interconversion between rotamers, the subsequent HAT pathway exhibits a rate constant of 1.5 × 10⁻¹³ cm³ molecule⁻¹ s⁻¹. In comparison, the RAF mechanism at both C2 and C3 sites proceeds with a higher rate constant of 2.1 × 10⁻¹³ cm³ molecule⁻¹ s⁻¹. These results indicate that radical addition to the double bond is kinetically more favorable than the hydrogen atom transfer pathway under the studied conditions. These calculated values provide critical insights into the competition between these pathways, clarifying the observed regioselectivity in the aforementioned mechanistic analysis.

Table 7. Rate constants (k_eck) for HHNQ + OH reaction at 298.15 K in the gas phase.

k eck (s−1)	k eck (cm3 molecule−1 s−1)	k eck (cm3 molecule−1 s−1)
cis–trans rotamer	HATtrans	RAF-C2	RAF-C3
2.5 × 103	1.5 × 10−13	2.1 × 10−13	2.1 × 10−13

4. Conclusions

In this study, density functional theory (DFT) calculations at the M06-2X/6-311++G(d,p) level, combined with the SMD solvation model, were employed to investigate the radical-scavenging behavior of 5-hydroxy-1,4-naphthoquinone derivatives (HXNQ) toward hydroxyl radicals (HO˙). The results demonstrate that environmental polarity significantly influences the energetics of the unsubstituted scaffold (X = H), with polar media stabilizing ionic species, lowering ionization energies, and enhancing electron affinities, thereby facilitating electron-transfer processes. Among the competing mechanisms, radical adduct formation (RAF) at the C2 C3 double bond was identified as the most favorable pathway, exhibiting near-barrierless activation energies (0.1–1.1 kcal mol⁻¹) and strong thermodynamic stabilization. In contrast, the hydrogen atom transfer (HAT) pathway is kinetically disfavored due to the presence of a strong O5–H⋯O4 intramolecular hydrogen bond, which stabilizes the cis conformation and imposes a substantial energetic penalty associated with cis–trans isomerization prior to hydrogen abstraction.

Aromaticity analyses based on the complementary HOMA and NICS descriptors reveal a close relationship between electronic delocalization and chemical reactivity. In particular, the radical anion HHNQ˙⁻ represents a unique electronic state in which partial aromaticity recovery occurs within the quinoid Ring A while the aromatic character of the benzenoid Ring B is largely preserved. This dual stabilization provides a structural and electronic basis for the favorable electron-transfer and radical-addition processes observed in the system. Overall, the results support the concept of aromaticity recovery through enhanced π-electron delocalization as a useful framework for understanding and predicting the reactivity of naphthoquinone derivatives and highlight ground-state aromaticity descriptors as efficient computational tools for evaluating radical-scavenging activity.

Conflicts of interest

There are no conflicts to declare.

Data availability

The data supporting this article have been included as part of the supplementary information (SI). Supplementary information: Cartesian coordinates and thermochemical energy corrections for the selected optimized species, optimized bond lengths, calculated HOMA and NICS aromaticity descriptors, the dataset used for the Hammett–HOMA correlation, and kinetic data for the reactions between HHNQ and HO˙. See DOI: https://doi.org/10.1039/d6ra05671g.