Theoretical studies on the spin trapping of the 2-chloro-5-hydroxy-1,4-benzoquinone radical by 5,5-dimethyl-1-pyrroline N-oxide (DMPO): the identification of the C–O bonding spin adduct

Abstract

The detection and identification of related radicals is crucial for the elucidation of the reaction mechanisms for metal-independent decomposition of hydroperoxides by halogenated quinones. In this study, the spin trapping of the 2-chloro-5-hydroxy-1,4-benzoquinone radical (CBQ) produced in the reaction of 2,5-dichloro-1,4-benzoquinone and t-butylhydroperoxide by 5,5-dimethyl-1-pyrroline N-oxide (DMPO) and its subsequent reaction processes have been systematically investigated at the B3LYP/6-311++G(d,p) level of theory in combination with the atoms in molecules (AIM) theory, natural bond orbital (NBO) theory, and ab initio molecular dynamics. It was found that DMPO and CBQ can not only form the C–C bonding spin adduct observed experimentally, but also can form the C–O bonding spin adduct. This point has been further tested by the spin trapping of the other halogenated CBQ radicals. After that, the keto–enol tautomerization occurs for the formed C–C bonding spin adduct, where the explicit water molecule plays an important catalytic role in assisting the proton transfer process. Subsequently, spontaneous proton transfer has been observed from the hydroxyl group of the CBQ fragment to the adjacent O atom of the DMPO fragment in the formation process of the oxidation state of the spin adduct. These results not only help deepen our understanding of the spin trapping mechanism of CBQ-type radicals by DMPO, but also can provide important clues to the clarification of the reaction mechanism between halogenated quinone and organic hydroperoxides.

---

1. Introduction

As a priority environmental contaminant, pentachlorophenol (PCP) has been widely used as a herbicide and insecticide in agriculture and industry due to its broad applications. It has been listed as a group 2B environmental carcinogen by the International Association for Research on Cancer.¹ More recently, PCP and its derivatives have been classified as persistent organic pollutants (POPs) by Stockholm Convention.² As one of the important genotoxic and carcinogenic quinoid metabolites of PCP, polyhalogenated quinones can create a variety of hazardous effects in vivo, such as acute hepatoxicity, nephrotoxicity, and carcinogenesis. Moreover, they have also been observed as reactive oxidation intermediates or products in the processes used to oxidize or destroy polychlorinated POPs in various chemical and enzymatic systems.^3–8

Recently, more studies have shown that halogenated quinones can react with the hydrogen peroxide or organic hydroperoxides to produce the hydroxyl radical, organic alkoxyl radicals, and quinone ketoxy radicals experimentally,^9–16 which can be used to partly explain the potential carcinogenicity of polyhalogenated aromatic environmental pollutants. Taking the reaction between 2,5-dichloro-1,4-benzoquinone (DCBQ) and t-butylhydroperoxide (t-BuOOH) for example,¹⁵ a nucleophilic attack of t-BuOOH to DCBQ occurs firstly, forming a chloro-t-butylperoxyl-1,4-benzoquinone intermediate. Subsequently, this unstable intermediate decomposes homolytically to produce t-butoxyl radical and O-centered CBQ radical, where the latter can be further isomerized to the C-centered CBQ radical. Moreover, the C-centered CBQ radical has been spin trapped by DMPO using the electron spin resonance (ESR) method experimentally,^16,17 providing strong evidence for the proposed reaction mechanism between halogenated quinones and organic hydroperoxides. Unfortunately, no corresponding spin adducts associated with the O-centered CBQ radical have been directly identified experimentally. Meanwhile, some uncertain questions are still needed to be addressed. For example, the nature of the CBQ radical, the structural features and bonding mechanism of the formed spin adducts, the thermodynamic and kinetic parameters in the spin-trapping process remain unclear as well as the subsequent reaction processes. Obviously, the solution of the above questions is important for the clarification of the reaction mechanism between halogenated quinones and organic hydroperoxides. On the other hand, some difficulties deriving form the active radicals, such as short half-life span, low steady-state concentration, and the poor stability of the formed spin adduct, limit the detection and identification of the radicals. Therefore, it is necessary to systematically investigate the spin-trapping process of the radical at the molecular level employing theoretical methods.

Nowadays, theoretical calculations based on the density functional theory (DFT) have increasingly become a powerful tool in studying the radicals trapping by spin traps. Taking the widely used DMPO spin trap for example, the characteristic ESR spectra of the formed DMPO-radical adducts can be used to detect and identify the specific radicals in chemical and biological systems experimentally. Theoretically, the spin trapping of HO˙, O₂˙⁻, HO₂˙, CO₂˙⁻, CO₃˙⁻, and SO_n˙⁻ (n = 2, 3, 4) radicals by DMPO have already been investigated employing the DFT methods.^18–28 The obtained results, such as the nature of the formed DMPO-radical adducts and the thermodynamic and kinetic parameters associated with the spin-trapping process, can not only provide new insights into the spin-trapping mechanism at the molecular level, but also can provide the theoretical guidance for the rational design of more efficient spin traps.

Therefore, to better elucidate the nature of the quinone radical CBQ produced in the reaction of halogenated quinone with organic hydroperoxide and its interaction mechanism with spin trap, the spin-trapping behavior of CBQ radical by DMPO and its subsequent reaction processes have been systematically investigated employing DFT method. Firstly, the formed possible spin adducts have been obtained. In addition to the C–C bonding adduct observed experimentally, the C–O bonding adduct has also been located for the first time. Similarly, the same is also true for the spin trapping of the other halogenated CBQ radicals. Secondly, the keto–enol tautomerization reactions for the spin adducts have been explored. The positive role of explicit water molecules in the assistance of the proton transfer has been observed. Additionally, spontaneous proton transfer phenomenon has been observed in the formation process of the oxidation state of the most stable C–C bonding adduct.

2. Computation details

All the geometries have been fully optimized at the B3LYP/6-311++G(d,p) level of theory. Here, the reliability and validity of the level of theory adopted has been verified by a lot of investigations on the various systems.^18–30 Moreover, vibrational frequency analysis has been performed at the same level of theory to identify the nature of the optimized structures. For the calculated transition states characterized by only one imaginary frequency, intrinsic reaction coordinate (IRC) calculations have been carried out to further verify that the transition states indeed connect the initial reactants and products.^31,32

To clarify the existence and the nature of the interactions between DMPO and CBQ radical, the atoms in molecules (AIM) theory has been carried out on the basis of the optimized geometries. According to the AIM theory,³³ the existence of the interatomic interaction is indicated by the presence of a bond critical point (BCP). The interaction strength can be estimated from the magnitude of the electron density (ρ_bcp) at the BCP. Similarly, the ring structures are characterized by the existence of a ring critical point (RCP). Moreover, the nature of the interatomic interaction can be reflected from the topological parameters at the BCP, such as the Laplacian of electron density (∇²ρ_bcp) and energy density (H_bcp). Generally, ∇²ρ_bcp < 0 and ∇²ρ_bcp > 0 suggest that the charge is concentrated as in covalent bonds (shared interaction) and depleted as in ionic bonds, H-bonds, and van der Waals interactions (closed-shell interaction), respectively. As for ∇²ρ_bcp > 0 and H_bcp < 0, the interaction (e.g., intermolecular H-bonding) is partly covalent in nature.^34–36

To investigate the bonding mechanism and electron transfer behavior during the formation process of the spin adducts, the natural bond orbital (NBO) theory analyses have been performed based on the optimized geometries.³⁷ Here, both the alpha and beta spin orbitals have been included for the open shell systems.

To evaluate the structural changes during the spin-trapping process, deformation energy has been defined as the energy difference between the monomer in the spin adduct and its isolated state.

To obtain the interaction strengths between DMPO and CBQ radical, the interaction energies are defined as the energy differences between the formed spin adducts and the corresponding monomers, which are further corrected by the zero-point vibrational energy and basis set superposition errors (BSSE). Here, the Boys–Bernardi counterpoise technique has been employed to evaluate the BSSE.³⁸

To assess the kinetic stabilities of the formed spin adducts, ab initio molecular dynamics has been performed at the BLYP/DNP level of theory.^39,40 Constant temperature simulations at 298.15 K are performed using the Nosé–Hoover chain method, where the total simulation time is 1.0 ps with a time step of 1 fs. The stabilities of the spin adducts can be predicted from the radial distribution functions (RDFs) of the contact distance between DMPO and CBQ radical fragments.

All the calculations have been performed using Gaussian 03 program⁴¹ except for the ab initio molecular dynamics.

3. Results and discussion

It is the first step to clarify the active sites of CBQ radical before studying its spin-trapping behavior by DMPO. As shown in Fig. 1, CBQ exhibits a planar geometry with C_s symmetry. The spin densities of the unpaired electron are mainly distributed on the C1, C2, C5, O11, and O12 atoms, where most of them are located on the C5 atom followed by O11 and O12 atoms. As can be seen from the electrostatic potential surface in Fig. 1 and the calculated natural charges in Table 1, three O atoms of CBQ possess more negative charges. Correspondingly, the C atoms directly attached to the O atoms have more positive charges. As for the DMPO, its active site is the C atom linked by a double bond to N atom, which has been confirmed by many studies.^18–28

Fig. 1 The spin density distributions (left) and electrostatic potential surface (right) for the optimized CBQ radical. The isodensity contours are 0.002 electron per bohr³.

Table 1 The natural charge populations for CBQ radical

C1	C2	C3	C4	C5	C6
−0.063	−0.245	0.443	0.353	−0.100	0.395

---

H7	H8	Cl9	O10	O11	O12
0.240	0.225	0.093	−0.458	−0.424	−0.459

---

3.1 The spin-trapping process of CBQ radical by DMPO

Based on the analyses of the spin densities and charge populations of CBQ radical, the possible spin adducts have been designed and fully optimized. As a result, six spin adducts have been located on their potential energy surface, which have been shown in Fig. 2. For the sake of simplicity, the symbol IMn(X) has been employed to stand for the formed spin adduct. Here, n and X refer to the number of the spin adducts and the active sites of CBQ radical, respectively. As mentioned below, the relative stabilities of the spin adducts decrease with the increasing of the value of n. For example, IM1(C5) stands for the most stable one among the formed spin adducts and the C5 atom of CBQ radical directly interacts with the DMPO.

Fig. 2 Molecular graphs of the optimized spin adducts, where the BCP and RCP are denoted as small red and yellow dots, respectively. The units of the selected distance are in Å.

3.1.1 The structural features and bonding mechanism of spin adducts. Overall, as shown in Fig. 2, the formed spin adducts can be clarified into the C–C and C–O bonding types according to the interaction modes between DMPO and CBQ radical.

As for the spin adduct IM1(C5), the active C atom of DMPO interacts with the C5 atom of CBQ to form the C5–C15 bond, which can be confirmed by the presence of the BCP. Moreover, as shown in Table 2, the negative ∇²ρ_bcp at the BCP of C5–C15 bond suggests that this bond should possess the covalent nature. Actually, this point is also reflected from the short distance of C5–C15 bond (1.568 Å). Similarly, the same is also true for the spin adducts IM5(C2) and IM6(C1). As for the IM2(O11), IM3(O12), and IM4(O10) adducts, new C–O covalent bond has been formed. Additionally, as shown in Fig. 2, the intermolecular C⋯H or O⋯H H-bond has been observed in the formed spin adducts except for those of IM3(O12) and IM6(C1). From the positive ∇²ρ_bcp and H_bcp at the BCP of these H-bonds, one can see that these H-bonds should be governed by the electrostatic interactions.

Table 2 The topological analyses for the formed spin adducts between DMPO and CBQ^a

Complexes	BCP	ρbcp	∇^2ρbcp	Vbcp	Gbcp	Hbcp
a The ρbcp, ∇^2ρbcp, Vbcp, Gbcp, and Hbcp is electron density, the Laplacian of the electron density, potential energy density, kinetic energy density, and energy density at the BCP, respectively. All the units are a.u.
IM1(C5)	O10–H29	0.0031	0.0101	−0.0016	0.0021	0.0005
	O11–H19	0.0082	0.0303	−0.0053	0.0064	0.0012
	C6–O31	0.0108	0.0375	−0.0071	0.0082	0.0012
	C5–C15	0.2207	−0.4379	−0.2155	0.0530	−0.1625
IM2(O11)	O11–C15	0.2363	−0.4220	−0.4753	0.1849	−0.2904
	H8–O31	0.0128	0.0417	−0.0079	0.0092	0.0013
IM3(O12)	O12–C15	0.2268	−0.4112	−0.4218	0.1595	−0.2623
	Cl9–H20	0.0106	0.0392	−0.0061	0.0080	0.0018
IM4(O10)	O11–H20	0.0215	0.0828	−0.0154	0.0180	0.0026
	O10–C15	0.1917	−0.2453	−0.3248	0.1317	−0.1931
IM5(C2)	C2–C15	0.1838	−0.2718	−0.1628	0.0474	−0.1154
	O10–H29	0.0076	0.0253	−0.0045	0.0054	0.0009
	C3–H17	0.0062	0.0228	−0.0034	0.0045	0.0012
	C4–H19	0.0065	0.0188	−0.0031	0.0039	0.0008
IM6(C1)	C1–C15	0.2238	−0.4489	−0.2177	0.0528	−0.1650
	O12–O31	0.0120	0.0442	−0.0084	0.0097	0.0013
	Cl9–H17	0.0075	0.0254	−0.0039	0.0051	0.0012

---

To further confirm the bonding mechanism between DMPO and radical, NBO analyses have been performed on the basis of the optimized geometries. As shown in Fig. 3 and S1 of the ESI,† the σ-bonds have been observed for those C–C and C–O bonds, which is consistent with the AIM analyses mentioned above.

Fig. 3 Schematic graphs of the NBO orbital associated with the interaction between DMPO and CBQ radical in the spin adduct IM1(C5) on the basis of the alpha orbitals. The isodensity contours are 0.004 electron per bohr³.

To explore the extent of the structural changes occurring in the spin-trapping processes, we have calculated the deformation energies for the DMPO and CBQ radical relative to their free sates. As presented in Table 3, both the DMPO and CBQ have undergone different structural changes depending on the specific interaction modes adopted in the adducts. In details, larger structural changes for DMPO take place in the formation processes of IM1(C5), IM2(O11), IM3(O12), and IM4(O10) spin adducts, where the calculated deformation energies range from 17.58 to 31.89 kcal mol⁻¹. On the contrary, larger structural changes have been observed for CBQ radical upon formation of spin adducts IM5(C2) and IM6(C1).

Table 3 Deformation energy of the DMPO and CBQ during the spin-trapping process^a

Monomer	IM1(C5)	IM2(O11)	IM3(O12)	IM4(O10)	IM5(C2)	IM6(C1)
a All the units are in kcal mol^−1.
CBQ	16.45	10.56	12.18	10.79	22.65	35.42
DMPO	31.89	21.78	20.90	17.58	22.05	28.27

---

To investigate the changes of the spin densities of CBQ radical upon spin-trapping, spin density analyses have been performed on the basis of the optimized adducts, which has been defined as the difference between the alpha and beta electrons. As shown in Fig. 4, the spin densities have been transferred from the CBQ radical fragment to the N and O atoms of the DMPO in all the adducts except for IM5(C2) and IM6(C1), reflecting the nature of the DMPO as a spin trap. As for the IM5(C2) and IM6(C1) adducts, few spin densities are still distributed on the original CBQ fragment, suggesting that the unpaired electron of CBQ radical has not been truly trapped by DMPO. Therefore, the spin trapping of CBQ by DMPO is sensitive to the specific interaction modes adopted between them.

Fig. 4 Spin density distribution for the spin adducts. The isodensity contours are 0.005 electron per bohr³.

To explore the electron redistribution behavior during the spin-trapping process, the electron density difference maps have been constructed for the three most stable spin adducts. As shown in Fig. 5, significant electron redistributions occur nearby the formed C–C or C–O bond between DMPO and CBQ as well as the N=O bond of DMPO fragment. Taking the formation of spin adduct IM1(C5) for example, the depleted electron density is mainly concentrated on the regions around the C5–C15 bond axis. While the increased electron density is mainly concentrated on the N atom of DMPO fragment and O atom of CBQ fragment. As a result, as presented in Table 4, slight electron transfer of 0.096 occurs from DMPO to CBQ. Similarly, the same phenomenon has also been observed for the other spin adducts although the magnitudes of the electron transfer are larger than that of IM1(C5). Therefore, DMPO should play the role of a reducing agent in the spin-trapping process.

Fig. 5 Electron density difference maps upon the formation of the spin adducts. The blue and purple regions represent the depleted and increased electron density, respectively. The isodensity contours are 0.005 electron per bohr³.

Table 4 Calculated net charges on the CBQ fragment in the spin-trapping process^a

IM1(C5)	IM2(O11)	IM3(012)	IM4(O10)	IM5(C2)	IM6(C1)
a Negative value stands for the electron transfer from DMPO to CBQ radical.
−0.096	−0.360	−0.364	−0.453	−0.244	−0.205

---

3.1.2 The thermodynamic and kinetic parameters of the spin-trapping process. As shown in Table 5, the calculated relative energies are 1.48, 11.46, 25.83, 35.68, and 47.67 kcal mol⁻¹ for IM2(O11), IM3(O12), IM4(O10), IM5(C2), and IM6(C1) adducts relative to that of IM1(C5), respectively. Therefore, the order of the relative stabilities among the formed spin adducts is: IM1(C5) > IM2(O11) > IM3(O12) > IM4(O10) > IM5(C2) > IM6(C1). Noted that the C–C bonding adduct has been detected by ESR method experimentally,¹⁷ which is consistent with the most stable spin adduct IM1(C5). Especially, the total energy and Gibbs free energy gaps between IM1(C5) and IM2(O11) are only 1.48 and 1.00 kcal mol⁻¹, indicating that both of them can coexist under normal condition. As for the other four spin adducts, larger relative energies suggest that the formation possibilities of them are very small. In addition, considering the spin density and population analyses mentioned above and the fact that no spin density distributes on the O10 atom of CBQ radical, one can see that the spin-trapping process of CBQ radical by DMPO should be the radical addition in nature, which is mainly controlled by the spin density distribution of radical.

Table 5 The relevant energy parameters and Gibbs free energy barriers during the formation processes of the spin adducts^a

Complexes	ΔERel	ΔEInter	ΔH	ΔG	ΔG*
a All the units are in kcal mol^−1.
IM1(C5)	0.00	−20.83	−23.36	−10.15	—
IM2(O11)	1.48	−20.02	−21.87	−9.15	5.66
IM3(O12)	11.46	−9.72	−11.79	0.37	9.87
IM4(O10)	25.83	4.36	2.59	15.20	17.19
IM5(C2)	35.68	14.80	12.37	25.74	27.21
IM6(C1)	47.67	27.57	24.52	37.52	38.29

---

Table 5 presents the calculated interaction energies between DMPO and CBQ in all the formed spin adducts. Obviously, the relative orders of the interaction energies are well consistent with the order of the relative stabilities mentioned above. Here, it should be noted that the interaction energies in the three most stable adducts are negative and the absolute values of IM1(C5) and IM2(O11) adducts are much larger than that of IM3(O12). As for the other three spin adducts, all of them have positive interaction energies, indicating that they are unstable with respect to those of the DMPO and CBQ. In addition, comparisons of the interaction energies and the amounts of the electron transfer during the spin-trapping process suggest that there is no direct relationship between them.

Thermodynamically, as presented in Table 5 and Fig. 6, the enthalpy and Gibbs free energy changes are negative in the formation processes of the IM1(C5) and IM2(O11) adducts. Therefore, the formations of them are exothermic processes and can occur spontaneously. On the contrary, for the formations of the rest adducts, all the Gibbs free energy changes are positive, suggesting that they are unfavorable to be formed thermodynamically. Moreover, the corresponding transition states have been investigated for the formations of these spin adducts. As shown in Fig. 7, five transition states have been located except for that of IM1(C5) adduct. To explore the formation process of the IM1(C5) adduct, energy scan has been performed along with the C–C contact distance on the basis of the optimized IM1(C5) adduct. As displayed in Fig. 8, no corresponding saddle point appears during the approach of DMPO to CBQ, indicating that the spin-trapping process is a barrierless process. As for the formations of the IM2(O11) and IM3(O12) adducts, the calculated free energy barriers are only 5.66 and 9.87 kcal mol⁻¹, respectively. However, high barriers ranging from 17.19 to 38.29 kcal mol⁻¹ have been observed for the formations of the rest spin adducts. Thus, besides the IM1(C5) adduct, the formation of the IM2(O11) adduct is also feasible thermodynamically and kinetically.

Fig. 6 The Gibbs free energy profile for the spin trapping of CBQ radical by DMPO.

Fig. 7 Optimized transition states in the formation processes of the spin adducts.

Fig. 8 The potential energy curve for IM1(C5) along the distances of its C5–C15 bond.

Moreover, to further investigate the kinetic stability of the IM2(O11) adduct, ab initio molecular dynamics has been performed on the basis of its optimized geometry. As shown in Fig. 9, the radial distribution function (RDF) for the C5–O11 bond formed between DMPO and CBQ is mainly distributed in the vicinity of 1.5 Å, indicating that the formed spin adduct is stable during the available time scale. Therefore, the C–O bonding adduct can be formed in the spin-trapping process. In other words, the O-centered CBQ radical can be also trapped by DMPO, providing the direct theoretical evidence for the proposed reaction mechanism between halogenated quinone and organic hydroperoxides experimentally. Unfortunately, the C–O bonding adduct has not been detected experimentally,^16,17 which may be due to its low steady-state concentration or short half-life. Thus, more sophisticated experiments are required to clarify this point.

Fig. 9 Radial distribution function of the C5–O11 bond for the IM2(O11) adduct.

As described above, the C–C σ-bond can be formed in the most stable spin adduct IM1(C5). Expectedly, its other isomers may exist through the rotation of the C–C bond. To confirm this point, relaxed potential energy profile has been constructed on the basis of the IM1(C5) adduct, where only the dihedral angle D(C4,C5,C15,N30) has been fixed in the scanning process. As displayed in Fig. 10, several minima have been located at the −160.0, −80.0, −40.0, 50.0, and 60.0° of dihedral angle. Here, the global minimum at the 60.0° of dihedral angle corresponds to the IM1(C5). Therefore, the different orientations between DMPO and CBQ have certain effects on the stability of the formed spin adduct for the C–C bonding mode.

Fig. 10 The relaxed potential energy curve for IM1(C5) adduct along the dihedral angle of D(C4,C5,C15,N30).

Additionally, the effects of the explicit water molecules on the formations of the C–C and C–O bonding spin adducts have been investigated systematically. Firstly, the interactions of one water molecule with DMPO and CBQ radical have been investigated, respectively. It was found that both the DMPO and CBQ radical have weak interactions with the water molecule through the intermolecular H-bonds, where the calculated interaction energies are −6.34 and −3.70 kcal mol⁻¹ at the B3LYP/6-311++G(d,p) level of theory. As displayed in Fig. S2 of the ESI,† the intermolecular H-bonds between water molecule and DMPO and CBQ are 1.826 and 2.042 Å, respectively. More importantly, the introduction of explicit water molecules does not influence the active sites of DMPO and CBQ, implying slight effects of water molecules on the spin trapping of CBQ by DMPO. To further confirm this point, two water molecules have been introduced in the formation process of the IM1(C5) and IM2(O11) spin adducts, where two water molecules interact with the DMPO and CBQ radical, respectively. As a result, the calculated C–C and C–O bonding distances between DMPO and CBQ have been changed slightly no more than 0.004 Å relative to those of the corresponding spin adducts in the absence of the water molecules. The interaction energies and the Gibbs free energy changes have been decreased by 3.36(3.04) and 3.95(3.33) kcal mol⁻¹ upon the introduction of the water molecules, where the data in parentheses refer to the results of the IM2(O11) spin adduct, therefore, the presence of the water molecules plays a slight negative role in the formation of the spin adducts overall.

3.1.3 The halogen effects. As mentioned above, besides the C–C bonding adducts, the C–O bonding adducts can be formed in the spin-trapping process of CBQ by DMPO. A question may arise, namely, is it true for the other CBQ derivatives (e.g., halogenated CBQ) since they can be produced in the reactions of the halogenated quinone with organic hydroperoxides^10–16? To clarify this point, the spin trapping of halogenated CBQ radicals by DMPO has been investigated on the basis of the interaction modes adopted in the adducts of IM1(C5) and IM2(O11). As shown in Table 6, all the Gibbs free energy changes and enthalpy changes are negative values in the spin-trapping processes, suggesting the feasibility of the spin trapping thermodynamically. Meanwhile, similar to the energy difference between IM1(C5) and IM2(O11) mentioned above, the small energy gaps between the formed C–C and C–O bonding adducts suggest that the C–O bonding adducts can be also formed. In particular, as for the trichloro-substituted CBQ radical, the formed C–O bonding adduct is more stable than that of the C–C bonding adduct, indicating that the former is more likely to be detected experimentally.

Table 6 The thermodynamic parameters in the spin trapping of halogenated CBQ by DMPO^a

Halogenated type	ΔG	ΔH	ΔE
a All the units are in kcal mol^−1. The data before and after slash refer to the results associated with the formations of the C–C and C–O bonding adducts, respectively. ΔE is the relative energy for the C–O bonding adduct relative to that of the corresponding C–O bonding adduct.
Monofluoro-	−10.60/−9.19	−23.73/−21.94	1.77
Monobromo-	−10.01/−9.07	−23.17/−21.84	1.33
Difluoro-	−11.04/−9.29	−24.32/−22.10	2.20
Dichloro-	−10.23/−9.05	−23.57/−21.93	1.62
Dibromo-	−9.78/−8.78	−23.23/−21.75	1.44
Trifluoro-	−7.24/−5.10	−20.90/−17.94	2.90
Trichloro-	−1.79/−2.75	−15.69/−15.94	−0.30
Tribromo-	−3.33/−1.81	−16.81/−15.37	1.57

---

3.2 The keto–enol tautomerization process of the spin adducts

Given the fact that the formed C–C and C–O bonding spin adducts are characterized by the keto features, the occurrence of the keto–enol tautomerization processes is expected through the proton transfer. To obtain the tautomerization behavior for these spin adducts, the proton transfer (PT) processes have been investigated on the basis of the three most stable adducts. For the sake of simplicity, the PTs from the C5 atom of CBQ to its adjacent O11 and O12 atoms have been denoted as the PT11 and PT12, respectively.

As shown in Fig. 11, for the tautomerization process of the IM1(C5), strong H-bonding interaction has been observed between the transferred proton and the O atom of the N=O group of DMPO fragment, where the corresponding H-bond distances are 1.649 and 1.624 Å in two enol forms. Further AIM analyses confirm that these H-bonds possess partly covalent nature as shown in Table S1 of the ESI.† Moreover, the electron density at the BCP of the C5–C15 bond linking the DMPO and CBQ fragments has been increased, implying the strengthening of the C5–C15 bond in the proton-transferred product. Actually, as given in Table 7, the enol form of the IM1(C5) adduct is more stable by about 3.11–5.33 kcal mol⁻¹ than that of the keto form. Moreover, the enthalpy and Gibbs free energy changes are negative, suggesting the feasibility of the proton transfer thermodynamically. Kinetically, the calculated free energy barriers are 50.64 and 48.23 kcal mol⁻¹ for the direct PT of PT11 and PT12 processes, respectively. Therefore, it is difficult to take place for the keto–enol tautomerization if no other factors are considered. In view of the fact that explicit water molecules exist in realistic conditions, we further consider the assistance of the PT with water molecules, where the corresponding proton-transferred products have been given in Fig. S3 of the ESI.† Similar to that of the water-assisted PT in glycinamide,⁴² the introduced water molecule plays the role of the proton acceptor and proton donor simultaneously. As a result, as shown in Table 7, the free energy barriers mentioned above have been decreased if water molecules are introduced. For example, the barrier has been decreased by 20.64 kcal mol⁻¹ when one water molecule is introduced to assist the PT12 process. Moreover, the barrier has been further decreased by 7.85 kcal mol⁻¹ to 19.74 kcal mol⁻¹ when a second water molecule is introduced. Thus, the positive catalytic role of water molecules should be highlighted in the assistance of the PT process. Moreover, the water-assisted proton transfer should proceed with a concerted mechanism since no zwitterionic complexes have been located in the tautomerization process.

Fig. 11 Molecular graphs of the keto–enol tautomerization products, where the BCP and RCP are denoted as small red and yellow dots, respectively.

Table 7 Calculated thermodynamic parameters and Gibbs free energy barriers during the keto–enol tautomerization process of the spin adducts^a

Keto–enol process	ΔE	ΔH	ΔG	ΔG*
a All the units are in kcal mol^−1.
IM1-PT11	−5.33	−5.18	−5.54	50.64
IM1-PT12	−3.11	−2.55	−3.38	48.23
IM1-PT11-1W	−9.94	−9.86	−10.06	30.22
IM1-PT12-1W	−4.95	−3.52	−5.68	27.59
IM1-PT11-2W	−8.57	−7.90	−8.77	21.94
IM1-PT12-2W	−3.91	−3.93	−3.87	19.74
IM2-PT11	74.78	73.53	75.38	69.03
IM2-PT12	50.23	49.79	50.40	55.32
IM3-PT11	48.22	48.14	48.41	68.30
IM3-PT12	77.01	76.44	77.61	78.71

---

On the contrary, for the tautomerization of IM2(O11) and IM3(O12), the proton-transferred products are less stable than those of the keto forms. As shown in Table 7, the enthalpy and Gibbs free energy changes are large positive values, indicating that the tautomerization processes are difficult to occur for them thermodynamically. Moreover, this point can be further confirmed by the high free energy barriers ranging from 55.32 to 78.71 kcal mol⁻¹. Thus, the subsequent proton transfer can not proceed for the C–O bonding adducts.

3.3 The oxidation state of the C–C bonding adduct

According to the experimental findings,^16,17 the oxidation state of the C–C bonding adduct has also been detected. However, its structural feature and formation process have not been reported to the best of our knowledge. Therefore, based on the IM1-PT11 geometry, we have investigated the oxidation state of the spin adduct through the loss of the H atom.

As shown in Fig. 12, two oxidated structures can be produced through the loss of the H atoms attached to the C15 atom of the DMPO fragment and hydroxyl O11 atom of the CBQ fragment. For the former, spontaneous proton transfer from the hydroxyl group to the O atom of the N=O group of the DMPO fragment has been observed in the oxidation process. Meanwhile, strong H-bond has been formed between the transferred proton and the original carbonyl O atom, where the corresponding H-bond distance is 1.463 Å. Moreover, as shown in Table S2 of the ESI,† this H-bond possesses covalent nature partly with the ∇²ρ_bcp > 0 and H_bcp < 0 at the BCP.

Fig. 12 Molecular graphs of the oxidated keto–enol tautomerization products based on the IM1-PT11, where the BCP and RCP are denoted as small red and yellow dots, respectively.

To confirm which bond is easier to cleavage in the oxidation process, the bond dissociation enthalpy (BDE) has been calculated for the C15–H and O11–H bonds. As shown in Table 8, the vertical and adiabatic BDEs for the C15–H bond are lower by 27.07 and 38.12 kcal mol⁻¹ than that of the O11–H bond, respectively. Therefore, the C–H bond should be easier to cleavage to produce the oxidation state, which is consistent with the experimental findings.^16,17 Additionally, significant structural changes are expected in the oxidation processes from the large difference between the vertical and adiabatic BDEs.

Table 8 The calculated BDEs of the selected C–H and O–H bonds in IM1-PT11 adduct^a

Bonds	BDEver	BDEadi	ΔBDE
a All the units are in kcal mol^−1.
C15–H	84.19	37.81	46.38
O11–H	111.26	75.93	35.33

---

4. Conclusions

In this study, the spin trapping of CBQ radical by DMPO and its subsequent reaction processes have been systematically investigated at the B3LYP/6-311++G(d,p) level of theory. The structural features and bonding mechanisms of the formed spin adducts as well as the thermodynamic and kinetic behavior in the spin-trapping process have been clarified. The main conclusions are as follows:

(1) The nature of the spin trapping of CBQ radical by DMPO is the radical addition reaction, which is mainly controlled by the spin density population of the radical rather than its charge population.

(2) Besides the C–C bonding adduct detected experimentally, the C–O bonding adduct has also been obtained. Similarly, it is also true for the spin trapping of halogenated CBQ radicals by DMPO. The identification of the C–O bonding adduct provides the direct theoretical evidence for the reaction mechanism between halogenated quinone and organic hydroperoxide proposed experimentally.

(3) Keto–enol tautomerization reaction can occur in the selected C–C bonding adduct. The important catalytic role of explicit water molecules should be highlighted in the assistance of the proton transfer with the concerted mechanism.

(4) The oxidation state of the C–C bonding adduct detected experimentally can be produced through the C–H cleavage, accompanying the spontaneous proton transfer from CBQ to DMPO moieties.

Expectedly, these findings can provide not only many microscopic details about the spin trapping of CBQ radical by DMPO, but also important clues to the elucidation of the reaction mechanism between halogenated quinone and organic hydroperoxides.

Acknowledgements

This work is supported by NSFC (21577076, 21303093, and 21003082) and the NSF of Shandong Province (ZR2014BM020). The State Key Laboratory of Environmental Chemistry and Ecotoxicology, Research Center for Eco-Environmental Sciences, Chinese Academy of Sciences (KF2013-05) is also acknowledged.

References

1. IARC monographs on the evaluation of carcinogenic risks to humans: occupational exposures in insecticide application, and some pesticides. Pentachlorophenol, 1991, 53, 371–402.
2. Stockholm Convention, Countries move forward on important issues for sustainable management of chemicals and waste, Geneva, 2015, http://chm.pops.int/Home/tabid/2121/Default.aspx.
3. J. L. Bolton, M. A. Trush, T. M. Penning, G. Dryhurst and T. J. Monks, Chem. Res. Toxicol., 2000, 13, 135–160.
4. Y. Song, B. A. Wagner, J. R. Witmer, H. J. Lehmler and G. R. Buettner, Proc. Natl. Acad. Sci. U. S. A., 2009, 106, 9725–9730.
5. B. Meunier, Science, 2002, 296, 270–271.
6. S. S. Gupta, M. Stadler, C. A. Noser, A. Ghosh, B. Steinhoff, D. Lenoir, C. P. Horwitz, K. W. Schramm and T. J. Collins, Science, 2002, 296, 326–328.
7. A. Sorokin, J. L. Seris and B. Meunier, Science, 1995, 268, 1163–1166.
8. Y. L. Zhao, F. Qin, J. M. Boyd, J. Anichina and X. F. Li, Anal. Chem., 2010, 82, 4599–4605.
9. D. H. Sarr, C. Kazunga, M. J. Charles, J. G. Pavlovich and M. D. Aitken, Environ. Sci. Technol., 1995, 29, 2735–2740.
10. B. Z. Zhu, N. Kitrossky and M. Chevion, Biochem. Biophys. Res. Commun., 2000, 270, 942–946.
11. B. Z. Zhu, H. Zhao, B. Kalyanaraman and B. Frei, Free Radical Biol. Med., 2002, 32, 465–473.
12. B. Z. Zhu, B. Kalyanaraman and G. B. Jiang, Proc. Natl. Acad. Sci. U. S. A., 2007, 104, 17575–17578.
13. B. Z. Zhu and G. Q. Shan, Chem. Res. Toxicol., 2009, 22, 969–977.
14. B. Z. Zhu, L. Mao, C. H. Huang, H. Qin, R. M. Fan, B. Kalyanaraman and J. G. Zhu, Proc. Natl. Acad. Sci. U. S. A., 2012, 109, 16046–16051.
15. B. Z. Zhu, H. T. Zhao, B. Kalyanaraman, J. Liu, G. Q. Shan, Y. G. Du and B. Frei, Proc. Natl. Acad. Sci. U. S. A., 2007, 104, 3698–3702.
16. B. Z. Zhu, G. Q. Shan, C. H. Huang, B. Kalyanaraman, L. Mao and Y. G. Du, Proc. Natl. Acad. Sci. U. S. A., 2009, 106, 11466–11471.
17. C. H. Huang, G. Q. Shan, L. Mao, B. Kalyanaraman, H. Qin, F. R. Ren and B. Z. Zhu, Chem. Commun., 2013, 49, 6436–6438.
18. F. A. Villamena, C. M. Hadad and J. L. Zweier, J. Am. Chem. Soc., 2004, 126, 1816–1829.
19. F. A. Villamena, C. M. Hadad and J. L. Zweier, J. Phys. Chem. A, 2005, 109, 1662–1674.
20. F. A. Villamena, J. K. Merle, C. M. Hadad and J. L. Zweier, J. Phys. Chem. A, 2005, 109, 6083–6088.
21. F. A. Villamena, J. K. Merle, C. M. Hadad and J. L. Zweier, J. Phys. Chem. A, 2005, 109, 6089–6098.
22. F. A. Villamena, E. J. Locigno, A. Rockenbauer, C. M. Hadad and J. L. Zweier, J. Phys. Chem. A, 2006, 110, 13253–13258.
23. F. A. Villamena, E. J. Locigno, A. Rockenbauer, C. M. Hadad and J. L. Zweier, J. Phys. Chem. A, 2007, 111, 384–391.
24. F. A. Villamena, J. K. Merle, C. M. Hadad and J. L. Zweier, J. Phys. Chem. A, 2007, 111, 9995–10001.
25. F. A. Villamena, Y. P. Liu and J. L. Zweier, J. Phys. Chem. A, 2008, 112, 12607–12615.
26. F. A. Villamena, J. Phys. Chem. A, 2010, 114, 1153–1160.
27. P. L. Zamora and F. A. Villamena, J. Phys. Chem. A, 2012, 116, 7210–7218.
28. B. Aguilera-Venegas and H. Speisky, J. Mol. Graphics Modell., 2014, 52, 57–70.
29. P. Li, Z. Li, Q. Sun, W. H. Wang, S. W. Bi, A. J. Du and Y. Zhao, ChemPhysChem, 2013, 14, 2737–2743.
30. P. Li, Z. Y. Ma, W. H. Wang, R. Song, Y. Z. Zhai, S. W. Bi, H. T. Sun and Y. X. Bu, Phys. Chem. Chem. Phys., 2011, 13, 5931–5939.
31. C. Gonzalez and H. B. Schlegel, J. Chem. Phys., 1989, 90, 2154–2161.
32. C. Gonzalez and H. B. Schlegel, J. Phys. Chem., 1990, 94, 5523–5527.
33. R. F. W. Bader, Atoms in molecules: a quantum theory, Oxford University Press, Oxford, UK, 1990.
34. L. F. Pacios, J. Phys. Chem. A, 2004, 108, 1177–1188.
35. I. Rozas, I. Alkorta and J. Elguero, J. Am. Chem. Soc., 2000, 122, 11154–11161.
36. W. D. Arnold and E. Oldfield, J. Am. Chem. Soc., 2000, 122, 12835–12841.
37. A. E. Reed, L. A. Curtiss and F. Weinhold, Chem. Rev., 1988, 88, 899–926.
38. S. F. Boys and F. Bernardi, Mol. Phys., 1970, 19, 553–566.
39. B. Delley, J. Chem. Phys., 1990, 92, 508–517.
40. B. Delley, J. Chem. Phys., 2000, 113, 7756–7764.
41. M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, J. A. Montgomery Jr, T. Vreven, K. N. Kudin, J. C. Burant, J. M. Millam, S. S. Iyengar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega, G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox, H. P. Hratchian, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, P. Y. Ayala, K. Morokuma, G. A. Voth, P. Salvador, J. J. Dannenberg, V. G. Zakrzewski, S. Dapprich, A. D. Daniels, M. C. Strain, O. Farkas, D. K. Malick, A. D. Rabuck, K. Raghavachari, J. B. Foresman, J. V. Ortiz, Q. Cui, A. G. Baboul, S. Clifford, J. Cioslowski, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng, A. Nanayakkara, M. Challacombe, P. M. W. Gill, B. Johnson, W. Chen, M. W. Wong, C. Gonzalez and J. A. Pople, Gaussian 03 (Revision E.01), Gaussian, Inc., Wallingford, CT, USA, 2004.
42. P. Li and Y. X. Bu, J. Phys. Chem. B, 2004, 108, 18088–18097.

---

Footnote

† Electronic supplementary information (ESI) available: Graphs of the NBO orbital interaction between DMPO and CBQ radical, optimized complexes of water molecule with CBQ, DMPO, IM1(C5), and IM2(O11), molecular graphs of the water-assisted keto–enol tautomerization products, topological analyses for the keto–enol tautomerization products of the three most stable spin adducts, topological parameters for the oxidation state of the IM1-PT11 spin adduct, and the Cartesian coordinates of the geometries including reactants, transition states, spin adducts, and the oxidated keto–enol tautomerization products. See DOI: 10.1039/c6ra07696c

---