138

THEORETICAL ESR SPECTRUM OF 1,3-

BENZOSEMIQUINONE RADICAL

V.Chiş1, Sanda Filip2, D.Ristoiu1, Raluca Marcu1, M. Oltean1,

L.David1, O.Cozar1

1Babeş-Bolyai University, Faculty of Physics, 1 Kogălniceanu, RO-400084 Cluj-Napoca, Romania 2University of Oradea, Faculty of Science, Str. Armatei Romane 5, RO-410087 Oradea, Romania

Abstract

Semiempirical, Ab Initio and hybrid density functional calculations are

used to calculate geometries, spin densities and isotropic hyperfine coupling

constants for the meta-benzosemiquinone anion radical. Three optimum

structures, with C2v(2B1), C2v(

2A2), and Cs(2A") symmetries have been located

by geometry optimization, the former being 5.17Kcal/mol higher in energy than

the second one. While UHF/3-21G method gives an energy difference between

the last two structures of 1.82Kcal/mol, a higher level of theory (B3LYP/EPR-

II) gives practically the same energy. The C2v(2A2) structure can be considered

as the transition state between two equivalent mirror image Cs structures, as

pointed out by Tripathi et al. [1]. The influence of the surrounding solvent

molecules by their dielectric properties on the calculated hyperfine coupling

124

constants of 1H and 13C nuclei has been considered by applying the Isodensity

Polarizable Continuum (IPCM) model. It is shown that, contrary to ortho-

benzosemiquinone radical, the solvent effects have a minor influence on the

isotropic hyperfine coupling constants of the meta- isomer. A very good

agreement between experiment and theory was obtained by improving the

quality of the grid used in numerical integration of electron density, in the

framework of Density Functional Theory. The present results suggest that the

unpaired spin density of the radical does not have an odd-alternant pattern like

that for phenoxyl or ortho-benzosemiquinone radicals.

Introduction

Quinones represent an important class of biomolecules,

being involved in a large number of biophysical processes such as

cellular respiration, blood clotting or aging. They have been also

recognized as microbial controlling agents [2]. The quinone-type

radicals represent important cofactors for electron transfer in

photosynthesis, acting as electron acceptors in the initial charge

separation process [3-6]. These relatively stable radicals are

formed by one-electron reduction of quinones or by one-electron

oxidation of quinols.

Their in vivo reduction of the quinone by redox proteins to

the benzosemiquinone (BSQ) anion radicals state has been

involved in both their anticancer function and their cytotoxic side

effects.

On the other hand, the study of transitional metal complexes

containing quinones has developed into active area of research.

125

From this kind of studies, it emerged that the quinone form of the

molecule does not readily bind to transition metals, but

semiquinones and catechols are much more able to form stable

complexes with first, second and third row transition metals [7].

A semiquinone is paramagnetic by virtue of the odd

number of electrons and thus appropriate to be studied by ESR

spectroscopy. Moreover, since redox processes responsible for the

production of semiquinone radicals occur without significant

structural changes, the ESR spectrum of a semiquinone radical

cold identify the parent quinone or quinol, as well.

This has made the ESR spectroscopy a very important tool

in the study of naturally occurring quinones and quinols. Indeed,

many experimental works have been reported on the hyperfine

structure of BSQ radicals [8-11]. Besides, vibrational

spectroscopies can also provide useful information on the

structure and properties of these radicals [1,12].

The ESR spectra provide the hyperfine coupling constants

of magnetic nuclei and the line pattern of the spectrum can

provide a good insight into the structure of these radicals.

However, the experimental data have to be compared with their

theoretical counterparts for a better understanding of the particular

properties of the detected radicals. Moreover, the analysis is often

complicated by a strong influence of the solvent on the observed

ESR spectra, and the hyperfine coupling constants observed in

solution frequently disagree with those predicated by quantum-

126

chemical calculations on free, isolated species. Theory may be of

assistance through comparisons of observed and computed

hyperfine coupling constants, which may lead to the plausible

assignments of the other properties.

In the last decade, methods based on Density Functional

Theory (DFT) became an alternative approach for calculating

hyperfine coupling constants (hfcc's) and this is due to

considerably less computational cost and memory requirements

than those of conventional correlated Ab Initio procedures [13-

16]. In addition, due to their lower computational cost and hence

to their ability to treat larger systems, DFT methods may be used

to obtain more realistic descriptions of the interactions between

radical systems and their surroundings by explicit consideration of

the latter.

Within the DFT framework, the hybrid density functionals

are increasingly being shown to provide highly accurate

descriptions of free radical properties such as isotropic and

anisotropic hyperfine coupling constants [15-17]. These

functionals include a gradient correction for both exchange and

correlation energy and differ essentially from the older functionals

in that some Hartree-Fock exchange is included. One of the

motivations for the introduction of these self-consistent hybrid

functionals is the hope to obtain with the same method and basis

set good geometrical parameters as well as thermochemical and

spectroscopic properties. This is particularly important for open

127

shell species for which experimental structures are often not

available.

128

Many theoretical studies are reported on

benzosemiquinone (BSQ) anion radicals. Tripathi and coworkers

[1] reported an unrestricted Hartree-Fock (UHF) Ab Initio study

on the calculated hyperfine structure of 1,3-BSQ radical, at

UHF/3-21G level of theory. The electronic structure of para-BSQ

and its ESR spectrum was investigated by O'Malley [18] and

Langaard and Spanget-

Larsen [19], while the

hydrogen bonding effects

on the properties of this

free radical was reported

by Chipman [20]. The

influence of the solvent

effects on the hyperfine

structure of ortho-BSQ

radical was investigated

by Langaard and Spanget-Larsen [19] and by Chiş et al. [21]. The

meta- isomer received by far much less interest from this

perspective so that in this work we used INDO, Ab Initio and DFT

molecular orbital calculations to examine its structure and ESR

spectrum. The structure of the radical is shown in Fig.1 together

with the atom numbering scheme used.

The hyperfine coupling constants of magnetic nuclei arises

from the Fermi contact interaction between the unpaired electron

and the nucleus [22] and are given by

Fig.1. Structure and atom numbering scheme of 1,3-Benzosemiquinone radical

129

Q(0)βgβgµ3

2a NNee0

(N)iso = (1)

where the isotropic hfcc aiso is in Hz, ge and gN are the electron

and nuclear g factors, µB and µN are the Bohr and the nuclear

magneton respectively. Q(0) is the net unpaired electron spin

density at the position of the nucleus N. Since the unpaired spin

density is usually expressed using the atomic units (e.g. in bohr-3)

and the hfcc’s are generally expressed in MHz (or Gauss), for

practical calculations, the equation (1) become [23]

Q(0)a

1h

10βgβgµ

32

(MHz)a30

-6

NNee0(N)iso = (2)

where h is the Planck constant and a0 is the first Bohr radius.

Where as 30

NnBeha

1gg µµ is a constant for a specific

nucleus, Q(0) has to be obtained from quantum theoretical

calculations, based on the equation

Ψ)ρ(rΨQ(0) N=

(2)

where ρ(rN) is the spin density operator evaluated at the position

of the nucleus N.

Computational Details

INDO, Ab Initio, and DFT methods have been used to

calculate the geometry and ESR spectrum of the radical. The basis

130

sets used represent two distinct groups: the Pople’s 4-31G and 6-

31G series up to 6-31+G(d) [24] and the new specially tailored for

hfcc’s calculations EPR-II and EPR-III basis sets of Barone

[15,16,25]. For Density Functional calculation the hybrid B3LYP

functional [26,27] has been chosen due to the fact that it proved its

ability for reproducing the free radical properties with high

accuracy [15,16,18,19,25,28,29]. The force constant matrices

calculated for all stationary points were checked to have no

negative eigenvalues in order to ensure that they are minima on

the potential energy hypersurface. Comparative hfcc’s

calculations were also performed using the semiempirical INDO

method in conjunction with optimized radical geometry at

different Ab Initio and DFT levels of theory. This method is

capable of calculating negative spin densities arising from

electron correlation since it uses unrestricted wave functions. It is

also able to account for spin polarization by including one-center

exchange integrals, so that unpaired spin arises naturally in the 1s

orbitals of hydrogen atoms attached to aromatic carbon atoms.

The INDO method is thus the least complex semiempirical

method available which is able to account in general for the

observed features of the ESR spectra of free radicals and it is

expected to work well only for radicals at or near their equilibrium

geometries and also, with no unusual bonding situations [30].

The Self-Consistent Reaction Field method (SCRF)

[31,32] (IPCM model) was used to model the long-range

131

environmental effects on hyperfine couplings. All the calculations

were performed using the GAUSSIAN 98W program package

[41].

Results and Discussion

1. Geometry

Three optimum structures have been located by geometry

optimizations. Two of them have C2v symmetry with 2B1 and 2A2

wave-function symmetries. Without any constraint on the

symmetry we obtained a third Cs geometrical structure with a

wave-function corresponding to 2A" symmetry. Similar results

were obtained by Tripathi et al. [1] by Ab Initio calculations at

UHF/3-21G level of theory. Our results obtained by the superior

B3LYP/EPR-II method indicate that the C2v(2B1) geometry is

5.17Kcal/mol higher in energy than C2v(2A2), while the latter has

practically the same energy as the Cs(2A"). The last two structures

may be regarded as two separate conjugated systems joined by CC

single bonds [1]. Moreover, the C2v(2A2) structure represents the

transition state between the two equivalent mirror image Cs

structures.

Selected bond lengths and angles for 1,3-BSQ radical in its

free (gas-phase) form, calculated on Cs symmetry at different

levels of theory with various basis sets, are given in Table 1.

132

To the best of our knowledge, there have not been crystal

structures reported either for 1,3-quinone or for 1,3-BSQ radical.

However, an X-ray structure has been reported for the related

molecule 1,2-benzoquinone (1,2-BQ) [34]. Our calculated

geometrical parameters are compared with experimental and

theoretical results for other similar molecules [18-20,24,35,36 ].

As one can see from Table 1, the bond lengths are predicted

slightly longer by DFT method than the traditional Ab Initio

approach. On closer examination of the bond lengths it is

observed that the C4-C5 and C5-C6 bonds are shorter than the C1-

Table 1. Calculated bond lengths and angles of 1,3-BSQ radical in its free form, on the Cs symmetry (bond lengths in Å and angles in degrees)

Geometrical parameter

UHF 4-31G

UHF 6-31+G(d)

UHF EPR II

UB3LYP 4-31G

UB3LYP 6 -31+G(d)

UB3LYP EPR II

C1 – O1 1.261 1.237 1.236 1.293 1.272 1.272

C1 – C2 1.403 1.413 1.418 1.415 1.421 1.428

C3 – C4 1.477 1.490 1.496 1.470 1.479 1.487

C4 – C5 1.387 1.391 1.396 1.389 1.393 1.399

C4 – H4 1.072 1.076 1.078 1.085 1.088 1.088

C5 – H5 1.075 1.078 1.081 1.088 1.091 1.091

C1C2C3 123.34 123.13 123.08 123.68 123.22 123.25

C2C3C4 116.95 116.89 116.90 116.35 116.60 116.53

C4C5C6 119.70 119.63 119.60 119.64 119.60 119.58

O1C1C6 116.90 116.94 116.79 117.89 117.76 117.56

O3C3C4 118.32 118.43 118.45 118.15 118.38 118.37

H4C4C3 116.44 116.50 116.47 116.14 116.38 116.28

H4C4C5 122.03 121.77 121.77 121.87 121.63 121.66

133

C2, C2-C3 and C3-C4 bonds. This suggests more delocalisation of

the electron distribution in the lower part of the ring compared

with the upper part (see Figure 1). One explanation for these

geometrical features is given by the classical resonance structures

of the radical which suggest a localization of the unpaired electron

mainly at C2, C4 and C6 positions. In this resonance forms there

is the greatest amount of double bond character for C4-C5 and

C5-C6 and the least for C3-C4, C6-C1, C1-C2, C2-C3. C-O

calculated bonds are closer to the experimentally observed C=O

double bond lengths in para-benzoquinone (1.225 Å) [18] or in

1,2-BQ (1.216Å) [35]. The two CO bonds may easily

communicate by direct π through-bond interaction via C1-C2-C3

group. The C-H bond lengths obtained at UHF level of theory are

generally shorter than those calculated by DFT method. The C-H

bond lengths of phenol for example, were reported to range from

1.083 Å to 1.087 Å [37] and we would expect that the C-H

distance of the 1,3-BSQ anion are similar because the radical’s

unpaired electron occupies primarily π atomic orbitals. Thus the

DFT calculated C-H distances are closer to those experimentally

determined C-H distance in closed-shell molecules and are likely

to be closer to the real C-H distance in the radical. No major

angular changes are found between the used methods or basis sets.

2. Isotropic Hyperfine Couplings

a) Experimental hyperfine coupling constants

134

The hyperfine coupling constants of 1,3-BSQ radical have been

measured in three different solvents, namely acetone, acetonitrile

and water. As easily can be seen in Table 2, were all the available

experimental data [1,8-10] are collected, the hfcc’s are not

sensitive to the impact of solvent. Thus, the greater AH4,H6

isotropic hfcc’s are between 11.73G (for acetonitrile) and 11.44G

(for water). The intermediate value AH5 value is ranging from

2.53G (for acetonitrile) to 2.43G (for water) and the smallest

proton isotropic hfcc correspond to H2 nucleus (0.55G for

acetonitrile and 0.68G for water).

b) Calculated data

For this kind of radicals the unpaired electron is located in a π-

type orbital and hence, unpaired spin density arises at the proton

positions via a delicate balance of various spin polarization

mechanism [33,38,39]. The single occupied molecular orbital

(SOMO) plot is shown in Figure 2. As seen, for the two lower

Table 2. Experimental hyperfine coupling constants of 1,3-BSQ radical in different solvents (absolute values, in Gauss)a)

Solvent Dielectric constant

AH2 AH4,H6 AH5 AC1,C3 AC2 AC4,C6 AC5

acetone 20.7 0.76 11.62 2.5 n.a. n.a. n.a. n.a. acetonitrile 36.6 0.55 11.73 2.53 n.a. n.a. n.a. n.a.

water 78.4 0.68 11.44 2.43 5.9 1.3 15.0 12.3

n.a. - not available a) 1H hfcc's from [1,8,9]; 13C hfcc's from [10]

Fig.2 The highest single occupied molecular orbitals for 1,3-BSQ radical at UHF/6-31+G(d) level contoured at 0.001 /au3. The radical orientation is as shown in Figure 1. Top: C2v symmetry, 2B1 state; bottom left: C2v symmetry, 2A2 state; bottom right: C2v symmetry, 2A" state

135

energy states 2A2 and 2A" the SOMO is bonding between C1, C2

and C3 and has minor contribution from the others carbon atoms.

In the case of C2v(2B1) wave-function, this orbital has major

contributions from pz atomic orbitals of C4, C6, C1, C3 and the

two oxygen atoms.

136

Table 3. Calculated hyperfine coupling constants of gas-phase 1,3-BSQ radical with Cs symmetry using different methods and basis sets.

UHF/ 6-31+G(d)

UHF/ EPR-II

B3LYP/ 6-31+G(d)

B3LYP/ EPR-II

B3LYP/ EPR-III

B3LYP/ 6-31+G(d)

grid ultrafinea)

B3LYP/ EPR-II

grid ultrafinea)

Energy (a.u.)

-379.280 -379.340 -381.530 -381.570 -381.680 -381.53 -381.57

<S2> 0.96 / 0.77 0.97 / 0.77

0.78 / 0.75 0.78 / 0.75 0.77 / 0.75 0.77/0.75 0.77/0.75

Dipole moment (debye)

6.02 5.80 5.29 4.96 5.06 5.26 4.98

AH2 -1.48 -1.51 -0.01 -0.03 -0.10 0.19 0.27

AH4,H6 -32.91 -31.32 -12.64 -12.32 -11.99 -12.61 -11.76

AH5 22.10 21.30 3.41 3.37 3.21 3.30 2.87

AC1,C3 -16.45 -19.28 -5.64 -7.11 -6.34 -5.51 -6.75

AC2 -0.46 -0.83 -1.98 -1.80 -1.73 -2.33 -2.50

AC4,C6 60.21 48.20 22.19 15.20 14.63 22.37 15.12

AC5 -53.18 -50.97 -13.02 -13.02 -12.42 -12.96 12.91

AINDOH2 1.12 1.09 2.16 2.10 2.36 - -

AINDOH4,H6 -15.15 -15.34 -14.36 -14.63 -13.70 - -

AINDOH5 7.19 7.35 6.68 7.08 6.45 - -

AINDOC1,C3 -11.3 -11.3 -12.16 -12.16 - - -

AINDOC2 -5.49 -5.44 -6.84 -6.73 - - -

AINDOC4,C6 35.86 36.23 33.60 34.18 - - -

AINDOC5 -24.29 -24.61 -23.04 -23.52 - - -

a) Single point calculations on the optimized geometry obtained by the corresponding method.

137

The optimized geometries of the radical are those given in

Table 1 and the computed hfcc’s for Cs symmetry are shown in

Table 3. AINDO means computed isotropic hyperfine coupling

constants by INDO method on the optimized geometries obtained

by corresponding methods. For C2v(2A2) structure we have

obtained identical results with those given in Table 3, while for 2B1 wave-function, theoretical results differ drastically from the

experimental data. Thus, the B3LYP/EPR-II calculated values are

in this case: -16.32G, 0.32G and -0.72G for H2, H4,6 and H5

respectively. Obviously, this isomer can not be responsible for the

experimental pattern of the ESR spectrum of 1,3-BSQ radical.

The third column in Table 3 contains the expectation value

of the spin operator S2. A high spin contamination of the

wavefunction, reflected in large <S2> values, can affect the

geometry and population analysis and significantly alter the spin

density. It is generally accepted [40] that the spin contamination is

negligible if the value <S2> differs from S(S+1) (i.e. 0.750 for

pure doublet states) by less than 10%.

It is well known that the UHF approach has the severe

drawback of providing wave functions that are not exact

eigenfunctions of the S2 operator because the description of the

doublet ground state of radicals is contamined by some

contributions of higher spin multiplicities (quartet, sextet, etc.).

This spin contamination can be annihilated by techniques already

implemented in the Gaussian 98W program package [39,41]. In

138

Table 3 the <S2> values are given before/after spin annihilation.

Even after removing the spin contamination, the UHF method

give very unsatisfactory hfcc’s, clearly due to the overestimation

of spin polarization, especially on H5, H4 and H6 protons. As

seen in Table 3, excepting C2 nucleus, the 13C hyperfine splittings

are also overestimated by UHF method.

On the other hand, the corresponding <S2> value obtained

by pure functionals in the unrestricted Kohn-Sham (UKS)

formalism upon which the DFT methods are based [42], is very

close to the theoretical value for the pure spin state. However, for

the hybrid B3LYP functional the spin contamination can not be

neglected due to the inclusion of the exact exchange energy given

by Hartree-Fock theory. Anyway, a substantial improving of the

calculated AH4,H6 and AH5 values is obtained in this case, values

which are predicted in qualitative agreement with the

corresponding experimental counterparts. Contrary to the UHF

formalism, DFT method underestimates the spin polarization of 1s

orbitals of the H2 atom.

The INDO method has been tested in calculating the hfcc’s

for the radical in gas phase optimized at different levels of theory.

As seen in Table 3, with this method a fairly good agreement was

obtained between the experimental data and calculated hfcc’s

values. Important discrepancies are noted however for H5 atom.

For 13C hyperfine splittings, even more important deviations are

noted.

139

c) Solvent Effects

In order to test the influence of environmental surrounding on the

hyperfine structure of 1,3-BSQ, we used the Isodensity

Polarizable Continuum Model (IPCM) [43]. The hfcc’s have been

calculated in single point runs for all the solvents for which

experimental data exist and the results are summarized in Table 4.

For these calculations we used the optimized B3LYP/EPR-II

geometry of the gas-phase radical in Cs(2A") symmetry. The

molecular structure of the radical was not optimized for each

different value of the dielectric constant of the solvent because it

was shown that the influence of the solvent field on the

geometrical parameters is minimal [19].

As shown in Fig.3, the solvent has a very minor influence on the

spin densities and hence on the hfcc's of the radical. The most

important changes in Mulliken spin densities are noted for C1 and

C3 atoms. For this reason we do not expect important differences

between the hfcc's of the gas-phase and solvated 1,3-BSQ radical.

Fig.3 Mulliken spin densities on atomic centers in gas phase and solvated (water) 1,3-BSQ radical.

140

For the IPCM solvation model we must provide the isodensity

value, an empirical parameter which characterize the isosurface of

the total molecular electron density. As seen in Table 4, the most

affected calculated hfcc by varying the isodensity parameter is

AH2.

However, especially for this hyperfine coupling, the agreement

between the experimental and calculated values is the most

unsatisfactory. It is worth to note that in the case of 1,2-BSQ an

important dependence of the calculated hfcc's on solvent effects

was noted [19,21]. In that case, experimental data was reproduced

only after an appropriate choice of the isodensity parameter.

138

Table 4. B3LYP/EPR-II calculated isotropic hyperfine couplings of 1,3-BSQ radical on B3LYP/EPR-II gas-phase geometry by using IPCM solvation model.

Solvent Acetone Acetonitrile Water

Isodensity Dipole

moment AH2 AH4,H6 AH5

Dipole moment

AH2 AH4,H6 AH5 Dipole

moment AH2 AH4,H6 AH5 AC1,C3 AC2 AC4,C6 AC5

(Debye) (G) (G) (G) (Debye) (G) (G) (G) (Debye) (G) (G) (G) (G) (G) (G) (G)

0.001 6.91 -0.08 -12.27 3.45 6.97 -0.08 -12.26 3.45 7.02 -0.08 -12.26 3.46 -6.40 -2.09 15.51 -13.39 0.002 7.36 -0.10 -12.23 3.47 7.44 -0.10 -12.22 3.47 7.50 -0.10 -12.21 3.47 -6.07 -2.18 15.44 -13.41 0.003 7.52 -0.11 -12.20 3.47 7.61 -0.11 -12.18 3.47 7.68 -0.11 -12.17 3.47 -5.88 -2.23 15.40 -13.42 0.004 7.39 -0.12 -12.19 3.48 7.48 -0.11 -12.17 3.48 7.55 -0.11 -12.16 3.48 -5.85 -2.20 15.45 -13.48 0.005 7.01 -0.13 -12.21 3.49 7.09 -0.13 -12.19 3.49 7.15 -0.13 -12.18 3.49 -5.97 -2.08 15.61 -13.65 0.006 6.49 -0.15 -12.23 3.50 6.55 -0.15 -12.22 3.50 6.60 -0.15 -12.21 3.50 -6.17 -1.93 15.81 -13.67 0.007 6.49 -0.15 -12.22 3.50 6.55 -0.15 -12.21 3.50 6.60 -0.15 -12.20 3.50 -6.13 -1.93 15.80 -13.67 0.008 6.48 -0.15 -12.22 3.49 6.55 -0.15 -12.21 3.50 6.60 -0.15 -12.20 3.50 -6.10 -1.93 15.79 -13.67 0.009 6.46 -0.15 -12.22 3.49 6.53 -0.15 -12.21 3.50 6.58 -0.15 -12.20 3.50 -6.08 -1.93 15.79 -13.68

138

For 1,4-BSQ instead, the experimental hyperfine couplings are very well

reproduced even for the gas-phase radical.

Due to the small spin density on the C2 atom, an important

perturbation is expected from the large spin density located on the

neighboring C1 and C3 atoms. Moreover, the strong variation of the spin

density in C1C2C3 region implies also a marked non-uniformity of the

electronic density. For this reason we calculated again the spin densities and

hfcc's of the radical by choosing the ultrafine grid for the integration of the

electron density. A significantly better agreement between experimental and

calculated data was obtained in this way. Thus, AH2 value is 0.27G, AH4,H6=-

11.76G and AH5=2.87G. It is important to note the change in sign for AH2

isotropic hyperfine coupling constant, whose positive value suggest a

negative spin density on C2 atom. These results show that for this radical

the unpaired spin density does not have an odd-alternant pattern, as it was

observed for phenoxyl [44] or 1,2-BSQ [19,21].

Conclusions

In the present study, geometry and hyperfine coupling constants for

1,2-benzosemiquinone anion radical were calculated through the use of

INDO, Ab Initio, and Density Functional methods.

The possible geometrical structures of this radical are C2v(2B1), C2v(

2A2) and

Cs(2A"), the former being 5.17Kcal/mol higher in energy than the others.

The C2v(2A2) structure may be regarded as a transition state between the two

equivalent mirror image Cs structures.

142

Based on the calculated hfcc's it is shown that C2v(2B1) structure can

not be responsible for the ESR spectrum of the radical due to the fact that in

this case theoretical results differ drastically form the experimental data. For

Cs and C2v(2A2) structures a very good agreement has been obtained

between theoretical and experimental data by using the ultrafine grid for

integration of electronic density. This fact suggest a marked non-uniformity

of the electron density for this radical.

Theoretical results show a different pattern of the unpaired spin

density distribution from that observed for phenoxyl or 1,2-benzo-

semiquinone radical.

It is also shown that solvent effects have minor influence on the

hyperfine coupling constants corresponding to the protons but also to 13C

nuclei of the radical.

References

1. G.N.R.Tripathi, D.M.Chipman, C.A.Miderski, H.F.Davis,

R.W.Fessenden, R.H. Schuler, J.Phys.Chem., 90,3968(1986)

2. S.Patai (Ed.), The Chemistry of Quinoid Compounds, John Wiley and

Sons, New York, 1974

3. B.Trumpower (Ed.), Functions of Quinones in Energy Converting

Systems, Academic, New York, 1982

4. M.Y.Okamura, G.Feher, Annu. Rev. Biochem., 61, 881(1992)

5. H.Ding, C.C.Moser, D.E.Robertson, M.K.Tokito, F.Daldal, P.L.Duttion,

Biochemistry, 34, 11606(1995]

6. J.P.Klinman, M.David, Annu. Rev. Biochem., 63, 299(1994)

7. C.G.Pierpont, C.W.Lange, Prog. Inorg. Chem., 41, 331(1994)

142

8. J.A.Pedersen, CRC Handbook of EPR Spectra from Quinones and

Quinols, CRC Press, Boca Raton, Florida, 1985

9. C.Jinot, K.P.Madden, R.H.Schuler, J.Phys.Chem., 90, 4979-

4981(1986)

10. K.P.Madden, H.J.McManus, R.H.Schuler, J.Phys.Chem., 86, 2926-

2929(1982)

11. J.Pilar, J.Phys.Chem., 74, 4029-4037(1970)

12. X.Zhao, H.Imahori, C.G.Zhan, Y.Sakata, S.Iwata, T.Kitagawa,

J.Phys.Chem. A, 101,622-631(1997)

13. V.Barone, C.Adamo, N.Russo, Chem.Phys.Lett., 5,212(1993)

14. L.A.Eriksson, P.L.Malkina, V.G.Malkin, D.R.Salahub, J.Chem.Phys,

100,5066(1994)

15. V.Barone, in: D.P.Chong (Ed.) Recent Advances in Density Functional

Methods, Part I , World Scientific Publishing, Singapore, 287(1995)

16. C.Adamo, V.Barone, A.Fortunelli, J.Chem.Phys., 102,384(1995)

17. V.G.Malkin, O.L.Malkina, L.A.Eriksson, D.R.Salahub, in:

J.M.Seminario and P.Politzer (Eds.) Modern Density Functional Theory:

A Tool for Chemistry, p.273, Elsevier Science, 1995

18. P.J.O'Malley, J.Phys.Chem., 101,6334(1997)

19. M.Langaard, J.Spanget-Larsen,J.Molec.Struct.(Theochem),

431,173(1998)

20. D.M.Chipman, J,Phys.Chem. A, 104,11816-11821(2000)

21. V.Chiş, A.Nemeş, L.David, O.Cozar, Studia UBB, Physica, 47(1)157-

170(2002)

22. J.A.Weil, J.R.Bolton, J.E.Wertz, Electron Paramagnetic Resonance,

Wiley, New York, 1994

23. V.Chiş, L.David, O.Cozar, A.Chiş, Rom.J.Phys., 48,429-4438(2003)

142

24. W.J.Hehre, L.Radom, P.v.R.Schleyer, J.A.Pople, “Ab Initio” Molecular

Orbital Theory , John Wiley & Sons, New York, 1986

25. N.Rega, M.Cossi, V.Barone, J.Chem.Phys., 105,11060(1996)

26. A.D.Becke, J.Chem.Phys., 98,5648(1993)

27. C.Lee, W.Yang, R.G.Parr, Phys.Rev.B, 37,785(1998)

28. S.Naumov, A.Barthel, J.Reinhold, J.Geimer, D.Beckert,

Phys.Chem.Chem.Phys., 2,4207(2000)

29. J.W.Gauld, L.A.Eriksson, L.Radom, J.Phys.Chem.A, 101,1352(1997)

30. J.A.Pople, D.L.Beveridge, Approximate Molecular Orbital Theory,

McGraw-Hill, New York, 1970

31. M.W.Wong, M.J.Frisch, K.B.Wiberg, J.Am.Chem.Soc.,

113,4776(1991)

32. M.Karelson, T.Tamm, M.C.Zerner, J.Phys.Chem., 97,11901(1993)

33. W. Gordy, Theory and Applications of Electron Spin Resonance, Wiley,

New York, 1980

34. A.L.Macdonald, J.Trotter, J.Chem.Soc.Perkin Trans., 2, 473(1973)

35. D.E.Wheeler, J.H.Rodriguez, J.K.McCusker, J.Phys.Chem. A,

103,4101-4112(1999)

36. C-G. Zhan, D.M.Chipman, J.Phys.Chem. A, 102,1230-1235(1998)

37. G.Portalone, G.Schultz, A.Domenicano, I.Hargittai, Chem.Phys.Lett.,

197,482(1992)

38. M.J.Cohen, D.P.Chong, Chem.Phys.Lett., 234,405(1995)

39. D.M.Chipman, Theor.Chim. Acta, 82, 93-115(1992)

40. M.W.Wong, L.Radom, J.Phys.Chem., 99,8582(1995)

41. Gaussian 98, Revision A.7, M. J. Frisch, G. W. Trucks, H. B. Schlegel,

G. E. Scuseria, M. A. Robb, J. R. Cheeseman, V. G. Zakrzewski, J. A.

Montgomery, Jr., R. E. Stratmann, J. C. Burant, S. Dapprich, J. M.

142

Millam, A. D. Daniels, K. N. Kudin, M. C. Strain, O. Farkas, J. Tomasi,

V. Barone, M. Cossi, R. Cammi, B. Mennucci, C. Pomelli, C. Adamo,

S. Clifford, J. Ochterski, G. A. Petersson, P. Y. Ayala, Q. Cui, K.

Morokuma, D. K. Malick, A. D. Rabuck, K. Raghavachari, J. B.

Foresman, J. Cioslowski, J. V. Ortiz, A. G. Baboul, B. B. Stefanov, G.

Liu, A. Liashenko, P. Piskorz, I. Komaromi, R. Gomperts, R. L. Martin,

D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng, A. Nanayakkara, C.

Gonzalez, M. Challacombe, P. M. W. Gill, B. Johnson, W. Chen, M.

W. Wong, J. L. Andres, C. Gonzalez, M. Head-Gordon, E. S. Replogle,

and J. A. Pople, Gaussian, Inc., Pittsburgh PA, (1998)

42. R.G.Parr, W.Yang, Density Functional Theory of atoms and molecules,

Oxford University Press, 1989

43. S.Miertus, S.Scrocco, J.Tomasi, Chem.Phys. 55,117(1981)

44. M.Lucarini, V.Mugnaini, G.F.Pedulli, M.Guerra, J.Am.Chem.Soc., 125,

8318-8329(2003)