Electron transfer reactions of extended o-, p-quinones - Voltammetric and EPR/ENDOR spectroscopic investigations

Article abstract of DOI:10.1039/a705464e

The electrochemical properties of four ortho-, para-extended quinones/quinonemethides ? 1-4 have been investigated using cyclic voltammetry (CV) and normal pulse voltammetry (NPV). The species are reduced by addition of two, four or six electrons to form

Full text of DOI:10.1039/a705464e

View Article Online / Journal Homepage / Table of Contents for this issue
Electron transfer reactions of extended o-, p-quinones—voltammetric
and EPR/ENDOR spectroscopic investigations
a
a
a,b
Jinkui Zhou, Michael Felderhoff, Natalja Smelkova (née Timoschkova),
b
,a
Leonid M. Gornastaev and Anton Rieker*
a
Institute of Organic Chemistry, University of Tübingen, Auf der Morgenstelle 18,
D-72076 Tübingen, Germany
Chair of Chemistry, Krasnoyarsk State Pedagogical University, Lebedevoy Str. 89,
b
6
60049 Krasnoyarsk, Russia
The electrochemical properties of four ortho-, para-extended quinones/quinonemethides† 1–4 have been
investigated using cyclic voltammetry (CV) and normal pulse voltammetry (NPV). The species are reduced
by addition of two, four or six electrons to form a dianion (1), tetraanions (2, 3) and a hexaanion (4), all
of which can be re-oxidized to the corresponding quinonoid species. These compounds, thus, exhibit
a multistep redox behaviour, acting as good (quasi)reversible electron acceptors with well separated
potential steps, some of them corresponding to a formal two-electron transfer. The radical anions
Ϫ
᝽
Ϫ
ؒ
3Ϫ
1
–4 and the radical trianion 3 , prepared electrochemically in the first and third reduction steps,
᝽
are persistent for several hours in the absence of air. They have been characterized by EPR and ENDOR
spectroscopy, revealing a delocalization of the odd electron. The areas of preferential spin residing,
ؒ
3Ϫ
Ϫ
however, are quite different in radical trianion 3
a strong influence of the charge state in these radical anions.
as compared to radical monoanion 3᝽
, indicating
Introduction
2
1
O
Quinones are indispensable in organic electrochemistry due to
O
3
4
1
A
their multistep redox properties and they play a substantial
6
6
′
2
1′
role in biological electron transfer chains. Like quinones, their
anion radicals attract interest in biology, medicine and chem-
5
′
5
B
2
′
1,3
4′
istry. Extended quinones, on the other hand, have found
increasing application as electron acceptors for the production
3′
4–8
1
of organic conducting materials. Recently, we reported the
electrochemical and spectroscopic properties of a series of
9
extended p-quinones. One important aspect of these investig-
5′
4′
3′
O
O
6′
1′
ations was to mark out the structural prerequisites for the use
of p-phenylene units as ferromagnetic couplers. With a similar
aim we investigated the ortho-, para-extended quinones/quin-
onemethides (1–4), where the internal steric strain and the s-cis
5
A
6
7
4
4
a
3
2′
C
B
1
8
a
2
O
8
O
O
2
O
α β
O
O
O
6′
1′
I
5
A
6
7
4
β-keto group (structural element I) should provide further
interesting properties. Such quinones are also of interest, be-
cause some of their partially reduced forerunners, i.e. hydroxy-
4a
3
2′
2′′
C
B
1
8
a
2
1′′
8
D
10
O
phenylquinones, are potential 5-lipoxygenase blockers.
6′′
Here, we describe the voltammetric properties of these com-
pounds using cyclic voltammetry (CV) and normal pulse vol-
tammetry (NPV), as well as the EPR/ENDOR characterization
of the radical species obtained on their reduction.
3
O
O
O
O
O
Experimental
4′′′
4′
C 1′′′
A
1′
4
Instrumentation
5
6
3
2
B
1
Elemental analyses were performed using a Carlo Erba Elem-
ental Analyzer-1106. IR spectra were recorded on a Perkin-
1
′′′′
1′′
D 4′′
4
′′′′ E
O
†
IUPAC names for o- and p-quinonemethide are 6-methylenecyclo-
hexa-2,4-dien-1-one and 4-methylenecyclohexa-2,5-dien-1-one.
4
J. Chem. Soc., Perkin Trans. 2, 1998
343

View Article Online
Elmer 281 spectrometer. NMR spectra were recorded on
1
13
Bruker AC-250 and WM 400 spectrometers. All H and
C
OH
OH
NMR chemical shifts are given relative to SiMe as internal
O
O
O
O
4
standard and for CDCl as solvent; the coupling constants (J)
3
i
are reported in Hz. Mass spectra were recorded with Finigan
MAT 711A or TSQ spectrometers. EPR and ENDOR spectra
were obtained with a Bruker ESP 300E spectrometer. For
g-factor measurements, the field gradients were corrected by
replacing the sample with a reference compound (2,6-di-tert-
butyl-4-tert-butoxyphenoxyl in benzene, g = 2.004 63).
Cl
OH
5
6
ii
iii
Computer simulations of EPR spectra were performed with
‘
programme PIP’. The EPR-analysis software was furnished
by Illinois EPR-Research Center, NIH division of research
resources grant No. RR 001811, by R. L. Belford, A. M.
Maurice, M. J. Nilges, University of Illinois, Urbana.
Electrochemical experiments were carried out at 25 ЊC under
argon using a BAS CV-50W electrochemical analyzer (Bio-
analytical Systems). For CV and NPV a glassy carbon disk
O
O
O
O
2
electrode with an electroactive area of 0.07 cm was used as
2
working electrode. The auxiliary electrode consisted of a Pt
Scheme 1 Reagents and conditions: i, K CO , MeOH–H O; ii, NaNO ,
2
3
2
2
wire. Ag/AgClO (0.01  in MeCN/0.1  NBu PF ) was used as
4
4
6
DMSO; iii, CAN
reference electrode and separated by two glass frits from the
Haber–Luggin capillary. All potentials given relate to this refer-
ence electrode. The supporting electrolyte consisted of a 0.1 
solution of NBu PF in pyridine. Pyridine was purified by dis-
4
3
4
(1 H, d, J 2.6, H-2Ј,6Ј), 7.84 (1 H, td, J 7.5, J 1.6, H-6), 7.92 (1
3
4
4
H, td, J 7.5, J 1.6, H-7), 8.04 (1 H, d, J 2.6, H-2Ј,6Ј), 8.25 (1
4
6
3
4
3
4
tilling it three times under argon after drying over KOH for
several weeks, and was then kept over 4 Å molecular sieves.
NBu PF was prepared from NBu Br and NH PF , recrystal-
H, dd, J 7.6, J 1.3, H-5), 8.30 (1 H, dd, J 7.8, J 1.3, H-8);
δ (62.9 MHz) 29.69 (18 Me), 36.40 (2 Me C); 128.41, 128.82,
C
3
129.00, 134.97, 136.31 (C-5, C-6, C-7, C-8, C-2Ј, C-6Ј), 132.54,
137.01, 137.51 (C-3, C-4a, C-8a), 147.67 (C-1Ј), 154.86, 155.05
(C-3Ј, C-5Ј), 180.40, 184.86, 185.03, 187.11 (C-1, C-2, C-4,
C-4Ј).
4
6
4
4
6
lized four times from EtOH and dried in vacuo at 110 ЊC for
8 h. The number of electrons transferred per molecule was
4
confirmed by controlled-potential electrolysis (CPE) at the
relevant potentials, chosen from the corresponding CVs. For
CPE, cylindrical Pt (10% iridium) gauze working and auxiliary
electrodes, separated by a porous glass frit, were used.
2,3-Bis(3,5-di-tert-butyl-4-oxocyclohexa-2,5-dien-1-ylidene)-
1,2,3,4-tetrahydronaphthalene-1,4-dione 3. Cerium() ammon-
ium nitrate [Ce(NH ) (NO ) ] (1.10 g, 2.0 mmol) in a mixture of
4
2
3 6
The anion radicals were generated in ca. 0.5 m solutions of
the quinones 1–4 in a CPE cell (see above) under argon at 25 ЊC.
Upon completion of the reduction (current drop at constant
potential) they were transferred to the EPR tubes in an argon
box. Intra muros generation (without potential control) gave
almost identical spectra. During the ENDOR measurements,
the electrolysis had to be interrupted.
5 ml of acetonitrile and 5 ml of water was added to 2,3-bis(3,5-
14
di-tert-butyl-4-hydroxyphenyl)-1,4-naphthoquinone (0.57 g,
1.0 mmol) in 6 ml of acetonitrile. After 5 min of stirring the
precipitate was filtered off and washed with 5 ml of methanol.
Yield 63%, mp 190–191 ЊC (Found: C, 81.12; H, 7.93. Calc. for
C H O : C, 80.85; H, 7.80%); δ (250.1 MHz) 1.16 (18 H, s,
3
8
44
4
H
t
t
4
Bu ), 1.32 (18 H, s, Bu ), 6.85 (2 H, d, J 2.6, Hquin), 8.11 (2 H, d,
4
J 2.6, Hquin), 7.80–7.85 (2 H, m, Harom), 8.18–8.23 (2 H, m,
Syntheses
Harom); δ (100.6 MHz) 29.33, 29.63 (12 Me), 35.71, 35.90 (4
C
2
,6-Di-tert-butyl-4-(3,5-di-tert-butyl-6-oxocyclohexa-2,4-
Me C), 127.29, 129.04, 134.90, 135.12, 137.84, 140.06, 151.78,
3
12
Ϫ1
dien-1-ylidene)cyclohexa-2,5-dien-1-one 1. Compound 1 was
prepared by oxidation of the corresponding biphenol, 3,3Ј,5,5Ј-
tetra-tert-butylbiphenyl-4,2Ј-diol,
δC(100.6 MHz) 29.17, 29.68, 29.72, 29.75 (12 Me), 35.28, 35.53,
5.78, 36.07 (4 Me C), 118.67, 128.67, 131.55, 131.74, 138.66,
40.31, 148.79, 148.90, 150.72, 151.36 (10 Carom), 186.61 (C-1),
91.23 (C-6Ј); m/z (EI) 410 (M ϩ 2, 5%), 393 (M Ϫ Me, 4),
51 (M Ϫ Bu , 31), 337 (M Ϫ Me᎐Bu , 30), 57 (Bu , 100), see
151.93 (18 Carom and Colefinic), 186.67, 186.75 (4 CO); νmax/cm
ϩ
(KBr) 1676, 1616 (C᎐O); m/z (FD) 564 (M ).
11,12
with PbO2 in toluene.
2,3,5,6-Tetrakis(3,5-di-tert-butyl-4-oxocyclohexa-2,5-dien-1-
ylidene)cyclohexane-1,4-dione 4. Compound 4 was prepared in
a two step synthesis starting with the substitution of chlorine
in tetrachloro-1,4-benzoquinone by 2,6-di-tert-butylphenol to
3
1
1
3
3
ϩ
ϩ
give 7, followed by oxidation of 7 with PbO in toluene (Scheme
2
ϩ
t
ϩ
t
t
2) as described below.
also ref. 13.
2,3,5,6-Tetrakis(3,5-di-tert-butyl-4-hydroxyphenyl)-1,4-benzo-
quinone 7.—Tetrachloro-1,4-benzoquinone (0.25 g, 1.0 mmol),
2,6-di-tert-butylphenol (0.82 g, 4.0 mmol) and potassium car-
bonate (0.56 g, 5.0 mmol) were dissolved in 10 ml of DMSO
and heated to 100 ЊC for 2 h. The solution was diluted with
50 ml of water and extracted three times with chloroform.
The organic layers were separated and dried over potassium
carbonate. After filtration and evaporation in vacuo, the residue
was chromatographed on silica gel with toluene. Yield 0.37 g
(40%), deep red crystals, mp >300 ЊC (Found: C, 79.99; H,
9.32. Calc. for C H O : C, 80.47; H, 9.15%); δ (400.1 MHz)
3
-(3,5-Di-tert-butyl-4-oxocyclohexa-2,5-dien-1-ylidene)-
1
,2,3,4-tetrahydronaphthalene-1,2,4-trione 2. Compound 2 had
10
been synthesized by Wurm, starting from 2-(3,5-di-tert-butyl-
-hydroxyphenyl)-3-chloro-1,4-naphthoquinone 5 via 2-(3,5-di-
tert-butyl-4-hydroxyphenyl)-3-hydroxy-1,4-naphthoquinone
Scheme 1) and oxidation of the latter with cerium() am-
monium nitrate [Ce(NH ) (NO ) , CAN]. Although not quoted
4
6
(
4
2
3 6
10
by Wurm, 5 had already been synthesized earlier by Russkikh
14
et al. We found that 5 can be directly converted to 2 by
hydrolysis and oxidation with sodium nitrite.
6
2
84
6
H
t
2
-(3,5-Di-tert-butyl-4-hydroxyphenyl)-3-chloro-1,4-naph-
1.27 (72 H, s, Bu ), 5.16 (4 H, br s, OH), 6.84 (8 H, s, Harom);
thoquinone 5 (0.4 g, 1.0 mmol) and NaNO2 (0.069 g, 1.0
mmol) in 3 ml of DMSO were stirred for 3 h at 25 ЊC. The
brown precipitate of 2 was filtered, washed with water and
recrystallized from hexane. Yield 0.27 g (71%), mp 224–226 ЊC
δ (100.6 MHz) 30.33 (24 Me), 34.18 (8 Me C), 125.19, 128.30,
C
3
134.98, 143.95, 153.58 (28 C
and Cquin), 187.61 (2 CO); νmax/
arom
Ϫ1
ϩ
cm (KBr) 3645 (OH), 1657 (C᎐O); m/z (FD) 924 (M ).
2,3,5,6-Tetrakis(3,5-di-tert-butyl-4-oxocyclohexa-2,5-dien-1-
ylidene)cyclohexane-1,4-dione 4.—Compound 4 was prepared
by oxidation of 7 (92 mg, 0.10 mmol) with a four-fold excess
(
Found: C, 76.73; H, 6.33. Calc. for C H O : C, 76.57; H,
24
24
4
t
t
6
.43%); δ (250.1 MHz) 1.29 (9 H, s, Bu ), 1.30 (9 H, s, Bu ), 7.79
H
3
44
J. Chem. Soc., Perkin Trans. 2, 1998

View Article Online
o
a
Table 1 Formal potentials (E /mV) and potential separation (∆E/mV)
o1
1
o2
2
o3
3
o4
4
Comp.
E
∆E
E
∆E
E
∆E
E
∆E
1
2
3
4
Ϫ830
Ϫ425
Ϫ650
Ϫ405
82
82
88
79
Ϫ1365
Ϫ1065
Ϫ1110
Ϫ763
84
83
89
77
b
Ϫ1955
130
90
85
Ϫ1700b
Ϫ2100b
Ϫ2190
160
180
Ϫ1180
a
o
r
p
o
Potentials vs. Ag/(0.01 ) AgClO electrode; E , the mean value of E (reduction peak potential) and Ep (oxidation peak potential); ∆E =
4
r
o b
Ep Ϫ E . Formal two-electron transfer.
p
O
OH
Cl
Cl
Cl
Cl
+
4
i
O
HO
OH
OH
O
ii
4
O
HO
Fig. 1 Cyclic voltammograms of 1–4 in pyridine solution; potential, vs.
7
Ϫ1
Ag/AgClO (0.01  in MeCN/0.1  NBu PF ); scan rate, 100 mV s
;
4
4
6
Scheme 2 Reagents and conditions: i, K CO –DMSO; ii, PbO –
(a) 1, (b) 2, (c) 3, (d) 4
2
3
2
Toluene
of PbO in toluene. After 30 min of stirring, PbO was filtered
2
2
off and the solvent removed in vacuo. Yield 77 mg (84%), deep
red crystals, mp 294 ЊC (Found: C, 80.80; H, 8.87. Calc. for
C H O : C, 80.83; H, 8.75%); δ (400.1 MHz) 1.17 (36 H, s,
6
2
80
6
H
t
t
4
Bu ), 1.36 (36 H, s, Bu ), 6.74 (4 H, d, J 2.5, Hquin), 8.35 (4 H, d,
J 2.5, Hquin); δ (100.6 MHz) 29.47, 29.71 (24 Me), 35.85, 36.17
4
C
(
8 Me C), 127.71, 129.12, 137.05, 140.50, 152.67, 152.68 (24
3
Ϫ1
Colefinic), 186.51 (4 CO), 187.01 (2 CO); νmax/cm (KBr) 1684
ϩ
(
sh), 1631, 1620 (C᎐O); m/z (EI) 920 (M ).
᎐
Results and discussion
Electrochemical investigation
Cyclic (CV) and normal pulse (NPV) voltammograms of 1–4
are shown in Figs. 1 and 2; the electrochemical data are listed in
Table 1.
Fig. 2 Normal pulse voltammograms of 1–4 in pyridine solution;
potential, vs. Ag/AgClO (0.01 M in MeCN/0.1 M NBu PF ); scan rate,
0 mV s ; pulse amplitude, 50 mV; pulse width, 50 ms; pulse period,
00 ms; (a) 1, (b) 2, (c) 3, (d) 4
The o-, p-diphenoquinone 1 shows two well separated reduc-
4
4
6
Ϫ1
2
2
tion (Pr , Pr ) and re-oxidation peaks (Po , Po ) in CV [Fig.
1
2
1
2
1
(a)]. NPV [Fig. 2(a)] reveals equal currents of the first (Pr1)
and the second peak (Pr ), indicating that the same number of
2
electrons are exchanged during each electron-transfer process.
The combined results from CV, NPV and CPE confirm that 1 is
reduced in the first step by one-electron transfer to a radical
anion (which can further be characterized by EPR and
ENDOR spectroscopy, see below). This may be re-oxidized to
the quinone state or further reduced by a second electron-
transfer to a dianion. The latter is re-oxidizable to 1 in two
successive one-electron transfers.
reduced by incorporation of two more electrons to give the
tetraanion, which can be oxidized back to the neutral state in
the corresponding re-oxidation peaks. Because the third and the
fourth reduction step apparently occur at potentials very close
together, the two peaks overlap, and only one peak is observed
for both reduction steps.
The two-fold extended 3 exhibits CV [Fig. 1(c)] and NPV
[Fig. 2(c)] curves different from those of 2 in that the fourth
peak couple of reduction (Pr ) and re-oxidation (Po ) can be
The CV of 2 [Fig. 1(b)] shows three peak couples of reduc-
4
4
tion (Pr –Pr ) and re-oxidation (Po –Po ). The currents of Pr
observed due to the larger difference between the third and
1
3
1
3
1
and Pr are approximately equal, but they are much smaller
fourth reduction potentials than in the case of 2. Interestingly,
2
o
than that of Pr . The NPV of 2 is shown in Fig. 2(b). Here, the
the average of the E values for the third and fourth electron-
3
o
currents of the first peak (Pr ) and the second peak (Pr ) are
transfer in 3 is close to the E value for the third peak couple of
1
2
equal, as in the case of 1, and their sum is approximately equal
2. The peak currents of the fourth couple of 3 are smaller than
those of the other couples, and the potential separation
between the reduction and re-oxidation peak is here much
to the current of the third peak (Pr ), indicating that in the first
3
and second reduction step 2 is reduced via the anion radical
15,16
(
which gives an EPR signal, see below) to the dianion in two
larger than in the case of the other couples
(Table 1). Such
17–19
9a
consecutive one-electron transfers. The dianion is further
effects are known,
and we have discussed the first effect in
J. Chem. Soc., Perkin Trans. 2, 1998
345

View Article Online
terms of a lower electron-transfer rate‡ and a high repellent
17
effect of the multiply charged anions towards the cathode.
The second effect, and especially the lower re-oxidation peak of
the fourth couple, might be due to a reversible chemical reaction.
16
In similar cases, dimerization of radical anions or a combin-
20
ation of radical anion and parent quinone have been postu-
lated. In the present case, reversible protonation–deprotonation
reactions involving the highly negatively charged anions, even
in pyridine (containing trace of water) seem to be more
reasonable.
CV and NPV spectra of 4 are shown in Figs. 1(d) and 2(d).
They resemble the voltammograms of 2, as far as regards the
reduction peaks Pr –Pr . The currents of Pr and Pr2 are
1
3
1
approximately the same, and each of them is only half of the
current of Pr , indicating that 4 undergoes two successive one-
3
electron reductions up to the dianion via a monoanion (which
is EPR-active, see below). Obviously, the dianion is further
reduced in a two-electron step to the tetraanion. A fourth two-
electron transfer peak couple, although less reversible, reveals
that the tetraanion is further reducible to a hexaanion. All
anion species can be re-oxidized to the neutral 4. The potential
separations between the third and fourth, and between the fifth
21
and sixth reduction step are very small, or even inverted, so
that each pair of processes exhibits only one peak. The current
of Pr is smaller than that of Pr , the reasons for this may be the
4
3
same as for the case of the fourth reduction peak of 3, discussed
above. Probably, there are again proton-transfer equilibria
coupled to the fifth and sixth electron-transfers, because the
fourth re-oxidation peak is relatively small when compared with
the corresponding reduction peak. This is reasonable since the
polyanions involved certainly are strong protophilic (basic)
species. Nevertheless, 4 displays six-step redox behaviour and is,
thus, capable of acting as an excellent electron acceptor.
Fig. 3Ϫ EPR spectra of the radical anions in pyridine solution at 298 K;
Ϫ
Ϫ
ؒ
3Ϫ
Ϫ
(a) 1
᝽ , (b) 2
, (c) 3᝽
, (d) 3
, (e) 4᝽
᝽
The positions of the CV peak potentials of 1–4 depend on
the scan rate. With increasing scan rate, the reduction peaks
r
p
(
E ) are shifted towards more negative potentials, whereas the
o
p
re-oxidation peaks (E ) move towards more positive poten-
tials. The resulting increase in the potential separations ∆E
p
o
r
(
=Ep Ϫ E ) indicates that the reduction of 1–4 is quasi-
p
15
reversible. From the structures of 1–4 and the experimental
results of CV and CPE measurements it can be derived that 1,
2
/3 and 4 are reducible with an overall two, four and six elec-
tron transfer to form a dianion, tetraanions and a hexaanion,
resepectively. Since all anion species can be re-oxidized up to
their neutral quinonoidal states, the electrochemical behaviour
of 1, 2/3 and 4 can be formally denoted as quasi-reversible EE,
EE(EE) and EE(EE)(EE) processes, respectively.
As can be seen from Table 1, the enlargement of the π-system
with an increasing number of keto groups changes the redox
potentials. The first reduction potentials become less negative
in the order 1 < 3 < 2 < 4, indicating that 2–4 possess a much
higher electron-accepting ability than 1, and 4 is the strongest
electron acceptor.
o
The ease of electron-uptake (expressed in E ) of the different
Fig. 4 ENDOR spectra of the radical anions in pyridine solution at
Ϫ
Ϫ
Ϫ
ؒ
3Ϫ
Ϫ
species of one extended quinone as well as of species with the
same charge but derived from different parent quinones reflects
the structural features of the respective compounds. This
information, although not complete, can be derived from the
magnetic resonance studies of the radical anions of 1–4.
233 K; (a) 1᝽
, (b) 2᝽ , (c) 3᝽
, (d) 3
, (e) 4᝽
spectra (Fig. 3) appeared in the g = 2.004 region, where the
signals of radical anions of quinones are usually observed. The
t
interaction of the odd electron with the protons of the Bu
groups could not be resolved by EPR spectroscopy. In addition,
ENDOR spectroscopy was used to characterize further the
EPR and ENDOR spectroscopy
Ϫ
Ϫ
ؒ
3Ϫ
Ϫ
Ϫ
ؒ
3Ϫ
The radical anions 1 –4 and radical trianion 3 should be
radicals 1 –4 and 3 . As shown in Fig. 4, the ENDOR
᝽
᝽
᝽
᝽
able to be electrochemically generated in the first and third
reduction steps. In pyridine solution at room temperature, the
species obtained proved to be persistent for several hours in
the absence of air, as shown by EPR investigation. The EPR
spectra at 233 K allow the resolution of the different proton
couplings. Even the interaction of the odd electron with the
t
protons of the Bu groups is resolvable by ENDOR spec-
troscopy. The hyperfine splitting constants (hfs) a and g tensor
H
values are listed in Table 2. The coupling constants (aH)
obtained at 233 K with ENDOR are approximately equal to
those obtained at room temperature with EPR spectroscopy. In
order to get further verification of the experimental results, a
‡
As stated correctly on p. 933 of ref. 9a; on p. 934, left column, in the
last sentence of the second paragraph, however, ks1 and ks2 should be
exchanged.
3
46
J. Chem. Soc., Perkin Trans. 2, 1998

View Article Online
Table 2 aH values
O
O
O
Radical
aH/G
Ϫ
a
c
c
c
O
O
1
᝽
Ϫ
1.05, 0.05, 0.16, 1.04
2
6
a
c
c
2
᝽
Ϫ
0.95, 0.10, 0.93
a a
b
b
c
c
c
c
2′
6′
3
0.64, 1.05, 0.65(2H), 1.05(2H), 0.06, 0.64, 1.10
b b b c c
᝽
ؒ
3Ϫ
3
0.12(2H), 0.25(4H), 0.65(2H), 0.12, 0.26, 0.65
a b c c
Ϫ
4᝽
0.43, 0.44 (8 H), 0.02, 0.43
a
c
b
O
Obtained by EPR at room temperature. Obtained by simulation.
Obtained by ENDOR at 233 K.
8
9
O
computer simulation of the more complicated EPR spectra of
O]–
Ϫ
Ϫ
ؒ
3Ϫ
the anion radicals 3᝽
/4
᝽
and 3 was carried out, furnishing
O
O
2′
very good agreement between experimental and simulated
curves. However, some assignments remain equivocal, since a
specific deuteriation was not undertaken. To make the discus-
sion easier, the rings in the species 1–4 are marked with the
letters A–E.
6′
′′
2
6′′
O]–
n
Let us start with the monoanions. The most straightforward
Ϫ
EPR spectrum seems to be that of radical anion 2 [Fig. 3(b)],
᝽
10
11
consisting of a triplet hfs with a = 0.95 G, which should be
H
assigned to the quinonoid ring protons H-2Ј,6Ј in moiety A.
This means that the two quinonoidal protons are equivalent
either accidentally or by rapid rotation (relative to the EPR
timescale) around the weakened exocyclic double bond. This
coupling can be found again in the ENDOR spectrum [Fig.
(a = 0.06 G) is observed. Therefore, it is more reasonable to
assume a delocalization of the free electron only to rings A, B
H
and D on a first-order treatment. Then, a = 1.10 and 0.64 G
H
should be assigned to the quinonoid protons H-2Ј,2Љ,6Ј,6Љ. Why
4
(b)], which reveals an additional small coupling expected for
these protons are magnetically non-equivalent in contrast to the
t
Ϫ
Ϫ
the Bu protons. The value of ca. 0.10 G is however too large
for Bu in the ortho-position to the quinonoid keto group. For
the anion radicals of para-extended quinones and the other
species of Table 1 the corresponding values are <0.06 G.
Thus, the splitting of 0.10 G may be due to a small coupling
of the free electron to protons in the naphthalene ring super-
imposed by an even smaller coupling caused by the Bu pro-
tons. Anyway, the observed pattern shows that the free
electron in 2 is mainly delocalized into moiety A and (pre-
sumably) B.
situation with 2 and 4 is not evident; presumably, 3 is more
᝽
᝽
t
rigid than 2 or 4. Interestingly, the average of the two coupling
9a
constants, aAV = 0.87 G, is just twice the a-value (0.43 G) for the
Ϫ
quinonoidal ring protons of 4 with a delocalization area of
᝽
Ϫ
about twice that of 3 , a fact which strongly supports the
᝽
above conclusions.
t
Ϫ
The EPR spectrum of 1 [Fig. 3(a)] surprisingly consists of
two broad lines (∆H = 1.14 G) with relative intensities 1:1, sug-
᝽
Ϫ
gesting a strong coupling of the odd electron to one ring proton
᝽
(a = 1.05 G) and relatively small couplings to other protons.
H
13
A simple EPR spectrum [Fig. 3(e)] is also observed for
Smaller signals at both wings may be due to C coupling. The
Ϫ
radical anion 4 : seven equidistant lines with a = 0.43 G. This
corresponding ENDOR [Fig. 4(a)] exhibits three line pairs,
᝽
H
t
indicates a coupling of the free electron with six equivalent
protons. Since there is no reason to assume that two ring pro-
tons should be different from six others and since the ENDOR
spectrum shows only one line pair (a = 0.43 G) in addition to
the line pair due to the Bu protons (a = 0.02 G), it is more
indicating an interaction with the protons of the Bu groups
(a = 0.05 G) and two types of aromatic protons. In analogy to
H
24
o-quinones it is reasonable to assign the largest coupling to
Ϫ
H-4Ј of the o-quinonoid part of 1 . The coupling constant
H
᝽
t
a = 0.16 G would then correspond to one or more of the
H
H
reasonable to assume equal coupling to all eight quinonoidal
ring protons. In principle, nine lines would then be expected,
however, the two lines at both wings are apparently so weak
that they cannot be observed. This was further verified by com-
puter simulation (eight equivalent protons) also showing only
seven lines on the same intensity scale. This means that the free
electron is delocalized over all rings A–E. The relatively small
remaining quinonoidal protons in both rings, which might be
accidentally magnetically equivalent. Since in the radical anion
of 3,3Ј,5,5Ј-tetra-tert-butyl-p-diphenoquinone 9 the ring pro-
9a
tons show a = 0.45 G, the above results, although being com-
H
patible with a delocalization of the free spin over the whole
π-system, suggest that the free electron is located mainly in the
o-quinonoidal moiety B of the molecule.
t
ؒ
3Ϫ
coupling constants for ring and Bu protons are in agreement
The only EPR-active multianion species is the trianion 3
.
with this assumption. Again, we are unsure, whether the eight
ring protons are either accidentally equivalent or whether the
rotation around the exocyclic double bonds is rapid. Similar
observations have been reported by West and co-workers in
the case of the radical anion of the 1,2-diquinocyclobutane-
Its EPR spectrum [Fig. 3(d)] with 21 lines is much more com-
Ϫ
plicated than that of 3 . The hfs constants aH obtained by
᝽
simulation (Table 2) are 0.12 (2 H), 0.25 (4 H) and 0.65 (2 H)
G. The ENDOR spectrum [Fig. 4(d)] also exhibits three line
pairs, confirming further the simulation results. In this case,
the coupling of the protons of the tert-butyl groups is too
weak to be observed. This is reasonable, if we assume a struc-
22
dione 8.
Ϫ
The EPR spectrum of radical anion 3 [Fig. 3(c)] exhibits
seven non-equidistant lines, the ENDOR spectrum [Fig. 4(c)]
᝽
2Ϫ
ture like 10 for the dianion 3 , in which the steric strain
between the two quinonoidal rings present in 3 is relieved due
to the twisting around the single bonds which now connect the
shows three line pairs, indicating that the odd electron is
coupled to two types of ring protons (a = 1.10 and 0.64 G) and
H
t
ؒ
3Ϫ
also interacts with the protons of the Bu groups (a = 0.06 G).
phenolate (A,D) and the quinone (B) rings. In 3 g, which can
be regarded as the semiquinone of 10, the coupling of the free
electron with the now equivalent protons H-2Ј,2Љ,6Ј,6Љ seems to
occur by π–σ interaction and is therefore small. For the same
H
The simulation of the EPR spectrum reveals a multiplicity of
two for each of the two coupling proton types. At first glance,
this pattern would suggest a coupling of the free electron with
the 2 × 2 protons of the naphthalene ring. However, the coup-
ling constants seem to be too large for a 1,4-naphthosemi-
t
reason, a coupling with the Bu protons is no longer possible.
Thus, we see the dramatic effect of the charge state which
strongly influences the spin density distributions and hence the
23
t
quinone and, moreover, a usual coupling with the Bu protons
J. Chem. Soc., Perkin Trans. 2, 1998
347

View Article Online
EPR spectra mainly via conformational changes; an additional
direct influence on the spin density can, however, not be
excluded.
2 D. Voet and J. G. Voet, Biochemistry, Wiley, New York, 1995.
3
J. Schreiber, C. Mottley, B. K. Binha, B. Kalyanaraman and R. P.
Mason, J. Am. Chem. Soc., 1987, 109, 348.
4
5
H. Vogler and M. C. Böhm, Theor. Chim. Acta, 1984, 66, 51.
Mc. R. Maxfield, A. N. Bloch and D. O. Cowan, J. Org. Chem.,
1
985, 50, 1789.
Conclusions
6
Y. Yamaguchi and M. Yokoyama, Chem. Mater., 1991, 3, 709.
The ortho-, para-extended quinones 2–4 are better electron
acceptors than the corresponding para-, para-extended com-
7 R. West, J. A. Jorgenson, K. L. Stearley and J. C. Calabrese, J. Chem.
Soc., Chem. Commun., 1991, 1234.
8 P. Boldt, D. Bruhnke, F. Gerson, M. Scholz, P. G. Jones and F. Bär,
Helv. Chim. Acta, 1993, 76, 1739.
9a
pounds or simple quinones. This is reflected in their more
positive redox potentials. In the series 1–4 the potential increases
roughly with the number of keto groups, where the non-
extended groups, as expected, exert a stronger effect. The posi-
tions of the individual peak-couples in CV seem to be mainly
determined by structural, especially conformational, features
responsible for electron conjugation and free spin delocaliz-
ation. This is nicely demonstrated by the EPR/ENDOR spectra
9
(a) J. Zhou and A. Rieker, J. Chem. Soc., Perkin Trans. 2, 1997, 931;
b) A. Rebmann, J. Zhou, P. Schuler, H. B. Stegmann and A. Rieker,
(
J. Chem. Res. (S), 1996, 318; (M), 1996, 1765; (c) A. Rebmann,
J. Zhou, P. Schuler, A. Rieker and H. B. Stegmann, J. Chem. Soc.,
Perkin Trans. 2, 1997, 1615.
1
1
1
0 G. Wurm, Arch. Pharm. (Weinheim, Ger.), 1991, 324, 491.
1 K. Omura, Tetrahedron, 1995, 51, 6901.
2 H. Kessler and A. Rieker, Tetrahedron Lett., 1966, 5257.
Ϫ
ؒ
3Ϫ
of the monoanion 3 and the trianion 3 . Presumably, the
᝽
13 J. Heiß, K.-P. Zeller and A. Rieker, Org. Mass Spectrom., 1969, 2,
effect of formal two-electron transfers observed in some cases,
should also be attributed to such changes (for a summary see
ref. 25). Generally, the spectra disclose the complex structure of
the radical anions. The species are obviously not planar, and
some protons maintain very low spin densities with unresolv-
able hfs. More detailed discussions will only be possible if struc-
tural determinations of at least some of the involved com-
pounds are known; we are currently investigating this possibil-
ity. If we compare the highest reduction potentials for 2, 3 and
1325.
14 S. A. Russkikh, L. S. Klimenko and E. P. Folin, Zh. Org. Khim., 1983,
1
9, 158.
1
1
1
5 R. S. Nicholson, Anal. Chem., 1965, 37, 1351.
6 J. Heinze, Angew. Chem., 1984, 96, 823.
7 S. Wawzonek, R. Berkey, E. W. Blaha and M. E. Runner,
J. Electrochem. Soc., 1956, 103, 456.
18 A. M. Bond, Modern Polarographic Methods in Analytical Chem-
istry, Marcel Dekker, New York and Basel, 1980, pp. 236–270.
9 G. C. Barker and A. W. Gardner, At. Energy Res. Establ. (GB)
AERE, C/R-2297, 1961; Chem. Abstr., 1961, 55, 24325a.
1
4
, we see that they are never lower than about Ϫ2.0–Ϫ2.2 V.
2
2
0 N. Gupta and H. Linschitz, J. Am. Chem. Soc., 1997, 119, 6384.
1 D. H. Evans and K. Hu, J. Chem. Soc., Faraday Trans., 1996, 92,
This shows that 4 especially is a good electron acceptor, which
can be used, if such a substance is needed, to allow successive
3
983.
(
reversible) electron uptake at different potentials, as in the case
of constructing polynary chips. Finally, to our knowledge,
is the first example of a quinoradialene derived from a [6]-
22 L. A. Wendling, S. K. Koster, J. E. Murray and R. West, J. Org.
Chem., 1977, 42, 1126.
3 M. R. Das, D. H. Connor, D. S. Leniart and J. H. Freed, J. Am.
Chem. Soc., 1970, 92, 2258.
4 S. Berger, P. Hertl and A. Rieker, in The Chemistry of the Quinonoid
Compounds, ed. S. Patai and Z. Rappoport, vol. II, Wiley,
Chichester, 1988, pp. 55–61.
2
4
radialene system 11 (n = 6). The corresponding [3]- and [4]-
radialene systems (11, n = 3 and n = 4, respectively) are known
and their redox activities well investigated by West and co-
2
22,26
workers
(for a related compound see ref. 27).
25 See ref. 1b, p. 732.
2
6 (a) K. Komatsu and R. West, J. Chem. Soc., Chem. Commun., 1976,
70; (b) R. West and D. C. Zecher, J. Am. Chem. Soc., 1970, 92, 155,
161; (c) J. L. Benham and R. West, J. Am. Chem. Soc., 1980, 102,
5
Acknowledgements
We wish to thank the Volkswagenstiftung (I/71009) and the
Land Baden-Württemberg for support of this work.
5
3
054; (d) D. E. Wellman and R. West, J. Am. Chem. Soc., 1984, 106,
55; (e) S. K. Koster and R. West, J. Org. Chem., 1975, 40, 2300.
2
7 S. Hauff and A. Rieker, Tetrahedron Lett., 1972, 1451.
References
Paper 7/05464E
Received 28th July 1997
Accepted 2nd October 1997
1
J. Q. Chambers, in The Chemistry of the Quinonoid Compounds,
a) ed. S. Patai, Wiley, London, 1974, p. 737; (b) ed. S. Patai and
Z. Rappoport, Wiley, Chichester, vol. II, 1988, p. 719.
(
3
48
J. Chem. Soc., Perkin Trans. 2, 1998