UV photolysis of 1,4-diaminobenzene in a low-temperature argon matrix to 2,5-cyclohexadiene-1,4-diimine via 4-aminoanilino radical

Article abstract of DOI:10.1021/jp022694z

UV photolysis of 1,4-diaminobenzene isolated in a low-temperature argon matrix has been investigated by Fourier transform infrared spectroscopy with the aid of the density-functional-theory calculation. Infrared bands of an intermediate produced from 1,4-

Full text of DOI:10.1021/jp022694z

J. Phys. Chem. A 2003, 107, 6725-6730
6725
UV Photolysis of 1,4-Diaminobenzene in a Low-Temperature Argon Matrix to
2,5-Cyclohexadiene-1,4-diimine via 4-Aminoanilino Radical
Nobuyuki Akai, Satoshi Kudoh, and Munetaka Nakata*
Graduate School of BASE (Bio-Applications and Systems Engineering), Tokyo UniVersity of Agriculture and
Technology, Naka-cho, Koganei, Tokyo 184-8588, Japan
ReceiVed: December 30, 2002; In Final Form: April 30, 2003
UV photolysis of 1,4-diaminobenzene isolated in a low-temperature argon matrix has been investigated by
Fourier transform infrared spectroscopy with the aid of the density-functional-theory calculation. Infrared
bands of an intermediate produced from 1,4-diaminobenzene upon UV irradiation (λ < 350 nm) are assigned
to a semiquinone-type radical, 4-aminoanilino radical. A final product produced from the 4-aminoanilino
radical upon shorter-wavelength irradiation (λ < 310 nm) is assigned as 2,5-cyclohexadiene-1,4-diimine.
Optimized structures of 1,4-diaminobenzene, 4-aminoanilino radical, and 2,5-cyclohexadiene-1,4-diimine are
compared with one another, resulting in changes of the π-conjugated system similar to those of the hydroquinone
and 1,4-benzoquinone system. In addition, trans-cis isomerization of 2,5-cyclohexadiene-1,4-diimine upon
UV irradiation is observed in photoequilibrium, where the cis/trans population ratio depends on the irradiation
wavelength.
1. Introduction
SCHEME 1
The oxidation-reduction system between hydroquinone and
1,4-benzoquinone has attracted the attention of many researchers
who study electron and/or proton transport in biological
systems.1 This oxidation-reduction system is applied to a
quinhydrone electrode to measure proton concentration in
solutions. A reaction intermediate between hydroquinone and
benzoquinone was identified as p-benzosemiquinone radical, as
shown in the Scheme 1.2-5 For example, Trinpathi and Schuler
observed a time-resolved resonance Raman spectrum of transient
species in the pulse radiolysis of hydroquinone solution and
identified the observed spectrum of p-benzosemiquinone radical
by a vibrational analysis.3 On the other hand, Ja¨ger and Norris
investigated laser flash photolysis of hydroquinone and benzo-
quinone by time-resolved electron spin resonance spectroscopy
and detected the signals of the semiquinone radical.5
SCHEME 2
In the present study, we have investigated a similar oxida-
tion-reduction system having a NH2 group instead of a OH
group, i.e., 1,4-diaminobenzene and 2,5-cyclohexadiene-1,4-
diimine system. By analogy with the hydroquinone and ben-
zoquinone system, 4-aninoanilino radical is expected to be a
reaction intermediate produced from 1,4-diaminobenzene by
dissociation of one hydrogen atom; another dissociation of one
hydrogen atom of the other amino group produces a final
product, 2,5-cyclohexadiene-1,4-diimine, where the π-conju-
gated system is quinone-type, as shown in Scheme 2.
Reports on the 1,4-diaminobenzene and 2,5-cyclohexadiene-
1,4-diimine system are fewer than those on the hydroquinone
and benzoquinone system, and the intermediate has not been
identified so far, because the intermediate and the final product
are both less stable than the corresponding species in the
hydroquinone and benzoquinone system. Thus, the purposes of
the present study are (i) to measure infrared spectra of the
photoreaction intermediates and final products by the low-
temperature matrix-isolation technique, (ii) to analyze the
observed bands by comparison with the calculated spectral
patterns obtained by the density-functional-theory (DFT) cal-
culation, and (iii) to elucidate the photoreaction mechanism of
the 1,4-diaminobenzene and 2,5-cyclohexadiene-1,4-diimine
system. In the present study, we show the first experimental
evidence for 4-aminoanilino radical, where the π-conjugated
system is similar to that of p-benzosemiquinone radical.
2. Experimental and Calculations
Since the vapor pressure of 1,4-diaminobenzene, purchased
form Tokyo Chemical Industry Co. Ltd., was too low to transfer
the vapor sufficiently into a glass cylinder, a small amount of
the solid was placed in a stainless steel pipe nozzle with a
heating system, on which pure argon gas (Nippon Sanso,
99.9999% purity) was flowed. The sample was heated to about
310 K, and the flow rate of rare gas was adjusted by a needle
valve to obtain sufficient isolation. The mixed gas was expanded
through a stainless steel pipe (o.d. 1/16 in.) and deposited in a
* To whom correspondence should be addressed. E-mail: necom@
cc.tuat.ac.jp. Phone: +81-42-388-7349. Fax: +81-42-388-7349.
10.1021/jp022694z CCC: $25.00 © 2003 American Chemical Society
Published on Web 08/05/2003

6726 J. Phys. Chem. A, Vol. 107, No. 34, 2003
Akai et al.
SCHEME 3
vacuum chamber on a CsI plate cooled by a closed-cycle helium
refrigerator (CTI Cryogenics, model M-22) to about 16 K. The
temperature of the CsI plate was controlled with a PID
(proportional plus integral plus derivative) action method and
changed at a rate of 1 K/min to anneal the matrix sample. UV/
vis light from a super-high-pressure mercury lamp was used
for photolysis of the sample. A water filter was used to remove
thermal reactions, and UV-36, UV-34, UV-32, UV-30, and UV-
28 short-wavelength cutoff filters and a U-330 band-path filter
(HOYA) were used to select the irradiation wavelength. Infrared
spectra of the matrix samples were measured with an FTIR
spectrophotometer (JEOL, model JIR-7000). The spectral
resolution was 0.5 cm-1, and the accumulation number was 64.
Other experimental details were reported elsewhere.6,7
DFT calculations were performed using the GAUSSIAN 98
program8 with the 6-31++G** basis set. Becke’s three-
parameter hybrid density functional,9 in combination with the
Lee-Yang-Parr correlation functional (B3LYP),10 was used
to optimize the geometrical structures.The open shell model was
used for the calculations of radicals.
Figure 1. Infrared spectrum of 1,4-diaminobenzene in an argon matrix
and calculated spectral pattern obtained by DFT/B3LYP/6-31++G**:
(a) observed spectrum (bands marked with “b” are due to a small
amount of water impurity); (b, c) calculated spectral patterns of the
trans and cis isomers, respectively. A scaling factor of 0.98 is used.
about 1 cm-1, which is much narrower than that previously
observed in the solid and solution. To distinguish the trans and
cis bands, we have also calculated the infrared spectral patterns
of the two isomers by the DFT/B3LYP/6-31++G** method.
As compared in Figure 1, the wavenumbers of the trans bands
measured in the region between 800 and 1500 cm-1 are found
to be consistent with the corresponding cis bands within 1 cm-1
.
One possibility to distinguish the isomer bands is to examine
the NH2 scissoring mode around 1600 cm-1, where one and
two bands should appear for the trans and cis isomers,
respectively. However, this spectral region is disturbed by a
small amount of water impurity, and thus, it is difficult to find
the second cis band. Another possibility to distinguish the isomer
3. Results and Discussion
bands is to examine the NH2 wagging mode around 600 cm-1
,
3.1. Conformation of the Reactant 1,4-Diaminobenzene.
Since the reactant 1,4-diaminobenzene is stable under normal
conditions, some spectroscopic data have been obtained in the
solid and solution in addition to theoretical data at various
calculation levels.11-15 There are two possible conformations
for 1,4-diaminobenzene, as shown in Scheme 3; the two
hydrogen atom pairs of the amino groups for the trans
conformation are on opposite sides of the molecular plane, while
those for the cis conformation are on the same side.
where one and two bands for the trans and cis isomers,
respectively, should appear as for the NH2 scissoring mode. One
strong band appearing at 688 cm-1 may be assigned to the
second NH2 wagging mode for the cis isomer, where the
corresponding calculated value is 641 cm-1. The difference
between the observation and the DFT calculation may be caused
by large anharmonicity of the wagging vibrational potential.
We could not observe the other NH2 wagging modes for the
trans, 611 cm-1, and cis, 604 cm-1, isomers, probably because
they are out of range of our spectral measurement. The weak
bands appearing at 763 and 678 cm-1 may be assigned to the
trans and cis isomers, respectively. Thus, we conclude that the
trans and cis isomers of 1,4-diaminobenzene coexist in the
matrix. The observed and calculated wavenumbers for 1,4-
diaminobenzene are summarized in Table 1 with the relative
intensities.
Whether both trans and cis conformations coexist has not
been clarified. For example, Tzeng and Narayanan compared
the experimental vibrational wavenumbers with the correspond-
ing theoretical values obtained by the Hartree-Fock/6-31+G*
method and temporarily assigned the observed bands to only
the trans conformation.14 On the other hand, Akalin and Akyu¨z
assumed in their normal-coordinate analysis of the infrared
spectra that the trans and cis isomers exist in the solid.15 Thus,
we first tried to distinguish the bands of each isomer, because
infrared spectra of matrix-isolated species are usually composed
3.2. Identification of the Photoreaction Intermediate
4-Aminoanilino Radical. When the matrix sample of 1,4-
diaminobenzene was exposed to UV light shorter than 350 nm
after the measurement of Figure 1, spectral changes due to
photoreaction were observed. Figure 2a shows a difference
spectrum between those measured before and after UV irradia-
tion through a UV-34 filter (λ > 330 nm) for 10 min. The
increasing and decreasing bands are ascribed to photoproducts
and the reactant, respectively. The observed increasing bands
are divided into two groups; intensities of the bands marked
with “×” are unchanged during annealing up to 28 K, while
those of the others decrease. This means that the former bands
are assignable to a stable photoproduct, while the latter bands
are assignable to a transient species which is unstable at matrix
temperature higher than 28 K. Figure 2b shows a difference
spectrum between those measured after and before the annealing,
of sharp bands with a bandwidth of a few cm-1
.
The energy difference between the more stable isomer, trans,
and the less stable isomer, cis, has been calculated to be 0.28
kJ mol-1 by the present DFT calculation, while the isomerization
barrier from trans to cis is 13.8 kJ mol-1. Since the population
ratio of the isomers at the deposition temperature is usually
maintained in a low-temperature matrix,16 the cis/trans popula-
tion ratio of 1,4-diaminobenzene in the matrix is estimated to
be about 0.9, assuming the Boltzmann distribution law at 310
K. This means that infrared bands of the two isomers can be
measured by the matrix-isolation technique, if the hydrogen atom
tunneling does not occur, unlike for hydroquinone.17
The observed matrix infrared spectrum of 1,4-diaminobenzene
is shown in Figure 1. The bandwidth of almost all the bands is

UV Photolysis of 1,4-Diaminobenzene
J. Phys. Chem. A, Vol. 107, No. 34, 2003 6727
TABLE 1: Observed and Calculated Wavenumbers (cm-1
and Their Relative Intensities of 1,4-Diaminobenzene
)
TABLE 2: Observed and Calculated Wavenumbers (cm-1
and Their Relative Intensities of 4-Aminoanilino Radical
)
obsd
calcd, trans
calcd, cis
obsd
calcd
ν
int
2.1
Va
int
sym
Va
int
sym
ν
int
Va
int
approximate assignmentb
3477
3467
3396
3387
3062
3059
3043
3034
3030
3018
3586
3586
3486
3485
3122
3119
3104
3103
0
3.8
0
2.5
0
6.9
0
6.8
bg
au
ag
bu
ag
au
bg
bu
3588
3588
3487
3487
3119
3116
3101
3101
4.9
0
b1
a2
b2
a1
a1
b1
a2
b2
3528
15.2 3634
42.4 3520
3366
13.2 V(NH2)
32.6 V(NH2)
2.7 V(N-H)
1.6 V(C-H)
6.4 V(C-H)
8.2 V(C-H)
7.6 V(C-H)
5.5
2.3
4.2
0.9
1.8
1.6
1.7
2.4
1.4
3429c
2.9
0.3
0.0
9.1
0
3148
3122
3109
3102
8.9
1614
1582
100
30.9 1598
1527
8.6 1478
9.6 1463
16.0 1340
3.5 1332
1318
1631 100
NH2 scissors
62.2 V(CdC), NH2 scissors
0.4 V(CdC), NH2 rocking
9.7 V(CdC)
1649
1630
1625
1602
1523
1458
1335
1331
1273
1258
1179
1135
0
23.5
0
ag
bu
ag
bg
bu
au
au
bg
ag
bu
ag
au
1648
1629
1625
1602
1523
1458
1335
1331
1272
1258
1179
1135
0.7
26.2
4.4
0
50.1
0.0
5.0
0
0.0
30.5
0.0
3.6
a1
b2
a1
a2
b2
b1
b1
a2
a1
b2
a1
b1
1472
1464
1335
1330
1616
16.8
9.8 V(CdC)
25.1 V(CdN)
0
2.7 V(C-C), â(N-H), NH2 rocking
1.9 â(C-H)
1521
100
38.3
0.0
3.8
0
1305
1157
1151
1109
1028
977c
26.6 1300
37.0 1162
2.3 1152
11.4 1107
3.3 1031
30.5 V(C-N)
1322
1312
2.9
1.9
22.4 â(C-H), â(N-H)
0.8 â(C-H), â(N-H)
12.7 â(C-H), NH2 rocking
1.8 NH2 rocking
0
1263
24.4
23.1
0
1.2
975
942
915
823
815
790
772
686
679
631
0.4 â(C-H)
1128
8.3
15.1
2.7
0.0 γ(C-H)
1083b
0.0 γ(C-H)
1096
1047
1002
915
899
840
811
789
760
699
646
638
611
0
0.3
0.0
0.0
0
0
16.8
0
bg
au
bu
au
ag
ag
bu
bg
bu
ag
bg
ag
bu
1095
1046
1002
916
900
841
810
790
763
687
646
641
604
0
0.4
0.0
0
0.1
1.1
19.8
0.0
0.0
3.8
0
a2
b1
b2
a2
b2
a1
a1
b1
b2
b2
a2
a1
b2
0.2 ring breathing
28.5 γ(C-H)
1032b
1010
6.2
0.9
828
791
776
701
695
633
13.0
0.4
6.3
10.1
2.2
2.1
9.7 γ(C-H), γ(N-H)
6.0 ring breathing
20.6 γ(N-H)
3.7 γ(C-C-C), γ(N-H)
0.3 â(C-C-C)
822
17.0
a A scaling factor of 0.98 is used. b The symbols V, â, and γ represent
stretching, in-plane bending, and out-of-plane bending, respectively.
c Splitting bands.
763c
678d
2.3
3.5
0.2
0
0
0
688d
e
32.9
34.4
100
100
hydrogen atom recombines with aminoanilino radical to produce
the original reactant 1,4-diaminobenzene at matrix temperature
higher than 28 K. One may have a question as to whether the
decreasing rate of the reactant bands is smaller than that
expected. This may be explained by the assumption that a small
amount of the reactant is pumped out during the annealing. The
intense band appearing at 1614 cm-1 is assignable to the NH2
scissoring mode of aminoanilino radical, and the 1350 cm-1
band to the C-N stretching mode. The observed and calculated
wavenumbers are summarized in Table 2 with the relative
intensities and approximate vibrational assignments.
a A scaling factor of 0.98 is used. b May be assigned to combination
bands. See the text. c Assigned to the trans isomer. d Assigned to the
cis isomer. e Out of the range of our spectral measurement.
Ernstbrunner et al. observed a resonance Raman spectrum
of a radical cation in the photolysis of 1,4-diaminobenzene
instead of 4-aminoanilino radical.11 Since they photolyzed the
sample in a protonic matrix at 77 K, they detected 1,4-
diaminobenzene radical cation stabilized by interaction with the
solutions, but not 4-aminoanilino radical. This cation is easily
produced in solutions of 1,4-diaminobenzene halide known as
Wu¨rster’s salts18-20 and produced by 187 nm laser irradiation
of a supersonic molecular beam.21 In contrast to this radical
cation, no experimental evidence for 4-aminoanilino radical has
been obtained till the present study.
To identify the unknown species marked with “×”, we
compared the observed spectrum with calculated spectral
patterns of 1,4-diaminobenzene radical cation and possible
photoproducts such as aniline radical cation, Dewar-type 1,4-
diaminobenzene, and so on. However, no calculated spectral
patterns were consistent with the unknown species bands.
Therefore, identification of the unknown species remains
unresolved in this work.
Figure 2. Difference spectra of 1,4-diaminobenzene: (a) between
spectra measured before and after UV irradiation through a UV-32 filter
for 10 min; (b) between spectra measured after and before annealing
up to 28 K after the irradiation (bands marked with “×” are due to an
unknown photoproduct); (c) calculated spectral pattern of 4-amino-
anilino radical obtained by DFT/B3LYP/6-31++G** using a scaling
factor of 0.98.
where the stable photoproduct bands disappear because their
intensities are unchanged during the annealing. Since the
calculated spectral pattern of 4-aminoanilino radical obtained
by the DFT calculation shown in Figure 2c reproduces the
observed spectrum shown in Figure 2b satisfactorily, we
conclude that the UV photolysis of 1,4-diaminobenzene yields
4-aminoanilino radical by dissociation of one hydrogen atom
of the amino groups. We assume that the photodetached
3.3. Photoisomerization of the Final Products, trans- and
cis-2,5-Cyclohexadiene-1,4-diimine.. When the matrix sample

6728 J. Phys. Chem. A, Vol. 107, No. 34, 2003
Akai et al.
Figure 4. Difference spectrum between those measured before and
after UV irradiation through a UV-34 filter for 10 min after measure-
ment of Figure 3: (a) observed spectrum, where a broad band marked
with an asterisk is due to the reactant 1,4-diaminobenzene appearing
artificially; (b) calculated spectral patterns of 2,5-cyclohexadiene-1,4-
diimine obtained by DFT/B3LYP/6-31++G** using a scaling factor
of 0.98. Trans bands are shown downside and cis bands upside.
Figure 3. Difference spectrum between those measured before and
after UV irradiation through a UV-28 filter for 20 min after the UV-
32 irradiation for 120 min: (a) observed spectrum; (b) calculated
spectral pattern obtained by DFT/B3LYP/6-31++G** using a scaling
factor of 0.98. A mixture of trans- and cis-2,5-cyclohexadiene-1,4-
diimine bands are shown upside, and 4-aminoanilino radical bands
downside. The bands marked with an asterisk in the observed spectrum
are due to 1,4-diaminobenzene remaining in the first photolysis.
irradiation, and trans-cis photoisomerization is induced by the
former irradiation. This spectral change is compared with the
calculated spectral pattern shown in Figure 4b. The calculated
pattern reproduces the observed pattern satisfactorily. The
observed and calculated wavenumbers and relative intensities
for trans- and cis-2,5-cyclohexadiene-1,4-diimine are sum-
marized in Table 3. The bands appearing in the region between
1500 and 600 cm-1 are consistent with the corresponding
calculated wavenumbers within 9 cm-1, except for the trans
band appearing at 1165 cm-1, which may be assigned to a
combination band or a splitting band due to Fermi resonance.
We have also found that the trans and cis isomers exist in a
photoequilibrium state in the matrix during irradiation and that
the cis population decreases as the irradiation wavelength
shortens. For example, the absorbance ratio between the cis band
at 1132 cm-1 and the trans band at 881 cm-1 is estimated to
be 0.68 ( 0.03 upon UV-34 irradiation (λ > 330 nm), while
the ratio decreases to 0.61 ( 0.03 upon UV-28 irradiation (λ >
270 nm). This finding implies that the electronic absorption of
the cis isomer slightly shifts to the shorter-wavelength side from
that of the trans isomer if the absorption coefficient of the cis
isomer is nearly equal to that of the trans isomer. Measuring
the absorption spectra for the cis and trans isomers separately
seems to be useful for confirmation of this assumption; however,
it has not been achieved.23
SCHEME 4
was exposed to the shorter-wavelength UV light, the infrared
spectra of the final products were observed. Figure 3a shows a
difference spectrum between those measured before and after
UV-28 irradiation (λ > 270 nm) for 20 min after the UV-32
irradiation (λ > 310 nm) for 120 min. The decreasing bands
are due to 4-aminoanilino radical and the reactants remaining
in the first photolysis, while the increasing bands are due to
the final products. To identify the final products, we have
performed DFT calculations of the most plausible candidate,
i.e., 2,5-cyclohexadiene-1,4-diimine, which is produced from
4-aminoanilino radical by dissociation of one hydrogen atom
of the other amino group.
This final product, 2,5-cyclohexadiene-1,4-diimine, is planar
and has two conformations, cis and trans, around the CdN
bonds, as shown in Scheme 4.
Although infrared and electronic absorption spectra of 2,5-
cyclohexadiene-1,4-diimine in film or solution were previously
measured,22-24 conformational analyses were not complete.24
The calculated spectral patterns of a mixture of trans- and cis-
2,5-cyclohexadiene-1,4-diimine obtained by the present study
are compared with the obtained spectrum in Figure 3. Since
the calculated pattern reproduces the observed difference
spectrum satisfactorily, we conclude that the final photoproduct
bands are assignable to a mixture of trans- and cis-2,5-
cyclohexadiene-1,4-diimine.
3.4. Optimized Geometries of 1,4-Diaminobenzene, 4-Ami-
noanilino Radical, and 2,5-Cyclohexadiene-1,4-diimine. From
the vibrational analyses of the observed infrared spectra with
the aid of the DFT calculation, we conclude that the π-conju-
gated system of 1,4-diaminobenzene changes to a quinone type
in 2,5-cyclohexadiene-1,4-diimine through a semiquinone type
in 4-aminoanilino radical. This conclusion is confirmed by the
calculations of the optimized geometries for 1,4-diaminoben-
zene, 4-aminoanilino radical, and 2,5-cyclohexadiene-1,4-di-
imine at the same DFT/B3LYP/6-31++G** level. The obtained
bond lengths are compared with one another in Figure 5. The
geometrical parameters of 1,4-diaminobenzene correspond to
the experimental data obtained by electron diffraction12 within
an error of 0.8%, which is also consistent with previous
calculations.12-14,25
The energy difference between the more stable isomer, trans,
and the less stable isomer, cis, is calculated to be 1.45 kJ mol-1
,
while the isomerization barrier is 116.7 kJ mol-1. When the
matrix sample was irradiated by the other wavelength UV light,
spectral changes due to photoisomerization between the trans
and cis isomers were observed. Figure 4a shows a difference
spectrum between those measured before and after UV-34
irradiation (λ > 330 nm) for 10 min after the U330 irradiation
(200 < λ <400 nm) for 180 min; 2,5-cyclohexadiene-1,4-
diimine is produced from 4-aminoanilino radical by the latter
The imino group of 4-aminoanilino radical is in the benzene
plane, while the amino group is bent out of the plane by 27.7°.
This value is smaller than that of 1,4-diaminobenzene, 40.5°.
The bond angle of H-N-H is 114.4° and 110.8° for 4-ami-
noanilino radical and 1,4-diaminobenzene, respectively. This

UV Photolysis of 1,4-Diaminobenzene
J. Phys. Chem. A, Vol. 107, No. 34, 2003 6729
TABLE 3: Observed and Calculated Wavenumbers (cm-1) and Their Relative Intensities of 2,5-Cyclohexadiene-1,4-diimine
trans isomer
calc
cis isomer
obsd
obsd
calcd
ν
int
νa
int
0
sym
ν
int
νa
int
3.2
3.9
2.3
0.1
8.7
2.8
0.1
sym
obsdb ν
3173
3374
3374
3148
3147
3112
3112
1671
1628
1611
1575
1429
1397
1362
1315
1171
1166
1129
1050
987
ag
bu
ag
bu
ag
bu
ag
bu
bu
ag
ag
bu
bu
ag
bu
ag
ag
bu
au
bg
bu
bg
au
au
ag
bu
bg
bg
ag
3373
3373
3156
3139
3121
3102
1672
1628
1612
1575
1429
1388
1375
1310
1172
1168
1128
1051
990
a1
b2
a1
b2
a1
b2
a1
b2
a1
a1
b2
a1
b2
b2
a1
a1
b2
a1
a2
a2
b2
b1
a2
b1
a1
b2
a2
b1
b2
6.5
0
4.6
0
7.4
0
1594c
63.1
89.4
2.6
0
1594c
1426
100
1594
1585
2.3
0.1
12.6
6.7
3.7
37.6
0.5
0.2
78.3
13.3
0
0
11.1
1433
1366
1221
1406
1364
43.8
18.8
39.2
11.0
0
1374
1315
33.3
88.9
1170
56.3
25.0
40.4
0
0
48.9
1.0
0
2.5
0
100
7.5
0
1165d
1132
100
1056c
84.4
1056c
1075/1072
1010
962
936
975
969
0
945
945
2.8
66.0
0
912
906
856
77.8
25.9
912
881
100
881
873
845
851
42.1
0.1
2.0
0
16.3
1.4
859/854
779
779
765
1.7
0
0
0
765
724
728
704
722
11.1
716
607
607
a A scaling factor of 0.98 is used. b Measured in a solid film.24 c Trans and cis bands are overlapped. d Assigned to a combination mode, overtone,
or Fermi coupling.
SCHEME 5
radical is slightly distorted, where the CdC lengths at the imino
group side are about 0.02 Å longer than those at the amino
group side. This distortion is also found in the structure of
p-benzosemiquinone radical in the hydroquinone and 1,4-
benzoquinone system.4
3.5. Photoreaction Mechanism of 1,4-Diaminobenzene. We
summarize the photoreaction mechanism of 1,4-diaminobenzene
in the low-temperature argon matrix as follows: (1) No reaction
occurs when the irradiation wavelength is longer than 350 nm.
(2) 4-Aminoanilino radical is produced from 1,4-diaminoben-
Figure 5. Optimized bond lengths of 1,4-diaminobenzene, 1,4-
diaminobenzene radical cation, 4-aminoanilino radical, and 2,5-
cyclohexadiene-1,4-diimine calculated at the DFT/B3LYP/6-31++G**
level. The numbers represent C-C, C-N, and N-H bond lengths in
angstroms.
zene when the irradiation wavelength is shorter than 350 nm.
(3) 2,5-Cyclohexadiene-1,4-diimine is produced from 4-ami-
noanilino radical when the irradiation wavelength is shorter than
310 nm. (4) Photoisomerization between trans- and cis-2,5-
cyclohexadiene-1,4-diimine occurs, where the trans isomer
changes to the cis isomer by longer-wavelength irradiation and
the cis isomer changes to the trans isomer by shorter-wavelength
irradiation. The photoreaction pathway from 1,4-diaminobenzene
to 2,5-cyclohexadiene-1,4-diimine via 4-aminoanilino radical
is shown in Scheme 5.
finding implies that the amino group of 4-aminoanilino radical
is in the middle between sp2 (120°) and sp3 (109.5°). On the
other hand, all the C-N and C-C bond lengths of 4-amino-
anilino radical are between the corresponding bond lengths of
1,4-diaminobenzene and 2,5-cyclohexadiene-1,4-diimine. For
example, the C-N bond lengths are calculated to be 1.410,
1.383/1.335, and 1.292 Å for 1,4-diaminobenzene, amino/imino
side of the 4-aminoanilino radical, and 2,5-cyclohexadiene-1,4-
diimine, respectively. The benzene ring of 4-aminoanilino
Acknowledgment. We thank Professors Kozo Kuchitsu and
Masao Takayanagi (BASE, Tokyo University of Agriculture &
Technology) for their helpful discussion.

6730 J. Phys. Chem. A, Vol. 107, No. 34, 2003
Akai et al.
(10) Lee, C.; Yang, W.; Parr, R. G. Phys. ReV. B 1988, 37, 785.
References and Notes
(11) Ernstbrunner, E. E.; Girling, R. B.; Grossman, W. E. L.; Williams,
K. P. J.; Hester, R. E. J. Raman Spectrosc. 1981, 10, 161.
(12) Colapietro, M.; Domenicano, A.; Portalone, G.; Schultz, G.;
Hargittai, I. J. Phys. Chem. 1987, 91, 1728.
(1) Morton, R. A., Ed. Biochemistry of Quinones; Academic Press:
New York, 1965.
(2) Tripathi, G. N. R. J. Chem. Phys. 1981, 74, 6044.
(3) Tripathi, G. N. R.; Schuler, R. H. J. Phys. Chem. 1987, 91, 5881.
(4) Mohandas, P.; Umapathy, S. J. Phys. Chem. 1997, 101, 4449.
(5) Ja¨ger, M.; Norris, J. R., Jr. J. Magn. Reson. 2001, 150, 26.
(6) Nakata, M.; Kudoh, S.; Takayanagi, M.; Ishibashi, T.; Kato, C. J.
Phys. Chem. 2000, 104, 11304.
(7) Kudoh, S.; Takayanagi, M.; Nakata, M. J. Photochem. Photobiol.
A 1999, 123, 25.
(13) Noto, R.; Leone, M.; Manna, G. L.; Bruge´, F.; Fornili, S. L.
THEOCHEM 1998, 422, 35.
(14) Tzeng, W. B.; Narayanan, K. THEOCHEM 1998, 434, 247.
(15) Akalin, E.; Akyu¨z, S. Vib. Spectrosc. 2000, 22, 3.
(16) Kudoh, S.; Takayanagi, M.; Nakata, M. Chem. Phys. Lett. 1998,
296, 329.
(17) Akai, N.; Kudoh, S.; Takayanagi, M.; Nakata, M. Chem. Phys. Lett.
(8) Gaussian98 (Revision A.6): Frisch, M. J.; Trucks, G. W.; Schlegel,
H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Zakrzewski, V. G.;
Montgomery, J. A.; Stratmann, R. E.; Burant, J. C.; Dapprich, S.; Millam,
J. M.; Daniels, A. D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J.;
Barone, V.; Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo, C.;
Clifford, S.; Ochterski, J.; Peterson, G. A.; Ayala, P. Y.; Cui, Q.; Morokuma,
K.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.;
Cioslowski, J.; Ortiz, J. V.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz,
P.; Komaromi, I.; Gomperts, R.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-
Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Gonzalez, C.; Challacombe,
M.; Gill, P. M. W.; Johnson, B. G.; Chen, W.; Wang, M. W.; Andres, J.
L.; Head-Gordon, M.; Replogle, E. S.; Pople, J. A., Gaussian, Inc.,
Pittsburgh, PA, 1998.
2002, 356, 133.
(18) Mayer, E.; Girling, R. B.; Hester, R. E. J. Chem. Soc., Chem.
Commun. 1973, 192.
(19) Yokoyama, K.; Maeda, S. Chem. Phys. Lett. 1977, 48, 59.
(20) Michaelis, L.; P. Schubert, M.; Granick, S. J. Am. Chem. Soc. 1939,
61, 1981.
(21) Ozeki, H.; Okuyama, K.; Takahashi, M.; Kimura, K. J. Phys. Chem.
1991, 95, 4308.
(22) Corbett, J. F. J. Chem. Soc., Perkin Trans. 2 1972, 999.
(23) Corbett, J. F. J. Chem. Soc. B 1969, 213.
(24) Prichard, F. E. Spectrochim. Acta 1964, 20, 925.
(25) Chipman, D. M.; Sun, Q.; Tripathi, G. N. R. J. Chem. Phys. 1992,
97, 8073.
(9) Becke, A. D. J. Chem. Phys. 1993, 98, 5648.