Direct and ketone-sensitized photoconversion of 1-nitro-9,10-anthraquinone to 1-amino-9,10-anthraquinone mediated by donor radicals

Article abstract of DOI:10.1016/j.chemphys.2009.12.004

The full photoreduction of 1-nitro-2-R-9,10-anthraquinone (R = H: N1, methyl: N2) was studied in benzene, acetonitrile and acetonitrile-water mixtures in the presence of 2-propanol and triethylamine (TEA). The major photoproduct is the fluorescing 1-amino-2-R-AQ (A1, A2). The quantum yield of full reduction increases with the donor concentration, approaching Φ_NH2 = 0.1. The intermediates involved are assigned on the basis of spectral and kinetic characteristics. The short-lived triplet state (≤20 ns) of N2 can be intercepted by 2-propanol or TEA, thereby forming the spectroscopically hidden donor radicals and the nitroAQ radicals which absorb at 400 and 540 nm; the latter band is due to the radical anion. The triplet state of N1 was not observed at room temperature, but the radical properties and decay in the nitrosoAQ are similar for N1 and N2. For donors in lower concentrations Φ_NH2 is strongly increased in the presence of benzophenone, acetophenone or acetone, approaching 0.22. The results under direct and sensitized conditions are compared and major dependences and the effects of mixtures of acetonitrile with water are outlined.

Full text of DOI:10.1016/j.chemphys.2009.12.004

Chemical Physics 368 (2010) 20–27
Contents lists available at ScienceDirect
Chemical Physics
journal homepage: www.elsevier.com/locate/chemphys
Direct and ketone-sensitized photoconversion of 1-nitro-9,10-anthraquinone
to 1-amino-9,10-anthraquinone mediated by donor radicals
*
Helmut Görner , Henry Gruen
Max-Planck-Institut für Bioanorganische Chemie, D-45413 Mülheim an der Ruhr, Germany
a r t i c l e i n f o
a b s t r a c t
Article history:
The full photoreduction of 1-nitro-2-R-9,10-anthraquinone (R = H: N1, methyl: N2) was studied in ben-
zene, acetonitrile and acetonitrile–water mixtures in the presence of 2-propanol and triethylamine (TEA).
The major photoproduct is the fluorescing 1-amino-2-R-AQ (A1, A2). The quantum yield of full reduction
Received 7 September 2009
In final form 4 December 2009
Available online 11 December 2009
increases with the donor concentration, approaching
U
NH2 ¼ 0:1. The intermediates involved are assigned
on the basis of spectral and kinetic characteristics. The short-lived triplet state (620 ns) of N2 can be
intercepted by 2-propanol or TEA, thereby forming the spectroscopically hidden donor radicals and the
nitroAQ radicals which absorb at 400 and 540 nm; the latter band is due to the radical anion. The triplet
state of N1 was not observed at room temperature, but the radical properties and decay in the nitrosoAQ
Keywords:
Nitroanthraquinone
Aminoanthraquinone
Photoreduction
Radical
are similar for N1 and N2. For donors in lower concentrations
UNH2 is strongly increased in the presence of
benzophenone, acetophenone or acetone, approaching 0.22. The results under direct and sensitized con-
ditions are compared and major dependences and the effects of mixtures of acetonitrile with water are
outlined.
Ó 2009 Elsevier B.V. All rights reserved.
1
. Introduction
tion. 1-Phenylethanol and ca. 0.01 mM acetophenone (as trace
impurity) function as donor and sensitizer, respectively. The quan-
The photochemical features of 9,10-anthraquinone (AQ) and
several derivatives in solution have been intensively studied [1–
6]. The key-intermediate in the photoreduction of AQs is the sem-
iquinone radical, the properties of which have been reviewed [17].
Two-photon ionization and subsequent electron transfer can take
place for the 1,5-disulfonate of AQ [6]. The photoreactivity of
AQs is lowered, when electron donating or accepting groups are
substituted. For 2-aminoAQ in 2-propanol the quantum yield of
reduction is close to zero, whereas it is 0.9 for the conversion of
tum yield (UNH2 ) of full conversion of N2 to A2, determined by pho-
tochemical means, is up to 0.2 in 1-phenylethanol [27]. Methanol,
ethanol or 2-propanol are attractive H-atom donors, whereas
amines, e.g. phenylethylamine and triethylamine (TEA), function
as electron donors. A plausible photoreduction sequence of 1-nit-
1
2
roAQ would be H-transfer from the donor (DH ) thereby forming
nitrosoAQ; then further reduction leads to hydroxylaminoAQ and
eventually to aminoAQ, Scheme 1. Radicals derived form 1-nitroAQ
and the (H-atom or electron) donors are intermediates. An alterna-
2
parent AQ to 9,10-dihydroanthraquinone (AQH ) [1–3]. The low
2 2
tive pathway via the 1-nitrodihydroAQ (O NAQH ), pathway b in
reactivity of aminoAQs is ascribed to a charge transfer process,
leading to fast deactivation into the ground state. The photopro-
cesses of aminoAQs have been extensively studied [4,18–21], but
concerning the reactions of nitroAQs only little is known [22–26].
The photoreduction of 1-nitro-2-R-AQ (R = H: N1, methyl: N2)
to the corresponding 1-aminoAQs (A1, A2) in benzene and acetoni-
trile in the presence of phenylethylamine and 1- or 2-phenyletha-
nol was the subject of a preceding study [27]. The triplet lifetime of
Scheme 2, had been rejected [27]. For conversion of 1-nitroAQ
and 2-nitroAQ in 2-propanol to NHOHAQs quantum yields of
0.0001 and 0.047 for kirr = 254 nm and 0.0001 and 0.0015 for
kirr = 365 nm have been reported, respectively [25].
The photoconversion of a nitroarene to the aminoarene [28–36]
is a rare process which requires six electrons. The formulation of a
self-consistent mechanism accounting for the complete reduction
by photochemical means has been achieved for nitrobenzene
derivatives in the presence of perchloric acid and 10-methyl-
9,10-dihydroacridine [32]. In another system, formate has been
employed as an efficient reducing agent of various nitroarenes
N2 is
The longer lifetime
causes an enhanced reactivity at a given lower donor concentra-
s
T
= 5–20 ns at room temperature and that of N1 is shorter.
s
T
due to the methyl group in N2 therefore
[
33]. The relatively high efficiency for nitroAQs is based on the
bi-functional AQ system, where both the 1-nitro group and one
of the carbonyl groups are intermediate H-atom acceptor sites
via a cascade-like mechanism.
*
Corresponding author. Tel.: +49 2083063593; fax: +49 2083063951.
E-mail address: goerner@mpi-muelheim.mpg.de (H. Görner).
0
301-0104/$ - see front matter Ó 2009 Elsevier B.V. All rights reserved.
doi:10.1016/j.chemphys.2009.12.004

H. Görner, H. Gruen / Chemical Physics 368 (2010) 20–27
21
Scheme 1.
O
•
N
OH
O
O
NO2
O
•
O
O
NO
•
H
+
H
-H O
2
+
pathway a
OH NO2
OH NO2
(NOAQ)
O
•
•
-H O
2
(
NO AQ)
+H
2
pathway b
(NO AQH )
2 2
O
OH
Scheme 2.
In the present work, the direct photolysis of N1 and N2, denoted
here as nitro(methyl)AQ, was studied in solution in the presence of
alcohols as H-donors or TEA as electron donor. Several intermedi-
ates were observed by time-resolved techniques. The second goal
was to arrive at an understanding of the photosensitized reduction.
For this purpose benzophenone, acetophenone or acetone were
used. The benzophenone/acetonitrile/alcohol system has been em-
ployed for reduction of cationic dyes [37,38]. The third goal aimed
at the reduction of 1-aminoAQ which is a subsequent photoreac-
tion of 1-nitroAQ.
2
. Experimental
The synthesized (A2 and N2) and the purchased (A1, N1, N2: Al-
drich, TCI) AQs are the same as used previously [27]. The other
compounds and the solvents were used as commercially available
(
Aldrich, Fluka, Merck) and checked for impurities. Acetonitrile
Fig. 1. Absorption spectra of N2 in argon-saturated acetonitrile in the presence of
.1 M TEA at 0, 10, 30 and 200 s irradiation at 313 nm (1–4, respectively), insets:
absorption at 480 nm as a function of time for 100 (s), 30 (d), 10 (e), 3 (N) and
mM (h) TEA.
0
was Uvasol quality, TEA was purified by distillation. Water was
from a Millipore (milli Q) system. The molar absorption coefficient
of N1 in acetonitrile is
1
3
ꢁ1
ꢁ1
e
329 = 4.6 ꢀ 10 M cm and that of A1 is
3
ꢁ1
ꢁ1
e
470 = 6.7 ꢀ 10 M cm . For photoconversion a 1000 W Xe–Hg
lamp and a monochromator was applied. Alternatively, irradiation
was performed with the 254 nm line of a low pressure Hg lamp or
a 250 W high pressure Hg lamp and a band-pass filter for 313 nm.
The UV–Vis absorption spectra were recorded on a diode array
spectrophotometer (HP, 8453). For various donor concentrations
the quantum yield of conversion was obtained, keeping the initial
absorbance at kirr = 313 nm constant. The quantum yields
U
the identity of the products was checked by HPLC analysis [27].
These analyses were performed on a reverse phase ODS-3 (Perfect
cence quantum yield (
= 0.9, as reference. Two excimer lasers (Lambda Physik, pulse
width of 20 ns and energy <100 mJ) were used for excitation at
08 and 248 nm (EMG 200, EMG 210 MSC); for simplicity, the re-
f
U ) using rhodamine 101 in methanol,
U
f
3
sults refer only to kexc = 308 nm. The absorption signals were mea-
sured with two digitizers (Tektronix 7912AD and 390AD) and an
Archimedes 440 computer for data handling. All measurements,
UH2 and
o
NH2 were determined using the aberchrome 540 actinometer and
except for phosphorescence, refer to 24 C.
ꢁ1
Target, 3
composed of 0.5% trifluoroacetic acid and either acetonitrile–water
1:5) or neat acetonitrile as eluent. The retention times were
= 18.5 and 18.2 min for N1 and A1, respectively. For A2 and N2
the method fails since the retention times of are the same under
our conditions, t = 18.1 min; the chromatograms could only be
l
m) column, 0.8 ml min , with mobile phase gradient,
3. Results and discussion
(
3.1. Iradiation of nitro(methyl)AQs
t
r
The absorption spectra of the two nitroAQs are similar, they
have a major peak at 250 nm, e.g. in acetonitrile, and a second
r
separated by different wavelengths. HydroxylaminoAQs were not
observed by HPLC. The experimental error in the quantum yield
determination is mostly ±15% and ±30% for values smaller than
one at k
N
= 330 nm. A colored photoproduct with a band centered
at k = 475 nm appears upon irradiation at 313 nm in the presence
A
of argon-saturated TEA. This is shown in Fig. 1 for N2. Based on
analysis by HPLC, the product of N1 is attributed to A1. Comparable
spectra were recorded using 2-propanol as donor. The absorption
0
.01. A spectrofluorimeter (Cary, eclipse) was employed to mea-
sure the fluorescence and phosphorescence spectra. The fluores-

2
2
H. Görner, H. Gruen / Chemical Physics 368 (2010) 20–27
Table 1
a
Quantum yield of photoconversion of nitroAQs.
Donor
-Phenylethanolb
Conc. (M)
N1:
U
NH2
N2:U
NH2
1
9
Neat
0.5
16 (Neat)
0.06
0.09
0.08
0.04
0.20
0.22
0.14
0.13
TEA
-Propanol
2
a
In argon-saturated acetonitrile, kirr = 313 nm.
Containing 0.01 M acetophenone, values taken from Ref. [27].
b
Table 2
a
Quantum yield of direct photoconversion of nitroAQs to aminoAQs.
Solvent
Donor
Conc. (M)
N1:
U
NH2
N2:U
NH2
Benzene
Acetonitrile
TEA
TEA
0.1
0.001
0.01
0.1
2
0.03
0.002
0.01
0.001
0.008
0.005
0.09
0.01
0.06 (0.07)b
0.02
Fig. 2. Absorption at 480 nm of N2 in argon-saturated acetonitrile in the presence
of 8 (d), 1 ( ), 0.3 (s) and 0.1 M (h) 2-propanol vs time; insets: absorption spectra
D
2-Propanol
2-Propanol
in the presence of 1 M 2-propanol at 0, 20, 50 and 200 s irradiation at 313 nm, 1–4,
respectively.
0.09
c
0.08c
MeCN–H
2
O (4:1)
2
a
b
c
In argon-saturated solution, kirr = 313 nm.
For kirr = 254 nm.
Same value at pH 3 or 8, see text for pH 12.
shown in Fig. 3 in a double logarithmic presentation. The concen-
max
2
tration, where
characteristic for the efficiency, e.g. [donor]1/2 is ca. 1 M for N2/
-propanol in acetonitrile solution. The NH2 values are similar in
argon-saturated benzene or acetonitrile (Table 2). NH2 of N2 in
U
NH2
=
U
is 50% of its maximum, [donor]1/2, is
NH
2
U
U
neat ethanol and methanol is 0.06 and 0.04, respectively. The pho-
tochemical reduction of 1- and 2-nitroAQs in alcohols has earlier
been studied by El’tsov and coworkers [23–25].
UNHOH values of
formation of 1-hydroxylaminoAQ upon irradiation of N1 in 2-pro-
ꢁ
4
panol were 10 for both kirr = 254 and 365 nm [25]. They are too
low and in our hands,
for N1.
UNH2 as a minimum for UNHOH is ca. 0.01
The presence of 10–30% water has no significant effect on
U
NH2
of N2 in 2-propanol at pH 3–8 (Table 2). At pH 12, however, the
product spectrum shows no 475 nm peak as expected for amino-
Fig. 3. Double-logarithmic plots of
UNH2 vs. the TEA (circles) and 2-propanol
(
triangles) concentration for N1 (full) and N2 (open) in argon-saturated acetonitrile.
(
methyl) AQs, but a broad band around 710 nm. This is attributed
to the deprotonated form of 1-hydroxylaminoAQ, in analogy to re-
sults with N1 in the presence of sodium methoxide [25]. Such a
710 nm band was also observed with aqueous TEA (not shown).
spectra of irradiated N2 in the presence of 2-propanol (Fig. 2, inset)
= 330 nm, k = 475 nm, are similar to the N1 case. Close inspec-
k
N
A
tion of the spectra indicates a narrowing of the band (and a small
shift to shorter wavelengths in some cases) as a function of the
irradiation time. We attribute this to the enrichment of the
hydroxylaminoAQ and the subsequent conversion to the slightly
different band of the aminoAQ. Note that nitrosoAQs are not ex-
pected to absorb at 475 nm. In the presence of oxygen, the photo-
induced 475 nm band does not appear at all. Prolonged irradiation
leads to a further species with maximum at 390 nm. They are 1-
3
.3. Sensitized reduction of nitro(methyl)AQs
The photoreduction of the nitroAQs is more efficient when ben-
zophenone, acetophenone or acetone were employed as sensitiz-
ers. The absorption spectra of the nitro(methyl)AQs in argon-
saturated acetonitrile at k < 330 nm are overlapped by those of
the sensitizer. The new colored photoproduct with a band centered
2 2 2 2
aminodihydroAQs, H NAQH and H NMeAQH in the cases of A1
at k
sensitizers. Examples are shown for N1 in the presence of TEA
Fig. 4, inset). Comparable spectra were recorded using alcohols
as donors. Recently, we found that NH2 is up to 0.2 (Table 1). 1-
A
= 475 nm upon irradiation at 313 nm is the same as without
and A2, respectively.
(
3.2. Quantum yield of full reduction
U
Phenylethanol and acetophenone, as already mentioned, function
as donor and sensitizer, respectively [27].
The absorption of the nitro(methyl)AQs at 475 nm increases as
a function of the time of irradiation at 313 nm. Plots of A480 for N2
as a function of the irradiation time are shown in the presence of
TEA (Fig. 1, inset) and 2-propanol (Fig. 2). The slope of the initial
linear part is a measure of the quantum yield, the values are com-
H
A new band centered at k = 390 nm appears upon further irra-
diation in the presence of 2-propanol, i.e. when N2 is completely
converted to the photoproduct A2. This band is due to 1-amino-
2-methyldihydroAQ [27]; the admission of air changes the
piled in Tables 1 and 2.
U
NH2 in neat argon-saturated benzene or
A
390 nm band back to that of A2 with k = 480 nm. A 390 nm peak
ꢁ
3
acetonitrile is below 10 and larger in mixtures with several alco-
hols, approaching
NH2 ¼ 0:04 for N1 in neat 2-propanol and 0.13
for N2. The dependences of NH on the donor concentrations are
and a thermal step (a decrease in A400 within a few minutes or less)
after admission of air could also be registered for N1 with 2-propa-
nol as donor (not shown). For TEA (10 mM) as donor the
U
U
2

H. Görner, H. Gruen / Chemical Physics 368 (2010) 20–27
23
Table 4
a
Quantum yield of photoreduction of aminoAQs.
Sensitizer
Noneb
Donor
TEA
A1:
U
H2
A2:U
H2
0.002
0.003
0.4
<0.002
<0.002
0.5
2-Propanol
Acetone
TEA
2-Propanol
0.6
0.5
Acetophenone
Benzophenone
TEA
TEA
0.4
0.6
0.4
0.5
2-Propanol
0.6
0.5
a
In argon-saturated acetonitrile, kirr = 313 nm, the donor concentration is 0.1 M
for alcohols and 5–9 mM for TEA.
b
Using concentrations of 10 M 2-propanol or 0.5 M TEA.
Fig. 4. Absorption at 480 nm of N1 in argon-saturated acetonitrile in the presence
of 0.1 M 2-propanol (open) and 10 mM TEA (full) with benzophenone (circles),
acetophenone (triangles) and acetone (squares); insets: absorption spectra for N1,
benzophenone and TEA at 20, 130, 200 and 400 s irradiation at 313 nm 1–4,
respectively.
Table 3
a
Quantum yield of sensitized photoconversion of nitroAQs to aminoAQs.
Sensitizer
Acetone
Donor
N1:
U
NH2
N2:U
NH2
2-Propanol
Methanol
TEA
0.06
0.09
0.1
0.12
0.09
0.05
0.12
0.10
0.20
0.22
Benzophenone
0.11 [0.1]b
0.12 [0.1]
0.17
2
-Propanol
Ethanol
Methanol
TEA
0.12
0.22
0.20
Acetophenone
Fig. 6. Plots of I
f
(kex = 390 nm, kem = 590 nm) as a function of time of irradiation at
2
-Propanol
3
13 nm of A2 in argon-saturated acetonitrile in the presence of benzophenone and
a
In argon-saturated acetonitrile, kirr = 313 nm. The donor concentration is 0.1 M
0.1 M 2-propanol (s) and after admission of air (d); insets: Fluorescence spectra
for alcohols and 5–9 mM for TEA, respectively.
under argon at 200 s (full) and after admission of air (broken).
b
In benzene.
3.4. Photoreduction of amino(methyl)AQs
The absorption spectrum of A1 in argon-saturated acetonitrile
has a major band centered at k
appears upon irradiation at 313 nm in the presence of 2-propanol
Fig. 5, insets) or TEA. The photoproduct is attributed to 1-amin-
A H
= 475 nm. A peak at k = 390 nm
(
odihydroAQ (see above). The spectra are similar to those for A2,
where the photoproduct is 1-amino-2-methyldihydroAQ. The
slope of the initial linear increase of A390 is taken as measure of
the quantum yield (UH2 ). Examples of the conversion to the dihy-
droAQs in the presence of TEA or 2-propanol are shown in Fig. 5,
the values are listed in Table 4. The quantum yield of formation
of 1-aminodihydroAQ in the presence of 2-propanol or TEA is
small,
UH2 < 0:01, and smaller for A2. The photochemical reduction
of A1 in the acetonitrile–alcohol mixtures is enhanced on addition
of benzophenone (<1 mM), acetophenone (5 mM) or acetone
(
0.3 M). It should be noted that the quantum yield of reduction
of A1 in acetone–water (1:1, vol) is = 0.06 at pH 3–11 [26].
The observed H2 values under sensitized conditions (Table 4)
U
d
U
Fig. 5. Absorption at 390 nm of A1 in argon-saturated acetonitrile in the presence
of 0.1 M 2-propanol (open) and 10 mM TEA (full) with benzophenone (circles),
acetophenone (triangles) and acetone (squares); insets: absorption spectra for A1,
benzophenone and 2-propanol at 0, 10 and 100 s irradiation at 313 nm 1–3,
respectively and after admission of air (4).
are much larger and only little smaller than unity, the maximum
attainable value.
3.5. Fluorescence of the photoproducts
Nitro(methyl)AQs show no fluorescence, in contrast to amin-
conversion to A2 is similar (Fig. 4, inset). The time-dependence is
shown in Fig. 4, the slope of the initial linear plot is taken as rela-
oAQs [18–21]. However, fluorescence appears upon conversion of
nitroAQs to the fully reduced aminoAQ photoproducts. The fluores-
tive
U
NH2 (see Table 3).
f
cence emission of A1 or A2 with maximum at k = 590–600 nm is a

2
4
H. Görner, H. Gruen / Chemical Physics 368 (2010) 20–27
Fig. 7. Transient absorption spectra of N2 in (a) air- and (b) argon-saturated
acetonitrile in the presence of 10 mM TEA at 20 ns (s), 1 s ( ), 10 s (h), 0.1 ms
d) and 1 ms (N) after the pulse; insets: kinetics at 500 nm.
Fig. 8. Transient absorption spectra of N2 in (a) air- and (b) argon-saturated 2-
propanol-acetonitrile (1:9) at 20 ns (s), 1 s ( ), 10 s (h), 0.1 ms (d), 1 ms (N)
and 0.1 s (ꢁ) after the pulse; insets: kinetics at 400 nm.
l
D
l
l
D
l
(
mirror image of the excitation spectrum and the latter resembles
that of the absorption. Representative examples are shown in
Fig. 6 (inset). The fluorescence quantum yield of A2 in acetonitrile
and 2-propanol (1:1) is 0.02, in agreement with the literature value
of
f
U = 0.02 in neat acetonitrile [20]. The observation of a green
fluorescence immediately behind the entrance window during
irradiation of the vigorously purged solution indicates reduction
at the two carboxy groups and rapid disappearance, possibly by
reaction with trace amounts of oxygen. The intensity (I
f
)
(kex = 390 nm, kem = 560 nm) of N2 in 2-propanol increases as a
function of the time of irradiation at 313 nm. Likewise, the fluores-
cence intensity of A2 in acetonitrile in the presence of 2-propanol
or TEA and kex ¼ 390 of A1 can be used to measure the photocon-
version to 1-amino-(2-methyl)dihydroAQs and the reversion upon
admission of air. The latter is only partly under direct excitation
and better under sensitized excitation. An almost complete revers-
ible case for A2/benzophenone/2-propanol is shown in Fig. 6.
Fig. 9. Transient absorption spectra of N1 in (a) air- and (b) argon-saturated
acetonitrile–water 3:1 in the presence of 2 M 2-propanol at 20 ns (s), 0.1 s (e),
s ( ), 10 s (h), 0.1 ms (d) and 10 ms (j) after the pulse; insets: kinetics at 400
and 500 nm.
l
Apparently,
f
U of the aminodihydroAQs is larger than of the amin-
1
l
D
l
oAQs, an appropriate fluorescence study would however, be far be-
yond the scope of this work.
3.6. Triplet state
3.7. Radicals
The short-lived transient of N2 with a lifetime of 620 ns at
In several solvents, e.g. alcohols, the triplet state of 1-nitro-2-
methylAQ is overlapped by a second long-lived transient with
maximum at k = 400 nm. Examples of the spectra and kinetics in
the presence of 2-propanol and TEA are shown for N2 in acetoni-
trile (Figs. 7 and 8, respectively). The long-lived species in the pres-
room temperature has been attributed to a triplet state [27]. No
triplet quenching by oxygen could be found. The triplet state of
N1 is too short-lived to be observed by our set-up upon excitation
at 308 nm. Generally, the photoreactions are initiated by intersys-
R
1
*
tem crossing, AQ-R is very short-lived, the triplet states of AQs
have a much longer lifetime, = 3–10 s [14,15]. The triplet of
ence of an alcohol is attributed to the conjugate acid of the radical
ꢀ
s
T
l
anion: O
2
NAQH . The spectra in the presence of TEA exhibit two
N2 appears at 400 nm in acetonitrile even in the presence of either
TEA (in low concentration) or 2-propanol (insets in Figs. 7a and 8a).
R
broad bands with maxima at k = 400 and 550 nm which are attrib-
uted to the radical anion. In this respect 1-nitroAQs are the same
AQ [13,17]. The donor radicals are non-detectable under our
conditions.
1ꢂR ꢁ AQ ! 3ꢂR ꢁ AQ
R ꢁ AQ þ h
m
!
ð1; 2Þ
Upon excitation at 308 nm of A2 in argon-saturated acetonitrile,
only a weak transient with maximum at 400 nm was found [27].
T
Owing to the triplet lifetime of s < 20 ns in argon-saturated
acetonitrile, a relatively high donor concentration is required for
effective radical formation. The half-lives (t1/2) are 1 ms – 1 s (Ta-
ble 5) and sensitive to trace amounts of oxygen. In air-saturated
acetonitrile the radical yield is virtually the same, but t1/2 is as
This is in agreement with the literature of low
Uisc for animoAQs
ꢁ
1
[
19–21]. The triplet energy was found to be 220 kJ mol ; the onset
o
of the phosphorescence peak in ethanol at ꢁ196 C is at 480 nm for
ꢁ1
each: A1, A2, N1 and N2. A value of 250 kJ mol has been reported
for A1 [19]. Energy transfer from triplet benzophenone in argon-sat-
urated acetonitrile showed only physical quenching, but no tran-
sient. This is in agreement with the short triplet lifetime of N1.
On the other hand, the triplet state of benzophenone is quenched
by N1, the rate constant in argon-saturated acetonitrile is
short as 3–5 ls (Figs. 7a and 8a). The reactivity of the radicals to-
ward scavening by oxygen in 1-phenylethanol was found to be
two orders of magnitude lower than in acetonitrile [27]. Examples
of the spectra and kinetics of N2 in the presence of aqueous 2-pro-
panol at pH 8 are shown in Fig. 9. The half-lives under air and argon
are t1/2 = 5
ls and ca. 0.5 s, respectively. The decay of the radical in
9
ꢁ1 ꢁ1
k
q
= 6 ꢀ 10 M
s .
the presence of water is slower (Table 5).

H. Görner, H. Gruen / Chemical Physics 368 (2010) 20–27
25
Table 5
a
Absorption maxima and half-live of the radicals of nitroAQs.
Solvent
Donor
Conc. (V)
k
R
(nm)
t1/2 (ms)
N1
N2
N1
N2
Benzene
Acetonitrile
TEA
TEA
0.1
0.1
4
8
0.1
2
420, 550
420, 550
400
390
420, 550
420,550
420, 550
400
2
4
1
10
30
4
20
1
15
40
>500
>500
1
2
-Phenylethanol
-Propanol
400
MeCN-H
2
O (3:1)
TEA, pH 13
-Propanol, pH 8
pH 12
420, 550
420, 500
420, 550
2
2
a
In argon-saturated acetonitrile, kexc = 308 nm.
The TT absorption spectrum of benzophenone in argon-satu-
soAQ. Alternatively, nitrosoAQ results from H-transfer via (4b).
DH reacts by step (5) with nitrosoAQ to form the corresponding ni-
troso-derived radical and termination (6) yields hydroxylaminoAQ.
ꢀ
rated acetonitrile is quenched by alcohols. The resulting ketyl rad-
ical reacts with N1 and N2, yielding a nitro(methyl)AQ radical with
maximum at 400 nm. The same nitroAQ radical was formed upon
excitation of acetophenone or acetone (not shown). The spectra
of A2 and TEA in argon-saturated acetonitrile are shown in
Fig. 10 for benzophenone-sensitized excitation conditions. Con-
cerning the role of water on the transients: a significant effect is
a delay of the radical termination (Table 5).
3ꢂ
ꢀ
ꢀꢁ
ꢀ
ꢀþ
O
2
NAQ þ DH
2
! O
2
NAQH =O
2
NAQ þ DH =DH
ð3Þ
2
ꢀ
2
ꢀ O
2
NAQH ! ONAQ þ O
2
NAQ þ H
2
O
ð4aÞ
ð4bÞ
ꢀ
ꢀ
O
2
NAQH þ DH
2
! ONAQ þ H O þ DH
2
ꢀ
ꢀ
DH þ ONAQ ! D þ ONAQH
ð5Þ
ð6Þ
ꢀ
2
ꢀ ONAQH ! HOHNAQ þ ONAQ
3
.8. Reaction scheme of direct photoreduction of nitro(methyl)AQ
ꢀ
ꢀ
HOHNAQ þ DH ! HOHNAQH þ D
ð7aÞ
ð7bÞ
The proposed mechanism of photoreduction of the nitro-
methyl)AQs is shown in Scheme 3. It involves the triplet state as
the only reactive excited state and triggers a series of reduction
steps via the donor radical. The first chemical step (3) is H-transfer
to this triplet state yielding the acceptor and donor radicals [39–
ꢀ
ꢀ
(
DH þ HOHNAQH ! HOHNAQH þ D
2
HOHNAQH2 ! H NAQ þ H O
2
2
ð8Þ
For the final two reduction steps the semiquinone radical is consid-
ered [27]. It is formed via step (7a) and converted into the product
via (7b) plus (8), i.e. a radical self – coupling reaction is not required.
This would explain why the reaction readily proceeds to the amino
stage, in contrast to molecules which lack an H-accepting group in
close proximity. The earlier proposed mechanism for N1 in 2-propa-
nol contains three photoinduced steps: the first one involves H-
transfer to an excited state of nitroAQ and leads to nitrosoAQ, the
second photon leads to hydroxylaminoAQ as second non-radical
species and the third photon leads to A1 [24,25]. However, the like-
lihood for the two light-induced steps to react efficiently with a do-
nor is not high. The radical is efficiently quenched by oxygen,
4
1]. Termination of the former radical, step (4a), yields the nitro-
ꢀ
ꢁ
ꢀ
thereby forming superoxide radicals ðO =O
2
H Þ which eventually
2
react to hydrogen peroxide with a rate constant of k
q
= (0.5–1) ꢀ
7
ꢁ1 ꢁ1
1
0 M
s . Therefore, when trace amounts of air were admitted,
nitrosoAQ could not be detected due to scavenging of the precursor
by oxygen. According to the proposed mechanism, a minimum of
three nitroAQ molecules in the triplet state is required for one
aminoAQ: they give nitrosoAQs, reactions (3) plus (4), and hydrox-
ylaminoAQs, reactions (5) plus (6), which reacts with a donor
Fig. 10. Transient absorption spectra of A2 in argon-saturated acetonitrile in the
presence of benzophenone and 10 mM TEA at 20 ns (s), 1 s ( ), 10 s (h), 0.1 ms
d) and 1 ms (N) after the pulse; insets: kinetics as indicated.
l
D
l
(
Scheme 3.

2
6
H. Görner, H. Gruen / Chemical Physics 368 (2010) 20–27
Scheme 4.
Scheme 5.
radical via (7) plus (8). The observed
non-sensitized conditions is much smaller than the maximum
attainable value, this being NH2 = 0.33.
U
NH2 (Tables 1 and 2) under
3.10. Reactions of photogenerated radicals with nitro(methyl)AQs
0
0
U
The radicals are efficiently formed via reactions (1 ), (2 ) and
0
(
9 ).
3
.9. Reactions of photoreduction of amino(methyl)AQs
ꢀ
ꢀ
ð90Þ
O
2
NAQ þ DH ! O
2
NAQH þ D
ꢀ
Owing to the fact that during irradiation both DH type radicals
Sequence (4)–(8) is similar to that of direct photolysis, Schemes 4
and 6. The rate constant of reduction step (9 ) can be estimated from
a half-life of t1/2 = 0.2–1 ms. The rate constant of H-transfer from
and aminoAQ molecules are present, further reduction steps (9)
plus (10) of 1-aminoAQ to 1-aminodihydroAQ have to be taken
into account, see Scheme 4. Thus prolonged irradiation of nitroAQs
forms aminoAQs and finally their dihydro forms. Admission of
oxygen, however, converts 1-aminodihydroAQ molecules back,
reaction (11), a process being in common for all AQs.
0
0
6
ꢁ1 ꢁ1
alcohols to triplet ketones is k = (1–2) ꢀ 10 M
s
for 2-propa-
2
0
nol [41] and smaller for methanol. For benzophenone/TEA k =
2
10
ꢁ1 ꢁ1
1
ꢀ 10
M
s
[42]. Therefore, the expected alcohol concentra-
tion for 50% quenching is 0.01–1 M, whereas [TEA] is 3–4 orders
of magnitude smaller. For acetone as sensitizer and 2-propanol as
donor only one radical is involved: 2-hydroxypropyl. This is differ-
ent for the other radical reactions because of the formation of ketyl
and alcohol radicals in equal amount. The benzophenone/donor sys-
tem has been employed for reduction of related dyes [37,38]. The
rate constant of radical termination in aqueous solution is
ꢀ
ꢀ
H
2
2
NAQ þ DH ! H
2
NAQH þ D
ð9Þ
ð10Þ
ð11Þ
ꢀ
ꢀ H
2
2
NAQH ! H
NAQH þ O
2
NAQH þ H
2
NAQ
2
H
2
! ! H
2
NAQ þ H
2 2
O
2
0
The sensitized reduction occurs via photoreaction (1 ), followed
0
by the ketone triplet reaction step (2 ) and reactions (9) and (10).
9
9
9
ꢁ1 ꢁ1
0
2.4 ꢀ 10 , 1.5 ꢀ 10 and 1.1 ꢀ 10 M
s
for methanol, ethanol
1ꢂSens ! Sens
3ꢂ
ð1 Þ
Sens þ h
m
!
2
and 2-propanol, respectively, and the rate constant of termination
ꢀ
ꢀꢁ
ꢀ
ꢀþ
0
3
ꢂ
Sens þ DH
! SensH =Sens þ DH =DH
ð2 Þ
9
ꢁ1 ꢁ1
2
of 2-propanol radicals in acetonitrile is 1 ꢀ 10 M
s
[41,43,44].
The radicals involved in the ketone/TEA system are shown in
Scheme 5.
3.11. Mechanistic aspects of the full photoreduction of
nitro(methyl)AQs
For A1 no triplet is populated on direct excitation and
UH2 is
ꢁ3
smaller than 10 even in the presence of an alcohol. This is due
to the absence of any semiquinone radical. A CT triplet state of
A1 which is inactive towards H-transfer and intramolecular hydro-
gen bonding, has been considered [20,21]. The proposed mecha-
nism of photoreduction of the amino(methyl)AQs requires an
initial radical, either from nitroAQ precursor, reaction (3), or a sen-
The transient absorption spectra and kinetics of the nitro-
(methyl)AQs are shown in Figs. 7–9 and the half-lives are compiled
in Table 5. They show the triplet state of N2 with a maximum
around 400 nm and
T
s = 10 ns and the detectable radicals with
one or two bands in the 350–600 nm range as long-lived interme-
diates. They react via steps (4a) (4b) in nitroso(methyl)AQ or can
be scavenged by oxygen. The radical of N1 is formed without ob-
servable triplet precursor.
The transient absorption signals after a few milliseconds are too
weak to trace the subsequent steps from the nitrosoAQ to the
aminoAQ. The photoreduction requires rigorous deoxygenation,
0
sitizer, step (2 ). An example with benzophenone as sensitizer is
shown in Fig. 10. The important step is H-transfer (9) from the ke-
tyl radical to the aminoAQ yielding the semiquinone radical. The
subsequent termination (10) leads to 1-aminodihydroAQ within
1
ms. Therefore, the resulting absorption is larger around 390 nm
and a bleaching occurs around 475 nm.

H. Görner, H. Gruen / Chemical Physics 368 (2010) 20–27
27
Scheme 6.
e.g. purging with argon (or nitrogen) since the
U
values are much
[9] I. Loeff, S. Goldstein, A. Treinin, H. Linschitz, J. Phys. Chem. 95 (1991) 4423.
10] K. Gollnik, S. Held, D.O. Mártire, S.E. Braslavsky, J. Photochem. Photobiol. A:
Chem. 69 (1992) 155.
11] I. Gutiérrez, S.G. Bertolotti, M.A. Biasutti, A.T. Soltermann, N.A. García, Can. J.
Chem. 75 (1992) 423.
12] A.E. Alegría, A. Ferrer, G. Santiago, E. Sepúlveda, W. Flores, J. Photochem.
Photobiol. A: Chem. 127 (1999) 57.
13] H. Görner, Photochem. Photobiol. 77 (2003) 171.
14] H. Görner, Photochem. Photobiol. Sci. 5 (2006) 1052.
15] H. Görner, Photochem. Photobiol. 82 (2006) 71.
16] L.R. Comini, S.C.N. Montoya, M. Sarmiento, J. Photochem. Photobiol. A: Chem.
[
[
[
lower when trace amounts of oxygen remain. The obvious reason is
scavenging of the radicals by oxygen which prevents radical termi-
nation. Radical termination (4) takes place in the ms range and the
remaining absorption is considerably reduced (Figs. 7–9). The re-
sults of continuous irradiation are in agreement.
[
[
[
[
4
. Conclusion
188 (2007) 185.
The photochemical reactions of 1-nitroAQs were studied in ace-
[17] S. Steenken, P. Neta, in: Z. Rappoport (Ed.), The Chemistry of Phenols, Wiley,
New York, 2003, p. 1107.
tonitrile in the presence of 2-propanol or other alcohols as H-do-
nors and of triethylamine as electron donor. The triplet state of
N2 is short-lived, but can be intercepted. Secondary radicals lead-
ing to the nitrosoAQ were observed for both N1 and N2 by flash
photolysis. The conversion to the corresponding hydroxylamino
form and eventually to A1/A2 is suggested to occur via nitroAQ
and donor radicals. The photoreduction to aminoAQs is a cas-
cade-like mechanism requiring six electron equivalents. Subse-
quent reduction to 9,10-dihydroxyaminoAQs takes place
efficiently. The photoconversion of various non-quinone nitro-
arenes is blocked at the stage of the hydroxylamino form and
diazaarenes are formed rather than the corresponding amino-
arenes. The full photoreduction of nitroAQs is due to the bi-func-
tional AQ system, where both the nitro group and one of the
carbonyl groups are intermediate H-acceptor sites.
[
18] K.J. Reszka, P. Bilski, C.F. Chignell, J.A. Hartley, N. Khan, R.L. Souhami, A.J.
Mendonca, J.W. Lown, J. Photochem. Photobiol. B: Biol. 15 (1992) 317.
[19] C. Müller, J. Schröder, J. Troe, J. Phys. Chem. B 110 (2006) 19820.
[
20] P. Dahiya, M. Kumbhakar, t. Mukherjee, H. Pal, J. Mol. Struct. 798 (2006)
0.
4
[
21] P. Dahiya, M. Kumbhakar, D.K. Maity, T. Mukherjee, A.B.R. Tripathi, N.
Chattopadhyay, H. Pal, J. Photochem. Photobiol. A: Chem. 181 (2006) 338.
22] R. Frank, G. Greiner, H. Rau, Phys. Chem. Chem. Phys. 1 (1999) 3481.
23] A.V. El’tsov, O.P. Studinskii, Y.K. Levental, M.V. Florinskaya, Zh. Org. Khim. 13
[
[
(
1977) 1061 (Engl. Transl. 13 (1977) 975).
[24] A.V. El’tsov, O.P. Studinskii, Y.K. Levental, Z. Org. Khim. 12 (1976) 1594 (Engl.
Transl. 12 (1976) 1594).
[
25] O.P. Studinskii, Y.K. Levental, A.V. El’tsov, Zh. Org. Khim. 14 (1978) 2150 (Engl.
Transl. 14 (1978) 1991).
[26] S.M. Tsui, W. Chu, Chemosphere 44 (2001) 17.
[
[
27] H. Gruen, H. Görner, Photochem. Photobiol. Sci. 7 (2008) 1344.
28] G.A. Russell, E.J. Geels, F.J. Smentowski, K.-Y. Chang, J. Reynolds, G. Kaupp, J.
Am. Chem. Soc. 89 (1967) 3821.
[29] E. Norambuena, C. Olea-Azar, A.M. Rufs, M.V. Encinas, Phys. Chem. Chem. Phys.
(2004) 1230.
6
[
30] D. Döpp, in: W.M. Horspool, P.-S. Song (Eds.), CRC Handbook of
Photochemistry and Photobiology, CRC Press, Boca Raton, 1995, p. 1019.
31] R. Nakagaki, K. Mutai, Bull. Chem. Soc. Jpn. 69 (1996) 261.
Acknowledgements
[
We thank Professor Wolfgang Lubitz for his support, Mrs. Gabri-
ele Schmitz and Mr. Horst Selbach and Leslie J. Currell for technical
assistance.
[32] S. Fukuzumi, Y. Tokuda, Bull. Chem. Soc. Jpn. 65 (1992) 831.
[
[
33] A. Cors, S.M. Bonesi, R. Erra-Balsells, Tetrahedron Lett. 49 (2008) 1555.
34] Y.G. He, A. Gahlmann, S.J. Feenstra, S.T. Park, A.H. Zewail, Chem. Asian J. 1
(
2006) 56.
[
[
35] C. Yu, B. Liu, L. Hu, J. Org. Chem. 66 (2001) 919.
36] A.C. Bhasikuttan, L.V. Shastri, A.V. Sapre, J. Photochem. Photobiol. A: Chem. 84
References
(
1994) 237.
[
[
37] S. Jockusch, H.-J. Timpe, C.-H. Fischer, W. Schnabel, J. Photochem. Photobiol. A:
Chem. 63 (1992) 217.
38] S. Jockusch, H.-J. Timpe, W. Schnabel, N.J. Turro, J. Photochem. Photobiol. A:
Chem. 96 (1996) 129.
39] A. Demeter, T. Bérces, J. Photochem. Photobiol. A: Chem. 46 (1989) 27.
40] A. Demeter, T. Bérces, J. Phys. Chem. 95 (1991) 1228.
41] C.V. Costa, M.A. Grela, M.S. Churio, J. Photochem. Photobiol. A: Chem. 99
[
[
1] J.M. Bruce, in: S. Patei (Ed.), The Chemistry of the Quinoid Compounds, Wiley,
New York, 1974, p. 465.
2] K. Maruyama, A. Osuka, in: S. Patei, Z. Rappoport (Eds.), The Chemistry of the
Quinoid Compounds, vol. 2, Wiley, New York, 1988, p. 737.
3] D. Schulte-Frohlinde, C. von Sonntag, Z. Phys. Chem. N.F. 44 (1965) 314.
4] A.N. Diaz, J. Photochem. Photobiol. A: Chem. 53 (1990) 141.
5] J.F. McKellar, Rad. Res. Rev. 3 (1971) 141.
6] M. Goez, V. Zubarev, J. Phys. Chem. A 103 (1999) 9605.
7] M. Lukeman, M.S. Xu, P. Wan, Chem. Commun. (2002) 136.
8] K.L. McGilvray, M.N. Chretien, M. Lukeman, J.C. Scaiano, Chem. Commun.
[
[
[
[
[
[
[
[
[
(
1996) 51.
[
[
[
42] H. Görner, Photochem. Photobiol. 82 (2006) 801.
43] P.S. Rao, E. Hayon, Biochim. Biophys. Acta 292 (1973) 516.
44] M.S. Alam, B.S.M. Rao, E. Janata, Rad. Phys. Chem. 67 (2003) 723.
(
2006) 4401.