Deactivation of the first excited singlet state of thiophenols

Article abstract of DOI:10.1039/b516924k

On the bases of picosecond and nanosecond laser flash photolysis with detection by emission and absorption spectroscopy, a quantitative description is given of all deactivation channels of the first excited singlet state of thiophenols ArSH(S₁) such as fluorescence, intersystem crossing (ISC), chemical dissociation into radicals, and radiation-less internal conversion (IC). For this purpose, the photolysis of thiophenol and its methyl-, methoxy-, and chloro-substituted derivatives was studied in solvents of increasing polarity: 1-chlorobutane, ethanol, and acetonitrile. The fluorescence lifetime of the thiophenols was found to range from some hundreds of picoseconds up to a few nanoseconds, correlating with fluorescence quantum yields between 0.001-0.040, at room temperature. Depending on the substitution pattern of the aromatic ring, the quantum yield of the S-H bond dissociation was found to be between 0.3-0.5, irrespective of the solvent polarity. In laser photolysis, no triplet formation of the investigated compounds could be observed neither by the direct way nor by subsequent sensitization with β-carotene. As a difference to the total, the radiation-less internal conversion (Φ_IC ≥ 0.5) was found to be the dominating process. the Owner Societies 2006.

Full text of DOI:10.1039/b516924k

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/pccp | Physical Chemistry Chemical Physics
Deactivation of the first excited singlet state of thiophenols
Yasser M. Riyad, Sergej Naumov, Ralf Hermann and Ortwin Brede*
Received 30th November 2005, Accepted 3rd February 2006
First published as an Advance Article on the web 23rd February 2006
DOI: 10.1039/b516924k
On the bases of picosecond and nanosecond laser flash photolysis with detection by emission and
absorption spectroscopy, a quantitative description is given of all deactivation channels of the
first excited singlet state of thiophenols ArSH(S1) such as fluorescence, intersystem crossing (ISC),
chemical dissociation into radicals, and radiation-less internal conversion (IC). For this purpose,
the photolysis of thiophenol and its methyl-, methoxy-, and chloro-substituted derivatives was
studied in solvents of increasing polarity: 1-chlorobutane, ethanol, and acetonitrile. The
fluorescence lifetime of the thiophenols was found to range from some hundreds of picoseconds
up to a few nanoseconds, correlating with fluorescence quantum yields between 0.001–0.040, at
room temperature. Depending on the substitution pattern of the aromatic ring, the quantum yield
of the S–H bond dissociation was found to be between 0.3–0.5, irrespective of the solvent
polarity. In laser photolysis, no triplet formation of the investigated compounds could be
observed neither by the direct way nor by subsequent sensitization with b-carotene. As a
difference to the total, the radiation-less internal conversion (FIC Z 0.5) was found to be the
dominating process.
Introduction
ArSH þ hv/2hv - [ ArSHꢀ þ esꢁolv
1
]
- ArSꢀ þ H1 þ eꢁsolv
ionization
ionization
(7a)
(7b)
Because of their antioxidant properties, the photophysics and
photochemistry of thiols are of great interest in various fields
of science and technology.1,2 In biological systems, the wide-
spread thiols play a prominent role. Because of their low
ionization potential and the relatively weak S–H bond, they
are active in radical repair and deactivation as well as in
cellular redox processes.3,4 Thiols are often thought to act as
protectors against ionizing radiation via their radical scaven-
ging activity. In addition, it was found that thiyl radicals also
cause biologically important chemical changes such as the
efficient cis–trans isomerisation of mono- and polyunsaturated
fatty acid residues in model membranes via a catalytic ac-
tion.5,6 The knowledge about light- and radiation-induced
transient reaction behaviour of aliphatic thiols is widely under-
stood, while that of aromatic thiols is limited.
ArSꢁ þ hv - ArSꢀ þ eꢁsolv
ArSH = thiophenol, E = excess heat.
Our experimental conditions were chosen in such a way that
self-quenching (reaction (6)), and a possible photoionization,
which occurs at higher laser power (reaction (7a)) or at high
pH-values (reaction (7b)) in aqueous solution, could be ne-
glected. Our interest was only focused on the remaining
reactions (1)–(5).
Generally, it is known that photoexcitation of aromatic
thiols in polar or non-polar solvents at room temperature
results in the formation of phenylthiyl radicals (ArSꢀ).7–9 Yet,
the excited state precursors of the phenylthiyl radical forma-
tion of aromatic thiols have not been characterized. Informa-
tion on the mechanism of the radical formation does not exist,
nor has any quantitative data of the deactivation channels of
the first excited singlet state of the aromatic thiols been
reported.
To establish what happens after a thiophenol (ArSH)
molecule has been excited by an UV photon with regard to
the relaxation of the first excited singlet state ArSH(S1), the
competing reactions (1)–(6) should be considered.
In a previous paper,10 we reported a quantitative description
of the deactivation channels of the first excited singlet state of
2- and 4-thiosalicylic acids. The main point of that study was
the first observation of a directly photogenerated triplet state
of those compounds. But, as it will be shown in this paper,
triplet states of thiophenols are not generated directly by
photoexcitation. The triplet states of these compounds, how-
ever, could be generated via pulse radiolysis by sensitization
experiments with compounds of higher triplet energy.11
In this paper, we give a comprehensive picture and a
quantitative description of all deactivation channels of the
first excited singlet state (S1) of various thiophenols generated
by mono-photonic excitation (see reactions (1)–(5)). Because
ArSH þ hv1 - ArSH(S1)
ArSH(S1) - ArSH(S0) þ hv2
- ArSH(S0) þ E
(1)
(2)
fluorescence
internal conversion (3)
intersystem crossing (4)
- ArSH(T1)
- ArSꢀ þ Hꢀ
þ ArSH - 2 ArSH(S0)
dissociation
(5)
(6)
self-quenching
University of Leipzig, Interdisciplinary Group Time-Resolved
Spectroscopy, Permoserstr. 15, D-4303 Leipzig, Germany. E-mail:
brede@mpgag.uni-leipzig.de; Fax: þ49-341-235-2317
ꢂc
This journal is the Owner Societies 2006
Phys. Chem. Chem. Phys., 2006, 8, 1697–1706 | 1697

View Article Online
of the complexity of the problem and the variety of influencing
factors, the treatment is based on the separate investigation of
the singlet relaxation processes. Therefore, we directly inves-
tigated all of the S1 deactivation pathways except for internal
conversion (reaction (3)) using the complementary methods of
laser photolysis and pulse radiolysis with detection by emis-
sion and absorption spectroscopy. The remaining radiation-
less internal conversion was then determined by a difference
calculation using the experimental data.
Electron pulse radiolysis. The liquid samples were irradiated
with high energy electron pulses (1 MeV, 12 ns duration)
generated by a pulse transformer type accelerator ELIT (In-
stitute of Nuclear Physics, Novosibirsk, Russia). The dose
delivered per pulse was measured with an electron dosimeter
and was usually between 50–100 Gy. Detection of the transient
species was performed using an optical absorption set-up
consisting of a pulsed xenon lamp (XPO 450, Osram), mono-
chromator (SpectraPro-500, Acton Research), R4220 photo-
multiplier (Hamamatsu Photonics) and 1 GHz digitizing
oscilloscope (TDS 640, Tektronix). Further details of this
equipment are given elsewhere.14
Experimental
Materials
All experiments were performed at room temperature.
Freshly prepared solutions were used flowing continuously
through a 5 mm or 10 mm quartz sample cell in laser
photolysis or pulse radiolysis, respectively. Prior to the laser
photolysis and pulse radiolysis experiments, the solutions were
deoxygenated by bubbling with purest grade N2 or saturated
with the desired gas for 15 min, and were used within 1 h.
Acetonitrile (AN), ethanol (EtOH), and 1-chlorobutane
(1-BuCl) of the highest spectroscopic grade were chosen as
solvents (99.9%, VWR). Benzophenone (99.9%), naphthalene
(99%) and b-carotene (95%) were obtained from Aldrich.
High grade phenol (99.5%) was obtained from Riedel–de
Haen. Water treated in a Millipore milli-Q plus system was
used for the experiments in aqueous solutions.
Quantum chemical approach. Dipole moments and S–H
bond dissociation energies for the singlet ground and
excited states of the parent thiophenol were calculated using
DFT B3LYP method15 with the 6-311þG(d, p) basis set
(Gaussian 03).16
The following thiophenols were investigated: thiophenol
(ArSH, Acros), 2-, 3-, and 4-thiocresol (2-, 3-, and 4-CH3-
ArSH, Aldrich), 2,4-, 2,6-, and 3,5-dimethylthiophenol (2,4-,
2,6-, 3,5-DMTP) (Acros), 2,4,6-trimethylthiophenol (2,4,6-
TMTP, Acros), 2-, 3-, and 4-methoxythiophenol (2-, 3-, and
4-MeO-ArSH, Acros), and 4-chlorothiophenol (4-Cl-ArSH,
Aldrich). All thiophenols were of the highest analytical grade
and partially further purified by recrystallization from an
alcohol–water mixture (1 : 2) or preparative chromatography.
Results and discussion
Steady state measurements
Fluorescence spectra of thiophenol were studied in polar
(acetonitrile and ethanol) and non-polar surroundings (1-
chlorobutane) as shown in Fig. 1. In each of these solvents,
the spectra consist of a broad and unstructured band. The
fluorescence maxima are given in Table 1. The steep decay on
the UV side is explained by the background Raman line
appearing in all solvents. For substituted thiophenols, an
analogous behaviour has been formed. Hence, it was found
that the fluorescence maxima of all aromatic thiols are around
lF (max) = 295–310 nm.
Apparatus and methods
The ground state optical absorption and fluorescence spectra
were recorded with an UV/VIS spectrophotometer (UV-2101
PC, Shimadzu) and a FluoroMax-2 (Instruments S. A., Jobin
Yvon-Spex), respectively.
Picosecond laser photolysis. The fluorescence lifetimes were
measured with an amplified (CPA) 10 Hz Ti:Sapphire femto-
second-UV laser system. After excitation with a 250 fs
(FWHM) laser pulse at 253 nm, a highly sensitive streak
camera C5680/M5676 (Hamamatsu Photonics) recorded the
fluorescence kinetics. A time resolution of 1 ps was achieved
for the emission studies by applying the deconvolution tech-
nique.12
Nanosecond laser flash photolysis. Real-time nanosecond
laser photolysis was applied with detection by absorption
spectroscopy. The solutions were photolyzed by the fourth
harmonics (266 nm) of a Quanta-Ray GCR-11 Nd:YAG laser
(Spectra Physics). Pulses of o3 ns duration (FWHM) with
energies of 0.5 mJ were used. The optical detection system
consisted of a pulsed xenon lamp (XBO 150, Osram), mono-
chromator (SpectraPro-275, Acton Research), R955 photo-
multiplier tube (Hamamatsu Photonics) or fast Si-photodiode
with 1 GHz amplification, and a 500 MHz digitizing oscillo-
scope (DSA 602 A, Tektronix). The laser power was mon-
itored for every pulse using a bypath with a fast Si photodiode.
The laser flash photolysis set-up is reported elsewhere in more
detail.13
Fig. 1 Corrected fluorescence spectra (lexc = 266 nm) of thiophenol
(0.4 mmol dmꢁ3) in 1-chlorobutane ( ꢃ ꢃ ꢃ ꢃ ), ethanol (– – –), and
acetonitrile (——) solutions at room temperature.
ꢂc
This journal is the Owner Societies 2006
1698 | Phys. Chem. Chem. Phys., 2006, 8, 1697–1706

View Article Online
Table 1 Spectral parameters of fluorescence and phenylthiyl radicals
of thiophenols
b
ArSꢀ
e/dm3 molꢁ1 cmꢁ1
lmax/nm
Substances
ArSH
lF(max)a/nm
310
295
460
300
460
300
470
310
500
320
460
300
480
320
520
300
510
310
480
300
480
300
470
320
480
2000
2200
2000
2100
2000
2200
6300
3500
3500
3000
3000
3500
2-Me-ArSH
3-Me-ArSH
4-Me-ArSH
2-MeO-ArSH
3-MeO-ArSH
4-MeO-ArSH
4-Cl-ArSH
2,4-DMTP
3,5-DMTP
2,6-DMTP
2,4,6-TMTP
a
300
300
310
315
305
315
325
310
300
296
305
The lF maxima are in polar solvents (EtOH and AN) at room
b
ꢀ
temperature. The e
values were calculated in a N2O-saturated
(ArS )
alkaline aqueous solution (pH 12) of thiophenols (2 mmol dmꢁ3) at
l
max in the visible region using sodium azide (0.02 mol dmꢁ3) by pulse
radiolysis. The wavelength accuracy is ꢄ1 nm.
Fluorescence kinetics
We determined the fluorescence lifetimes, tF, by recording the
fluorescence decay using a streak camera for the S0 ’ S1
transition of thiophenols in 1-BuCl, EtOH and AN.
As shown in Fig. 2, the fluorescence kinetics of thiophenol is
described by a bi-exponential function. One component decays
rapidly in a few tens of ps (a tS2 = 30–50 ps), and the other
decays in longer times (tS1 = 0.58–2.70 ns (see also Table 2).
For substituted thiophenols, a similar behaviour is ob-
served. The corresponding fluorescence lifetimes are given in
Table 2. The only exception is the fluorescence decay of 2-
methoxythiophenol, which can be fitted by a single exponen-
tial function in all of the solvents used (see Fig. 3). The
lifetimes were found to be nearly constant around B40 ps,
irrespective to the solvent polarity.
Fig. 2 Experimental fluorescence decay (—) (lexc = 253 nm) of
thiophenol (0.4 mmol dmꢁ3) in (A) 1-BuCl, (B) EtOH, and (C) AN
solutions. The bi-exponental fit curves (J) describe the fluorescence
kinetics.
Obviously, the values of the short component (tS2) in EtOH,
and AN are close to that in 1-BuCl. But for the major
component (tS1) a more pronounced difference is observed.
We assume that S0 ’ S2 transition is too rapid and therefore it
occurs without any chemical consequence. Despite of a few
exceptions, the following conclusions can be drawn from the
data given in Table 2:
Due to the use of a cut-off filter (l = 308 nm), the
contributions of the 290 nm emission caused by the quartz
cell and the scattering light due to the excitation pulse are
excluded from consideration. Nevertheless, the fluorescence
decay profiles measured within a narrow emission wavelength
window did not show the expected single exponential decay.
The short component of the decay behaviour might be either
be attributed to the spectral shifts which appear when the
timescales for relaxation and emission are comparable17 or to
a S0 ’ S2 transition. Using different cut-off filters (308, 340,
and 390 nm), we found that the intensity of the short compo-
nent (tS2) was strongly decreased. This suggests a S0 ’ S2
relaxation.
(i) For a fixed thiophenol type, the fluorescence lifetime
always increases with rising solvent polarity (characterized by
the dielectric constant).
(ii) At first inspection, a mono-substitution with an electron-
donating group (methyl- or methoxy) enhances the fluores-
cence lifetime.
(iii) Compared with m- or p-thiocresol, a methyl group in
the ortho position enhances the fluorescence lifetime strongly.
(iv) Mono-substitution in the para position only slightly
affects the fluorescence lifetimes.
ꢂc
This journal is the Owner Societies 2006
Phys. Chem. Chem. Phys., 2006, 8, 1697–1706 | 1699

View Article Online
Table 2 Quantum yields (F) of all deactivation channels and fluor-
escence kinetics of the first excited singlet state of thiophenols in
1-BuCl, EtOH, and AN at room temperature. The indices of F refer
to fluorescence (F), triplet (T), dissociation (D), and radiation-less
internal conversion (IC), tS1 and tS2 are the fluorescence lifetimes of S1
and S2, respectively
1-BuCl (e = 7 dm3 molꢁ1 cmꢁ1 a
)
b
tF/ns
tS1
tS2
0.03
0.12
—
—
—
—
FF
FT
FD
FIC
ArSH
0.003
0.006
0.006
0.0006
0.006
0.020
0.001
0.0006
0.005
0.004
0.007
0.003
0
0
0
0
0
0
0
0
0
0
0
0
0.30
0.45
0.40
0.40
0.40
0.45
0.40
0.40
0.40
0.40
0.40
0.50
0.697
0.544
0.594
0.594
0.594
0.530
0.599
0.599
0.595
0.596
0.593
0.497
0.58
3.10
0.30
0.58
0.03
1.40
2.10
1.2
1.50
2.35
3.20
1.70
2-CH3-ArSH
3-CH3-ArSH
4-CH3-ArSH
2-MeO-ArSH
3-MeO-ArSH
4-MeO-ArSH
4-Cl-ArSH
2,4-DMTP
2,6-DMTP
3,5-DMTP
2,4,6-TMTP
Fig. 3 Experimental fluorescence decay of 2-methoxythiophenol (0.4
mmol dmꢁ3) in 1-BuCl ( ꢃ ꢃ ꢃ ꢃ ), EtOH (– – –), and AN (——)
solutions.
0.27
0.01
0.04
0.03
0.18
0.02
Indeed, the lifetimes (tS1) increase in the following order:
t3,5-DMTP > t2,6-DMTP > t2,4,6-DMTP > t2,4-DMTP
(viii) For 2,4,6-TMTP, the fluorescence lifetimes are quite
.
EtOH (e = 24 dm3 molꢁ1 cmꢁ1 a
)
tF/nsb
similar in 1-BuCl, EtOH, and AN (tS1 = 1.7 ns).
Hence, the fluorescence behaviour of aromatics depends not
only on the chemical and physical properties of the surround-
ings,18 but also on the nature and position of the substituent.
Because of the very complex factors of influence (protic and
aprotic media, polarity, viscosity, and dipole moment pheno-
mena), any satisfying interpretation is difficult.
tS1
tS2
FF
FT
FD
FIC
ArSH
0.004
0.017
0.011
0.002
0.011
0.03
0.002
0.002
0.01
0
0
0
0
0
0
0
0
0
0
0
0
0.30
0.30
0.30
0.30
0.50
0.40
0.70
0.30
0.30
0.30
0.45
0.55
0.696
0.683
0.689
0.698
0.489
0.570
0.298
0.698
0.690
0.690
0.540
0.443
1.20
3.50
0.31
1.35
0.04
2.00
1.12
1.25
0.50
2.70
2.90
1.70
0.05
0.03
0.03
0.05
—
0.05
0.04
2-CH3-ArSH
3-CH3-ArSH
4-CH3-ArSH
2-MeO-ArSH
3-MeO-ArSH
4-MeO-ArSH
4-Cl-ArSH
2,4-DMTP
2,6-DMTP
3,5-DMTP
2,4,6-TMTP
On the basis of our data, general solvent–solute excited state
interactions19 as well as specific interactions were considered.
However, we are unable to distinguish between the influences
of these effects because of the small polarity of aromatic thiols.
On the other hand, it seems that the nature of the chromo-
phore mainly affects the fluorescence lifetimes.
—
0.03
0.03
0.03
0.01
0.01
0.01
0.007
AN (e = 38 dm3 molꢁ1 cmꢁ1 a
)
Generally, it seems that electron-donating groups induce
stabilization of the first excited singlet state, whereas electron-
withdrawing groups do the opposite. Surprisingly, the 2-MeO-
ArSH exhibits an extremely short fluorescence lifetime (tF
between 30–40 ps). This might be caused by the low stability of
the first excited singlet state S1 and points to a more efficient
non-radiative relaxation. This peculiar influence might be
attributed to the interaction between the –SH group and an
oxygen lone electron pair of the methoxy substituent or the
formation of something like a hydrogen bond.20 Such an effect
may lead to a destabilization of the first excited singlet state
and increases considerably the yield of the non-radiative
internal conversion deactivation channel.
tF/nsb
tS1
tS2
FF
FT
FD
FIC
ArSH
0.004
0.013
0.013
0.001
0.010
0.037
0.002
0.001
0.009
0.009
0.012
0.007
0
0
0
0
0
0
0
0
0
0
0
0
0.30
0.40
0.40
0.50
0.80
0.50
0.60
0.40
0.50
0.40
0.50
0.50
0.696
0.587
0.587
0.499
0.190
0.463
0.398
0.599
0.491
0.591
0.488
0.493
2.70
5.20
0.35
1.20
0.04
2.60
3.00
1.90
1.00
2.90
3.60
1.70
0.03
0.04
—
0.03
—
0.24
0.02
0.01
0.03
0.03
0.02
0.02
2-CH3-ArSH
3-CH3-ArSH
4-CH3-ArSH
2-MeO-ArSH
3-MeO-ArSH
4-MeO-ArSH
4-Cl-ArSH
2,4-DMTP
2,6-DMTP
3,5-DMTP
2,4,6-TMTP
a
b
The yields are within ꢄ10% accuracy. The error of tF is about
Fluorescence quantum yields (UF)
ꢄ10%.
The fluorescence quantum yields of the investigated com-
pounds were determined by steady state fluorescence measure-
ments using phenol in cyclohexane, ethanol and acetonitrile as
standards (FF = 0.083, 0.16 and 0.19, respectively).13 The
fluorescence quantum yield FF was determined with the same
optical density for the phenol and the investigated aromatic
thiols at lexc = 266 nm, i.e., OD266 nm (ArOH) = OD266 nm
(ArSH). Under these optically matched conditions, FF was
determined for diluted solutions (r0.5 mmol dmꢁ3) by
(v) Surprisingly, a methoxy group in the ortho (neighbour-
ing) position to the –SH group causes extremely short lifetimes
(tF between 30–40 ps).
(vi) A chloro-substituent in the para position seems to
slightly enhance the fluorescence lifetime.
(vii) The fluorescence lifetimes of di-, and trimethyl-substi-
tuted thiophenols are longer than those of the thiophenol.
ꢂc
This journal is the Owner Societies 2006
1700 | Phys. Chem. Chem. Phys., 2006, 8, 1697–1706

View Article Online
comparing the integrals of the reference and sample substance
over the whole emission range.21,22
The calculated fluorescence quantum yields in the three
solvents 1-chlorobutane, ethanol and acetonitrile with an error
limit within ꢄ10% are given in Table 2. It is obvious that
aromatic thiols fluoresce with a low quantum yield (in the
range between 1 ꢅ 10ꢁ3 and 4 ꢅ 10ꢁ2). For thiophenol, the FF
values are comparable in polar solvents, but slightly higher
than those in non-polar surroundings. For substituted thio-
phenols, the same trend is observed but the increase in the
quantum yields from non-polar to polar surroundings is more
significant (Table 2). In contrast, FF values of all para-
substituted thiophenols (–CH3, –OCH3, and –Cl) are in the
range between 1 ꢅ 10ꢁ3 and 2 ꢅ 10ꢁ3, irrespective of the
solvent polarity.
Hence, the short fluorescence lifetimes as well as the low
quantum yields and the large difference between the radiative
(t0 = tF/FF) and experimental (tF) lifetimes of the investi-
gated compounds indicate that the radiative channel is of
minor importance and accounts for less than 4% of the S1
decay. This means that radiation-less and/or photodissocia-
tion channels play the main role in the deactivation of the first
excited singlet state.
A thorough analysis of the deactivation mechanism should
entail a detailed study of all non-radiative processes. Accord-
ing to whether the transition is spin-allowed or spin-forbidden,
the radiation-less transitions are known as internal conversion
(IC) or intersystem crossing (ISC), both in connection with
vibrational relaxation (VR). In the next sections, the radia-
tion-less channels and the dissociation will be characterized
quantitatively.
Flash photolysis, transient absorption
Fig. 4 Transient absorption spectra obtained after laser photolysis of
In oxygen-free saturated 1-chlorobutane solution, the time-
dependent transient absorption spectra obtained after 266 nm
thiophenol purged with N2 in (A) 1-chlorobutane (0.4 mmol dmꢁ3
)
and (B) acetonitrile (0.6 mmol dmꢁ3) solutions taken 0.5 (K), 5 (m),
and 15 (’) ms after the pulse (laser power = 0.7 mJ). The insets show
time profiles of N2 (—) and O2 ( ꢃ ꢃ ꢃ ꢃ ) saturated solutions at char-
acteristic wavelengths.
laser flash photolysis of thiophenol (ArSH, 0.4 mmol dmꢁ3
)
are shown in Fig. 4A. The initial spectrum measured 500 ns
after the pulse consists of three bands, a sharp and intense one
with a maximum at 295 nm, and two moderate and broad
bands in the 340–380 and 440–490 nm regions.
Analyzing the time profiles given as insets of Fig. 4A, the
long-lived species observed in the UV (295 nm) and visible
(450–490 nm) regions is attributed to the phenylthiyl radical
(ArSꢀ) which is formed by the homolytic S–H bond fission of
the first excited singlet state of thiophenol ArSH(S1) (cf.
reaction (5)). The absorption maxima of the phenylthiyl
radical agree in laser photolysis7–9,23–26 and pulse radiolysis
measurements.27–30
(340–380 nm) reveals that the species at 360 nm is much
longer-lived than the phenylthiyl radicals. Moreover, we as-
sume that the cyclohexadienyl-type radical decays by a second
order process (reaction (10)).
2 ArSꢀ - ArSSAr k8 E 2 ꢅ 109 dm3 molꢁ1 sꢁ1
(8)
The product obtained 15 ms after the pulse with lmax o 310
nm (see Fig. 4A) is attributed to molecular disulfide resulting
from the recombination of phenylthiyl radicals (reaction (8)).
Indeed, thiyl radicals decay with rates close to the diffusion
control.29–31 In addition to the disulfide product, cyclohexa-
dienyl-type radicals (H-adduct), could be the responsible for
the band observed in the 340–380 nm region. The H-adduct is
formed upon the reaction of the dissociative hydrogen radicals
(cf. reaction (5)) with the benzene ring of thiophenol as shown
by reaction (9). The analysis of the time profiles in this region
ð9Þ
ð10Þ
In oxygen saturated solution, it was found that the transient
observed at 360 nm is slightly quenched, whereas those at 290
ꢂc
This journal is the Owner Societies 2006
Phys. Chem. Chem. Phys., 2006, 8, 1697–1706 | 1701

View Article Online
and 460 nm are not affected (see insets of Fig. 4A). These
results support the identification of phenylthiyl radicals which
are insensitive toward oxygen.24,32–34 Therefore, the entire
absorption spectra are dominated by the more stable phenyl-
thiyl radical and its disulfide product.
The laser photolysis of thiophenol in EtOH and AN shows a
similar trend to that seen for the 1-BuCl solution under similar
conditions (see Fig. 4B for an example in AN). Moreover, the
solvent change from non-polar (1-BuCl) to polar (EtOH or
AN) only causes a very minor bathochromic shift for the
absorption maxima of ArSꢀ (r5 nm). Hence, the absorption
maxima of the phenylthiyl radicals derived from thiophenols
taken independently in various solvents are given in Table 1.
Laser photolysis experiments of the substituted thiophenols
were carried out in the same manner and under similar
conditions to the experiments carried out for thiophenol.
Although, a similar trend to that seen for thiophenol was
observed, the spectra showed more pronounced absorption
maxima (see Fig. 5). Depending on the substituent nature and
its position, it can be seen that the maxima of the phenylthiyl
radicals and cyclohexadienyl-type radicals are slightly shifted
to longer wavelengths (see also Table 1). The maxima of the
phenylthiyl radicals derived from the substituted thiophenols
agree well with literature.7,23–27,29,30
In order to verify the presence of a cyclohexadienyl-type
radical, pulse radiolysis experiments were carried out in acidic
aqueous solutions (pH = 1.5) under a nitrogen atmosphere.
Under these conditions the solvated electrons are converted to
ꢀH via reaction with a proton (reaction (12)). Then, ꢀH and
ꢀOH abstract an H atom from ArS–H (reaction (13)) as well as
add to the aromatic ring (reaction (14)), with the formation of
cyclohexadienyl-type radicals, cf. Fig. 6.
ð11Þ
eꢁsolv þ H3O1 - Hꢀ þ H2O
(12)
ꢀOH/Hꢀ þ ArSH - H2O/H2 þ ArSꢀ
(13)
ð14Þ
It should be taken into account that cyclohexadienyl-type
radicals decay under high acidic conditions (pH 1.5) into
phenylthiyl type radicals, reaction (15).35 Therefore, one can
rationalize the difference in the kinetics of the cyclohexadienyl-
type radicals measured in neutral (see Fig. 4 and 5) and acidic
(see Fig. 6) solutions.
Fig. 5 Transients absorption spectra obtained after laser photolysis
of various substituted thiophenols: (A) 4-MeO-ArSH (0.5 mmol
dmꢁ3), (B) 4-Cl-ArSH, and (C) 2,4-DMTP (0.3 mmol dmꢁ3) in N2-
purged acetonitrile solutions taken 0.5 (K), 5 (m) and 15 (’) ms after
the laser pulse (laser energy = 0.7 mJ). The insets show time profiles at
characteristic wavelengths.
known as El-Sayed’s rules where transitions between identical
electronically excited states [such as 1(n, p*) 2 3(n, p*) in
thiophenol] are forbidden.36–38 Nevertheless, the thiophenol
triplet formation could be identified and characterized indir-
ectly via pulse radiolysis by the energy transfer from the triplet
state of the solvent benzene.11
ð15Þ
Intersystem crossing
In the laser photolysis of all of the aromatic thiophenols, no
triplet formation could be observed, neither directly nor by
subsequent sensitization experiments with b-carotene. This
behaviour can be explained by the selection criteria for ISC
An exception to this, and in contrast to recent findings on
thiophenol and its derivatives, the photo-induced first excited
triplet state of the 2- and 4-thiosalicyclic acids is directly
observed by its T1–Tn absorption spectra and characterized
ꢂc
This journal is the Owner Societies 2006
1702 | Phys. Chem. Chem. Phys., 2006, 8, 1697–1706

View Article Online
sponding self absorption which closed the window at the UV
border.
The transient absorption spectra of the substituted phe-
nylthiyl radicals were determined in the same manner as those
for thiophenol and under the same experimental conditions of
dose and concentration. Their corresponding extinction coeffi-
cients are given in Table 1. Fig. 7(b–d) shows the absorption
spectra of some substituted phenylthiyl radicals (2-, 3-, and 4-
thiocresol). The radical spectra in the visible region agree well
with the laser photolysis results.
Quantum yields of the phenylthiyl radicals (UD). The radical
quantum yields of the S–H bond photodissociation of the
ArSH(S1) were determined by nanosecond laser photolysis
using the benzophenone triplet, BP(T1), as an external stan-
dard. Using the known benzophenone triplet parameters,
Fig. 6 Transient absorption spectra obtained after pulse radiolysis of
acidic (pH 1.5) aqueous solution of 1 mmol dmꢁ3 thiophenol purged
with N2 taken 1 (K) and 2 (m) ms after the pulse.
FT(BP) = 1 and eT = 6250 dm3 molꢁ1 cmꢁ1 44
and the
,
determined eArSꢀ, the FD values were derived for N2-purged
diluted solutions (r0.5 mmol dmꢁ3) by comparing the con-
centration of reference (BP(T1)) and sample phenylthiyl radi-
cal (ArSꢀ) transients. Further details of the method are
reported in our previous paper.10 The radical quantum yields
of thiophenol and its substituted compounds were thus deter-
mined for solutions in 1-chlorobutane, ethanol and aceto-
nitrile (cf. Table 2). The measurements were performed at
low laser energy, B0.6 mJ per pulse, to avoid two-photon
ionization. In the case of thiophenol, the radical yields in
1-chlorobutane, ethanol, and acetonitrile are comparable and
relatively high (FD E 0.30). It is obvious that thiyl radical
formation shows no influence toward solvent polarity.
For substituted thiophenols, the radical quantum yields in
various solvents are given in Table 2. Overall, the following
conclusions can be drawn:
by sensitization experiments.10 The reason for this could be the
presence of the carboxyl group in these molecules. The first
excited singlet state of the thiosalicylic acids is a 1(p, p*) state,
and as there is a degenerate 3(n, p*) state, this allows the
intersystem crossing from the 1(p, p*) state to the 3(p, p*) state
3
via (n, p*) state.39,40
Dissociation
Extinction coefficient of the thiyl radical (eR). In order to
verify the spectra of the phenylthiyl radicals as well as to
estimate their yields, pulse radiolysis studies of the thiophenols
were performed in aqueous alkaline (pH = 12) solution under
an N2O atmosphere in the presence of sodium azide. Under
ꢀ
these conditions the solvated electrons are converted to OH,
see reaction (16), with subsequent generation of azide radicals,
reaction (17). Then the azide radicals react with the thiolate
anions leading to the quantitative formation of thiyl radicals
and azide anions, reaction (18). For aromatic thiols, the rate
constant k18 was calculated to be in the range (4–6) ꢅ 109 dm3
molꢁ1 sꢁ1. From the kinetic data, we found no evidence for the
formation of disulfide radical anions, previously observed for
aliphatic thiols.41,42
(i) Surprisingly, the high radical quantum yields of the
substituted thiophenols reflect an pronounced influence of
the substituents on the homolytic S–H bond fission of
ArSH(S1).
(ii) The radical yields of 4-chloro, and mono-, di-, and
trimethylthiophenol are comparable, irrespective of the sol-
vent polarity.
(iii) The radical yields of 2- and 4-methoxythiophenol are
extremely high, particularly in polar solvents. In contrast, a
methoxy group in a meta position shows only a small effect.
Because of the inherently weak S–H bond dissociation
energy (BDE) of thiophenols, which has less strength than
those of the O–H bonds in phenols, the differences in BDE’s of
thiophenols are small.45 Therefore, the effect of different
substituents and their positions on the homolytic S–H bond
is less pronounced. Nevertheless, the strength of S–H bonds
depends to some extent on the nature of the substituent. This
was supported by quantum chemical estimations of the
BDE.15,16 As expected, it was found that the S–H BDE of
the ground and first excited singlet states of unsubstituted
thiophenol are 79 and 27 kcal molꢁ1, respectively. This
demonstrates the weakness of the S–H bond of ArS–H(S1).
As a result, a considerable amount of the S1 excitation energy
ought to deactivate via the photodissociation channel.
The comparable radical quantum yields of mono-, di-, and
trimethylated thiophenol can be attributed to the weak
eꢁsolv þ N2O - OHꢀ þ OHꢁ þ N2
(16)
(17)
ꢁ
OHꢀ þ N3 - OHꢁ þ N3
ꢀ
ArSꢁ þ N3 - ArSꢀþN3 k18=5 ꢅ 109 dm3 molꢁ1 sꢁ1(18)
ꢁ
ꢀ
Pulse radiolysis of a N2O-saturated alkaline solution contain-
ing 0.02 mol dmꢁ3 of sodium azide leads to the generation of
ꢀ
the azide radical (N3 , lmax = 275 nm) as shown in Fig. 7a. As
expected, the transient absorption spectrum monitored by
pulse radiolysis of a solution containing 2 mmol dmꢁ3 of
thiophenol at pH = 12 (Fig. 7a) exhibits a peak at lmax = 460
nm. This band is attributed to the phenyꢀlthiyl radical obtained
according to reaction (18). The eArS (see Table 1) was
determined by comparing the phenylthiyl and azide absorp-
ꢀ
tion maxima (measured separately) using e(N3
dm3 molꢁ1 s
)275 nm = 2300
.
ꢁ1 43 Unfortunately, we could not see the absorp-
tion maxima of the phenylthiyl radical in the UV region
because of the high thiophenol concentration and the corre-
ꢂc
This journal is the Owner Societies 2006
Phys. Chem. Chem. Phys., 2006, 8, 1697–1706 | 1703

View Article Online
Fig. 7 Maximum transient optical absorption spectra obtained by pulse radiolysis of N2O-purged aqueous alkaline solutions (pH = 12) of: (a)
0.02 mol dmꢁ3 NaN3 only (J), and (a)–(d) thiophenol and its methyl derivatives (2 mmol dmꢁ3) in the presence of 0.02 mol dmꢁ3 NaN3 (K).
influence of the electron donating methyl group, which has a
small effect on the S–H bond homolysis. In other words,
neither the number of methyl group nor the position has
significant influence on the S–H bond and this may explain
why their radical yields are comparable.
no dependence of the radical yield on the specific or general
properties of the solvents.
Internal conversion
The remaining process is the radiation-less deactivation of the
first excited singlet (S1) to the singlet ground state (S0), called
internal conversion. In contrast to the other deactivation
channels studied, there is no direct and easy way to measure
the efficiency of the internal conversion. Using the data
directly determined for the competing ArSH(S1) deactivation
processes, the quantum yield difference to the total gives the
fraction of the internal conversion, reaction (19).
A strong electron donating methoxy group at ortho and
para positions causes higher radical quantum yields. This is
because a methoxy group in these positions leads to an extra
stability of the thiyl radical by decreasing the spin density at
the sulfur which leads to a sharp reduction of the BDE (S–H)
value. As a result, the S–H homolysis becomes much easier
and their corresponding radicals are much more stable than
the other substituents. On the other hand, a methoxy group in
the meta position shows only a minor or no effect on the
radical yield. Thiophenols with electron-withdrawing substi-
tuents, such as 2- and 4-thiosalicylic acids,10 exhibit much
smaller radical quantum yields than those of thiophenols with
electron-donating substituents. Therefore, the dissociation
process seems to be influenced mainly by the electronic pattern
of the substituent on the aromatic ring. In other words, there is
FIC = 1 – (FF þ FT þ FD)
(19)
The corresponding data are given in Table 2 together with the
experimentally determined quantum yields of the previously
described deactivation channels calculated for 1-chlorobutane,
ethanol and acetonitrile solutions.
It can be seen that the thiophenol singlet state relaxation via
IC is independent of the solvent nature. This indicates the
ꢂc
This journal is the Owner Societies 2006
1704 | Phys. Chem. Chem. Phys., 2006, 8, 1697–1706

View Article Online
weak dependence of photophysical and photochemical proper-
ties of thiophenols on the general and specific effects of various
solvents.
ences in the deactivation mechanisms of the first excited singlet
states of thiophenol and phenol analogs.13
Practically no solvent effects were observed on the spectral
properties or the kinetics of the first excited singlet state of
thiophenols, ArSH(S1). Moreover, it seems that the stability of
ArSH(S1) is mainly determined by the electronic pattern of the
substituents. Electron-withdrawing substituents cause shorter
fluorescence lifetimes whereas electron-donating groups cause
longer fluorescence lifetimes.
Analyzing the effects of substitution at the aromatic moiety,
all these various substances show comparable photophysical
characteristics. The FIC values of thiophenol derivatives range
from 0.5–0.7. On the basis of the data given in Table 2 (taking
singlet lifetime tS1 as a crude measure for characterizing
radiation-less deactivation), the following can be concluded:
(i) The FIC values are comparable and larger than 0.50 (with
few exceptions).
In the direct photolysis of all aromatic thiols (with the
exception of 2- and 4-thiosalicylic acids10), no triplet forma-
tion could be observed, either directly or by subsequent
sensitization. Independently, the thiophenol triplet formation
could be identified and characterized indirectly via pulse
radiolysis by the direct energy transfer from the triplet state
of the solvent benzene.11
(ii) An exception is represented by the methoxy group in
ortho and para positions, particularly in polar solvents, which
have efficient photochemical deactivations.
(iii) Mono- and dimethyl substituents, irrespective of their
positions on the aromatic moiety, and a chloro substituent in a
para position have comparable quantum yields. In contrast,
2,4,6-trimethylsubstituted thiophenol shows relatively low FIC
values suggesting that photodissociation is more prominent.
Generally, the overall IC process can be understood in terms
of the rate controlling resonance interaction (RI) between the
S1 and the S0 vibrational levels followed by vibrational
relaxation (VR). In liquids, the VR from an excited state,
e.g., from S1 to the vibrational ground state S0, is very rapid
Finally, the internal conversion, which is unaffected by the
solvent nature, plays dominant role in the deactivation
mechanism, i.e., FIC Z 0.5 (with some exceptions). The
exceptions are 2- and 4-methoxythiophenol and 2,4,6-tri-
methylthiophenol because of the efficient photochemical de-
activation process.
(kVR ꢅ 1013
s )
ꢁ1 46 and the excess energy is converted into heat
References
through collisions with solvent molecules. Because of the
observed weak solvent–solute interactions, this suggests that
RI is of minor importance. On the other hand, VR determines
the internal conversion pathway of the non-radiative relaxa-
tion to a high extent.
1 G. Scott, Atmospheric Oxidation and Antioxidants, Elsevier, Am-
sterdam, 1993, vol. I and II.
2 K.-D. Asmus, Methods Enzymol., 1990, 186, 168.
3 C. von Sonntag and H.-P. Schuchmann, Methods Enzymol., 1994,
233, 47.
4 P. Wardman and C. von Sonntag, Methods Enzymol., 1995, 251,
31.
5 J. Schwinn, H. Sprinz, K. Droßler, S. Leistner and O. Brede, Int. J.
Radiat. Biol., 1998, 74, 359.
6 H. Sprinz, J. Schwinn, S. Naumov and O. Brede, Biochim. Biophys.
Acta, 2000, 1483, 91.
7 M. M. Alam and O. Ito, J. Org. Chem., 1999, 64, 1285, and
citations therein.
8 M. Nakamura, O. Ito and M. Matsuda, J. Am. Chem. Soc., 1980,
102, 698.
9 F. C. Thyrion, J. Phys. Chem., 1973, 77, 1478.
10 Y. M. Riyad, R. Hermann and O. Brede, J. Phys. Chem. A, 2005,
109, 6464.
Usually IC efficiency is directly related to the energy gap
(DE) between the S1 and S0 states and becomes increasingly
unfavorable with increasing energy separation.47 Here, the
energy gap between the S1 and the S0 states is B120 kcal
molꢁ1, but a high IC efficiency is favored together with
photodissociation. This reflects a weaker energy gap depen-
dence than that demanded by the ‘‘energy gap law’’. A surface
crossing of the S1 and S0 surfaces near the n = 0 level of S1
could possibly be a reason for this.47
11 Y. M. Riyad, R. Hermann and O. Brede, Chem. Phys. Lett., 2003,
376, 776.
12 T. Haupl, C. Windolph, T. Jochum, O. Brede and R. Hermann,
Chem. Phys. Lett., 1997, 280, 520.
Conclusions
For the first time, this paper gives a comprehensive picture and
quantitative information about the deactivation mechanism of
the first excited singlet state of aromatic thiols. Because no
intersystem crossing is taking place, the mechanism is reduced
to the singlet system only. Excitation brings the molecule from
the ground state S0 to a vibrationally excited level of S1.
According to Kasha’s rule,48 the relaxation to the first electro-
13 R. Hermann, G. R. Mahalaxmi, T. Jochum, S. Naumov and
O. Brede, J. Phys. Chem. A, 2002, 106, 2379.
14 O. Brede, H. Orthner, V. E. Zubarev and R. Hermann, J. Phys.
Chem., 1996, 100, 7097.
15 (a) A. D. Becke, J. Chem. Phys., 1993, 98, 5648; (b) A. D. Becke, J.
Chem. Phys., 1996, 104, 1040.
16 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A.
Robb, J. R. Cheeseman, V. G. Zakrzewski, J. A. Jr Montgomery,
R. E. Stratmann, J. C. Burant, S. Dapprich, J. M. Millam, A.
Daniels, K. N. Kudin, M. C. Strain, O. Farkas, J. Tomasi, V.
Barone, M. Cossi, R. Cammi, B. Mennucci, C. Pomelli, C. Adamo,
S. Clifford, J. Ochterski, G. A. Petersson, P. Y. Ayala, Q. Cui,
K. Morokuma, P. Salvador, J. J. Dannenberg, D. K. Malick, A. D.
Rabuck, K. Raghavachari, J. B. Foresman, J. Cioslowski, J. V.
Ortiz, A. G. Baboul, B. B. Stefanov, G. Liu, A. Liashenko, P.
Piskorz, I. Komaromi, R. Gomperts, R. L. Martin, D. J. Fox, T.
Keith, M. A. Al-Laham, C. Y. Peng, A. Nanayakkara, M.
Challacombe, P. M. W. Gill, B. Johnson, W. Chen, M. W. Wong,
J. L. Andres, C. Gonzalez, M. Head-Gordon, E. S. Replogle and
J. A. Pople, GAUSSIAN 03, (Revision B.02), Gaussian Inc.,
Pittsburgh, PA, 2003.
nically excited level S1 takes place extremely rapidly (1013 ꢁ1).
s
Therefore, the ArSH(S1) level is the central point where the
deactivation processes (2) to (5) branch out.
With the exception of IC, all other deactivation reaction
channels (fluorescence (2) and dissociation (5)) were identified
and quantitatively characterized. Consequently, this also en-
abled the quantitative description of radiation-less IC (3) by a
difference calculation. Hence, a complete picture of the decay
channels of the first excited singlet state of aromatic thiols is
given. A comparison shows that there are significant differ-
ꢂc
This journal is the Owner Societies 2006
Phys. Chem. Chem. Phys., 2006, 8, 1697–1706 | 1705

View Article Online
17 G. Beddard, Rep. Prog. Phys., 1993, 56, 63.
18 C. Reichardt, Solvents and Solvent Effects in Organic Chemistry,
VCH Verlagsgeselschaft, Weinheim, 1990.
19 J. R. Lakowicz, Principles of Fluorescence Spectroscopy, Plenum
Press, New York, 1983.
20 E. Fujimaki, A. Fujii, T. Ebata and N. J. Mikami, J. Chem. Phys.,
1999, 110, 4238.
21 J. N. Demas and G. A. Crosby, J. Phys. Chem., 1971, 75, 991.
22 R. A. Lampert, S. R. Meech, J. Metealfe and D. Philips, Chem.
Phys. Lett., 1983, 94, 137.
23 M. M. Alam, M. Fujitsuka, A. Watanabe and O. Ito, J. Phys.
Chem. A, 1998, 102, 1338.
24 O. Ito and M. Matsuda, J. Am. Chem. Soc., 1979, 101, 1815.
25 O. Ito, Res. Chem. Intermed., 1995, 21, 69.
26 H. Yoshizawa, O. Ito and M. Matsuda, Br. Polymer J., 1988, 20,
441.
32 M. M. Alam, A. Watanabe and O. Ito, J. Org. Chem., 1995, 60,
3440.
33 M. M. Alam, H. Konami, A. Watanabe and O. Ito, J. Chem. Soc.,
Perkin Trans. 2, 1996, 263, 268.
34 O. Ito, in S-Centered Radicals, ed. Z. B. Alfassi, John Wiley &
Sons, New York, 1999, p. 193.
35 K. Sehested and E. J. Hart, J. Phys. Chem., 1975, 79, 1639.
36 M. A. El-Sayed, J. Chem. Phys., 1962, 36, 573.
37 M. A. El-Sayed, J. Chem. Phys., 1963, 38, 2834.
38 M. A. El-Sayed, J. Chem. Phys., 1964, 41, 2462.
39 H. Baba and M. Kitamura, J. Mol. Spectrosc., 1972, 4, 302.
40 J. C. Baum, J. Am. Chem. Soc., 1980, 102, 716.
41 M. Z. Hoffman and E. Hayon, J. Phys. Chem., 1973, 77, 990.
42 R. Zhao, J. Lind, G. Merenyi and T. E. Eriksen, J. Am. Chem.
Soc., 1994, 116, 12010.
43 E. Hayon and M. Simic, J. Am. Chem. Soc., 1970, 92, 7486.
44 Handbook of Photochemistry, ed. S. L. Murov, I. Carmichael and
G. L. Hug, Marcel Dekker, New York, 1993.
27 D. A. Armstrong, Q. Sun and R. H. Schuler, J. Phys. Chem., 1996,
100, 9892.
28 M. Bonifacic, J. Weiss, S. A. Chaudhri and K.-D. Asmus, J. Phys.
Chem., 1985, 89, 3910.
29 R. Hermann, G. Dey, S. Naumov and O. Brede, Phys. Chem.
Chem. Phys., 2000, 2, 1213.
30 G. N. R. Tripathi, Q. Sun, D. A. Armstrong, D. M. Chipman and
R. H. Schuler, J. Phys. Chem., 1992, 96, 5344.
31 O. Ito and M. Matsuda, J. Am. Chem. Soc., 1981, 103
5871.
45 F. G. Bordwell, Z. M. Zhang, A. V. Satish and J. P. Cheng, J. Am.
Chem. Soc., 1994, 116, 6605.
46 M. Klessinger and J. Michl, Lichtabsorption und Photochemie
organischer Molekul, VCH Verlagsgeselschaft, Weinheim, Ger-
many, 1969.
¨
47 N. J. Turro, Modern Molecular Photochemistry, University Science
Books, Sausalito, CA, 1991.
48 M. Kasha, Discuss. Faraday Soc., 1950, 9, 14.
ꢂc
This journal is the Owner Societies 2006
1706 | Phys. Chem. Chem. Phys., 2006, 8, 1697–1706