Energetics of photocyclization of polyphenylamines and assignment of the intermediate: A time-resolved photoacoustic calorimetric study

Article abstract of DOI:10.1039/b104228a

The photocyclization reactions of diphenylamine (DPA), N-methyldiphenylamine (MDPA) and triphenylamine (TPA) have been studied in methanol solvent through nanosecond/microsecond flash photolysis and time-resolved photoacoustic calorimetry (TRPAC). It is confirmed for all the amines that the long-lived dihydrocarbazole intermediate (¹XDHC⁰) is formed through a cascade of processes. However, DPA behaves differently from the other two members of the series so far as the rates of the individual processes are concerned. The quantum yield of triplet formation and the quantum yield of the photocyclization of the amine triplet have been estimated for the DPA system. In conjugation with the experiments, theoretical calculations, employing AM1-SCI method, have been performed to propose a structure for the intermediate. For all the amines, the intermediate has been proposed to be of zwitterionic nature in the ground state. The enthalpies of formation of the intermediate (¹XDHC⁰) from MDPA and TPA have been determined experimentally to be around 51 kcal mol^-1 and the values agree well with the calculated ones.

Full text of DOI:10.1039/b104228a

View Article Online / Journal Homepage / Table of Contents for this issue
Energetics of photocyclization of polyphenylamines and assignment of
the intermediate: A time-resolved photoacoustic calorimetric study
Nitin Chattopadhyay,*ab Carlos Serpa,a Luis G. Arnauta and Sebastiao J. Formosinhoa
a Department of Chemistry, Coimbra University, Coimbra, 3049, Portugal
b Department of Chemistry, Jadavpur University, Calcutta, 700 032, India.
E-mail: pcnitin=yahoo.com
Received 14th May 2001, Accepted 4th July 2001
First published as an Advance Article on the web 13th August 2001
The photocyclization reactions of diphenylamine (DPA), N-methyldiphenylamine (MDPA) and triphenylamine
(TPA) have been studied in methanol solvent through nanosecond/microsecond Ñash photolysis and
time-resolved photoacoustic calorimetry (TRPAC). It is conÐrmed for all the amines that the long-lived
dihydrocarbazole intermediate (1XDHC0) is formed through a cascade of processes. However, DPA behaves
di†erently from the other two members of the series so far as the rates of the individual processes are
concerned. The quantum yield of triplet formation and the quantum yield of the photocyclization of the amine
triplet have been estimated for the DPA system. In conjugation with the experiments, theoretical calculations,
employing AM1-SCI method, have been performed to propose a structure for the intermediate. For all the
amines, the intermediate has been proposed to be of zwitterionic nature in the ground state. The enthalpies of
formation of the intermediate (1XDHC0) from MDPA and TPA have been determined experimentally to be
around 51 kcal mol~1 and the values agree well with the calculated ones.
have not ruled out the possibility of the biradical structure.4h7
Introduction
Again, Shizuka et al. proposed that the ““610 nmÏÏ transient
di†ers from the intermediate 1XDHC0 for DPA.9,10 Only
recently,13 we have established experimentally that, for TPA,
the long-lived intermediate is not a biradical. From a com-
parison of the experimental and the calculated parameters of
TPA, we have further established that the intermediate is of
zwitterionic character with a partial positive charge on the
nitrogen atom. In this report, we intend to present compre-
hensively our experimental and theoretical results on all three
members of the series. It is noted that MDPA and TPA
behave very similarly with regard to their reactivities.
However, DPA di†ers in terms of reaction rates and reaction
quantum yields.
Since the works of Parker and Barnes,1 the oxidative photo-
cyclization of arylamine/s to carbazole/s has been the subject
of extensive research.2h14 Upon irradiation, diphenylamine
(DPA), N-methyldiphenylamine (MDPA) and triphenylamine
(TPA) undergo an oxidative photocyclization reaction leading
to the formation of the corresponding carbazoles as the Ðnal
product. The most challenging part of the studies of these
reactions is to Ðnd a plausible reaction mechanism for the
complex multi-step reactions and to assign the nature and
structure of the long-lived intermediates (XDHC),
unsubstituted or substituted dihydrocarbazole, depending on
the amine. The authors have employed mostly the Ñash
photolytic technique to identify the di†erent intermediates and
to determine the rates of the individual reactions through
which the total photoprocess proceeds. Although the multi-
step reaction mechanism as proposed by Rahn et al.3 is
accepted, in general, as the most plausible pathway for the
photocyclization process, there is still a lack of completeness
in the studies, particularly, in respect of the energetics of the
reactions. It has been observed that the photophysical param-
eters of DPA di†er remarkably from those of MDPA and
TPA, leading to a marked di†erence in the rate constants of
the individual steps of the photocyclization reaction for DPA
compared with those of the other two members of the series.
So far as the energetics of the reaction is concerned, we have
only two reports in the literature, one from Suzuki et al. for
Experimental
DPA, MDPA and TPA were of the best quality products
available from Aldrich. DPA and TPA were puriÐed by
vacuum sublimation followed by recrystallization from
ethanol. MDPA was used as received after checking its purity
spectroscopically. Spectroscopic grade methanol (Merck) was
used as received. 2-Hydroxybenzophenone (HBP, Aldrich)
was used as a reference for the TRPAC experiments as
before.13
A Shimadzu UV-2100 spectrophotometer and a Spex Flu-
orolog 2 spectroÑuorimeter were used for the absorption and
Ñuorescence measurements respectively. Although reported
elsewhere,13 we will reproduce the main points regarding the
instrumentation. This is particularly important since there are
some modiÐcations for the DPA system.
DPA using
a
time-resolved thermal lensing (TRTL)
technique11 and the other from our group for TPA using
time-resolved photoacoustic calorimetry (TRPAC).13
There is a long-standing controversy regarding the nature
of the long-lived intermediate that has a typical broad TÈT
absorption band at around 610 nm. While Suzuki et al. pro-
posed a biradical structure for it in its ground state,11 Grell-
mann et al. proposed a zwitterionic structure, although they
The Ñash photolysis set-up comprises a Spectra Physics
Quanta-Ray GCR-130 Nd-YAG laser, an Applied Photo-
physics LKS 60 laser Ñash photolysis spectrometer and a
Hewlett-Packard InÐnium Oscilloscope (500 MHz, 1 Gsa
s~1). For MDPA and TPA, the samples were excited with the
3690
Phys. Chem. Chem. Phys., 2001, 3, 3690È3695
This journal is ( The Owner Societies 2001
DOI: 10.1039/b104228a

View Article Online
third harmonic (355 nm, 8 ns FWHM) of the laser. Since there
is hardly any absorption of DPA at 355 nm, we used the
fourth harmonic (266 nm) of the laser. The monitoring light
was produced by a 150 W pulsed Xe lamp. The detection of
the transient species in the range 300È650 nm was made with
a Hamamatsu photomultiplier (model 1P-28). The sample
solutions were prepared with absorbance within 0.1È0.2 at the
excitation wavelength and were degassed with dry nitrogen
prior to their use in the Ñash experiments.
The photoacoustic calorimetry apparatus follows the front-
face irradiation design described elsewhere.13h15 The solutions
were pumped through a 0.11 mm thick cell at a Ñow rate of 1
mL min~1 with an SSI chromatography pump (model 300)
and irradiated with an unfocussed (area of the laser beam at
the cell surface being B0.5 cm2) N laser (PL 2300 from PTI)
2
working at a frequency of 2 Hz. The laser energy was a little
less than 1 mJ pulse~1. The acoustic waves generated by the
non-radiative processes following light absorption in the cell
were detected with a high frequency Panametrics transducer,
Scheme 1 Di†erent steps in the photoreaction of the amines.
which eventually forms the corresponding long-lived ground
state species (1XDHC0) (step III). 3XDHC* has a TÈT absorp-
tion band peaking at around 420 nm while 1XDHC0 has a
very broad absorption with a maximum at around 610 nm. In
degassed methanolic solution and at ambient temperature,
3A* has lifetimes of 20 and 50 ns for MDPA and TPA respec-
tively. For both these amines, 3XDHC* has a much longer
lifetime than that of 3A*. The lifetimes of 3XDHC* have been
determined to be 580 and 380 ns respectively for the two
amines. It is interesting to note that the rates of the individual
processes (step II and step III) are grossly comparable for
MDPA and TPA. However, the rate of step II is an order of
magnitude faster than step III for these two amine systems. In
sharp contrast, these rates are markedly di†erent for DPA
under similar experimental conditions. 3DPA* is much
longer-lived than 3MDPA* and 3TPA* and now the rates of
step II and step III are of the same order of magnitude. This
makes the accurate determination of the rate constants for the
individual steps difficult. There is also the problem of overlap
of the absorption bands responsible for the di†erent species.
Being aware of the fact that the 610 nm absorption of
1XDHC0 is contaminated by the 530 nm absorption band of
3A*, we monitored the growth of the 1XDHC0 at 640 nm to
minimize the 3A* contamination. From our Ñash photolysis
experiment, we determined the lifetimes of 3A* and 3XDHC*
for DPA as 800 and 1300 ns respectively in degassed methanol
solution. Although, the DPA triplet decay time, as determined
by us, is supported by others,3,11 for the intermediate triplet,
the data di†ers from the literature.3,11 For all the amines,
1XDHC0, responsible for the 610 nm absorption, is quite
long-lived, in the range of hundreds of microseconds.
1XDHC0 Ðnally gives the photoproduct, the carbazole deriv-
ative. For our TRPAC experiments this species is sufficiently
long-lived to serve as the dump. The kinetic data for step II
and step III, determined from our Ñash photolysis experi-
ments, have been summarized in Table 1.
preampliÐed with
a Panametrics ultrasonic preampliÐer
(model 5676), captured by the transient recorder (Tektronix
DSA 601) and transferred to a PC for data analysis. For
MDPA and TPA we used a 2.25 MHz transducer and for
DPA, because of the slow reaction rates, we used a 0.5 MHz
transducer. For each of the sample, reference and pure solvent
an average of 100 acoustic waves were collected. Four sets of
averaged waves for each of the sample, reference and solvent
were used for data analysis at a given laser intensity, and four
laser intensities were employed in each experiment. The di†er-
ent laser intensities used in the experiment were obtained by
interposing Ðlters with transmissions in the range 30 to 100%.
As oxygen was found to have a strong e†ect on the PAC
signal, the experiments were performed under an atmosphere
of constant purging with solvent-saturated N . Before going
2
through the TRPAC we conÐrmed that the PAC signal was
linear with the concentration of the sample, at least up to a
concentration of the solution with absorbance \ 0.3
(correlation coefficient P0.99). However, for the TRPAC
experiments we used sample solutions with absorbance ^0.1.
The absorbances of the reference solutions were the same as
those of the sample solutions within the limit of the experi-
mental error.
Although, less precise than ab initio calculations involving
extended basis sets with extensive conÐguration interaction
(CI), the semi-empirical molecular orbital methods have
already established their utility in explaining the structures,
energetics and reactivities of di†erent molecular systems in dif-
ferent electronic states. These methods are particularly useful
for large molecular systems. The methods provide acceptable
approximations to give results quite close to the experimental
Ðndings.13,16h20 In the present calculations, we have used the
commercial package, Hyperchem 5.01 (Hypercube Inc.,
Canada). The energies of the various electronic states of the
di†erent species (the amines and di†erent possible structures
of the intermediate, 1XDHC0) have been calculated for the
optimized structures using the AM1-SCI method.
Photoacoustic calorimetry
Quantum yield of triplet amine formation. In TRPAC the
acoustic wave, generated by the heat released in the non-
radiative processes following electronic excitation, is analysed.
The experimental wave (E-wave) of the sample is compared
with that of the pressure transducer (T-wave). The T-wave is
obtained with the calorimetric reference (in our case, HBP), in
the same solvent absorbing the same fraction of light as the
sample and releasing it as thermal energy in a time much
shorter than the transducer oscillation frequency. The phase
and amplitude di†erences between the T- and E-waves allow
simultaneous determination of the thermal energy released by
the transients (E ) and their lifetimes (s ). Typical background-
Results and discussion
Flash photolysis
The Ñash photolysis and the TRPAC experiments substantiate
the following scheme (Scheme 1) as proposed by Rahn et al.3
Results of our Ñash photolysis experiments on the amines in
methanol corroborate the literature data in general with some
exceptions for DPA in terms of the reaction rate constants. In
general, the photoexcited amine (1A*) forms its triplet (3A*) in
the Ðrst step (step I). The amine triplet has a broad TÈT
absorption band with a maximum at around 530 nm. 3A*
then forms the triplet of the intermediate (3XDHC*) (step II)
i
i
corrected sample and reference signals for MDPA in methanol
are shown in Fig. 1.
Phys. Chem. Chem. Phys., 2001, 3, 3690È3695
3691

View Article Online
Table 1 Rate constants (s~1) of the individual steps for the photoreaction of the amines in degassed methanolic solution measured by Ñash
photolysis
Rate constantsa/s~1
Reaction
Step II
Monitoring process
DPA
MDPA
TPA
Decay of 3A* at 530 nm
1.3 ] 106
1.2 ] 106
7.7 ] 105
8.0 ] 105
5.0 ] 107
5.4 ] 107
1.7 ] 106
1.9 ] 106
2.0 ] 107
2.0 ] 107
2.6 ] 106
2.6 ] 106
Growth of 3XDHC* at 420 nm
Decay of 3XDHC* at 420 nm
Growth of 1XDHC0 at 610 nmb
Step III
a The data incorporates a 10% error associated with the method. b For DPA, the monitoring wavelength was 640 nm to minimize contamination
from 3A* absorption.
For MDPA and TPA, we interpret the waves of the N -
saturated samples with three sequential exponents, following
In spite of the fact that the photoreaction (from photoexcit-
ed state to 1XDHC0) of DPA is very similar to those of
MDPA and TPA, involving three stages, we could not analyse
our TRPAC waves with three exponents. This is probably
because of the slowness and also the proximity of the rates of
step II and step III. Because of the long lifetime of 3A*, there
should be a considerable overlap of the time windows for the
thermal energies to evolve from the two individual steps. This
makes it hard to resolve the total heat released into the indi-
vidual components corresponding to the two steps. Consider-
ing the thermal energies coming out from these two steps
together, we interpreted the photoacoustic waves with two
sequential exponents, using the values q \ 1.0 ns and q \
2
the Ñash photolysis results. The Ðrst one for the formation of
the 3A* (step I), the second one corresponding to the forma-
tion of the intermediate in its triplet state (3XDHC*) (step II)
and the third one for its decay to 1XDHC0 (step III) as
described in Scheme 1. Each decay step is described by two
parameters: the lifetime of the transient (q) and the fraction of
thermal energy released (/) within that time window. Convol-
ution of the reference waves with the parameters of the kinetic
model for the decay of the transient species gives the calcu-
lated E-wave. The appropriateness of the kinetic model and its
parameters in describing the observed E-wave can be evalu-
ated by the di†erence (termed as residual) between the ampli-
tudes and phases of the observed (E-) and calculated (C-)
waves at each decay time (Fig. 1). For all the amines, the for-
mation of 3A*, following the excitation, is faster than the time
resolution of our experimental set-up, and we arbitrarily set
1
2
1300 ns.
The fractions of laser energy released by each system were
measured at four laser intensities. The Ðrst fraction was found
to vary with laser intensity. We plotted this fraction as a func-
tion of laser intensity and obtained linear correlation coeffi-
cients greater than 0.96. The di†erence was assigned to
transientÈtransient absorption and was corrected by extrapo-
lating the fraction of energy released to zero laser intensity.
From this laser intensity corrected value of the Ðrst fraction of
the lifetime of the Ðrst exponential decay to q \ 1 ns. Smaller
1
values of q do not change the other parameters in the decon-
1
volution. For MDPA and TPA, the second and third expo-
nential components are set from the q and q values as
2
3
determined from the Ñash experiments corresponding to step
II and step III and given in Table 1. We have also veriÐed the
time windows for the individual steps independently by Ðxing
the released heat, the quantum yields of triplet formation (/ )
T
were determined for all the amines from eqn. (1).14
/ \ M(1 [ / )E [ / E N/E
hv
/ being the laser intensity corrected value of the Ðrst fraction
(1)
q
and q and allowing q to be adjusted by the MarquardtÏs
T
0
F
S
T
1
3
2
algorithm employed in the deconvolution process. The q
2
0
values that we get for good Ðts correspond well with the Ñash
results.
of the released heat, / , the Ñuorescence quantum yield, E
,
F
hv
the energy of the exciting radiation and E and E , the energy
S
T
of the Ðrst excited singlet and the lowest triplet states. For our
purpose, we have used the well established literature values of
the energies of the Ðrst excited singlet (E ) and the lowest
S
triplet (E ) states and also Ñuorescence quantum yields (/ ).
T
F
The determined / and the values of the parameters we used
T
for the determination are given in Table 2.
It is clear from the table that although the triplet yield (/ )
T
is almost unity for MDPA and TPA in methanol, it is much
less for DPA in the same solvent. The / values for the former
T
two amines agree well with the literature reports.8,21,23 It is
pertinent to mention here that unlike MDPA and TPA, the
literature data for the triplet yield (/ ) of DPA is quite scat-
T
tered, ranging from 0.32 to 0.89.3,23h25 The / value deter-
T
mined by our TRPAC technique is lower than that reported
Table 2 Triplet quantum yield (/ ) determined from TRPAC and
T
the literature values of the energies of the S (E ) and T (E ) states
Fig. 1 Typical sample photoacoustic wave (E-wave, ÈÈ ), reference
wave (T-wave, (È È È), calculated wave (C-wave, É É É) and the residual
(È É È É È). The calculated wave, C-wave, was obtained with three
sequential decays with lifetimes q \ 1 ns, q \ 20 ns and q \ 580 ns.
1
S
1
T
and the Ñuorescence quantum yields (/ ) of the three amines used for
the determination of / . The energies are in kcal mol~1
F
T
Parameter
DPA
MDPA
TPA
1
2
3
Residual \ (C-wave) [ (E-wave). For details see text. The E- and
T-waves were corrected for the background signal and normalized.
The normalization factor is the reciprocal of the largest absolute value
of the T-wave. The sample (MDPA), reference (HBP) and solvent
(methanol) data were obtained under the following experimental con-
/
T
0.57 ^ 0.06
88.9a
72.0b
0.95 ^ 0.06
90.2b
70.3b
0.04b,d
0.98 ^ 0.06
86.5a
E
s
E
/
70.0a,b
T
0.05c
0.05a
F
ditions: irradiation at 337 nm of N -saturated solutions with a Ðlter
2
with 72% transmitance; absorbance of 0.10 for both sample and refer-
a Ref. 21. b Ref. 8. c Ref. 2. d Ref. 22.
ence solutions; solution Ñow rate of 1 mL min~1.
3692
Phys. Chem. Chem. Phys., 2001, 3, 3690È3695

View Article Online
(2)
by Rahn et al.3 (0.86 in methanol) but closer to the values
reported by Lamola and Hammond23 (0.38 in benzene) and
by Sveshnikova and Snegov24 (0.47 in 1-butanol). As the Ñuo-
rescence quantum yield of DPA is of the same magnitude as
/ ] / \ 1
ISC
R
3A*, initially forms the intermediate in the triplet state
(3XDHC*), the latter then passes to its stable form (1XDHC0)
(steps II and III respectively in Scheme 1). However, since
there is no other way of deactivation of 3XDHC* but its
decay to 1XDHC0, we can safely consider the quantum yield
of step III as unity. Again, considering the thermal energy
received by the transducer in the second time window (for a
bi-exponential treatment of the acoustic waves in the case of
DPA) as the sum of the heat released in steps II and III, one
gets the following equation from the energy balance condition.
that for MDPA and TPA, a lower / value for the DPA com-
T
pared to the values for the other two systems points to an
additional reaction pathway from the S state of DPA that is
1
absent in the other two systems. Although, we did not notice
any absorption representative of a new species (undeÐned so
far for the system) within the wavelength range of our Ñash
photolysis experiments, probably photoionization of DPA
through the dissociation of the hydrogen attached to the
nitrogen is responsible for the lower value of / for DPA.
/E \ / E ] / (E [ E
)
(3)
T
hv
ISC
T
R
T
INT
where E , E and E
ation, 3A* and 1XDHC0 respectively and / is the fraction of
heat released during the second time window. With an experi-
are the energies of the exciting radi-
hv
T
INT
Enthalpies and quantum yields of reaction. It is known that
at room temperature and in degassed solutions, the cycliza-
tion reaction for the formation of 3XDHC* from 3A* is quan-
titative (/ ^ 1.0) for MDPA and TPA.7 So, E and E (refer
mental value of 0.60 for /, 84.82 kcal mol~1 for E , 72.0 kcal
hv
mol~1 for E (see Table 2) and 51.39 kcal mol~1 for E
T
INT
R
(from the calculation, see Table 5), we estimate the value of /
R
2
3
to be 0.41. Although this value is not so far from the value of
to Scheme 1) were estimated from the other two fractions of
0.37 reported by Suzuki et al.,11 this di†erence is responsible
heat released in the other two exponents (E \ / E , E \
2
2 hv
3
for the deviation in their *H value for DPA from our experi-
/ E ).13,14 After obtaining the energies involved in each of
3 hv
R
mentally determined values for MDPA and TPA.
the three steps, it is easy to determine the enthalpy of reaction
(*H ) for the formation of the stable intermediate, 1XDHC0,
R
from the ground state amines. Table 3 presents the energies
involved in step II (E ) and step III (E ) during the photo-
Assignment of the stable intermediate 1XDHC0. Apart from
the energetics of the photocyclization, the other interesting
aspect of the study of this reaction with these amines lies in
establishing the nature of the stable intermediate, 1XDHC0,
responsible for the 610 nm absorption. From their obser-
vation with the DPA system, that 3XDHC* lies closer to its
S state and not its S state, Suzuki et al. tentatively proposed
that the intermediate might have a biradical character in the
ground state.11 However, there is no direct experimental evi-
dence in favor of the proposition. In a recent report we have
established from both experiment and theoretical calcu-
lation,13 at least for TPA, that 1XDHC0 is not a biradical but
has a zwitterionic character. We have also applied the same
techniques to the other two members, viz., DPA and MDPA,
of the series.
From the experimental point of view, a biradical with a
very long lifetime should be very sensitive to quenching by
spin trapping agents like 5,5-dimethyl-1-pyrroline-N-oxide
(DMPO) and N-tert-butyl-a-phenylnitrone (PBN).26,27 For
all the amines in the present study, 1XDHC0 has a very long
lifetime, of the order of hundreds of microseconds. However,
our Ñash experiments with both the spin traps, viz., DMPO
and PBN, reÑected no change in the lifetime of the 610 nm
band corresponding to 1XDHC0. We have further tried with a
conjugated triene, viz., 1,6-diphenyl-1,3,5-hexa-triene, to see
whether there is any quenching in the lifetime of 1XDHC0
since the biradicals are susceptible to addition reactions with
these substrates. The negative results with all our quenching
experiments clearly negate the biradical nature of the interme-
diate in its ground state.
To assign the structure of the intermediate 1XDHC0, we
have calculated the energies of the electronic states for all rea-
sonable structures. The one that explains the experimental
Ðndings best is the trans isomer (with the two central hydro-
gens in a trans arrangement) with a zwitterionic character.
The skeleton is shown in Fig. 2. Although Linschitz and Grell-
mann also proposed a zwitterionic nature for the intermediate
without any experimental or theoretical support, they conjec-
tured that the two central hydrogens would be in the cis con-
Ðguration.5 For all the three systems, however, we have found
that the trans isomer is more stable than the cis isomer.
It is revealed that in the ground state the nitrogen atom
acquires a reasonable positive charge while all the carbon
2
3
production of 1XDHC0 and also *H , as determined by
R
TRPAC, for both MDPA and TPA.
Table 3 reveals that the enthalpy of reaction (*H ) for the
R
formation of 1XDHC0 in methanol solvent is nearly the same
(D51 kcal mol~1) for both MDPA and TPA. As will be dis-
cussed in a forthcoming section dealing with the theoretical
results, there is very good agreement between the experimen-
tal and calculated enthalpies of reaction for the formation of
the stable intermediate from MDPA and TPA. This gives
strong support for our calculations. For DPA, however,
0
1
Suzuki et al. have reported *H to be 62.1 kcal mol~1.11 A
R
large di†erence in *H for DPA compared with the other two
R
members of the series is rather surprising. Our calculation also
does not support a large di†erence in the *H value for DPA
R
from those for MDPA and TPA. Since there are no literature
data for / for the DPA system for a cross-check, it seems
R
that an error in their estimated / possibly led to an error in
R
the estimation of *H . For DPA, we could not determine E
and E individually for two reasons. Firstly, as already dis-
cussed, we could not resolve the thermal energies coming out
R
2
3
of the two steps and, secondly, the quantum yield of the cycli-
zation reaction (/ ) is not known. After establishing the valid-
R
ity of the method of calculation to calculate *H for MDPA
R
and TPA, we have extended the same method of calculation
to the DPA system and using the calculated value of *H , we
R
have determined / for DPA in the following way.
R
Considering that in degassed solution, 3A* deactivates via
two routes (i) returning to the ground state of the amine
(DPA) with a quantum yield of / and (ii) forming the inter-
ISC
mediate, with a quantum yield of / , one can write the fol-
lowing relation.
R
Table 3 The amount of energy involved in step II (E ) and step III
2
(E ) during the photoreaction of MDPA and TPA (see Scheme I) and
3
the enthalpy of reaction (*H ) for the formation of 1XDHC0 in meth-
R
anol solvent. The energies are in kcal mol~1
Parameters
MDPA
TPA
E
2
*H
13.8 ^ 1.0
5.5 ^ 0.6
51.0 ^ 1.5
14.9 ^ 1.0
3.9 ^ 0.6
51.2 ^ 1.5
E
3
R
atoms, particularly, C , C and C (or C , C and C
)
5
3
1
8
10 12
Phys. Chem. Chem. Phys., 2001, 3, 3690È3695 3693

View Article Online
tion observed for the intermediates with the calculated
S ] S energies and a very good agreement between the
0
1
experimental reaction enthalpies (Table 3) with the calculated
ones (Table 5) enable us to unhesitatingly assign the structure
of the intermediate to a zwitterionic one with a partial posi-
tive charge on the nitrogen centre as described above.
Conclusion
Fig. 2 Skeleton of the ground state structure of the intermediate,
1XDHC0. Only the carbon atoms are numbered. 14 represents H for
DPA, methyl carbon for MDPA and phenyl carbon for TPA. Charges
on the atoms have been dropped in the Ðgure as the negative charge
has been calculated to be delocalized (see Table 4).
We conÐrm the proposition of Rahn et al.3 that for DPA,
MDPA and TPA the long lived intermediate 1XDHC0 is
formed through a cascade of processes during the photoreac-
tion of the amines. TRPAC study reveals that the formation of
the intermediate from the amine is an endothermic process,
the endothermicity being D51 kcal mol~1 in methanol solu-
tion. For DPA, we have determined the quantum yield of for-
mation of the amine triplet and the quantum yield of the
cyclization reaction. From the good agreement between the
calculated and the experimental results, the structure of the
intermediate (1XDHC0) has been assigned for all the amine
systems. The present work establishes that for all members of
the present series, the intermediate is zwitterionic in the
ground state with a partial positive charge on the nitrogen
centre.
acquire partial negative charges. Table 4 gives the net charges
on the atoms comprising the skeleton of the intermediate for
DPA, MDPA and TPA.
We have simulated the electronic spectrum of the interme-
diate for all the amine systems under study. All of them show
a strong absorption (oscillator strength, f B 0.33) within the
range 633.3 to 643.6 nm corresponding to the 0È0 transition
of S ] S . Experimentally, a strong absorption band with a
0
1
broad maximum in the range 610È620 nm is observed for all
the systems, justifying the acceptability of our assigned struc-
ture for the transient. We have also calculated the ground
state energy of the amines from their optimized geometries.
From the di†erence in the ground state energies (or in the
enthalpy of formation) of the amines and the intermediate
Acknowledgements
Financial supports from PRAXIS/PCEX/QUI/0108/96
(European Union) and Fundacao para a Ciencia e a Tecnolo-
gia (grant BD/18362/98) are gratefully acknowledged. The
authors sincerely thank Professor H. D. Burrows for helpful
discussions.
(1XDHC0) we have estimated the enthalpy of reaction (*H )
R
for the formation of 1XDHC0 from the amines during the
photoreaction.
Table 5 lists the calculated values of *H as well as the
R
absorption energies (S ] S ) corresponding to the interme-
0
1
diate for all the amines. A comparison of the strong absorp-
References
Table 4 Net charges on di†erent atoms of the 1XDHC0 skeleton
1
2
C. A. Parker and W. J. Barnes, Analyst, 1957, 82, 606.
E. J. Bowen and J. H. D. Eland, Proc. Chem. Soc., L ondon, 1963,
202.
Net charge
DPA
Atom (see Fig. 2)
MDPA
TPA
3
4
5
6
7
8
9
R. Rahn, J. Schroeder, J. Troe and K. H. Grellmann, J. Phys.
C
1
[0.154 924
[0.129 259
[0.153 060
[0.085 703
[0.223 373
[0.017 471
[0.017 318
[0.223 551
[0.085 876
[0.152 892
[0.129 303
[0.155 040
]0.050 622
]0.243 295
[0.156 289
[0.128 447
[0.158 045
[0.081 198
[0.228 071
[0.015 094
[0.016 304
[0.220 944
[0.082 062
[0.156 744
[0.129 236
[0.155 487
]0.095 292
]0.000 918
[0.157 976
[0.127 901
[0.162 153
[0.080 966
[0.213 062
[0.016 082
[0.015 868
[0.213 423
[0.080 634
[0.162 351
[0.127 501
[0.158 232
]0.131 026
[0.115 392
Chem., 1989, 93, 7841.
C
2
K. H. Grellmann, G. M. Sherman and H. Linschitz, J. Am. Chem.
Soc., 1963, 85, 1881.
C
C
3
4
H. Linschitz and K. H. Grellmann, J. Am. Chem. Soc., 1964, 86,
C
5
303.
C
6
K. H. Grellmann, W. Kuhnle, H. Weller and T. Wol†, J. Am.
Chem. Soc., 1981, 103, 6889.
C
7
C
8
E. W. Forster, K. H. Grellmann and H. Linschitz, J. Am. Chem.
Soc., 1973, 95, 3108.
C
9
C
10
C
11
C
12
N
H. Shizuka, Y. Takayama, I. Tanaka and T. Morita, J. Am.
Chem. Soc., 1970, 92, 7270.
H. Shizuka, Y. Takayama, T. Morita, S. Matsumoto and I.
Tanaka, J. Am. Chem. Soc., 1971, 93, 5987.
C
a
14
10 H. Shizuka, Y. Takayama, I. Tanaka and T. Morita, J. Am.
Chem. Soc., 1972, 92, 727.
a C represents H for DPA, methyl carbon for MDPA and phenyl
14
carbon for TPA. For numbering see Fig. 2.
11 T. Suzuki, Y. Kajii, K. Shibuya and K. Obi, Bull. Chem. Soc. Jpn.,
1992, 65, 1084.
12 D. Sur, P. Purkayastha and N. Chattopadhyay, J. Photochem.
Photobiol. A, 2000, 134, 17.
Table 5 Calculated values of *H (kcal mol~1), S ] S absorption
13 N. Chattopadhyay, C. Serpa, P. Purkayastha, L. G. Arnaut and
S. J. Formosinho, Phys. Chem. Chem. Phys., 2001, 3, 70.
14 M. Pineiro, A. L. Carvalho, M. M. Pereira, A. M. dÏA. R. Gon-
salves, L. G. Arnaut and S. J. Formosinho, Chem. Eur. J., 1998, 4,
2299.
R
0
1
position (nm) and corresponding oscillator strength ( f ) for the three
amines
Parameter
DPA
MDPA
TPA
15 L. G. Arnaut, R. A. Cadwell, J. E. Elbert and L. A. Melton, Rev.
*H
51.39
633.3
0.32
48.94
645.0
0.32
50.71
643.6
0.33
R
Sci. Instrum., 1992, 63, 5381.
S ] S absorptiona
0
1
16 T. Arthen-Engeland, T. Bultmann, N. P. Earnsting, M. A. Rodri-
guez and W. Thiel, Chem. Phys., 1992, 163, 43.
Oscillator strengthb ( f )
17 J. Catalan, F. Fabero, M. S. Guijarro, R. M. Claramunt, M. D.
Santa Maria, M. C. Foces-Foces, F. H. Cano, J. Elguero and R.
Sastre, J. Am. Chem. Soc., 1990, 112, 747.
a The calculated absorption corresponds to the 0È0 transitions only,
while the observed absorption spectrum leads also to transitions to
higher vibrational levels. b For all the amines the calculated S ] S
transition has reasonably high oscillator strength, justifying the strong
absorption observed for the species in this wavelength region.
0
1
18 B. Dick, J. Phys. Chem., 1990, 94, 5752.
19 P. Purkayastha, P. K. Bhattacharyya, S. C. Bera and N. Chatto-
padhyay, Phys. Chem. Chem. Phys., 1999, 1, 3253.
3694
Phys. Chem. Chem. Phys., 2001, 3, 3690È3695

View Article Online
20 P. Purkayastha and N. Chattopadhyay, Phys. Chem. Chem. Phys.,
2000, 2, 203.
24 E. B. Sveshnikova and M. I. Snegov, Opt. Spectrosc., 1979, 29,
265.
21 S. L. Murov, I. Carmichael and G. L. Hug, Handbook of Photo-
chemistry, Marcel Dekker, New York, 2nd edn., 1993, Table 1.
22 M. F. Budyka, O. D. Laukhina and T. N. Gavrishova, J. Photo-
chem. Photobiol. A, 2000, 134, 143.
23 A. A. Lamola and G. S. Hammond, J. Chem. Phys., 1965, 43,
2129.
25 H. J. Haink and J. R. Huber, Chem. Phys. L ett., 1976, 74, 117.
26 T. J. Kemp, Prog. React. Kinet. Mech., 1999, 24, 287.
27 B. Halliwell and J. M. C. Gutteridge, Free Radicals in Biology
and Medicine, Oxford University Press, London, 3rd edn., 1998,
ch. 5.
Phys. Chem. Chem. Phys., 2001, 3, 3690È3695
3695