Non-vertical energy transfer to oximes: Role of structural changes

Article abstract of DOI:10.1039/b104163k

Energy transfer to aromatic ketone oximes can exhibit non-vertical behaviour depending on the planarity of ground state equilibrium geometry, a fact discussed in terms of structural changes of ground and triplet states.

Full text of DOI:10.1039/b104163k

View Article Online / Journal Homepage / Table of Contents for this issue
Non-vertical energy transfer to oximes: role of structural changes
Jacques Lalevee,a Xavier Allonas,*a Frederic Louerat,a Jean-Pierre Fouassier,a Hideki Tachi,b
Akihito Izumitani,b Masamitsu Shiraib and Masahiro Tsunookab
a Department of General Photochemistry, UMR no 7525, Ecole Nationale Superieure de Chimie,
3, rue Alfred W erner, 68093 Mulhouse Cedex, France
b Department of Applied Chemistry, Osaka Prefecture University, 1-1 Gakuen-cho, Sakai, Osaka,
599-8531, Japan
Received 10th May 2001, Accepted 30th May 2001
First published as an Advance Article on the web 19th June 2001
Energy transfer to aromatic ketone oximes can exhibit non-
vertical behaviour depending on the planarity of ground state
equilibrium geometry, a fact discussed in terms of structural
changes of ground and triplet states.
The behaviour of oximes in energy transfer processes can be
studied by tripletÈtriplet sensitization experiments using
donors with well deÐned triplet energies.9 It can be seen from
Fig. 1 that for high values of the donor triplet energy ED , k is
T
q
nearly di†usion-controlled. When ED decreases, the reaction
T
becomes more endergonic and k decreases. This is consistent
q
Non-vertical energy transfer (NVET) has long been an excit-
ing topic for photochemists. Despite the great variety of views
on this e†ect,1h7 there is now widespread acceptance for struc-
tural changes during the reaction, although no quantitative
evaluation of the energetic factor governing this behaviour has
ever been published. Most of the studies only deal with model
p systems such as styrene or stilbene derivatives. We have
recently reported on energy transfer experiments with O-
acyloximes derivatives that can be used in photosensitized
crosslinking reactions, suggesting that a NVET behaviour is
dependent on the structure.8 In order to demonstrate this
NVET process, energy transfer experiments are carried out, in
the present paper, on a set of carefully selected oximes¤ which
appeared as excellent model compounds exhibiting di†erent
degrees of Ñexibility (Scheme 1). The results are compared to
semi-empirical calculations that were made to obtain the equi-
librium structures of the molecules.
with an energy transfer reaction between the donor and the
oxime. An estimate of the triplet energy level of oxime EO can
T
be derived from the plot of k vs. ED by using the Sandros
q
T
relationship.10 Attempts to Ðt the results by this treatment
were successful in the case of FLO and to a lesser extent for
XTO, both molecules exhibiting vertical energy transfer char-
acter due to the strong decrease of the rate constants in the
endergonic region (Fig. 1). On the contrary, the rate constantÏs
decrease is less pronounced for APO and BPO, exhibiting a
marked NVET character. Such a behaviour could be properly
described by a more sophisticated model taking into account
the intrinsic molecular reorganization. Although the Marcus
theory has been proved to describe energy transfer accurately
for intramolecular11 or incarcerated systems,12 the description
of di†usion-controlled reactions resists that theory. An asymp-
totic free energy relationship, that has been shown to yield
good results6 was used. k is given by:
q
k
A
[ *Gj
RT
B
d
k \
with
k
\ k0 exp
q
en
en
k
k
A
*G
RT
B
~d
en
1 ]
] exp
(1)
where k stands for the di†usion rate constant, k for the rate
d
~d
constant of the encounter complex dissociation, *G \ EO
T
[ ED and k0 is the preexponential factor for the energy trans-
T
en
Scheme 1
¤ All oximes were synthesized from their parent ketones using the fol-
lowing procedure. The ketones (5 mmol), hydroxylammonium sulfate
(3 mmol) and sodium hydroxide (6 mmol) in 1 ml of ethanol and 2.5
ml of water were stirred and reÑuxed for 2 h. The reaction mixture
was extracted with ether and dried with magnesium sulfate. The ether
was removed under vacuum and the oximes were crystallized from
toluene. Pyridine was used instead of ethanol in the synthesis of XTO.
The laser Ñash photolysis system is based on a nanosecond optical
parametric oscillator (Sunlite, Continuum) pumped by a Nd : Yag
laser that generates narrowband radiation in the visible and the near
infrared spectral region. After frequency doubling, the wavelength is
tunable continuously from 225 up to 1800 nm with an energy about
50 mJ at 550 nm. The analyzing system (LP900, Edinburgh
Instruments) used a 450 W pulsed xenon arc lamp, a CzernyÈTurner
monochromator and a fast photomultiplier. Quenching experiments
were performed in degassed benzene (spectroscopic grade, Fluka) with
an optical density of 0.1 at the excitation wavelength.
Fig. 1 Plot of log k vs. ED for triplet energy transfer from donors to
q
T
XTO (=), FLO (K), APO (L) and BPO (…) in benzene. The plots
Ðtting the experimental data were obtained with the Sandros model
for XTO and FLO and by using eqn. (1) for APO and BPO.
DOI: 10.1039/b104163k
Phys. Chem. Chem. Phys., 2001, 3, 2721È2722
This journal is ( The Owner Societies 2001
2721

View Article Online
Table 1 Parameter values derived from the best Ðts and results of
molecular modeling
single electron in p orbital of C3 can undergo stabilization
through delocalization with the neighbouring phenyl rings. In
z
order to achieve this special stabilization, the phenyl groups
FLO
229
XTO
222
APO
È
BPO
È
and p orbital of C3 have to rearrange, thus leading to more
z
planar geometry, with an accompanying increase in reorgani-
Sandros
zation energy. In this framework, the NOH reorganization
leads to the same energy, and the di†erence in *Gj(0) for the
oximes can arise from Ñattening of the carbon skeleton. For
example, it can be seen that the p system in the triplet state is
generally more planar than in the ground state (b in Table 1).
Then, the energy involved in the simple rearrangement of the
carbon skeleton will strongly depend on the oxime studied,
and can be expected to increase with increasing b. This e†ect
is quantitatively in agreement with the change of *Gj(0): the
planar system of FLO and to a lesser extent XTO do not
undergo extensive rearrangement during the reaction, thus
leading to a low value of *Gj(0). On the contrary, APO and
BPO can undergo large reorganization, thus leading to high
values of *Gj(0). Unfortunately, due to the large number of
atoms, accurate semi-empirical calculations in the triplet state
are difficult.
EO/kJ mol~1
T
AgmonÈLevine
EO/kJ mol~1
221
2.0
5.5
216
1.0
6.6
240
1.2
14.8
227
1.3
17.3
T
k0 /1010 M~1 s~1
en
*Gj(0)/kJ mol~1
E
/kJ mol~1
tor
0.67
1.34
3.14
4.83
S
0
b/¡
a/¡
0
180
19.8
179
42
180
49.9
179.4
T
1
b/¡
a/¡
6.4
93.8
8.1
92.8
11.7
91.5
21.1
95.5
Recently, a more reÐned concept has been proposed,7 that
reduces the energy gap between ground and excited state:
thermal activation of torsion modes in the ground state allows
the molecule to adopt a more planar geometry, decreasing the
energetic gap between spectroscopic and relaxed triplet state.
Therefore, the energetic parameter governing the Ñexibility is
fer reaction, related to the adiabaticity of the reaction. *Gj is
given by:
*Gj(0) A A *G ln 2BB
*Gj \ *G ]
ln 1 ] exp [
(2)
ln 2
*Gj(0)
where *Gj(0) is the reorganizational intrinsic barrier related
to the changes in nuclear position. Fig. 1 shows that eqn. (1)
Ðts fairly well the experimental results for APO and BPO,
thus leading to the determination of EO , k0 and *Gj(0) for
related to a torsional energy (E ). The latter represents the
tor
ground state energy di†erence between the relaxed geometry
and the p system with b corresponding to the triplet state
geometry. It can be seen from Fig. 2 that the same trend exists
T
en
each oxime investigated (Table 1). Beyond the variation of EO ,
between *Gj(0) and the calculated E , a fact supporting the
T
tor
obviously depending on the orbital involved, the *Gj(0)
torsional mechanism.
values were found to vary strongly. Indeed, the results gath-
ered in Table 1 show that *Gj(0) is 5.5 kJ mol~1 for FLO
and increases up to 17.3 kJ mol~1 for BPO.
From both these viewpoints, it is clear that the Ñexibility of
the oxime structure strongly inÑuences the behaviour of the
energy transfer process. Although the present semi-empirical
calculations underline the role of the single bond torsion
modes of the ground state on the non-vertical character of the
reaction, it is not possible to estimate its inÑuence with respect
to the whole molecular reorganization. However, in this study
the role of the molecular structure was clearly evidenced and a
Therefore, the *Gj(0) value was taken as a relative measure
of the non-vertical character.6 To gain additonal insight into
NVET, the geometry changes accompanying the passage from
S
to T were analyzed by molecular calculation using AM1
0
1
hamiltonian13 (with the RHF and UHF approximation for
the ground and the excited state respectively). The results are
compiled in Table 1. The general behaviour found for oximes
is similar to that of oleÐns: the transition leads to a more
planar geometry of the p system in the triplet state. The latter
case is characterized by a lower value of the dihedral angle
b(C1C2C3N4) and a twist a(C2C3N4O5) of about 90¡ of the
OH group giving a biradical character to the excited state.14
At Ðrst sight, one would say this rotation of the NÈOH
group in the triplet state should require the same reorganiza-
tion energy for all the oximes studied, knowing that solvent
reorganization can be left out15 and that there is no bridge
nor steric hindrance for the compounds used. Experimental
*Gj(0) values show, however, that this crude approach is not
comprehensive. Indeed, in the spectroscopic triplet state, the
possible speciÐc stabilization of the single electron in the p
carbon orbital by the nearby phenyl rings is suggested. These
z
results provide a semi-quantitative picture of how the molecu-
lar structure a†ects the energy transfer. In a further study, a
purposely synthesized set of acetophenone oxime derivatives
will be studied through energy transfer experiments and high
level ab initio calculations.
References
1
2
G. S. Hammond and J. Saltiel, J. Am. Chem. Soc., 1963, 85, 2516.
W. G. Herkstroeter and G. S. Hammond, J. Am. Chem. Soc.,
1966, 88, 4769.
3
4
5
A. Bylina, Chem. Phys. L ett., 1968, 1, 509.
S. Yamauchi and T. Azumi, J. Am. Chem. Soc., 1973, 95, 2709.
S. Monti, E. Gardini, P. Bortolus and E. Amouyal, Chem. Phys.
L ett., 1977, 77, 115.
V. Balzani, F. Bolletta and F. Scandola, J. Am. Chem. Soc., 1980,
102, 2152.
M. K. Davies, A. A. Gorman, I. Hamblett and D. J. Unett, J.
Photochem. Photobiol., A: Chem., 1995, 88, 5.
X. Allonas, J. Lalevee, J. P. Fouassier, H. Tachi, M. Shirai and
M. Tsunooka, Chem. L ett., 2000, 1090.
S. L. Murov, I. Carmichael and G. L. Hug, Handbook of Photo-
chemistry, Marcel Dekker, New York, 1993.
6
7
8
9
10 K. Sandros, Acta Chem. Scand., 1964, 18, 2355.
11 M. E. Sigman and G. L. Closs, J. Phys. Chem., 1991, 95, 5012.
12 J. Parola, F. Pina, E. Ferreira, M. Meastri and V. Balzani, J. Am.
Chem. Soc., 1996, 118, 11610.
13 MOPAC ver. 6.0, J.J.P. Stewart, QCPE 455.
14 K. Segawa, O. Kikuchi, T. Arai and K. Tokumaru, T HEO-
CHEM, 1995, 343, 133.
15 I. Place, A. Farran, K. Deshayes and P. Piotrowiak, J. Am. Chem.
Soc., 1998, 120, 12626.
Fig. 2 Plot of *Gj(0) vs. the torsion energy E required by the
tor
ground state p-system to reach triplet state geometry.
2722
Phys. Chem. Chem. Phys., 2001, 3, 2721È2722