Photoinduced proton transfer in a pyridine based polymer gel

Article abstract of DOI:10.1021/jp104277r

We describe an experimental and theoretical consideration of photoexcited proton transfer in a poly(4-vinyl pyridine)/pyridine gel. Evidence was found for two states of a multiple state process analyzed by DFT modeling. According to the latter, following irradiation at 385 nm, the proton donor is the CH group of the polymer main chain and the proton acceptor is the nitrogen of the polymaric pyridine side chain. Proton transfer is made possible through the assistance of a mobile pyridine solvent molecule acting as a transfer vehicle. Proton transfer promotes both a geometrical rearrangement of the vinyl side chain as well as electronic density redistribution. The photoproduct intermediate - the hydrogen-bonded complex between the protonated solvent pyridine molecule and the deprotonated polymeric pyridine side chain - is identified by its Curie law magnetic susceptibility, ESR spectrum, and fluorescence lifetime measurements. The proton transfer from the nitrogen of the solvent pyridine molecule to the pyridine side chain nitrogen, producing pyridinium, is a thermodynamically favorable relaxation process and occurs without an energy barrier. The protonation of nitrogen on the polymeric side chain was detected by solid state NMR spectroscopy performed on a ¹⁵N-polymer enriched gel. The calculations and experimental data suggest a central role for the gel solvent molecule as a catalytic agent and proton transfer vehicle. The process suggested by DFT modeling may have relevance for photosensitive devices in part due to the fact that we have been able to show that long-lived paramagnetism may be included among the inducible properties of soft polymer gels. ? 2010 American Chemical Society.

Full text of DOI:10.1021/jp104277r

10728
J. Phys. Chem. B 2010, 114, 10728–10733
Photoinduced Proton Transfer in a Pyridine Based Polymer Gel
Evgenia Vaganova,*,† Ellen Wachtel,‡ Gregory Leitus,‡ David Danovich,§ Stepan Lesnichin,|
Ilja G. Shenderovich,| Hans-Henrich Limbach,| and Shlomo Yitzchaik*,†
Institute of Chemistry and the Farkas Center for Light-Induced Processes, The Hebrew UniVersity of
Jerusalem, 91904, Jerusalem, Israel, Chemical Research Support Unit, The Weizmann Institute of Science,
76100, RehoVot, Israel, Institute of Chemistry and the Lisa Meitner Center for Computational Quantum
Chemistry, The Hebrew UniVersity of Jerusalem, 91904, Jerusalem, Israel, and Department of Chemistry and
Biochemistry, Free UniVersity Berlin, 14195, Berlin, Germany
ReceiVed: May 11, 2010; ReVised Manuscript ReceiVed: June 23, 2010
We describe an experimental and theoretical consideration of photoexcited proton transfer in a poly(4-vinyl
pyridine)/pyridine gel. Evidence was found for two states of a multiple state process analyzed by DFT modeling.
According to the latter, following irradiation at 385 nm, the proton donor is the CH group of the polymer
main chain and the proton acceptor is the nitrogen of the polymeric pyridine side chain. Proton transfer is
made possible through the assistance of a mobile pyridine solvent molecule acting as a transfer vehicle.
Proton transfer promotes both a geometrical rearrangement of the vinyl side chain as well as electronic density
redistribution. The photoproduct intermediatesthe hydrogen-bonded complex between the protonated solvent
pyridine molecule and the deprotonated polymeric pyridine side chainsis identified by its Curie law magnetic
susceptibility, ESR spectrum, and fluorescence lifetime measurements. The proton transfer from the nitrogen
of the solvent pyridine molecule to the pyridine side chain nitrogen, producing pyridinium, is a
thermodynamically favorable relaxation process and occurs without an energy barrier. The protonation of
nitrogen on the polymeric side chain was detected by solid state NMR spectroscopy performed on a 15N-
polymer enriched gel. The calculations and experimental data suggest a central role for the gel solvent molecule
as a catalytic agent and proton transfer vehicle. The process suggested by DFT modeling may have relevance
for photosensitive devices in part due to the fact that we have been able to show that long-lived paramagnetism
may be included among the inducible properties of soft polymer gels.
Introduction
The unusual light-sensitive properties of a gel comprising the
polymer poly(4-vinyl pyridine) in liquid pyridine (P4VPy/Py)
have been characterized during the past few years: reversible,
photoinduced conductivity which increased by as much as 50%
under ultraviolet irradiation as compared to the dark state14 and
reversible changes of the ultraviolet induced emission spectrum
from bright blue at 440 nm to multiple red-shifted emissions.15
Proton transfer was suggested to play a central role in the
photoinduced behavior of the poly(4-vinyl pyridine)/pyridine
gel;14 however, the mechanism could not be unambiguously
identified.
Here, we present a combined experimental and theoretical
investigation of the photochemical reaction. Evidence for an
intermediate paramagnetic state photoproduct, an H-bonded
complex between the pyridine side chain and liquid pyridine,
predicted by DFT modeling, is obtained from magnetic sus-
ceptibility measurements and ESR spectroscopy. A triplet-singlet
relaxation process results in the formation of protonated pyridine
side chains, detected using 15N NMR spectroscopy. The calcula-
tions and experimental data suggest a central role for the gel
solvent molecule as a catalytic agent and proton transfer vehicle.
Soft polymer gels1 constitute a particularly flexible state of
matter in which the tuning of molecular electronic and/or
geometrical properties is markedly enhanced by collective
effects. Consequently, mechanical, physical, and chemical
properties can be readily modified by a wide variety of factors,
including electromagnetic radiation, acoustic and electric fields,
temperature, composition, and pH.2-7 Polymer gels undergoing
structural changes induced by ultraviolet, visible, or near-infrared
radiation have been proposed as candidates for light-sensitive
memory devices, artificial muscles, and optical switches.8-10
Currently, there is no comprehensive theory which describes
the behavior of gels. Many properties are under investigation
including the volume fraction of gel cross-linking points,11 solid
friction between moving segments of the polymer chain and
cross-linking gel points,12 and the concentration dependence of
the polymer diffusion coefficient.13 While intrinsic gel properties
are broadly responsible for flexibility and reversibility, the
particular combination of polymer, solvent, and external factors
renders each system unique in its activity.
Experimental Methods
* To whom correspondence should be addressed. Phone: 972 2658 6971.
Fax: 972 2658 5319. E-mail: gv@cc.huji.ac.il, sy@cc.huji.ac.il.
† Institute of Chemistry and the Farkas Center for Light-Induced
Processes, The Hebrew University of Jerusalem.
Materials and Methods. Gel Preparation. Poly(4-vinyl
pyridine) (P4VPy, MW 50 000, Polyscience), was dried in a
vacuum chamber for 1 month and then dissolved in dry pyridine
(g99.9%, Aldrich). The residual water content, according to
1H NMR measurements of 0.2 mg of polymer in 0.75 mL of
deuterated benzene (99.96% D), is below the sensitivity of the
‡ The Weizmann Institute of Science.
§ Institute of Chemistry and the Lisa Meitner Center for Computational
Quantum Chemistry, The Hebrew University of Jerusalem.
| Free University Berlin.
10.1021/jp104277r  2010 American Chemical Society
Published on Web 07/29/2010

Photoinduced Proton Transfer
J. Phys. Chem. B, Vol. 114, No. 33, 2010 10729
SCHEME 1: Synthesis of 15N-Poly(4-vinyl pyridine)
technique. The molar ratio of side chain units to free solvent
molecules was 1:1. The gel samples were irradiated for 1.0-2.0
h using a calibrated xenon light source (L7810, Hamamatsu) at
λ ) 385 nm (fluorescence filter 386/23, BrightLine). Following
irradiation, the samples were dried at room temperature in a
vacuum chamber (10-3 atm, ShelLab) for 1.0-1.5 h, in order
to reduce the concentration of free pyridine. Drying contributed
to the stabilization of some of the photoinduced properties.
15N labeled P4VPy was synthesized with a molecular weight
of 200 000 (95% 15N-enriched) (Scheme 1 and Supporting
Information, Synthesis of 15N 4-vinyl pyridine and poly(4-vinyl
pyridine)). Gel samples were prepared with the labeled polymer
and liquid pyridine as described above.
Photoinduced Emission. Photoluminescence spectra were
measured on a Shimadzu RF-5301PC spectrofluorimeter with
a resolution of 0.2 nm and with a xenon lamp light source, 5.1
mW/cm2 at 385 nm. Emission lifetime measurements were made
with a homemade device which included a frequency-tripled
Nd:YAG laser, λ ) 355 nm, a pulse duration of 5 ns
(Continuum), and a photomultiplier tube PMT R 928 (Hamamat-
su).
DFT modeling. We used the hybrid Hartree-Fock density
functional theory (DFT) B3LYP (the Becke three-parameter
Lee-Yang-Parr) method16 as implemented in the Gaussian 03
program17 in both its restricted (for singlet states) and unre-
stricted (for triplet states) forms in conjunction with Dunning’s
triple-ꢀ correlation consistent basis set (cc-pVTZ).18 This
allowed us to model the optimal structure, charge, and spin
distributions of the pyridine side chain in the ground and excited
states. The geometries of the molecules in the gas phase were
fully optimized for all states without symmetry constraints, and
local minima were characterized by frequency calculations
(Supporting Information, Tables S1 and S2). Zero point
vibrational energy (ZPE) corrections were taken into account
in all calculations, and the results presented below include this
correction.
SQUID MPMS XP field shielded magnetometer under both zero
field cooled (ZFC) and field cooled (FC) conditions. Each
measurement was repeated four times.
Electron Spin Resonance Spectroscopy. Experiments were
performed with the Bruker ESR spectrometer ELEXSYS 500.
The temperature was adjusted with a control unit (ER 4131VT)
with an accuracy of (1 K. For g-factor calculations, the signal
of DPPH (g ) 2.0037) was used as a standard. The microwave
power was 20 mW, and the modulation amplitude was 10-4 T.
Solid State 15N-NMR Spectroscopy. CP NMR experiments
were performed under MAS conditions with continuous proton
decoupling. MAS spinning speeds were of the order of 5 kHz.
The 90° pulse width was 4 µs, and the recycle time was 3 s.
Variation of the CP contact times between 2 and 8 ms did not
affect the spectra. All spectra were recorded on a Varian CMX
300 NMR spectrometer operating at 7 T and are referenced to
external solid 15NH4Cl (95% 15N-enriched).
Results
After two weeks of storage in the dark, the dry poly(4-vinyl
pyridine) dissolved in dry liquid pyridine stabilized in the form
of a homogeneous, amorphous gel, as shown by X-ray powder
diffraction measurements.15 The gel samples were irradiated for
1.0-2.0 h. Following irradiation, the samples were dried at room
temperature in a vacuum chamber for 1.0-1.5 h, in order to
reduce the concentration of liquid pyridine. The irradiated and
vacuum-dried gel displays emission at 504 nm, the intensity of
which increases with the time of irradiation at 385 nm (Figure
1A). The relatively long lifetime (∼600 ns) measured at room
temperature is evidence that the photoexcited electronic state
is triplet rather than singlet in character (Figure 1B).
In order to gain insight into the possible location of the
unpaired electronic spins, unrestricted B3LYP calculations were
applied. To the best of our knowledge, the polymer forms a
photosensitive system only in the presence of liquid pyridine.19
The effect of the interaction of the parent molecule with the
solvent was modeled by adding a free pyridine molecule to the
Magnetic Susceptibility Measurements. The DC magnetic
moment was measured at 125 K using a Quantum Design Co.

10730 J. Phys. Chem. B, Vol. 114, No. 33, 2010
Vaganova et al.
Figure 1. (A) Emission spectra of a thin layer (100 µm) of a P4VPy/Py gel irradiated at 385 ( 10 nm (xenon lamp, 5.1 mW/cm2) for 0 min
(dashed lines), 15 min (dotted lines), or 30 min (solid lines). Maximum emission is at 504 nm following excitation at 440 nm. (B) Decay of the
emission at 504 nm following 1 h of irradiation and 1 h of evacuation (black curve); the red curve is the fit to monoexponential decay.
calculation. As can be seen below, the pyridine molecule serves
as a vehicle for a proton transfer reaction similar to that
described as class 4 by Kasha.20 In the initial stage, the H11
atom is bound to the C11 atom and the free pyridine molecule
is located near the pyridine side chain. The structure of the
pyridine side chain in the presence of free pyridine and with a
second pyridine side chain as the terminal group on the main
chain was optimized on the singlet potential energy surface
(Figure 2A, 1). At the optimized singlet state geometry, the
H11-N2 bond length is 3.417 Å and the singlet-triplet vertical
vert
excitation energy ∆Ε is 398.74 kJ/mol (Figure 2B). Optimized
s-t
model parameters are included in the Supporting Information.
Because the transfer of the H-atom (H11) from C11 of the main
chain to N1 of the side chain pyridine is endothermic (-211.88
kJ/mol), such a transfer is not favorable on the singlet potential
energy surface. Further optimization of 1 on the triplet potential
energy surface (taking as a starting point the optimized geometry
of the singlet state) leads to molecular structure 1a (Figure 2A).
Molecular structure 1a is formed by hydrogen bonding of the
protonated solvent pyridine to the nitrogen atom of the pyridine
side chain. The reaction on the triplet potential energy surface
is exothermic (about 205 kJ/mol), and therefore, such a proton
transfer reaction is thermodynamically favorable. We emphasize
that 1 is not a stationary point (local minimum) on the triplet
potential energy surface. Our attempts to locate a local minimum
in addition to 1a or a transition state for the transformation from
1 to 1a on the triplet potential energy surface were unsuccessful.
We take that to mean that the structural change of 1 to 1a (Figure
2B) occurs without a barrier. The electronic ground state of 1a
is a triplet (T1) with a triplet-singlet vertical excitation energy
Figure 2. (A) Molecule 1 interacting with free pyridine following DFT
optimization (molecular structure parameters listed in Tables S1 and
S2 of the Supporting Information); bond lengths are in Å. (B) Energy
level diagram for the three states of molecular structure 1 calculated
using the B3LYP/cc-pVTZ method as implemented in the Gaussian
03 program. The placing of the energy levels on the vertical axis is
not to scale. Energies are in kJ/mol.
vert
of ∆Ε )13.81 kJ/mol. To evaluate the influence, if any, of
t-s
the adjacent side chain group on the photochemical reaction,
we replaced the nearest neighbor side chain by a H-atom (see
the Supporting Information). The results of the calculations for
vert
this molecule are similar to those obtained above (∆Ε ) 21.34
t-s
kJ/mol). Therefore, we conclude that the nature of the terminal
group R on the C11-CH2-CH2R chain has little influence on
the properties of the molecules studied in both the singlet and
triplet states.
safely applied to such a system. The Mulliken atomic charge
and spin density distribution, calculated by the B3LYP/cc-pVTZ
method, are presented in Table 1. We present values only for
C11, N2, H11, and the polymer non-hydrogen side chain atoms.
Because molecule 1a is in a triplet state, the total spin of the
molecule is equal to 2. The values presented clearly show that
the unpaired spin density in the triplet state of 1a is primarily
located on C11 (∼0.7) and on the polymer pyridine ring (∼0.65).
The remainder of the unpaired spin density (∼0.65) is delocal-
The molecular conformation of 1a is strongly polar with a
dipole moment of 11.81 D in S0 and 6.48 D in T1. This indicates
a potentially high sensitivity to an applied static electric field.14
In the UB3LYP calculations of the triplet state, the spin
2
ˆ
contamination is very small (S is around 2.02) which means
that the contribution of undesired spin states to the overall wave
function is almost negligible and the UB3LYP method can be

Photoinduced Proton Transfer
J. Phys. Chem. B, Vol. 114, No. 33, 2010 10731
TABLE 1: Mulliken Atomic Charges (in Electrons) for
Singlet and Triplet States of Molecules 1 and 1a and
Mulliken Atomic Spin Density Distributions for the Triplet
State of Molecule 1a Calculated Using the B3LYP/cc-pVTZ
Methoda
singlet state
triplet state
charge distribution charge distribution spin distribution
atom
1
1a
1a
1a
N1
C2
C3
C4
C5
C6
C11
H11
N2
-0.154
-0.100
-0.079
-0.020
-0.108
-0.095
-0.062
0.087
-0.269
-0.105
-0.130
0.068
-0.123
-0.102
-0.029
0.206
-0.196
-0.083
-0.104
0.064
-0.098
-0.080
0.027
0.182
-0.077
0.194
-0.114
0.194
-0.077
0.660
-0.008
0.249
0.193
-0.063
-0.168
-0.015
Figure 3. Temperature dependence of the molar magnetic susceptibility
of a poly(4-vinyl pyridine)/pyridine gel sample: (black squares)
following 2 h of irradiation at 385 nm (2.4 µW/cm2 xenon lamp) and
1 h of evacuation, FC 4 × 104 Am-1; (red circles) following irradiation
and evacuation, FC 1.2 × 105 Am-1; (red line) Curie law fit; (blue
triangles) following 1 h of evacuation, FC 4 × 104 Am-1; (maroon
triangles) following evacuation, FC 1.2 × 105 Am-1. Inset A: Field
dependence of the magnetic moment of an irradiated, evacuated gel at
10 (black squares) and 300 K (red circles). Inset B: Field dependence
of the magnetic moment of an evacuated gel at 10 (black squares) and
300 K (red circles).
a Values are presented for C11, N2, H11, and polymer side chain
atoms only. The total charge of molecules 1 and 1a (see Figure 2)
equals 0; the total spin of molecule 1a in the triplet equals 2.
ized over all other parts of the molecule (two pyridine rings,
C2H5 and CH2 groups) and is not presented in the table. The T1
state of molecular structure 1a may be long-lived, since it is in
fact the ground state on the triplet potential energy surface. It
is known that hydrogen bonding may increase the lifetime of a
T1 state,21 and that more than one electronic state is usually
involved in excited state proton transfer reactions.21,22 On the
other hand, due to the fact that the energy difference between
the triplet and singlet energy surfaces for 1a is relatively small
(13.81 kJ/mol), spin inversion can occur and there is a
probability for transferring the system on the singlet potential
energy surface to the much more thermodynamically stable
pyridinium polymer ion 1b. In order for spin inversion (SI) to
take place, the singlet and triplet potential energy surfaces should
have a crossing point.23 The minimum energy crossing point
(MECP) was located with the B3LYP/cc-pVTZ level of DFT
using the program developed by Harvey at al.24 (Cartesian
coordinates of the crossing point are given in Table S1 of the
Supporting Information.) It lies just 1.59 kJ/mol above the triplet
minimum of 1a and 12.22 kJ/mol below the energy of 1a on
the singlet potential energy surface. It is possible to estimate
the probability for a crossover at the spin inversion junction
using the Landay-Zener treatment (see ref 23 for a discussion
of this treatment). The probability is proportional to the spin
orbit coupling matrix element between the two spin states, the
effective velocity of passing through the crossing point, and
the difference in the slopes of the two intersecting surfaces at
the crossing point. Such calculations will be initiated in future
work. The singlet state of 1b is about 167.36 kJ/mol more stable
than the triplet state of 1a (Figure 2A, 1b). In 1b, the aromatic
structure of the pyridine side chain has been converted to a
conjugated structure of the quinoid type in its protonated form.
While the initial, nonprotonated side chain is rotated by 60°
with respect to the alkane chain, the protonated pyridine side
chain is almost parallel to the direction of the alkane chain.
To search experimentally for evidence of the triplet state
intermediate predicted by the DFT modeling, we first measured
the magnetic susceptibility of the irradiated and evacuated gel
as a function of both temperature and magnetic field. Typical
results for the molar susceptibility measurements (formula
weight ) 105.139 g/mol) obtained at 4 × 104 and 1.2 × 105
Am-1 for both irradiated and nonirradiated samples are presented
in Figure 3. The zero field cooled (ZFC) temperature dependence
of the susceptibility coincides with the field cooled (FC)
dependence. Taking into account a temperature independent
diamagnetic contribution to the susceptibility, ꢁ0 ) -7.799(0)
× 10-9 m3/kg, both curves can be satisfactorily described by
Curie’s law (ꢁ ) C/(T - Θ)), where C is the Curie constant
and Θ is the Curie-Weiss temperature. Fitting with the
Levenberg-Marquart procedure gives C ) 1.11(3) × 10-6
emu*K/g and Θ ) -4.04 K. The nonirradiated samples
demonstrated only temperature independent susceptibility (Fig-
ure 3) which comes from the orbital magnetism of the filled
atomic electronic shells, and the bonding molecular orbitals
together with the holder contribution. Increasing the time of
irradiation increased the relative amount of the paramagnetic
component: magnetic moments of 0.11 and 0.20 µB/mol for 1.0
and 2.0 h of irradiation, respectively. The paramagnetic proper-
ties are not permanent. After 10-15 h at 10 K, only the
diamagnetic contribution to the susceptibility was observed.
Material that had been irradiated but not vacuum-dried did not
display paramagnetism. The fact that magnetic susceptibility
measurements of the irradiated (2 h) gel gave a Curie constant
corresponding to an effective magnetic moment of ∼0.2 µB/
mol of irradiated material shows that, on average, an unpaired
electron is present on one out of five vinyl pyridine monomers.
Paramagnetic behavior was also probed with electron spin
resonance (ESR) measurements. The ESR signal at 125 K of a
gel sample immediately following 1 h of irradiation and 1 h of
vacuum drying is presented in Figure 4. The irradiated sample
displays a prominent resonance with a g-factor of 2.199 and
line width of 0.0199 T. A second weaker resonance was
observed with a g-factor of 1.914 and aline width of 0.0050 T.
g-factor calibration was based on DPPH (g ) 2.0037). After
5 h in the dark at room temperature (Figure 4), the low field
resonance is no longer observed. No low field resonance was
observed for the gel sample which had undergone vacuum
drying but not irradiation. Material that had been irradiated but
not vacuum-dried also did not display a low field resonance.
The high field resonance is identified (Figure 4) as being due

10732 J. Phys. Chem. B, Vol. 114, No. 33, 2010
Vaganova et al.
Figure 5. Room temperature 15N CP MAS NMR spectra of (a) gel
prepared with 15N labeled P4VPy and nonlabeled bulk pyridine without
irradiation and (b) gel prepared with 15N labeled P4VPy and nonlabeled
bulk pyridine after 2 h of irradiation at 385 nm and 10 min of measuring
time. Mw of the polymer: 200 000 Da. The variation of the CP contact
times between 2 and 8 ms did not affect the spectra. All spectra are
referenced to external solid 15NH4Cl (95% 15N-enriched). (The arrows
denote the resonance peaks.)
Figure 4. Low temperature (125 K) X-band ESR spectra of (A) poly(4-
vinyl pyridine)/pyridine gel (weight 0.077 g) following 1 h of irradiation
at 385 nm (2.4 µW/cm2 xenon lamp) and 1 h under a vacuum (10-3
atm). (B) The same sample as for part A after 5 h at room temperature
under Al foil. (C) The empty glass ESR tube. Experiments were
performed with the Bruker ESR spectrometer ELEXSYS 500. The
temperature was adjusted with a control unit (ER 4131VT) with an
accuracy of (1 K. Microwave power 20 mW; modulation amplitude
10-4 T.
to those obtained for neat frozen pyridine and collidine, 278
and 268 ppm, respectively.26 Irradiation gave rise to an
additional weak signal, high field shifted with respect to the
first by 115 ppm. Such a shift is characteristic of protonated
pyridines such as the pyridinium cation, 172 ppm, and the
collidinium cation, 148 ppm.24 The protonation is reversible.
The intensity of the signal attributed to the protonated pyridine
side chains decreases with time and disappeared after 1 h. It is
known that hydrogen bonding may increase the lifetime of a
tautomer state.21
to magnetic impurities in the empty glass ESR tube. We identify
the low field resonance as being due to uncompensated spins
of the P4VPy/Py gel.
The low field resonance in Figure 4 is typical of a single
unpaired spin, rather than a triplet species, and displays both a
large line width and an unusually large value for the g-factor.
At this point, we do not have an explanation for the latter even
considering strong spin-orbit interactions. The large line width
may have several causes. It may be accounted for by dipolar
coupling of paramagnetic centers randomly oriented in a frozen
matrix which produces inhomogeneous line broadening. How-
ever, when the gel is irradiated for 15 min instead of 1 h, with
1 h of vacuum drying, the same broad ESR signal is observed
but with much lower intensity. This indicates that a high
concentration of paramagnetic centers is not responsible for the
large line width. Anisotropies of g-factors and/or hyperfine
interactions are alternative causes. As far as the single spin
character of the resonance is concerned, such behavior has
indeed been observed previously for a long-lived, photoexcited
triplet biradical intermediate when one of the unpaired electrons
is well localized and the second is significantly delocalized.25
As described above, the DFT modeling predicts such a distribu-
tion of unpaired electrons for the P4VPy/Py system studied here.
We also note that the time scale probed by the measurement
may be responsible for some of the ESR features described here.
A solid sample of CuSO4 ·5H2O of known weight was used
as a standard for estimation of the concentration of paramagnetic
centers by double integration of the ESR signal. The result is
∼7 × 1018 electrons/g with an accuracy of (20%.
Conclusion
We have presented experimental evidence in support of a
possible mechanism for photoexcited proton transfer in a poly(4-
vinyl pyridine)/pyridine gel. According to DFT modeling,
following irradiation at 385 nm, the proton donor is the CH
group of the polymer main chain and the proton acceptor is the
nitrogen of the polymeric pyridine side chain. Proton transfer
is made possible through the assistance of the mobile pyridine
solvent molecule acting as a transfer vehicle. This process may
be relevant for photosensitive devices as well as being consid-
ered as a possible model for photochemical processes in
biological systems. Nevertheless, in order to prove that the
different chemical species of the multistep photochemical
reaction are related via specific chemical processes, it will be
necessary to determine minimum energy paths as well as the
magnitude of the spin orbit coupling. We do plan to undertake
this type of calculation at a later date. On the basis of our
experimental results, we also conclude that long-lived para-
magnetism may be included among the inducible properties of
soft polymer gels.
Acknowledgment. We thank Dr. Lev Weiner of the Weiz-
mann Institute of Science for the ESR measurements and Prof.
Uri Banin and Ph.D. student Amit Sitt of the Hebrew University
of Jerusalem for the luminescence lifetime measurements. E.V.
gratefully acknowledges financial support from the Israel
Ministry for Immigrant Absorption. This work was supported
in part by the EC through contract FP6-029192 (E.V. and S.Y.).
The presence of the more thermodynamically stable proto-
nated pyridine side chains, as predicted by DFT modeling
(molecule 1b, Figure 2), was detected at room temperature in
the irradiated 95% 15N-enriched P4VPy/Py gel. The room
temperature CP MAS solid state 15N NMR spectra were
measured both before and after 1 h of irradiation at 385 nm
(Figure 5). Prior to irradiation, one peak due to the chain
pyridine is observed at 270 ppm. The chemical shift is similar
Supporting Information Available: The Materials and
Methods section and supporting figures and tables. This

Photoinduced Proton Transfer
J. Phys. Chem. B, Vol. 114, No. 33, 2010 10733
(13) Wan, W.; Whittenburg, S. L. Macromolecules 1986, 19, 925–927.
(14) Berestetsky, N.; et al. J. Phys. Chem. B 2008, 112, 3662–3667.
(15) Vaganova, E.; Meshulam, G.; Kotler, Z.; Rozenberg, M; Yitzchaik,
S. J. Fluoresc. 2000, 10, 81–88.
(16) Kohn, W.; Holthausen, M. C. A Chemist’s Guide to Density
Functional Theory; Wiley-VCH: New York, 2001.
information is available free of charge via the Internet at http://
pubs.acs.org.
References and Notes
(1) De Gennes, P. G. Macromolecules 1976, 9, 587–598.
(2) Johlin, J. M. Science 1932, 75, 462.
(3) Tanaka, T.; Nishio, I.; Sun, S.-T.; Ueno-Nishio, S. Science 1982,
218, 467–469.
(4) Tokita, M.; Tanaka, T. Science 1982, 253, 1121–1123.
(5) Hu, Z.; Zhang, X.; Li, Y. Science 1995, 269, 525–527.
(6) Yashin, V. V.; Balazs, A. C. Science 2006, 314, 798–801.
(7) Siegel, R. A.; Firestone, B A. Macromolecules 1998, 21, 3254–
3259.
(8) Mamada, A.; Tanaka, T.; Kungwatchakun, D.; Irie, D. Macromol-
ecules 1990, 23, 1517–1519.
(17) Frisch, M. J. Gaussian 03, revision C.02; Gaussian, Inc.: Walling-
ford, CT, 2004.
(18) Dunning, T. H., Jr. J. Chem. Phys. 1989, 90, 1007–1023.
(19) Vaganova, E.; Yitschaik, S. Acta Polym. 1998, 49, 636–641.
(20) Kasha, M. J. Chem. Soc., Faraday Trans. 2 1986, 82, 2379–2392.
(21) Eichen, Y.; et al. J. Chem. Soc., Chem. Commun. 1995, 713–714.
(22) Godfrey, T. S.; Porter, G.; Suppan, P. Discuss. Faraday Soc. 1965,
39, 194–199.
(23) Danovich, D.; Shaik, S. J. Am. Chem. Soc. 1997, 119, 1773–1786.
(24) Harvey, J. N.; Aschi, M.; Schwarz, H.; Koch, W. Theor. Chem.
Acc. 1998, 99, 95–99.
(9) Suzuki, A.; Tanaka, T. Nature 1990, 346, 345–347.
(10) Juodkazis, S.; et al. Nature 2000, 408, 178.
(11) Li, Y.; Sun, Z.; Su, Z.; Shi, T.; An, L J. Chem. Phys. 2005, 122,
194909-1–194909-6.
(25) Sajimon, M. C.; Ramaiah, D.; Suresh, C. H.; Adam, W.; Lewis,
F. D.; George, M. V. J. Am. Chem. Soc. 2007, 129, 9439–9445.
(26) Lorente, P.; et al. Magn. Reson. Chem. 2001, 39, S18–S29.
(12) Burlatsky, S.; Deutch, J. Science 1993, 260, 1782–1784.
JP104277R