Pyridine as proton acceptor in the concerted proton electron transfer oxidation of phenol

Article abstract of DOI:10.1039/c1ob05090g

Taking pyridine as a prototypal example of biologically important nitrogen bases involved in proton-coupled electron transfers, it is shown with the example of the photochemically triggered oxidation of phenol by Ru ^III(bpy)₃ that th

Full text of DOI:10.1039/c1ob05090g

View Article Online / Journal Homepage / Table of Contents for this issue
Organic &
Biomolecular
Chemistry
Dynamic Article Links
Cite this: Org. Biomol. Chem., 2011, 9, 4064
www.rsc.org/obc
PAPER
Pyridine as proton acceptor in the concerted proton electron transfer
oxidation of phenol†
Julien Bonin, Cyrille Costentin, Marc Robert and Jean-Michel Save´ant*
Received 17th January 2011, Accepted 29th March 2011
DOI: 10.1039/c1ob05090g
Taking pyridine as a prototypal example of biologically important nitrogen bases involved in
proton-coupled electron transfers, it is shown with the example of the photochemically triggered
oxidation of phenol by RuIII(bpy)3 that this proton acceptor partakes in a concerted pathway whose
kinetic characteristics can be extracted from the overall kinetic response. The treatment of these data,
implemented by the results of a parallel study carried out in heavy water, allowed the determination of
the intrinsic kinetic characteristics of this proton acceptor. Comparison of the reorganization energies
and of the pre-exponential factors previously derived for hydrogen phosphate and water (in water) as
proton acceptors suggests that, in the case of pyridine, the proton charge is delocalized over a primary
shell of water molecules firmly bound to the pyridinium cation.
Introduction
Homogeneous as well as heterogeneous oxidation of phenols has
often been taken as a prototypal example in the mechanism
analysis of proton coupled electron transfer (PCET) reactions,1
an issue that attracts considerable current attention from a
fundamental point of view and also in relation with the role of
phenol oxidation in natural and artificial processes.2
One of the main thrusts of kinetic and mechanistic analysis of
PCET reactions is to establish whether or not the reaction follows
a concerted pathway (CPET) or one of the two stepwise pathways,
PET (proton transfer first, followed by electron transfer) or EPT
(electron transfer first, followed by proton transfer) as sketched
in Scheme 1. Concerted processes indeed have the advantage of
Scheme 1
skipping high energy intermediates, and thus fully benefit from
the increase of driving force deriving from protonation, provided
the kinetic price that may have to be paid consequently is not too
high.
We have recently examined in detail the intrinsic characteristics
of water (in the solvent water) as opposed to those of a more
conventional base, hydrogen phosphate, as proton acceptors in
the photochemically triggered oxidation of phenol by RuIII(bpy)3.3
The main conclusions reached then were as follows. Hydrogen
phosphate is a more efficient proton acceptor simply because
it offers a better driving force (7.2 in terms of pH units).
However, when the intrinsic characteristics, i.e., the characteristics
at zero-driving force are examined, water (in water) appears as a
more efficient proton acceptor. The intrinsic characteristics are
essentially the function of two parameters: the reorganization
energy of the solvent resulting from the oxidative production of the
proton and the pre-exponential factor, which reflects the coupling
between the electronic states in the transition state and therefore
the efficiency of proton transfer at this stage. The two parameters
governing the intrinsic characteristics of water (in water) are both
more favorable than in the case of phosphate. The reorganization
energy is smaller, revealing that the proton charge is delocalized
over a larger volume and the analysis of the pre-exponential factor
indicates that the phenol–proton acceptor couple is flexible in
line with a Grotthus-type mechanism for the delocalization of the
charge over the hosting water cluster.
Laboratoire d’Electrochimie Mole´culaire, UMR CNRS 7591, Universite´
Paris Diderot, Baˆt. Lavoisier, 15 rue Jean-Antoine de Baif, 75205 Paris
Cedex 13, France. E-mail: saveant@univ-paris-diderot.fr; Fax: +33 1 57 27
87 88; Tel: +33 1 57 27 87 95
† This work is dedicated to the memory of Athel Beckwith, a teacher and
scientist from whom we learned how to study chemistry by example. His
pioneering advances in radical chemistry laid the foundation for much of
the current radical clock methodology.
In this study, hydrogen phosphate was essentially used as a
point of comparison to investigate the special character of water
(in water) as a proton acceptor in a CPET reaction. The interest in
4064 | Org. Biomol. Chem., 2011, 9, 4064–4069
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
deriving the intrinsic properties of this particular proton acceptor
also resides in the comparison with other proton acceptors, besides
water, which may not only exert a different driving force but are
chemically different so that one may expect different properties
at zero driving force in terms of reorganization energies and pre-
exponential factors. Among the various proton acceptors one may
think of, a particular mention should be made of nitrogen bases in
view of their ubiquitous role in natural processes. Their frequent
involvement in H-bonded structures is a favorable factor for the
occurrence of a CPET mechanism because the distance between
the proton and the base corresponding to an H-bond is about
the right distance for easy proton tunneling. As a representative
of the family of H-bonding nitrogen bases, we selected pyridine
as the proton acceptor for the present analysis of the kinetics of
photochemically triggered oxidation of phenol by RuIII(bpy)3.
Fig. 2 Variation of the pseudo-second order phenol oxidation rate
constant with temperature in 5 ¥ 10-5 M equimolar solutions of pyridine
and (H or D) pyridinium ion (see Experimental Section) at 20 ◦C, in
light (squares in a, pH = 5.4) and heavy (circles in b, pD = 5.7) water
compared with the variation of the pseudo-second order phenol oxidation
rate constant with temperature in the absence of pyridine at pH = 2 (upward
triangles in a) and pD = 2.4 (downward triangles in b).
Results and discussion
Two series of laser flash experiments (see Experimental Section)
were carried out. In the first series, the phenol oxidation second
order rate constant was measured at a fixed temperature, namely
20 ◦C, as a function of the pyridine concentration at a pH (or
pD) equal to the pK of pyridine both in light and heavy water
(Fig. 1). As expected, raising the pyridine concentration results,
after an initial plateau, in an increase of the pseudo-second order
rate constant. It would seem that the variations in D2O exhibit a
shallow dip before the final rise of the rate constant. However, the
scatter in the experimental data points (Dlog k = 0.2) is such that
it does not allow one to consider this feature as significant.
At small concentrations of pyridine where the contribution of
the Py-CPET pathway is deemed to be negligible, one expects that
the rate constant should equal the value found in pure water for
the water-CPET pathway, as for example the value found at pH
2. This is indeed the case with H2O (Fig. 2a), indicating that at a
pH equal to the pyridine pK in light water (5.2),4 the OH--PET
pathway is not operating just because OH- concentration is too
low. The situation is different in D2O (Fig. 2b) because the pyridine
pK is higher (5.7),4 opening a narrow but significant route for an
OD--PET pathway.
It is noted that this contribution increases upon going to lower
temperatures, in line with a parallel increase of the pK in D2O.4
It follows that the third-order rate constant for the Py-CPET
pathway may be extracted from the raw data shown in Fig. 3 using
eqn (1) in the case of H2O:
k = kHH
+ kPHy −CPET Py
[ ]
(1)
O−CPET
2
and eqn (2) in the case of D2O:
(2)
Fig. 1 Variation of the pseudo-second order phenol oxidation rate
constant with the concentration of pyridine in equimolar solutions of
pyridine and (H or D) pyridinium ion (see Experimental Section) at 20 ◦C,
in light (squares) and heavy (circles) water. The left-hand data points were
obtained in the absence of pyridine.
In another series of experiments, the temperature was varied at
fixed values of the pyridine concentration both in light and heavy
water (Fig. 2), at a pH (or pD) equal to the pK of pyridine both in
light and heavy water.
As with phosphate,3 the overall oxidation rate is the sum of three
contributions, corresponding respectively to a concerted pathway
in which the solvent, H2O or D2O, is the proton acceptor that
we name H2O (D2O)-CPET, a “PET” pathway (see Scheme 1) in
which OH- is the proton acceptor, and—the object of the present
study—a CPET pathway in which the purposely added base, here
pyridine, functions as the proton acceptor (Py-CPET pathway).
Fig. 3 Variation of the pseudo-second order phenol oxidation rate
constant with temperature in equimolar solutions of pyridine and (H or
D) pyridinium ion (see Experimental Section) at 20 ◦C, in light (squares)
and heavy (circles) water at fixed equal concentrations of pyridine and H-
or D-pyridinium cation (M): 2.5 ¥ 10-2 (a), 1.5 (b).
This journal is
The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 4064–4069 | 4065
©

View Article Online
Thus leading to the Arrhenius plots displayed in Fig. 4, and
finally to:
1105 15
(
)
ln k H
= 18.1 0.05 −
( )
(
)
Py −CPET
T
and:
1680 15
(
)
ln k D
= 19.15 0.05 −
( )
(
)
Py −CPET
T
Fig. 4 a: Arrhenius plot for the third-order rate constant of the Py-CPET
pathway in H2O derived from the square-symbol data in Fig. 2 and 3
according to eqn (1). b: Arrhenius plot for the third-order rate constant
of the Py-CPET pathway in D2O derived from the circle-symbol data in
Fig. 2 and 3 according to eqn (2).
Fig. 5 Potential energy curves for the reorganization of the heavy atoms
of the system, including solvent molecules (parabolae) and for the proton
displacement concerted with electron transfer (upper insert). The symbols
are defined in Scheme 1 and in the text.
Uncertainties are obtained from a statistical treatment of the
data as detailed in reference 3.
These Arrhenius plots may be analyzed as follows in the same
way as previously developed for hydrogen phosphate as the proton
acceptor,3 based on a model whose main features, summarized in
Fig. 5, derive from a double application of the Born–Oppenheimer
approximation. The transition state is defined by the intersection
of the potential energy profiles of the reactant and product systems
toward the heavy-atom coordinate (parabolae in Fig. 5). We may
therefore start from a Marcus-type expression (eqn (3)) of the rate
constant:
term, Stherm, given by equation (10), contains the contributions
of various thermodynamical parameters of the reactions and of
work terms. The last term, Sdyc (equation (11)) derives from the
dynamical coupling of the two states represented in the upper
insert of Fig. 5: as sketched by the spring at the bottom of
the barrier, proton tunneling is more efficient when the distance
between the phenolic oxygen and the proton acceptor is smaller
than its equilibrium value and vice-versa. Sdyc thus involves the
force constant of the spring and the attenuation factor, b, of the
exponential decay of the coupling of the two vibronic states with
distance.
In the intercept (equation (8) in Table 1), the main term
(equation (12)) is the value of the pre-exponential factor for the
equilibrium distance between the phenolic oxygen and the proton
acceptor, Zeq. The term resulting from thermodynamics factors,
Itherm, is given by equation (13). The dynamical coupling term,
Idyc, (equation (14)) involves, as Sdyc, the force constant of the
spring and the attenuation factor, b, of the exponential decay of
the coupling of the two states with distance.
2
⎡
⎢
⎢
⎣
⎤
⎥
⎥
⎦
⎛
⎜
⎜
⎝
⎞
⎟
⎟
⎟
⎟
⎡
⎢
⎣
⎤
⎥
w R
RT
l
4RT
DG0 −w R +w P
⎜
⎢
(3)
k = Z exp −
exp −
1+
⎜
⎢
⎥
⎦
l
⎠ ⎥
Z is the pre-exponential factor. In the Franck–Condon exponen-
tial term, l is the reorganization energy and DG0, the reaction
standard free energy. wR and wP are the work terms required
to bring the reactants and products respectively, from infinite
separation to reacting distance.
Arrhenius plots such as those in Fig. 4a and 4b are obtained
by linearization of eqn (3) around the middle of the temperature
range, Tm (313 K in our case). Although, Z and l are the main
terms in the expression of the Arrhenius intercept (I) and slope
(S), respectively:
All ingredients of a full analysis of the reaction kinetics are
gathered together in Table 1 so as to avoid browsing the whole
paper in search of the pertinent equations and parameters.
The thermodynamic parameters necessary for the kinetic anal-
ysis were obtained from the following standard free energies and
their variations with temperature:
S
lnk = I −
T
other terms should also be taken into account. Both S and I
contain three terms as listed in Table 1.
In the slope (equation (7) in Table 1), reorg is an increasing
function of l given by equation (9) in Table 1. The second
0,CPET
Py,H(D)
DG
= DGH0,CPET O) −RT ln10pK Py,H(D)
O(D
2
2
4066 | Org. Biomol. Chem., 2011, 9, 4064–4069
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
Table 1 Analysis of the thermodynamics and kinetics of the H2O-CPET, PO4H2--CPET and Py-CPET oxidation of phenolg
b
5 c
a Tm = 313 K. approximating d, the distance between the two reactants (Scheme 2), by the value, 7 A. zR or P is the product of the charges in the
˚
reactant and product system respectively. eS, the solvent dielectric constant, varies with temperature: d(1/eS)/dT = 6.213 ¥ 10-5 K-1.6 lox: self-exchange
d
ds 2
reorganization energy for the RuIII/II couple equated to 0.57 eV.5 e for the definition of s and r, see Scheme 2;
: amplitude of the variation of s; f :
force constant of the harmonic oscillator of the H-bond between PhOH and B; Ceq: coupling constant between the two electronic states in the transition
f
state at equilibrium distance between PhOH and B; nn: nuclear frequency. b is the attenuation factor of the exponential decay of the coupling of the
two states with distance. g Uncertainties are obtained from errors on the Arrhenius plots linear regression as detailed in reference 3 and assuming an
uncertainty of 0.02 eV on the reorganization energy with pyridine.
This journal is
The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 4064–4069 | 4067
©

View Article Online
where the first term, the standard free energy involved in the H2O
(D2O)-CPET pathway, and its variations with temperature have
been previously derived3 and the second term is obtained from
literature data recalled in Fig. 6.4
Fig. 6 Variation of the pyridine pKs in H2O (squares) and D2O (circles).
The work terms are assumed to be exclusively from electrostatic
origin and to be the same in H2O and D2O. Their equivalent
enthalpies and entropies are calculated according to the equations
given in Table 1 where their values are also listed.
The treatment of data summarized in Table 1 was the same for
parameter values listed in the last column of the Table. As dis-
cussed earlier,3 the reorganization energy, lCPET , is mostly related
to solvent reorganization, which can be grossly approximated by:7
3
(eV)
lCPET
≈
a(Å)
It follows that the solvation radius, a (see Scheme 2) is
significantly larger for pyridine than for phosphate (approximately
˚
˚
5.7 A vs. 3.5 A). This observation suggests the involvement of a
primary shell of water molecules, solidly bound to the pyridinium
cation. A possible arrangement of these water molecules close to
the ortho and para positions of pyridine is sketched in Scheme 2.
The relatively small values found for the pre-exponential factor,
Z, similar to those found for hydrogen phosphate and definitely
smaller than for water, indeed fall in line with a strong association
of these water molecules, in contrast with the Grotthus-type
proton displacement taking place when water (in water) is the
proton acceptor. This is also in agreement with the observation of
a kinetic isotope effect independent of temperature.
Experimental
Ultrapure water (18.2 MX cm) from a MilliQ (Millipore) pu-
rification system or deuterium oxide (Euriso-Top, 99.9%) were
employed to prepare the samples. Ru(bpy)3,Cl2·6H2O (Fluka,
Technical Grade), phenol (Fluka, Ultra, ≥99.5%), pyridine (Fluka,
≥99.8%) were used without further purification. Methylviologen
dichloride hydrate (Aldrich, 98%) was recrystallized from ethanol
and dried overnight in vacuum before use.
The experimental setup used for the flash-quench measurements
has been described previously in detail as well as the photophysics
of the system.8
Scheme 2
6 to 8 mJ cm-2) via an OPO (Continuum, SLOPO Plus) pumped
by a frequency-tripled Nd:YAG laser (Continuum, Surelite II-
10). Perpendicular analyzing light provided by a 450 W ozone-
free pulsed xenon lamp (Osram XBO) was collected into a
spectrograph. Kinetics at 450 nm for Ru2+/Ru2+* and at 605 nm
for MV∑+ were then measured thanks to a photomultiplier tube
(Hamamatsu, R928) linked to a 100 MHz oscilloscope (Tektronix,
TDS 3012C). Samples typically contained 50 mM Ru complex,
40 mM MV2+, 25 mM phenol and different concentrations of
The transient absorption measurements were carried out with
a laser flash photolysis spectrometer (Edinburgh Instruments,
LP920-KS). The solutions were excited at 460 nm (5 ns pulse,
4068 | Org. Biomol. Chem., 2011, 9, 4064–4069
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
pyridine. Precise temperature of the sample port was set by a
Peltier effect controller (Quantum Northwest TC125). The control
and the synchronization of the whole setup were ensured by the
Edinburgh Instruments L900 software.
at zero driving force) kinetic characteristics of this proton acceptor
in the same way as done previously with hydrogen phosphate and
water (in water). Comparison of the reorganization energies and of
the pre-exponential factors suggests that, in the case of pyridine,
the proton charge is delocalized over a primary shell of water
molecules firmly bound to the pyridinium cation.
The pH of the H2O sample containing pyridine was adjusted
at the value of the pyridine pK with concentrated HCl or NaOH
and controlled with a Hanna, pH 210 pHmeter equipped with a
microelectrode (6 mm, Bioblock Scientific) at 20 ◦C. The same was
done in D2O, the pD being obtained by adding 0.4 to the indication
of the glass electrode.9 Equimolar solutions of pyridine and (H or
D) pyridinium ion could thus be obtained in this way both in
H2O and D2O at 20 ◦C. The pyridine pK in H2O changes with
temperature,4 but the pH and the pK remains practically equal.
This also applies in D2O.
Acknowledgements
Financial support from Agence Nationale de la Recherche (Pro-
gramme Blanc PROTOCOLE) is gratefully acknowledged.
Notes and references
All samples were purged with argon for 15 min prior to
measurement.
1 Chem. Rev. special issue (number 12) on proton-coupled electron
transfer, 2010, 110.
2 C. Costentin, M. Robert and J.-M. Save´ant, Phys. Chem. Chem. Phys.,
2010, 12, 11179.
Conclusions
3 J. Bonin, C. Costentin, C. Louault, M. Robert and J.-M. Save´ant, J. Am.
Chem. Soc., DOI: 10.1021/ja110935c.
4 I. R. Bellobono and P. Beltrame, J. Chem. Soc. B, 1969, 620.
5 N. Sutin and B. S. Brunschwig, ACS Symp. Ser., 1982, 198, 105.
6 CRC Handbook of Chemistry and Physics, 88th edn (ed. D. R. Lide) p.
6–3 (CRC Press, 2007).
7 J.-M. Save´ant, Elements of Molecular and Biomolecular Electrochemistry:
An Electrochemical Approach to Electron Transfer Chemistry; John Wiley
& Sons: Hoboken, NJ, 2006. Chap. 1.
8 J. Bonin, C. Costentin, C. Louault, M. Robert, M. Routier and J.-M.
Save´ant, Proc. Natl. Acad. Sci. U. S. A., 2010, 107, 3367.
9 A. Krezel and W. Bal, J. Inorg. Biochem., 2004, 98, 161.
Taking pyridine as a prototypal example of biologically important
nitrogen bases involved in proton-coupled electron transfer, we
have shown with the example of the photochemically triggered
oxidation of phenol by RuIII(bpy)3 that this proton acceptor
partakes in a concerted pathway whose kinetic characteristics can
be extracted from the overall kinetic response. The treatment of
these data, implemented by the results of a parallel study carried
out in heavy water, allowed the determination of the intrinsic (i.e.
This journal is
The Royal Society of Chemistry 2011
Org. Biomol. Chem., 2011, 9, 4064–4069 | 4069
©