Solvent-induced hydrogen tunnelling in ascorbate proton-coupled electron transfers

Article abstract of DOI:10.1016/j.tetlet.2011.01.142

The over-the-barrier proton-coupled electron transfer interaction of an ascorbate monoanion with a hexacyanoferrate(III) ion in water entered into a tunnelling regime in water-1,4-dioxane, water-ethanol and water-acetonitrile mixed solvents.

Full text of DOI:10.1016/j.tetlet.2011.01.142

Tetrahedron Letters 52 (2011) 1757–1761
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Solvent-induced hydrogen tunnelling in ascorbate proton-coupled
electron transfers
⇑
Ana Karkovi c´ , Cvijeta Jakobuši c´ Brala, Viktor Pilepi c´ , Stanko Urši c´
Faculty of Pharmacy and Biochemistry, University of Zagreb, A. Kova cˇ i ´c a 1, 10000 Zagreb, Croatia
a r t i c l e i n f o
a b s t r a c t
Article history:
The over-the-barrier proton-coupled electron transfer interaction of an ascorbate monoanion with a
hexacyanoferrate(III) ion in water entered into a tunnelling regime in water–1,4-dioxane, water–ethanol
and water–acetonitrile mixed solvents.
Received 16 December 2010
Revised 18 January 2011
Accepted 28 January 2011
Available online 4 February 2011
Ó 2011 Elsevier Ltd. All rights reserved.
Keywords:
Proton-coupled electron transfer
Kinetic isotope effect
Hydrogen tunnelling
Dynamics
Solvent
Ascorbate
Quantum-mechanical hydrogen tunnelling is of remarkable sig-
the electron and proton transfer simultaneously between the same
nificance for chemistry and biochemistry.1
–19
The role of hydrogen
donor and acceptor
21,22,35
are also included in the framework.
tunnelling in such fundamental chemical reactions as hydrogen
However, the concerted transfer of a proton and an electron where
there is a single chemical reaction step, direct coupling of the elec-
tron and proton in the transfer is the elementary characteristic of
an authentic PCET reaction.
transfers is well documented3
has been shown to take place via tunnelling.2
enzymatic H-transfer reactions, including those where there is
,4,17–19
and enzymatic C–H activation
–10,13,14,16
Many non-
N–H, S–H and O–H activation1
7,18
also involve hydrogen tunnelling.
We report here the observation of solvent-induced tunnelling in
the oxidation of an ascorbate monoanion with hexacyanoferrate(III)
The tunnelling in a condensed phase has sometimes revealed
somewhat ‘exotic’ features. Some examples are the appearance of
3
6–38
37
ions
(Scheme 1). This PCET reaction in ‘pure’ water does not
3
7
‘
colossal’ kinetic isotope effects (KIEs) ranging up to k
H
/k
D
= 455
exhibit hydrogen tunnelling. Very unexpectedly, on the addition
of only ꢀ1 M of an organic co-solvent, the reaction exhibited hydro-
gen tunnelling as demonstrated by the markedly changed isotopic
ratios of the Arrhenius pre-factors that are well beyond the semi-
1
8a
at room temperature
(in the case of N–H–O transfer) in water–
acetonitrile solution, and the rearrangement of matrix-isolated
hydroxymethylene at 11 K into formaldehyde19 which involves
H-transfer from oxygen to carbon by pure tunnelling through a
large energy barrier.
classical limits of 0.5–1.4 for the A
H
/A
D
ratio in a hydrogen transfer
1
,3,4,39–41
process,
of activation (DD
the semiclassical value of 5.1 kJ/mol for the difference between
as well as the isotopic differences in the enthalpies
à
Proton-coupled electron transfer (PCET) reactions play a key
role in a wide range of biological, biochemical and chemical
r 2 2
H ) between D O and H O, which are greater than
2
0–34
D
H
systems,
and solar fuels,
processes in biology would not be possible without the coupling
including those related to artificial photosynthesis
zero-point energies Eo ꢁ E for the dissociation of an O–H bond
o
2
3–27
33,34
1,3,4,38–40
and nanostructures and interfaces.
Many
and are indicative of hydrogen tunnelling in the reaction.
(Table 1). To the best of our knowledge, the observation of solvent-
induced tunnelling is unprecedented. Our findings are as follows:
(i) The oxidation of ascorbate monoanions with hexacyanofer-
rate(III) ions (Scheme 1) in various water–organic co-solvent
mixtures involves a PCET process. This is true for the water–
acetonitrile, water–1,4-dioxane, water–ethanol and water–acetone
solvents used in this Letter (see Table 1 and Supplementary data
(SI)). Thermochemical analysis of the corresponding consecutive
(PT/ET or ET/PT) reactions (see also the Supplementary data)2
of proton and electron motion.2
2,24
From a theoretical viewpoint,
both sequential electron transfer followed by proton transfer (ET/
PT) or vice versa (PT/ET), and the concerted transfer of these parti-
3
5
cles could be viewed within a unified theoretical framework of
PCET reactions. Hydrogen atom transfer (HAT) reactions, in which
⇑
Corresponding author. Tel.: +385 1 4818 306; fax: +385 1 4856 201.
E-mail address: stu@pharma.hr (S. Urši c´ ).
1,38
0
040-4039/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2011.01.142

1
758
A. Karkovi ´c et al. / Tetrahedron Letters 52 (2011) 1757–1761
OH
O
OH
-
O
O
O
kHAsc
-
HO
HO
3
-
4
-
+
+
[Fe(CN)6]
+ [Fe(CN)6]
+ H
•
-
O
O
-
O
OH
•
-
HAsc
Asc
Scheme 1. HAsc = ascorbate monoanion, Asc = ascorbyl radical anion.38
ꢁ
Åꢁ
Table 1
Activation parameters and kinetic isotope effects in the reaction shown in Scheme 1 in various mixed solvents38
Solventa,b,c,d
k
HAsc-/Mꢁ1
s
ꢁ1
KIE
D
H /kJ mol
à
ꢁ1
D
S /J K mol
à
ꢁ1
ꢁ1
DDHàd/kJ molꢁ1
H D
A /A
H
2
H
2
H
2
H
2
H
2
H
2
H
2
H
2
H
2
H
2
H
2
H
2
H
2
H
2
O–MeCN (1:1 v/v)
18.9 (0.1)
8.25 (0.09)
6.55 (0.05)
6.70 (0.27)
8.37 (0.16)
24.8 (0.3)
37.6 (0.2)
26.4 (0.2)
19.7 (0.1)
ꢁ137.3 (1.0)
ꢁ99.0 (0.7)
ꢁ136.0 (0.7)
ꢁ146.0 (0.3)
6.9 (0.4)
0.49 (0.09)
O–MeCN (1:1 v/v), 0.1 M Na–acetate
O–MeCN (1:1 v/v), 0.1 M NaCl
O–1,4-dioxane (1:1 v/v)b
O–1,4-dioxane (1:1 v/v), 0.3 M Na–acetate
O–1,4-dioxane (1:1 v/v), 0.3 M NaCl
O–EtOH (1:1 v/v)c
O–acetone (1:1 v/v)
O–MeCN (0.75:0.25 v/v)
O–MeCN (0.95:0.05 v/v)
O–1,4-dioxane (0.95:0.05 v/v)
O–EtOH (0.95:0.05 v/v)
141.5 (0.8)
149.5 (1.9)
8.2 (0.1)
282.9 (4.0)
274.9 (2.3)
11.3 (0.1)
6.7 (0.1)
22.8 (0.1)
52.2 (1.0)
47.2 (0.1)
53.6 (0.1)
46.2(0.8)
71.0 (1.0)
12.7 (0.4)
11.9 (0.4)
7.1 (0.4)
0.046 (0.008)
0.065 (0.010)
0.32 (0.04)
7.90 (0.22)
8.59 (0.13)
7.81 (0.17)
5.61 (0.16)
5.35 (0.06)
5.42 (0.02)
5.33 (0.17)
4.60 (0.06)
O–acetone (0.95:0.05 v/v)
e
O
19.1 (0.2)
ꢁ142.7 (0.7)
3.9 (0.4)
0.97 (0.15)
a
b
c
At 298 K, unless otherwise noted. Rate constants were determined as described in the Supplementary data.38 I = 0.0023, unless otherwise noted.
At 293 K.
I = 0.0015.
d
e
à
DD
H
(D,H).
Data taken from Ref. 37.
O
shows that the free energy barrier (
D
r
G ) for initial PT in the
are consistent with hydrogen tunnelling being an important reac-
tion channel at relatively high organic co-solvent content, at least
when 1,4-dioxane, ethanol or acetonitrile (and likely acetone) are
used, but also suggest entering into a tunnelling regime even at
ca. 1 M of the added organic co-solvent, at least in the case of
sequential PT/ET in, for example, water–acetonitrile (1:1) would
be at least 72.5 kJ/mol or even 76.5 kJ/mol. This would be at least
7
kJ/mol ‘up-hill’ with respect to the observed38 activation barrier
of 65.8 kJ/mol in the reaction in this solvent, at an ionic strength
I = 0.0023. As the activation barrier cannot be smaller than the
H D
water–acetonitrile. We do not have the A /A value for the
à
O
43
ground-state barrier, that is,
D
G P
D
r
G , this consecutive path-
water–acetone systems, but it seems reasonable to suppose, tak-
ing into account that the observed changes of the KIEs on going
from water to water–acetone mixtures are very similar to the cor-
responding changes in the other solvents used, that the basic phys-
ical picture in this case could be similar to the cases of the other
three organic co-solvents.
way should be ruled out. The corresponding free energy barrier
for the initial ET in the sequential ET/PT would be significantly
3
8
greater than 31 kJ/mol, the free energy barrier in water, but this
sequential pathway should be ruled out since a kinetic isotope ef-
H D
fect, (k /k = 8.2) was observed in the reaction (Table 1; the kinet-
ics were determined spectrophotometrically by monitoring the
decrease of absorbance of hexacyanoferrate(III) ions38). That is,
the observed kinetic isotope effect (KIE) clearly indicates the
involvement of proton transfer in the rate-controlling step, whilst
in the initial ET the intermediate protonated ascorbyl radical
would be formed; this process, of course, should not involve any
proton transfer. Additionally, in the hypothetical sequential ET/PT
process, the subsequent proton transfer from the protonated ascor-
(iii) The kinetic isotope effect between the ascorbate monoan-
ion and its 2-OD derivative in the reaction changes significantly
on the addition of only 5% of the organic co-solvent into the water
H D
/k
= 4.60 in ‘pure’ water solution,37
reaction solution; whilst k
values of /k = 5.61 (water–acetonitrile 0.95:0.05 v/v),
= 5.42 (water–ethanol 0.95:0.05 v/v), k /k = 5.35 (water–1,4-
dioxane 0.95:0.05 v/v) and k /k = 5.33 (water–acetone 0.95:0.05
k
H
D
H
k /
k
D
H
D
H
D
4
4
v/v) were observed (Table 1). Moreover, sizeable changes of the
KIE have been observed in the 1:1 v/v water–organic co-solvent
4
2
byl radical should be fast (the pK
a
of this radical is ꢁ0.45 ) and
should not be rate-controlling. Taken together, the obtained evi-
dence strongly suggests a concerted, proton-coupled electron
transfer in the reaction. Similar considerations apply in the cases
systems; here, k
(water–ethanol 1:1),
20 °C) and k /k = 8.59 (water–acetone 1:1). This observation is
consistent with an increase in the donor–acceptor distance for
/k
H D
H D
= 8.24 (water–acetonitrile 1:1), k /k = 7.90
k
H
/k = 8.37 (water–1,4-dioxane 1:1, at
D
H
D
3
8
of the other solvents used in this study (see Supplementary data ).
ii) The Arrhenius pre-factor A /A which is 0.97 in the reaction
in water, a typical value for an over-the-barrier reaction, changed
to A /A = 0.49 in water–acetonitrile (1:1 v/v), to A /A = 0.065 in
water–ethanol (1:1 v/v) and to A /A = 0.046 in water–1,4-dioxane
1:1 v/v), that is, well beyond the semiclassical limits of 0.5–1.4 for
the A /A ratio in a (over-the-barrier) hydrogen transfer process
Table 1). Moreover, on the addition of only 5% v/v of acetonitrile
to the water solution, the reaction immediately entered into a tun-
nelling regime. Very surprisingly, an isotopic ratio of A /A = 0.32
in water–acetonitrile (0.95:0.05 v/v) was observed. These findings
1
4,45
(
H
D
the transfer of a proton in a PCET
process; the donor–acceptor
3
7
distance in this reaction should also depend on the separation of
the redox partners, the hexacyanoferrate(III) anion and the ascor-
bate anion, due to electrostatic interactions (repulsive forces) be-
tween the negatively charged ions (see below).
H
D
H
D
H
D
(
H
D
(iv) The water molecule should be part of the activated complex
in the reaction as is evident from the proton inventory4
6–48
ob-
(
served in the reaction in water–acetonitrile 1:1. Inspection of the
4
9
H
D
proton inventory revealed a concave curved dependence of kobs
versus xD2O. This observation should be consistent with more than

A. Karkovi ´c et al. / Tetrahedron Letters 52 (2011) 1757–1761
1759
one proton being involved in the transition state,46–49 since proton
transfer from the ascorbate 2-OH group directly to hexacyanofer-
rate (followed by subsequent fast transfer of the proton to the bulk
under the conditions employed) would produce a linear proton
A
= 0.046 in the water–1,4-dioxane solvent, which is less polar
than the water–ethanol solvent, suggesting a decrease of the A
with an expected increase in the initial donor/acceptor separa-
D
H
/
A
D
1
4,45
tion,
reminiscent of the case of variation of the ratio of Arrhe-
4
6–48
inventory.
Therefore, the observed concave curved proton
nius pre-factors versus the donor–acceptor distance obtained in
the study of hydrogen tunnelling in soybean lipoxygenase-1.7
At first sight, the observation of the greatly increased, but very sim-
ilar KIEs on going from water to the mixed solvents (Table 1) could
contradict the expectation of an increase in the KIE with the above-
inferred increase of the donor–acceptor distance in the reaction.
However, as demonstrated by Pudney et al., the KIE can increase
as the donor–acceptor distance decreases because the force con-
stant for a compressive mode can be increased. The result will be
the relative invariance of the KIE as was obtained by Pudney
et al. in the H-tunnelling reaction of morphinone reductase,
and is what in fact we observed in the case of ascorbate oxidation.
In addition, the observed strong temperature dependence of the
KIE (Table 1 and Fig. 1) also points to the role of the compressive
modes coincident with the reaction coordinate, needed for efficient
–9
inventory should be consistent with a molecule of water being in-
volved in the proton transfer during the reaction.
(
v) Neither the observed rate constant nor the observed KIE in
the reaction were changed in the presence of 0.1 M of acetate
5
0
ions, a potential proton acceptor in 1:1 acetonitrile–water sys-
tem, and in the presence of 0.3 M acetate in water–1,4-dioxane
1
4
3
8
(
1:1) (Table 1). This observation suggests that acetate ions do
not act as proton acceptors during the reaction. It would appear
highly unexpected that 0.3 M acetate (being anionic yet 6.5 pK
a
1
4
units more basic than water) does not compete with the water
for the transferring proton in the reaction if the proton transfers
to the bulk water; this unexpected observation suggests that the
proton transfer from the 2-OH of ascorbate to a molecule(s) of
water could be more energetically favourable than the proton
transfer to an acetate ion. As unidirectional2
1,51,52
PCET can gener-
tunnelling.
3–15,59
ally be more energetically favourable than bidirectional PCET, the
absence of catalysis in the presence of acetate ions implies a con-
figuration where the water molecule can be found between the re-
dox partners. Taken together with the other evidence obtained,
the role of the configuration seems to be critical to the tunnelling
phenomena observed in this reaction.
Generally, the expectation is that barrier height is critically
important as well as barrier width in determining the probability
1
,14,15,17
of tunnelling in a chemical reaction.
There is an obvious in-
5
3
crease in the reaction barrier (see Table 1) in the case of the 1:1
3
7
mixed solvents (the barrier in water is 62.3 kJ ). However, we also
observed the entering into a tunnelling regime in the reaction in
the case (0.95:0.05 v/v water–acetonitrile system) where the in-
crease in the reaction barrier was only ꢀ1 kJ, which could be quite
puzzling if dynamical aspects were not be taken into account. Per-
haps, the observation we obtained, that only 1 M of the co-solvent
can have such a dynamic effect, is the most intriguing result in this
Letter. In our view, the dynamics of the compressive modes coinci-
dent with the reaction coordinate could be coupled with the sol-
vent environment dynamics, which in turn should be strongly
determined by the ‘integrity’ of the solvent shell around the reac-
tive configuration consisting of the redox pair and the water mol-
ecule. Hence, involvement of the molecule of acetonitrile might
have a significant effect on the dynamical features of the solvent
shell, and subsequently on the dynamics of the coupled compres-
sive modes coincident with the reaction coordinate for H-transfer.
In addition, we have reported recently that ascorbate is in-
In principle, the observed phenomena can be explained using
the Bell tunnelling correction model.1 However, an explanation
that involves dynamical aspects in optimizing the inter-nuclear
distance for nuclear tunnelling, and the full range of the isotopic
,3
differences between the Arrhenius pre-factors from A
and the temperature independent KIEs that have been observed
in numerous enzymatic reactions) to /A
ꢂ 1 values and
temperature dependent KIEs has been supplied by invoking a
/A
H D
ꢂ 1
(
A
D
H
Marcus-like tunnelling model.3
–15,55–57
Initially, within the
framework of the model, hydrogen tunnelling is considered to be
environmentally coupled with the protein dynamics, that is, it is
assumed that protein/substrate motions can modulate hydrogen
3
–15,55
58
tunnelling.
Furthermore, Johannissen et al. showed that a
modification of the model can be also applied to a system that does
1
4
not involve a protein environment, whilst Pudney et al. have
emphasized that compressive modes (the donor–acceptor distance
fluctuations coincident and coupled with the reaction coordinate,
6
1
volved in hydrogen tunnelling in the reaction with the 2,2,6,6-
‘
promoting modes’) may be a feature of both solution and enzyme
systems and may also lead to an increase of the KIE when the do-
nor–acceptor distance decreases (due to the increase of the force
constant for a compressive mode).
3
In our view, an initial elucidation of the observed solvent-in-
duced tunnelling in the reaction could also be traced within the
above framework of the Marcus-like tunnelling model. As men-
tioned, the absence of catalysis by the acetate ion and the observed
proton inventory pointed to a transition configuration consisting of
the redox partners and a molecule of water, suitable for the unidi-
rectional PCET process. Some insight into the dynamical aspects
may come from a consideration of the factors that could influence
the initial donor/acceptor separation in the configuration. It seems
reasonable to assume the donor–acceptor distance (between the
ascorbate 2-OH and the water molecule in the structure) will,
among others, depend on the solvent polarity (as the separation
of the negative charged redox partners due to repulsive interac-
tions should also depend on the solvent polarity). The expectation
is that the initial donor/acceptor separation would increase in the
transition structure with decreasing solvent polarity. Indeed, the
1
-
0
1
.0031
0.0032
0.0033
0.0034
0.0035
0.0036
-
1
(1/T ) / K
Figure 1. Arrhenius plots for the reaction of ascorbate and hexacyanoferrate(III)
ions in MeCN–H O (d) and MeCN–D O (s) 1:1 v/v mixtures at ionic strength
O (j) and EtOD–D O (h) 1:1 v/v mixtures at ionic
observed isotope effects on the Arrhenius pre-factor, A
in the more polar water–acetonitrile solvent is greater than A
= 0.065 in the less polar water–ethanol solvent, and A
H D
/A = 0.49
2
2
H
/
/
I = 0.0023, and in EtOH–H
strength I = 0.0015.
2
2
A
D
H

1
760
A. Karkovi ´c et al. / Tetrahedron Letters 52 (2011) 1757–1761
tetramethylpiperidin-1-oxyl (TEMPO) radical in water, possibly
suggesting that ascorbate also uses tunnelling in the PCET pro-
cesses in biological systems. The present results support the infer-
ence in so far as the mixed solvents used in this Letter could be
more similar to a biological environment than ‘pure’ water alone.
37. Jakobuši c´ Brala, C.; Pilepi c´ , V.; Sajenko, I.; Karkovi c´ , A.; Urši c´ , S. submitted to J.
Phys. Chem. A..
38. Kinetic and spectroscopic evidence and all other relevant details are given in
the Supplementary data.
39. Stern, M. J.; Weston, R. E. J. J. Chem. Phys. 1974, 60, 2808–2814.
4
4
4
0. Kohen, A. Prog. React. Kinet. Mech. 2003, 28, 119–156.
1. Romesberg, F. E.; Schowen, R. L. Adv. Phys. Org. Chem. 2004, 39, 27–77.
2. Bielski, B. H. J.; Allen, O.; Schwarz, H. A. J. Am. Chem. Soc. 1981, 103, 3516–3518.
and references cited therein.
Acknowledgement
4
3. Acetone absorbs strongly in the range we used to determine the ascorbate
concentration, which prevents application of the method described (see in
We thank the Croatian Ministry of Science and Technology for
support (Contract 006-0063082-0354).
38
Supplementary data ) for the precise determination of ascorbate. The rate
constants and KIEs were therefore based on the stoichiometric concentration
of ascorbate in the water-acetone solvent used.
4
4. Uncorrected values. The values corrected for the light water content in the
Supplementary data
38
reaction solution are 3–5% greater (see in Supplementary data ).
4
4
4
4
5. Decornez, H.; Hammes-Schiffer, S. J. Phys. Chem. A 2000, 104, 9370–9384.
6. Kresge, A. J. Pure Appl. Chem. 1965, 8, 243–258.
7. Schowen, R. L. Prog. Phys. Org. Chem. 1972, 9, 275–332.
8. Kresge, A. J.; O’Ferrall, R. A.; Powell, M. F. In Isotopes in Organic Chemistry;
Buncel, E., Lee, L. L., Eds.; Elsevier: New York, 1987; Vol. 7,. Chapter 4.
Supplementary data (experimental data: Kinetics and UV spec-
tra) associated with this article can be found, in the online version,
at doi:10.1016/j.tetlet.2011.01.142.
49. All the points in the dependence of kobs versus xD2O related to the H
2
O–D
mixtures deviate from the linear plot significantly. The deviations from
linearity are always for 3–6 values of (standard deviation of the mean values
2
O
38
References and notes
r
of the rate constants measured) in excess. At present, there is a lack of an
appropriate model to assess fractionation factors in this case.
1
2
3
.
.
.
Bell, R. P. The Tunnel Effect in Chemistry; Chapman & Hall: London, 1980.
Kohen, A.; Klinman, J. P. Acc. Chem. Res. 1998, 31, 397–404.
Kohen, A. In Isotope Effects in Chemistry and Biology; Kohen, A., Limbach, H. H.,
Eds.; Taylor & Francis, CRC Press: New York, 2006; pp 744–764.
Kohen, A. In Hydrogen-Transfer Reactions; Hynes, J. T., Klinman, J. P., Limbach,
H.-H., Schowen, R. L., Eds.; Wiley-VCH: Weinheim, 2007; Vol. 4, pp 1311–1339.
Sen, A.; Kohen, A. In Quantum Effects in Enzyme Kinetics; Scrutton, N. S.,
Allemann, R. K., Eds.; Royal Society of Chemistry, 2009; pp 165–181.
Klinman, J. P. Biochim. Biophys. Acta 2006, 1757, 981–987.
50. The rate constant for the reaction is changed (for the reasons we have
+
discussed elsewhere on the addition of the cation/salt, Na (NaCl) in this
instance, in water–acetonitrile (see Table 1). However, the observed rate
constant remained the same when Na-acetate (instead of NaCl) was present in
the reaction solution.
4
5
.
.
51. Carra, C.; Iordanova, N.; Hammes-Schiffer, S. J. Am. Chem. Soc. 2003, 125,
10429–10436.
52. Sjödin, M.; Irebo, T.; Utas, J. E.; Lind, J.; Merényi, G.; Åkermark, B.;
Hammarström, L. J. Am. Chem. Soc. 2006, 128, 13076–13083.
6
7
.
.
Meyer, M. P.; Tomchick, D. R.; Klinman, J. P. Proc. Natl. Acad. Sci. U.S.A. 2008, 105,
1
146–1151.
53. The following reasons support the proposal: (i) Any involvement of acetate as a
proton acceptor in the PCET process must lead to a change of the observed rate
constants in the reaction. The observed rate constants are, however, exactly the
same in the presence of 0.3 M acetate and in the absence of acetate ions. (ii) It
would be highly unexpected that acetate (being an anion, yet is 6.5 pKa units
more basic than water) does not compete with the water for transferring the
proton in the reaction if the proton transfers to the bulk water. (iii) The
observed proton inventory indicates the water molecule should be part of the
activated complex in the reaction. (iv) The water molecule acting as a proton
acceptor may therefore be found between the redox partners. This
8
9
1
.
.
Klinman, J. P. Chem. Phys. Lett. 2009, 471, 179–193.
Nagel, Z. D.; Klinman, J. P. Nat. Chem. Biol. 2009, 5, 543–550.
0. Sutcliffe, M. J.; Masgrau, L.; Roujenikova, A.; Johannissen, L. O.; Hothi, P.;
Basran, J.; Ranaghan, K. E.; Mulholland, A. J.; Leys, D.; Scrutton, N. S. Philos.
Trans. R. Soc. London, Ser. B 2006, 361, 1375–1386.
1
1
1
1
1
1. Johannissen, L. O.; Hay, S.; Scrutton, N. S.; Sutcliffe, M. J. J. Phys. Chem. B 2007,
1
11, 2631–2638.
2. Hay, S.; Sutcliffe, M. J.; Scrutton, N. S. Proc. Natl. Acad. Sci. U.S.A. 2007, 104, 507–
12.
3. Pang, J.; Hay, S.; Scrutton, N. S.; Sutcliffe, M. J. J. Am. Chem. Soc. 2008, 130, 7092–
097.
5
configuration can be appropriate for the unidirectional PCET. The advantage
of the unidirectional2
1,51,52
PCET arises from a lowering of the activation barrier
7
4. Pudney, C. R.; Johannissen, L. O.; Sutcliffe, M. J.; Hay, S.; Scrutton, N. S. J. Am.
Chem. Soc. 2010, 132, 11329–11335.
5. Hay, S.; Johannissen, L. O.; Sutcliffe, M. J.; Scrutton, N. S. Biophys. J. 2010, 98,
due to smaller reorganizational energy k, as compared with a bidirectional
PCET;2
1
,
5
1
,
5
2
the activated complex could resemble a solvent-shared ion pair
5
4
(taking into regard repulsion between negatively charged redox partners). The
ascorbate 2-OH is necessary oriented towards the hexacyanoferrate ion as any
other orientation would lead to an unfavourable bidirectional PCET. (v) A
consequence of the (potential) involvement of acetate ions between the redox
partners (instead of the water molecule) would be greater separation between
the redox partners which should diminish the (electronic) couplings between
them, and thus decrease the rate of the PCET reaction, since the rate of PCET
reactions depend strongly on the couplings. These points could explain why
acetate does not compete with water for the ascorbate 2-OH proton in the
PCET process.
1
21–128.
1
1
6. Hammes-Schiffer, S. Acc. Chem. Res. 2006, 39, 93–100.
7. Wu, A.; Mader, E. A.; Datta, A.; Hrovat, D. A.; Borden, W. T.; Mayer, J. M. J. Am.
Chem. Soc. 2009, 131, 11985–11987.
1
8. (a) Huynh, M. H. H.; Meyer, T. J. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 13138–
1
3
3141; (b) Huynh, M. H. H.; White, P. S.; Meyer, T. J. Angew. Chem., Int. Ed. 2000,
9, 4101–4104.
1
9. Schreiner, P. R.; Reisenauer, H. P.; Pickard, F. C.; Simmonett, A. C.; Allen, W. D.;
Matyus, E.; Csaszar, A. G. Nature 2008, 453, 906–909.
2
2
2
2
0. Cukier, R. I.; Nocera, D. G. Annu. Rev. Phys. Chem. 1998, 49, 337–369.
1. Mayer, J. M. Annu. Rev. Phys. Chem. 2004, 55, 363–390.
2. Huynh, M. H. H.; Meyer, T. J. Chem. Rev. 2007, 107, 5004–5064.
3. Hammarström, L.; Styring, S. Philos. Trans. R. Soc. London, Ser. B 2008, 363,
54. Conway, B. E. Ionic Hydration in Chemistry and Biophysics; Elsevier Scientific
Publishing: Amsterdam, 1981.
55. Knapp et al.56 adapted the model presented by Kuznetsov and Ulstrup57 giving
the expression:
1
283–1291.
2
2
4. Reece, S. Y.; Nocera, D. G. Annu. Rev. Biochem. 2009, 78, 673–699.
5. Concepcion, J. J.; Jurss, J. W.; Brennaman, M. K.; Hoertz, P. G.; Patrocinio, A. O.
T.; Iha, N. Y. M.; Templeton, J. T.; Meyer, T. J. Acc. Chem. Res. 2009, 42, 1954–
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
X
X
1
2
ꢃ
2
2
ꢁðD
G
þE ibþkÞ =ð4kRTÞ
3
v
k
tunn
¼
P
t
jVel
j
4p =kRT ꢀh exp
2
p
t
w
1
965.
2
2
6. Hammes-Schiffer, S. Acc. Chem. Res. 2009, 42, 1881–1889.
ꢄ ðF:C:termÞt;w
7. Magnuson, A.; Anderlund, M.; Johansson, O.; Lindblad, P.; Lomoth, R.; Polivka,
T.; Ott, S.; Stensjö, K.; Styring, S.; Sundström, V.; Hammarström, L. Acc. Chem.
Res. 2009, 42, 1899–1909.
8. Gagliardi, C. J.; Westlake, B. C.; Kent, C. A.; Paul, J. J.; Papanikolas, J. M.; Meyer,
T. J. Coord. Chem. Rev. 2010, 254, 2459–2471.
where ktunn is the rate constant related to tunnelling. The first term
describes electronic coupling and is isotope-independent; the second is
2
an environmental energy term relating the reorganizational energy, k,
O
and the driving force of the reaction,
D
r
G . Evib is the vibrational energy
29. Costentin, C.; Robert, M.; Saveant, J. M. Acc. Chem. Res. 2010, 43, 1019–1029.
30. Hsieh, C. C.; Jiang, C. M.; Chou, P. T. Acc. Chem. Res. 2010, 43, 1364–1374.
31. Warren, J. J.; Mayer, J. M. Proc. Natl. Acad. Sci. U.S.A. 2010, 107, 5282–5287.
32. Hazra, A.; Soudackov, A. V.; Hammes-Schiffer, S. J. Phys. Chem. C 2010, 114,
difference between product and reactant. The contribution of hydrogen
stretching to the rate due to a vibration-level specific Frank–Condon
nuclear overlap along the hydrogen coordinate is contained in the F.C.
term, which determines the tunnelling probability of hydrogen (⁄, R
and T are Planck’s constant divided by 2p, the gas constant and absolute
temperature, respectively). This model could explain the whole range
1
2319–12332.
3
3
3. Zhang, W.; Rosendahl, S. M.; Burgess, I. J. J. Phys. Chem. C 2010, 114, 2738–2745.
4. Venkataraman, C.; Soudackov, A. V.; Hammes-Schiffer, S. J. Phys. Chem. C 2010,
114, 487–496.
of the phenomena observed, that is, the full range of the observed
3
3
5. Hammes-Schiffer, S.; Soudackov, A. V. J. Phys. Chem. B 2008, 112, 14108–14123.
and references cited therein.
6. Kagayama, N.; Sekiguchi, M.; Inada, Y.; Takagi, H. D.; Funahashi, S. Inorg. Chem.
isotopic differences between the Arrhenius pre-factors from A
to A /A
ꢂ 1 as well as the temperature independent KIEs. A very
important feature of this model takes into account motions along the
H
/A
D
ꢂ 1
D
H
1
994, 33, 1881–1885. and references cited therein.

A. Karkovi ´c et al. / Tetrahedron Letters 52 (2011) 1757–1761
1761
59. Edwards et al.60 have pointed out that the temperature dependence of KIEs
H-transfer coordinate (compressive modes, promoting modes) that could
lead to a distance sampling (gating), and consequently to an increased
rate when the initial distance between the H donor and acceptor is too
long to support efficient tunnelling. The basis for this expectation is the
strong distance dependence of proton coupling, hence, it is assumed that
environmental oscillations can alter the donor–acceptor distance and the
could increase as the frequency of the compressive mode decreases (this
frequency could decrease as the donor–acceptor distance increases7
,8,14
)
which is expected here, for instance, in the case of the water–1,4-dioxane
system, where a stronger temperature dependence of the KIE was observed
than in the case of water–acetonitrile.
6
6
0. Edwards, S. J.; Soudackov, A. V.; Hammes-Schiffer, S. J. Phys. Chem. A 2009, 113,
hydrogen overlap.
2
117–2126.
1. Sajenko, I.; Pilepic, V.; Jakobušic Brala, C.; Uršic, S. J. Phys. Chem. A 2010, 114,
423–3430.
5
5
5
6. Knapp, M. J.; Rickert, K.; Klinman, J. P. J. Am. Chem. Soc. 2002, 124, 3865–3874.
7. Kuznetsov, A. M.; Ulstrup, J. Can. J. Chem. 1999, 77, 1085–1096.
8. Johannissen, L. O.; Irebo, T.; Sjödin, M.; Johansson, O.; Hammarström, L. J. Phys.
Chem. B 2009, 113, 16214–16225.
´
´
´
3