Multi PCET in symmetrically substituted benzimidazoles

Article abstract of DOI:10.1039/d1sc03782j

Proton-coupled electron transfer (PCET) reactions depend on the hydrogen-bond connectivity between sites of proton donors and acceptors. The 2-(2′-hydroxyphenyl) benzimidazole (BIP) based systems, which mimic the natural TyrZ-His190 pair of Photosystem II

Full text of DOI:10.1039/d1sc03782j

Chemical
Science
View Article Online
View Journal
EDGE ARTICLE
Multi PCET in symmetrically substituted
benzimidazoles†
Cite this: DOI: 10.1039/d1sc03782j
a
b
a
a
All publication charges for this article Emmanuel Odella,
Maxim Secor,
Mackenna Elliott,
Thomas L. Groy,
have been paid for by the Royal Society
a
Sharon Hammes-Schiffer *b and Ana L. Moore
a
Thomas A. Moore,
*
of Chemistry
Proton-coupled electron transfer (PCET) reactions depend on the hydrogen-bond connectivity between
0
sites of proton donors and acceptors. The 2-(2 -hydroxyphenyl) benzimidazole (BIP) based systems,
Z
which mimic the natural Tyr -His190 pair of Photosystem II, have been useful for understanding the
associated PCET process triggered by one-electron oxidation of the phenol. Substitution of the
benzimidazole by an appropriate terminal proton acceptor (TPA) group allows for two-proton
translocations. However, the prototropic properties of substituted benzimidazole rings and rotation
around the bond linking the phenol and the benzimidazole can lead to isomers that interrupt the
intramolecular hydrogen-bonded network and thereby prevent a second proton translocation. Herein,
a strategic symmetrization of a benzimidazole based system with two identical TPAs yields an
uninterrupted network of intramolecular hydrogen bonds regardless of the isomeric form. NMR data
confirms the presence of
a single isomeric form in the disubstituted system but not in the
monosubstituted system in certain solvents. Infrared spectroelectrochemistry demonstrates
a two-proton transfer process associated with the oxidation of the phenol occurring at a lower redox
potential in the disubstituted system relative to its monosubstituted analogue. Computational studies
support these findings and show that the disubstituted system stabilizes the oxidized two-proton transfer
product through the formation of a bifurcated hydrogen bond. Considering the prototropic properties of
the benzimidazole heterocycle in the context of multiple PCET will improve the next generation of
novel, bioinspired constructs built by concatenated units of benzimidazoles, thus allowing proton
translocations at nanoscale length.
Received 10th July 2021
Accepted 21st August 2021
DOI: 10.1039/d1sc03782j
rsc.li/chemical-science
a collection of articial models capable of mimicking the
Introduction
1
–3
crucial role of the Tyr
Z
-His190 pair.
Z
Nature employs a tyrosine Z-histidine (Tyr -His190) pair as
A benzimidazole attached covalently to a phenol derivative
a redox relay that couples proton transfer to the redox process (BIP 1, Fig. 1A) is a versatile platform that closely emulates
between the reaction-center P680 and the water oxidizing several aspects of the Tyr -His190 pair and its unique PCET
catalyst in photosystem II (PSII). The high efficiency in this event (the benzimidazole models His190 and the phenol
Z
4
–6
redox process relies on the coupled movement of both protons models Tyr
). Variants of this model have been reported in
Z
7
–9
and electrons, which imparts an energetic advantage over the which BIP is covalently linked to a high potential porphyrin.
single electron transfer process by lowering the activation In our recent work, a low or barrierless pathway for PCET with
1
barrier. This reaction, better known as proton-coupled electron the concerted process occurring in ꢀ100 fs has been identied
10
transfer (PCET), has been experimentally and theoretically using 2D electronic vibrational spectroscopy.
studied for years, inspiring many researchers to design
Upon electrooxidation of 1, a reversible one-electron coupled
2,11
with one-proton transfer (E1PT) occurs.
This proton
displacement between the phenol and the proximal benzimid-
azole nitrogen takes place in a well-dened and strong intra-
molecular hydrogen bond connecting both groups. An
expansion in the proton translocation distance was successfully
achieved by proton wires constructed by covalently intercalating
a series of benzimidazole units between the oxidation center
a
School of Molecular Sciences, Arizona State University, Tempe, Arizona 85287-1604,
USA. E-mail: amoore@asu.edu
b
Department of Chemistry, Yale University, New Haven, Connecticut 06520-8107, USA.
E-mail: sharon.hammes-schiffer@yale.edu
†
Electronic supplementary information (ESI) available: Materials and methods,
synthesis and structural characterization, nuclear magnetic resonance data,
crystal structure and X-ray data, and computational methods. CCDC 2094827. electron transfer from the phenol is coupled to up to four
For ESI and crystallographic data in CIF or other electronic format see DOI:
0.1039/d1sc03782j
(the phenol) and the terminal proton acceptor (TPA), in which
proton translocations that span a total distance of ꢀ16 A˚ .
12
1
©
2021 The Author(s). Published by the Royal Society of Chemistry
Chem. Sci.

View Article Online
Chemical Science
Edge Article
trans-enol forms over the intramolecularly hydrogen bonded
15,18,20,21
closed cis-enol.
Conformational isomers are energetically accessible to the
bioinspired constructs that signicantly hinder the PCET
process. Specically, rotamers associated with the bond
between the phenol and benzimidazole moieties ip the system
over and remove the possibility of multi-proton translocation.
Indeed, the annular 1,3-tautomerization facilitates the rotation
around this bond by stabilizing the system with the formation
of the hydrogen bond between the phenol and the lone pair of
the other nitrogen atom of the benzimidazole. Therefore, these
structures are interconvertible via tautomerism followed by
rotation along the bond connecting the phenol and benzimid-
azole moieties (Fig. 1B).
An example of multiple PCET (MPCET) has been demon-
strated by placing at the 7-position of the benzimidazole
22
a substituted phenylimine as the TPA (2, Fig. 1A). However, in
this system, only one of the isomers contains the necessary,
uninterrupted series of intramolecular hydrogen bonds for
facilitating a one-electron, two-proton transfer (E2PT) process.
Furthermore, this tautomerization/rotation process creates
potential interruptions in proton wires of arbitrary length by
Fig. 1 (A) Molecular structures of compounds 1–3. (B) 1,3-Annular
tautomerization/rotation process in BIP monosubstituted at 7-position converting EnPT (n represents the number of protons, n > 1)
(
top) with a terminal proton acceptor (TPA) and disubstituted at 4,7 systems to interrupted isomers so that only EnPT processes with
positions (bottom) with an identical TPA.
n less than the number of protons are possible. One strategy to
eliminate the effects of tautomerization/rotation is by alkylating
the imidazole nitrogen. However, this removes the NH group,
which is a participant in the second proton transfer process.
Another approach is a symmetrization of the benzimidazole
moiety by adding the same substitution at the 4 and 7 positions,
as illustrated in Fig. 1B. In this case, the annular 1,3-tautome-
rization is not eliminated, but the resulting species from the
proton exchange have equivalent structures with identical
intramolecular hydrogen bonds. This removes the issues
caused by tautomerization/rotation processes by making every
species capable of forming the two intramolecular hydrogen
bonds required for the E2PT process. The incorporation of
a second, identical TPA on BIP (3, Fig. 1A) yielding a system
invariant to this tautomerization/rotation is described here.
The success of these biologically inspired proton wires relies
on the uninterrupted sequence of intramolecular hydrogen
bonds, which creates a network over which protons move
reversibly between acidic and basic sites within the molecule via
a Grotthuss-type mechanism. Furthermore, a clear example of
the consequences associated with the disruption of a hydrogen-
bonded network has been described in our recent work. The
absence of a crucial hydrogen bond in the network causes
a pronounced change in the overall reversibility and thermo-
13
dynamics of the PCET process. The discontinuity of the
hydrogen-bonded network, and its potential consequences on
PCET, is also evidenced in a less explored feature—the intrinsic
prototropic property (annular 1,3 tautomerism) of benzimid-
14–16
azole.
BIP substituted at any of the remaining benzimid-
Results and discussion
Design and synthesis
azole positions (4 to 7 positions, see numbering in 1, Fig. 1A)
consists of a mixture of isomers, whose presence can be clearly
1
7–19
detected by NMR spectroscopy under certain conditions.
It
The synthesis and structural characterization of molecules 1
0
22,23
is known that systems based on 2-(2 -hydroxyphenyl)benz-
imidazole (BIP without tert-butyl groups) may exist in three
different enol conformers and a keto tautomeric struc-
and 2 have been previously reported.
Molecule 3 was ob-
tained following a seven-step synthetic strategy shown in
Scheme 1. The synthesis begins with cyanation of the
commercially available 4,7-dibromo-2,1,3-benzothiadiazole (A).
Subsequently, sulfur extrusion by a reductive reaction of the
dicyano derivative (B) employing the heterogeneous catalytic
15,18,20,21
ture.
Among them, the closed cis-enol species, featuring
an intramolecular hydrogen bond between the phenol and the
lone pair of the benzimidazole nitrogen, is the most stable
conformation in the ground state in non-polar or low polarity
solvents. However, the dielectric constant and hydrogen
bonding ability of the medium play an important role in the
population of the closed cis-enol conformer relative to the other
species in solution. Solvents acting as hydrogen bond donors or
acceptors stabilize preferentially the keto, open cis-enol, and
NaBH
dicyanobenzene (C).
the resulting aromatic diamino compound and 3,5-di-tert-butyl-
-hydroxybenzaldehyde in nitrobenzene results in the BIP-4,7
disubstituted platform (D). Alkaline hydrolysis of D produces
the dicarboxylic acid derivative E. Reduction of E using LiAlH
4
/CoCl
2
$6H
2
O
system
gives
1,2-diamino-3,6-
24,25
The condensation reaction between
2
4
affords the diol (F), which is then oxidized to the dialdehyde (G)
Chem. Sci.
© 2021 The Author(s). Published by the Royal Society of Chemistry

View Article Online
Edge Article
Chemical Science
Scheme 1 Synthetic strategy detailing reaction conditions, yields, and intermediate molecules for preparation of 3.
with activated MnO . The nal step includes the formation of shis in CDCl , the hydrogen bond between the phenolic
2
3
the imine linkage using compound G and two equivalents of p- proton and the nitrogen lone pair of the benzimidazole in 3 is
22,26
anisidine (4-methoxyaniline) following a known procedure.
slightly weaker than that for 2, whereas the hydrogen bond
The complete synthetic procedure and structural characteriza- between the benzimidazole NH and the imine N is slightly
tion are provided in the ESI section.†
stronger in the case of the 4,7 disubstituted molecule (3). The
Structural characterization
Crystallographic data is useful to establish how the hydrogen-
bonded network is extended within the molecule. A nearly
planar molecular framework across the phenol, the benzimid-
azole proton relay, and one of the TPA branches in 3 is observed,
with dihedral angles between the phenol and benzimidazole
ꢁ
moieties of 1.3 , and between the benzimidazole and the azo-
ꢁ
˚
methine group of 3.6 (Fig. 2). The fairly short O/N (2.61 A) and
˚
N/N (2.83 A) distances are indicative of strong hydrogen bonds
between the phenolic proton and the nitrogen lone pair of the
benzimidazole, as well as the benzimidazole NH and the imine
nitrogen of the TPA at the 7-postion. It is worth comparing the
O/N and N/N distances found in the X-ray data of 3 with its
monosubstituted analogue 2 (Fig. 2 and Table 1).
For the disubstituted molecule 3, the O/N distance is
˚
˚
comparable to that obtained for 2 (2.61 A vs. 2.57 A). The N/N
distance for 3 is also comparable to that of 2. Geometry opti-
mizations were performed with density functional theory (DFT)
2
7–31
using the B3LYP-D3(BJ) functional
set
and a 6-31G** basis
32–35
(see details in the ESI†). The resulting optimized struc-
tures support the crystallographic data. The agreement in
structure between experiment and computation is highlighted
in selected bonds and angles (see Table 1).
The existence of strong intramolecular hydrogen bonds and
the differences noticed between 2 and 3 are well-reected in
1
their H NMR spectra in dry non-polar, non-protic solvents such
as CDCl and CD Cl . Resonances of benzimidazole NH (dNH)
and phenolic (dOH) protons above 11.00 ppm and 13.00 ppm,
3
2
2
Fig. 2 Crystal structures of 2 (top) and 3 (bottom). Carbon atoms are
shown in gray, oxygen atoms in red, nitrogen atoms in light violet, and
hydrogen atoms in white. Thermal ellipsoids are drawn at the 50%
respectively, are characteristic in BIP derivatives featuring
22,36
strong hydrogen-bonded networks.
Based on the chemical probability level. The crystal structure of 2 is adapted from ref. 22.
©
2021 The Author(s). Published by the Royal Society of Chemistry
Chem. Sci.

View Article Online
Chemical Science
Edge Article
Table 1 Structural and electrochemical data for 2 and 3
2
3
a
b
b
a
b
b
O/N distance ( A˚ )
N/N distance ( A˚ )
Phenol-benzimidazole torsional angle ( )
2.57 , 2.57
2.61 , 2.58
a
a
2.92 , 2.84
2.83 , 2.83
ꢁ
a
b
b
a
b
b
3.4 , 2.4
1.9 , 0.1
1.3 , 3.3
3.6 , 0.6
ꢁ
a
a
Benzimidazole-imine linkage torsional angle ( )
c
d
NH (ppm)
OH (ppm)
11.84
13.30
11.97
13.14
0.85
0.87
64
c
d
d
e
Calculated E1/2 (V vs. SCE)
Experimental E1/2 (V vs. SCE)
0.95
f
0.95
130
0.97
g
DE
/i
p
g(mV)
i
c
a
0.91
a
b
c 1
d
Determined by X-ray diffraction measurements. Determined by DFT calculations. H NMR data in CDCl
3
.
Each computed redox potential
assumes the maximum number of intramolecular proton transfers. The redox potentials corresponding to the intermediate proton transfer
e
process (E1PT) are given in Table S13. The potential for 2 in CH Cl was used as the reference for all calculated potentials associated with an
2 2
f
E2PT product, and therefore it agrees with the experimental value by construction. The potential of the pseudoreference electrode was
determined using the ferrocenium/ferrocene redox couple as an internal standard and adjusting to the saturated calomel electrode (SCE) scale
with E1/2 taken to be 0.46 V vs. SCE in CH Cl ). Measured at 100 mV s .
2 2
g
ꢂ1
(
2 2
strength of these intramolecular hydrogen bonds, qualitatively CD Cl (Fig. S15†) certainly allow the distinction of each TPA
described by the dNH and dOH values, is consistent with the X-ray branch and are in agreement with the conformation adopted by
crystallographic distances found for both compounds (see Fig. 2 the molecule in the crystallographic unit cell (see Fig. 2). The
and Table 1). Interestingly, the X-ray data show that the p-OCH
phenylimine substituent at the 4-position is rotated 180 rela- bond (H in Fig. 4B) interacts through space with the aromatic
azomethine proton of the imine linkage exhibiting a hydrogen
3
ꢁ
a
tive to the same substituent at the 7-position (see Fig. 2, TPA protons of both the benzimidazole (H
hydrogen-bonded with the benzimidazole NH), and this ring (H ), whereas the azomethine proton of the TPA at the 4-
conformation remains in both CDCl and CD Cl solutions.
6 3
) and the p-OCH phenyl
c
3
2
2
Analysis of computationally obtained free energies reveal that
this conformation represents ꢀ72% of the relevant, thermody-
namically accessible isomers in the neutral state (vide infra).
The proportion of isomers in solution originated by the annular
1,3-tautomerism and rotation of the bond linking the phenol
and benzimidazole moieties is deeply affected by both the
dielectric constant of the solvent and its capability to interact
specically by hydrogen bonding with the molecule, among
15,18,20,21
other factors.
In the case of 2, the isomer featuring the
fully hydrogen-bonded network (see Fig. 1A) is the most stable
1
conformation in CDCl3 solution. Only one set of H NMR
22
resonances lines is accordingly detected. However, the switch
of CDCl for acetone-d alters the equilibrium between isomers,
3
6
and the presence of a minor population (isomeric ratio: 1 : 0.20)
stabilized by specic solvent–solute interactions is noticed
(Fig. 3). The formation of intermolecular hydrogen bond inter-
actions between the solvent (hydrogen bond acceptor) and
either the free NH or OH (hydrogen bond donors) in the open
cis-enol form in 2 could decrease the rate of inter-conversion
between the two tautomeric species when acetone-d is used
6
as solvent. The opposite is shown for 3, where only one species
in solution is detected independently of the nature of the
solvent, as a consequence of the additional substitution at the 4-
position with an identical TPA group (Fig. 3).
1
3 2 2 6
HNMR spectra of 3 in CDCl , CD Cl and acetone-d show
that both TPA groups are magnetically nonequivalent (see full
structural characterization in the ESI†). The phenol rotation
and annular 1,3-tautomerism are slow on the NMR timescale
because of the strong hydrogen-bonded network along the
1
Fig. 3 H NMR spectra of 2 (purple) and 3 (red) in acetone-d
playing resonances in the downfield (top), aromatic (middle), and
aliphatic (bottom) regions. The resonances of both OCH groups
dOCH ) in 3 appear at different ppm. Signals of the minor isomer of 2
3
(Fig. 4) and are highlighted (*).
6
dis-
3
(
3
molecule. NOESY experiments performed in CDCl
Chem. Sci.
© 2021 The Author(s). Published by the Royal Society of Chemistry

View Article Online
Edge Article
Chemical Science
3
Fig. 4 (A) NOESY of 3 in CDCl showing the downfield/aromatic
regions of the spectrum. (B) Expansion of the selected area (dotted
rectangle) in A. The arrows shown in the molecular structure describe
Fig. 5 Rotamers, tautomers, and protonation states of 2 (A) and 3 (B).
There are 11 and 15 possible arrangements of 2 and 3, respectively. The
blue markers indicate possible positions of the proton initially on the
phenol (before oxidation) within the hydrogen-bonded network. The
orange markers indicate possible positions of the proton initially on
benzimidazole (before oxidation) within the hydrogen-bonded
the interactions through the space of the azomethine protons (H
) of both TPA branches (color coded).
a
and
H
b
position (Hb in Fig. 4B) only shows a cross peak with the network.
aromatic proton of the p-OCH phenyl ring (H
0
).
3
c
only 0.02% of 3 is initially in a rotamer state that cannot facil-
itate two-proton translocation. These calculations of the relative
populations of the various states reveal an advantage of disub-
stituted, symmetrized bioinspired proton wires in terms of
decreasing the populations of conformations that are not
conducive to the E2PT process. Although compound 2 is mostly
congured to facilitate E2PT (98.6%), thermally accessible
rotamers could be a much greater obstacle in larger systems
where more than two proton transfers require a consistent
uninterrupted hydrogen-bonded network over multiple benz-
imidazole units. Longer proton wires typically have more
possible rotamers lacking the ability to facilitate further proton
transfer. As demonstrated here, the disubstituted system is
populated 99.98% by rotamers with uninterrupted hydrogen-
bonded networks allowing multiple proton transfers to occur.
Computational isomer analysis
Geometry optimizations and free energy calculations of 2 and 3
were performed at the DFT level of theory using the conductor-
like polarizable continuum model (CPCM)
3
7,38
to represent
CH Cl solvent. Both 2 and 3 have four rotamer species, each
2
2
with various possible protonation states, as shown in Fig. 5A
and B, respectively. All the possibilities shown in these gures
were found to correspond to local minima on the potential
energy surface. The relative free energies for all minima in both
the oxidized and neutral states of 2 and 3 are reported in Tables
S11 and S12,† respectively. The Boltzmann population for each
state was calculated using
ꢂDGi
kBT
e
P
i
¼
(1)
ꢂDGj
kBT
P
e
j
Electrochemical studies
th
where DG
i B
is the relative free energy of the i state, k is the Cyclic voltammetry (CV) was used to gain insight into the
Boltzmann constant, T is the temperature, and the summation thermodynamics of the PCET process. All potentials are given
+
in the denominator is over all possible states.
relative to the ferrocenium/ferrocene (Fc /Fc) redox couple and
In 2, the neutral system prior to oxidation is 98.6% rotamer C are adjusted to the saturated calomel electrode (SCE) scale
with both protons in position 1 and 1.2% rotamer B with both (Table 1). Redox potentials were calculated using DFT and
39
protons in position 1 (see Fig. 5A). In total, about 1.4% of 2 is methods previously benchmarked (see ESI†). The CV of 3 is
initially in a rotamer state that cannot facilitate two-proton shown in Fig. 6 and compared to that of 1, 2 and G, a precursor
translocation. In 3, the neutral system prior to oxidation is of 3 (see Scheme 1).
72.1% rotamer C with both protons in position 1 and 26.9%
Previous studies have shown that 2 exhibits a quasi-
rotamer A with both protons in position 1 (see Fig. 5B). Thus, reversible one-electron oxidation process with a peak-to-peak
©
2021 The Author(s). Published by the Royal Society of Chemistry
Chem. Sci.

View Article Online
Chemical Science
Edge Article
analysis of rotamers and protonation states of 3 suggests that, at
equilibrium aer E2PT, the system is ꢀ97% in rotamer A with
both protons in position 2 (see Fig. 7B). Thus, the oxidized
product of 3 is stabilized by an extra intramolecular hydrogen
bond relative to 2 (see Fig. 7), providing an explanation for the
difference observed in the redox potentials.
A bifurcated hydrogen bond would be formed in compound
3
aer an E2PT process. The new NH bond formed during
proton transfer from the phenoxyl radical to the benzimidazole
nitrogen in 3 now has two hydrogen bond acceptors, i.e., the
oxygen of the phenol group from which it came, and the
unprotonated nitrogen of the TPA at the 4-position. Based on
the DFT geometry optimizations, the N/O and N/N distances
˚
˚
associated with these hydrogen bonds are 2.65 A and 2.86 A,
˚
˚
respectively, for the E1PT product and 2.67 A and 2.86 A,
respectively, for the E2PT product. These hydrogen bonds
stabilize both the E1PT and E2PT products of compound 3.
Bifurcated hydrogen bonds have been studied experimen-
40
tally and theoretically and have been shown to stabilize PCET
in substituted phenol systems designed to mimic the tyrosine
. WE: glassy carbon. Pseudo RE: Ag radical redox system of PSII. The bifurcated hydrogen bond in
Fig. 6 CVs of G (blue), unsubstituted BIP 1 (black), 2 (purple) and 3
(
red). Concentration: 1 mM of the indicated BIPs, 0.1 M TBAPF
6
sup-
41
porting electrolyte in dry CH Cl
2
2
wire (ferrocene as internal reference). CE: Pt wire. Scan rate, 100 mV
compound 3 demonstrates similar character. The DFT calcula-
tions indicate that the E0PT, E1PT, and E2PT processes (i.e.,
oxidation with no proton transfer, one proton transfer, and two
proton transfers, see Table S13†) are all more thermodynami-
cally favorable for 3 than for 2. The calculations also indicate
that the E2PT process is thermodynamically favorable for both 2
and 3 (i.e., the E2PT product is more stable than the E0PT and
E1PT products for both compounds). These computational
results are consistent with the experimental observation that
both compounds undergo E2PT and that 2 (single TPA) has
a higher redox potential than 3 (double TPA). In the case of G,
the two electron-withdrawing aldehyde groups at the 4 and 7
positions cause an anodic shi of the E1/2 compared to 1 (Fig. 6,
ꢂ1
s
.
separation (DE ) of 130 mV in CH Cl , while the chemical
p
2
2
reversibility is evidenced by the high cathodic to anodic peak
ꢂ1 22
current ratio (i /i ¼ 0.97, at a scan rate of 100 mV s ). For 2,
c
a
the experimental midpoint potential (E1/2, taken as the average
of the anodic and cathodic peaks) of the phenoxyl radical/
phenol redox couple is 0.95 V vs. SCE (see Table 1). For this
molecule, the oxidation of the phenol is coupled with the
intramolecular transfer of both protons by an E2PT process,
leading to the protonation of the TPA and formation of the
phenoxyl radical as the oxidized product.
The electrochemical performance of 3 is clearly affected by
the presence of an additional TPA. The DE is reduced by
22
DE1/2 ¼ 210 mV). In this case, the lower pK of the protonated
a
aldehyde group prevents the formation of a local minimum
associated with the E2PT product (see Table S13†).
p
a factor of two, which suggests faster electrode kinetics
compared with 2. The chemical reversibility of 3 is also indi-
Analysis of the computational results provides further
insight into the impact of the bifurcated hydrogen bond and the
second TPA substituent in 3. Comparison of the free energies of
the E2PT product of 3 for rotamer A, which has a bifurcated
hydrogen bond, and rotamer C, which does not, indicates that
cated by the i
c
/i
a
ratio approaching unity (i
c
/i
a
¼ 0.91, at
ꢂ1
100 mV s , see Table 1). Furthermore, an appreciable cathodic
shi is observed (DE1/2 ¼ 80 mV, see Table 1) relative to 2,
indicating that the phenol becomes easier to oxidize when the
BIP has identical TPAs capable of accepting a second proton
when the phenoxyl radical is formed. The computational
the bifurcated hydrogen bond provides a stabilization of
ꢂ1
3
.25 kcal mol (Table S12†). Moreover, the E2PT redox poten-
tial of 3 is 0.17 V lower for rotamer A than rotamer C, mainly due
to the stabilization provided by the bifurcated hydrogen bond in
the oxidized species. Comparison of the E2PT redox potentials
of rotamer C for 2 and 3 indicates that the addition of the
second TPA increases the redox potential by 0.03 V. Thus, the
effect of the bifurcated hydrogen bond is much larger than the
effect of the substituent on the E2PT redox potential.
Infrared spectroelectrochemistry studies
Fig.
8
displays the experimental infrared spectroelec-
Fig. 7 (A) Oxidized E2PT product of 2 lacking a stabilizing, bifurcated
hydrogen bond. (B) Oxidized E2PT product of 3 containing a stabi-
lizing, bifurcated hydrogen bond.
trochemistry (IRSEC) spectra of 3 recorded under oxidative
conditions generating increasing amounts of the phenoxyl
Chem. Sci.
© 2021 The Author(s). Published by the Royal Society of Chemistry

View Article Online
Edge Article
Chemical Science
ꢂ1
ꢂ1
Fig. 8 IRSEC spectra of 3 recorded under electro-oxidative conditions in (A) the middle (1700–1450 cm ) and (B) the high (3500–3200 cm
)
2 2 6
frequency regions. Solvent: dry CH Cl , 0.1 M TBAPF . Characteristic bands displaying changes under polarization are indicated with upward and
downward grey arrows.
ꢂ1
22
radical from the neutral phenol species. DFT normal-mode frequency of this mode is situated at 3394 cm for 2; the
analysis of 3 has been employed for the assignment of slight shi of the nNH is in agreement with the downeld shi of
1
selected vibrational modes (Fig. S17 and Table S14†). Substan- the dNH observed in the H NMR (see Table 1). Upon phenol
ꢂ1
tial changes in the absorbance and band positions associated oxidation, the band at 3390 cm decreases and a new band at
ꢂ1
with the oxidation are indicated with arrows. It is interesting to 3344 cm arises, which is assigned to the nNH of the proximal
point out the existence of well-dened isosbestic points across benzimidazole NH. In the case of the monosubstituted
the spectra, suggesting that the neutral species are being analogue 2, the nNH in the oxidized species is centered at higher
ꢂ1 22
progressively converted into the oxidized species. The band at frequency (3360 cm ), suggesting that the decrease in the nNH
ꢂ1
1
614 cm in the neutral species is assigned to the stretching observed for oxidized 3 is most likely due to the bifurcated
C]N of the imine linkage (Fig. 8A), with a similar frequency of hydrogen bond.
22
that observed for the monosubstituted 2. The decrease in the
intensity of this band upon electrooxidation indicates that the
neutral species of 3 is being converted into its oxidized form.
Conclusions
Some of the new bands shown in the spectra are characteristic The effect on PCET due to prototropic tautomerism of
of vibrational modes affected by protonation aer PCET. In benzimidazole-phenol based systems has been investigated.
ꢂ1
particular, the new band at 1651 cm corresponds to the large Substitution of the benzimidazole core with two identical TPAs
component of the protonated imine group stretching, which is removes the existence of isomeric species possessing inter-
a distinctive IR marker displayed by other BIP derivatives rupted hydrogen-bonded networks that prevent the system from
12,22,42
bearing a substituted imine group as the TPA.
This is two intramolecular proton transfers corresponding to an E2PT
evidence that an E2PT product is being formed upon phenol process. Incorporation of the second TPA in the bioinspired
oxidation. Computational spectra (Fig. S17†) corroborate construct not only ensures an uninterrupted hydrogen-bonded
protonation of the TPA in the oxidized species. Specically, the network but also causes a cathodic shi of the phenoxyl
ꢂ1
peak at 1651 cm appears in the simulated IRSEC of the E2PT radical/phenol redox couple relative to its monosubstituted
product but does not appear in the simulated IRSEC of the E1PT analogue. The formation of a bifurcated hydrogen bond, only
product (Fig. S17B and A,† respectively), conrming that the possible in the disubstituted molecule, contributes to the rela-
experimental IRSEC corresponds to the E2PT product. tive stabilization of the E2PT product and explains the decrease
ꢂ1
Computed IRSEC spectra also show a band at 1593 cm
in redox potential. The current ndings, which combine theo-
assigned to a breathing mode of the TPA, which includes retical and experimental approaches, are crucial for future
a bending motion of the protonated azomethine. This band designs of proton wires having multiple translocating protons
ꢂ1
corresponds to the band at 1591 cm in the experimental triggered by a single electron transfer in a Grotthuss-type
ꢂ1
spectra. The band growing at 1560 cm
in the computed process over an uninterrupted hydrogen-bonded network
spectra, associated with the in-plane bending of both trans- including many benzimidazole units.
ferred protons, could explain the appearance of the band at
ꢂ1
1
558 cm in the experimental spectra.
Data availability
Evidence of proton translocation to form the E2PT product is
also observed in the high frequency region, where the NH All the details of the experimental and theoretical part are in the
2
2,23
stretching mode (nNH) is detected (Fig. 8B).
In the neutral ESI.† The crystal structure data have been deposited in the
ꢂ1
species, the nNH for 3 appears at 3390 cm , whereas the CCDC.
©
2021 The Author(s). Published by the Royal Society of Chemistry
Chem. Sci.

View Article Online
Chemical Science
Edge Article
1
1 E. Odella, T. A. Moore and A. L. Moore, Photosynth. Res., DOI:
Author contributions
10.1007/s11120-020-00815-x.
EO and ME performed the synthesis and characterization of the 12 E. Odella, B. L. Wadsworth, S. J. Mora, J. J. Goings,
compounds. MS performed computational work. TLG per-
formed X-ray diffraction characterization. All authors analysed
and interpreted the obtained results. EO and MS prepared
M. T. Huynh, D. Gust, T. A. Moore, G. F. Moore,
S. Hammes-Schiffer and A. L. Moore, J. Am. Chem. Soc.,
2019, 141, 14057–14061.

gures for publication. EO and MS wrote the manuscript with 13 W. D. Guerra, E. Odella, M. Secor, J. J. Goings, M. N. Urrutia,
input from TAM, SH-S and ALM. TAM, SH-S and ALM super-
vised and coordinated the project.
B. L. Wadsworth, M. Gervaldo, L. E. Sereno, T. A. Moore,
G. F. Moore, S. Hammes-Schiffer and A. L. Moore, J. Am.
Chem. Soc., 2020, 142, 21842–21851.
1
4 F. Su, Z. Sun, W. Su and X. Liang, J. Mol. Struct., 2018, 1173,
Conflicts of interest
690–696.
1
1
5 H. K. Sinha and S. K. Dogra, Chem. Phys., 1986, 102, 337–347.
6 A. R. Katritzky and J. M. Lagowski, in Prototropic
Tautomerism of Heteroaromatic Compounds: IV. Five-
Membered Rings with Two or More Hetero Atoms, ed. A. R.
Katritzky, Academic Press, 1963, vol. 2, pp. 27–81.
There are no conicts to declare.
Acknowledgements
This research was supported by the U.S. Department of Energy,
Office of Science, Office of Basic Energy Sciences, under Award
DE-FG02-03ER15393. The theoretical portion of this research
was supported as part of the Center for Molecular Electro-
catalysis, an Energy Frontier Research Center, funded by the
U.S. Department of Energy, Office of Science, Basic Energy
Sciences.
1
1
1
7 V. Sridharan, S. Saravanan, S. Muthusubramanian and
S. Sivasubramanian, Magn. Reson. Chem., 2005, 43, 551–556.
8 M. For ´e s, M. Duran, M. Sol `a , M. Orozco and F. J. Luque, J.
Phys. Chem. A, 1999, 103, 4525–4532.
´
´
9 C. I. Nieto, P. Cabildo, M. A. Garcıa, R. M. Claramunt,
I. Alkorta and J. Elguero, Beilstein J. Org. Chem., 2014, 10,
1620–1629.
2
0 K. Das, N. Sarkar, A. K. Ghosh, D. Majumdar, D. N. Nath and
K. Bhattacharyya, J. Phys. Chem., 1994, 98, 9126–9132.
1 K. Das, N. Sarkar, D. Majumdar and K. Bhattacharyya, Chem.
Phys. Lett., 1992, 198, 443–448.
Notes and references
2
1
2
3
R. Tyburski, T. Liu, S. D. Glover and L. Hammarstr ¨o m, J. Am.
Chem. Soc., 2021, 143, 560–576.
S. J. Mora, E. Odella, G. F. Moore, D. Gust, T. A. Moore and
A. L. Moore, Acc. Chem. Res., 2018, 51, 445–453.
D. R. Weinberg, C. J. Gagliardi, J. F. Hull, C. F. Murphy,
C. A. Kent, B. C. Westlake, A. Paul, D. H. Ess,
22 E. Odella, S. J. Mora, B. L. Wadsworth, M. T. Huynh,
J. J. Goings, P. A. Liddell, T. L. Groy, M. Gervaldo,
L. E. Sereno, D. Gust, T. A. Moore, G. F. Moore,
S. Hammes-Schiffer and A. L. Moore, J. Am. Chem. Soc.,
2018, 140, 15450–15460.
D. G. McCafferty and T. J. Meyer, Chem. Rev., 2012, 112, 23 M. T. Huynh, S. J. Mora, M. Villalba, M. E. Tejeda-Ferrari,
4
016–4093.
P. A. Liddell, B. R. Cherry, A.-L. Teillout, C. W. Machan,
C. P. Kubiak, D. Gust, T. A. Moore, S. Hammes-Schiffer
and A. L. Moore, ACS Cent. Sci., 2017, 3, 372–380.
4
5
6
7
8
9
L. Benisvy, R. Bittl, E. Bothe, C. D. Garner, J. McMaster,
S. Ross, C. Teutloff and F. Neese, Angew. Chem., Int. Ed.,
2005, 44, 5314–5317.
24 B. A. D. Neto, A. S. Lopes, M. W u¨ st, V. E. U. Costa, G. Ebeling
and J. Dupont, Tetrahedron Lett., 2005, 46, 6843–6846.
J. Mathias, L. Benisvy and F. Thomas, Chem.–Eur. J., 2012, 25 J. M. Aslan, D. J. Boston and F. M. MacDonnell, Chem.–Eur.
8, 5416–5429. J., 2015, 21, 17314–17323.
G. F. Moore, M. Hambourger, G. Kodis, W. Michl, D. Gust, 26 S. Morales, F. G. Guijarro, J. L. Garc ´ı a Ruano and M. B. Cid, J.
M. Orio, O. Jarjayes, B. Baptiste, C. Philouze, C. Duboc,
1
T. A. Moore and A. L. Moore, J. Phys. Chem. B, 2010, 114,
4450–14457.
S. J. Mora, D. A. Heredia, E. Odella, U. Vrudhula, D. Gust,
Am. Chem. Soc., 2014, 136, 1082–1089.
27 C. Lee, W. Yang and R. G. Parr, Phys. Rev. B: Condens. Matter
Mater. Phys., 1988, 37, 785–789.
1
T. A. Moore and A. L. Moore, J. Porphyrins Phthalocyanines, 28 A. D. Becke, J. Chem. Phys., 1993, 98, 5648–5652.
2019, 23, 1336–1345.
29 P. J. Stephens, F. J. Devlin, C. S. Ashvar, C. F. Chabalowski
W. D. Guerra, E. Odella, M. N. Urrutia, P. A. Liddell,
and M. J. Frisch, Faraday Discuss., 1994, 99, 103–119.
T. A. Moore and A. L. Moore, J. Porphyrins Phthalocyanines, 30 S. Grimme, J. Antony, S. Ehrlich and H. Krieg, J. Chem. Phys.,
021, 25, 674–682.
2010, 132, 154104.
G. F. Moore, M. Hambourger, M. Gervaldo, O. G. Poluektov, 31 S. Grimme, S. Ehrlich and L. Goerigk, J. Comput. Chem.,
2
T. Rajh, D. Gust, T. A. Moore and A. L. Moore, J. Am. Chem.
Soc., 2008, 130, 10466–10467.
0 Y. Yoneda, S. J. Mora, J. Shee, B. L. Wadsworth,
2011, 32, 1456–1465.
32 R. Ditcheld, W. J. Hehre and J. A. Pople, J. Chem. Phys.,
1971, 54, 724–728.
1
E. A. Arsenault, D. Hait, G. Kodis, D. Gust, G. F. Moore, 33 W. J. Hehre, R. Ditcheld and J. A. Pople, J. Chem. Phys.,
A. L. Moore, M. Head-Gordon, T. A. Moore and
1972, 56, 2257–2261.
G. R. Fleming, J. Am. Chem. Soc., 2021, 143, 3104–3112.
Chem. Sci.
© 2021 The Author(s). Published by the Royal Society of Chemistry

View Article Online
Edge Article
Chemical Science
3
4 P. C. Hariharan and J. A. Pople, Theor. Chim. Acta, 1973, 28, 39 B. H. Solis and S. Hammes-Schiffer, Inorg. Chem., 2014, 53,
13–222. 6427–6443.
5 M. M. Francl, W. J. Pietro, W. J. Hehre, J. S. Binkley, 40 I. Rozas, I. Alkorta and J. Elguero, J. Phys. Chem. A, 1998, 102,
2
3
M. S. Gordon, D. J. DeFrees and J. A. Pople, J. Chem. Phys.,
982, 77, 3654–3665.
6 T. F. Markle, I. J. Rhile, A. G. DiPasquale and J. M. Mayer,
9925–9932.
1
41 F. Thomas, O. Jarjayes, H. Jamet, S. Hamman, E. Saint-
Aman, C. Duboc and J.-L. Pierre, Angew. Chem., Int. Ed.,
2004, 43, 594–597.
3
3
3
Proc. Natl. Acad. Sci. U. S. A., 2008, 105, 8185–8190.
7 V. Barone and M. Cossi, J. Phys. Chem. A, 1998, 102, 1995– 42 E. Odella, S. J. Mora, B. L. Wadsworth, J. J. Goings,
2001.
M. A. Gervaldo, L. E. Sereno, T. L. Groy, D. Gust,
T. A. Moore, G. F. Moore, S. Hammes-Schiffer and
A. L. Moore, Chem. Sci., 2020, 11, 3820–3828.
8 M. Cossi, N. Rega, G. Scalmani and V. Barone, J. Comput.
Chem., 2003, 24, 669–681.
©
2021 The Author(s). Published by the Royal Society of Chemistry
Chem. Sci.