Isomerization mechanism in hydrazone-based rotary switches: Lateral shift, rotation, or tautomerization?

Article abstract of DOI:10.1021/ja200699v

Two intramolecularly hydrogen-bonded arylhydrazone (aryl = phenyl or naphthyl) molecular switches have been synthesized, and their full and reversible switching between the E and Z configurations have been demonstrated. These chemically controlled configu

Full text of DOI:10.1021/ja200699v

ARTICLE
pubs.acs.org/JACS
Isomerization Mechanism in Hydrazone-Based Rotary Switches:
Lateral Shift, Rotation, or Tautomerization?
||
Shainaz M. Landge,†, Ekatarina Tkatchouk,‡ Diego Benítez,‡ Don Antoine Lanfranchi,§ Mourad Elhabiri,§,^
William A. Goddard, III,‡ and Ivan Aprahamian*,†
†Department of Chemistry, Dartmouth College, Hanover, New Hampshire 03755, United States
‡Materials and Process Simulation Center, California Institute of Technology, Pasadena, California 91125, United States
§Laboratoire de Chimie Bioorganique et Mꢀedicinale, Universitꢀe de Strasbourg, ECPM, UMR 7509 CNRS-UdS, 25, rue Becquerel, 67200
Strasbourg, France
^Laboratoire de Physico-Chimie Bioinorganique, Universitꢀe de Strasbourg, Institut de Chimie de Strasbourg, ECPM, UMR 7177 CNRS-
UdS, 25, rue Becquerel, 67200 Strasbourg, France
Department of Chemistry, Georgia Southern University, Statesboro, Georgia 30458, United States
S Supporting Information
b
ABSTRACT: Two intramolecularly hydrogen-bonded arylhy-
drazone (aryl = phenyl or naphthyl) molecular switches have
been synthesized, and their full and reversible switching be-
tween the E and Z configurations have been demonstrated.
These chemically controlled configurational rotary switches
exist primarily as the E isomer at equilibrium and can be
switched to the protonated Z configuration (Z-Hþ) by the addition of trifluoroacetic acid. The protonation of the pyridine moiety
in the switch induces a rotation around the hydrazone CdN double bond, leading to isomerization. Treating Z-Hþ with base
(K2CO3) yields a mixture of E and “metastable” Z isomers. The latter thermally equilibrates to reinstate the initial isomer ratio. The
rate of the Z f E isomerization process showed small changes as a function of solvent polarity, indicating that the isomerization
might be going through the inversion mechanism (nonpolar transition state). However, the plot of the logarithm of the rate constant
k vs the Dimroth parameter (ET) gave a linear fit, demonstrating the involvement of a polar transition state (rotation mechanism).
These two seemingly contradicting kinetic data were not enough to determine whether the isomerization mechanism goes through the
rotation or inversion pathways. The highly negative entropy values obtained for both the forward (E f Z-Hþ) and backward (Z f
E) processes strongly suggest that the isomerization involves a polarized transition state that is highly organized (possibly involving a
highdegree ofsolvent organization), and hence itproceeds viaa rotation mechanismasopposedto inversion. Computationsof the Z T
E isomerization using density functional theory (DFT) at the M06/cc-pVTZ level and natural bond orbital (NBO) wave
function analyses have shown that the favorable isomerization mechanism in these hydrogen-bonded systems is hydrazoneꢀazo
tautomerization followed by rotation around a CꢀN single bond, as opposed to the more common rotation mechanism around the
CdN double bond.
’ INTRODUCTION
changes when activated chemically, whereas photochemical
modulation causes configurational12 changes. There seems to
be a significant lag in the area of chemically controlled config-
urational rotary switches and motors; bridging this gap may lead
to novel functional systems and materials.13
The hydrazone functional group is versatile and has found
extensive uses in medicine,14 dynamic combinatorial chemistry
(DCC),15 foldamers,16 and molecular tweezers,17 to name a few.
One important facet of hydrazone chemistry that has not been
exploited thus far is the fact that upon appropriate substitution
they can exist in solution as a mixture of isomers.18 1,2,3-
Tricarbonyl-2-arylhydrazones (TCAHs)19 for example can be
Molecular switches, motors and machines hold enormous
promise for the advancement of nanotechnology.1 The study and
synthesis of such “smart” molecules that can mimic biological2
and macroscopic counterparts has led to the development of a
diverse array of molecular systems, such as switches,3 shuttles,4
muscles,5 ratchets,6 walkers,7 robots,8 memory devices,9 etc., that
can be activated using various external stimuli, e.g., chemical,
electrochemical and photochemical. The induced motion can be
rotary in nature or linear (translation) depending on the targeted
application.1 Synthetic rotary switches10 are of particular interest
as they resemble rotary motors found in nature (e.g., ATPases).2
Such synthetic systems have given rise to the only known
examples to date of synthetic motors that can undergo unidirec-
tional motion.11 These rotary systems undergo conformational
Received: January 24, 2011
Published: May 17, 2011
r
2011 American Chemical Society
9812
dx.doi.org/10.1021/ja200699v J. Am. Chem. Soc. 2011, 133, 9812–9823
|

Journal of the American Chemical Society
ARTICLE
Scheme 1. Possible Mechanism for the Acid-Catalyzed Tautomerization in TCAHs; Similar Scheme Can Be Drawn for the Base-
Catalyzed Process
Scheme 2. Mechanism of E T Z Isomerization of Imines and
Scheme 3. Synthesis of 1 and 2
Hydrazones
present in solution as a pair of intramolecularly H-bonded
isomers (Scheme 1).20 This H-bond causes a diagnostic low-
1
field H NMR signal for the NH proton whose chemical shift
depends on the type of carbonyl group to which it is attached.
Hence, the different isomers can be readily distinguished using
1H NMR spectroscopy. TCAHs can be equilibrated in solution
by the addition of a catalytic amount of acid or base
(Scheme 1).20 The isomerization mechanism in such instances
consists of a hydrazoneꢀazo tautomerization21 followed by
rotation around a CꢀN single bond that leads to a change in
the relative position of the substituents around the CdN double
bond (i.e., E f Z isomerization);20d however, this equilibration
process is neither selective nor optimal and only affords a mixture
of isomers (e.g., it can proceed from pure E to a mixture of E and
Z isomers) that cannot be controllably switched back to the
initial state.
The uncatalyzed isomerization around the CdN double bond
(Scheme 2) is known18,22 to proceed via two different mechan-
isms: rotation and inversion. The rotation mechanism proceeds
through a polar transition state, and results in the rotation of the
substituents around the CdN double bond. The inversion
mechanism (also known as lateral shift) in contrast, occurs
through an inversion around the imine nitrogen, and goes
through a nonpolar transition state. Thus, the isomerization
rates in the rotation mechanism show dependence on solvent
polarity, whereas in the inversion mechanism no such effects are
observed. This divergence in properties has been historically
used in discerning the nature of the isomerization mechanism in
various hydrazones and imines.
What is certain is that more experimentation needs to be
conducted in order to fully understand the E T Z isomerization
mechanism in hydrazones, as this will eventually help us to
finetune and control the switching process in arylhydrazone-
based switches. Moreover, discerning the type of isomerization
mechanism will be important as far as applications are concerned;
the switching trajectory in each mechanism is different, rotation
vs a pendulum-type motion (inversion), making each suitable for
a different function.1
We have previously28 shown that it is possible to convert the
inefficient TCAH switches into reliable ones by replacing one of
the carbonyl groups with a pyridine ring. Having a pyridine group
as a proton acceptor and an ester moiety as a secondary H-bond
acceptor attached to the imine carbon improved the control over
the isomerization process. This structural change led to the
bridging of the gap between conformational and configurational
rotary switches and the development of chemically driven
configurational rotary switches. Here we provide an experimental
and computational account of the synthesis, characterization, pH
switching,29 and isomerization mechanism of two hydrazone-
based rotary switches having different aryl stators; phenyl (1) and
naphthyl (2).
’ RESULTS AND DISCUSSION
Synthesis. Compounds 1 and 228a were synthesized using a
straightforward procedure (Scheme 3). First ethyl-2-pyridyl
acetate was treated with sodium acetate in an ethanolꢀwater
mixture. In a second pot, the aromatic amine was treated with
sodium nitrite and hydrochloric acid at 0 ꢀC to create a
diazonium salt. Both these solutions were mixed together at
0ꢀ5 ꢀC and then brought to room temperature over a period of
1 h. The reaction mixture was stirred overnight during which the
product precipitated out from the alcoholꢀwater medium. The
crude solid compound was then filtered from the mother liquor
The isomerization mechanism has been studied in detail for
imines,23 and to a lesser extent for hydrazones,24,25 and in both
cases inversion is considered to be the common mechanistic
pathway.18,22,23,25 However, in most cases the studies focus on
the rates of the isomerization process, and less attention is given to
the contributions of enthalpy and entropy to the activation
barrier; this can lead to misleading conclusions, as attested by
contradictory reports found in the literature.26,27 These studies
put the generality of the inversion mechanism into question.
9813
dx.doi.org/10.1021/ja200699v |J. Am. Chem. Soc. 2011, 133, 9812–9823

Journal of the American Chemical Society
ARTICLE
Figure 1. ORTEP drawing (50% probability ellipsoids) of 1-E and 2-E. The protons are placed in calculated positions, except for the hydrazone N(2)H
proton.
and subjected to column chromatography (SiO2, CH2Cl2/hex-
anes (1:2), respectively) to yield 1 (78%) and 2 (75%) as orange
powders.
Structural Characterization. Crystals of the phenyl derivative
(1) were obtained by slow evaporation of methanol, resulting in
fine, orange needlelike crystals. A similar procedure using a
mixture of CH2Cl2 and hexanes (1:1) resulted in the orange-
red needlelike crystals of 2. Both derivatives crystallize in the E-
configuration (Figure 1).
The phenyl derivative (1-E) prefers to pack in a dimeric form,
as seen across the a-axis (X-ray S1a in the Supporting Informa-
tion [SI]) such that the ester oxygen of one molecule and the
methyl protons of another are in close proximity with each other.
We can see from the crystal structures (Figure 1) that there is an
downfield compared with 1-E. One should also take into account
the fact that the naphthyl group is a better electron-withdrawing
group (EWG), making the NH proton more acidic, the H-bond
stronger, and the structure more planar and hence more
conjugated.33 Moreover, resonance-assisted hydrogen bonding
(RAHB)34 comes into play in these systems, making the direct
analysis of the difference in the chemical shift of the NH proton in
1
1-E and 2-E complicated. The H NMR spectrum of 2 also
contains a small signal at 12.80 ppm stemming from 2-Z. The
isomer ratio for 2, as deduced from integration, is 97:3 which is
higher than in 1. This slight difference in isomer ratio stems from
the above-mentioned electronic effects that stabilize 2-E more
than 1-E. One-dimensional (1D) NOESY (Figure S4 in the SI)
clearly shows that the compound adopts the same rotameric form
in solution as was observed in the solid state.35
intramolecular H-bond (NꢀH N: 1.997(1) Å, 130.88(10)ꢀ
3 3 3
in 1-E and 1.918(1) Å, 133.70(10)ꢀ in 2-E) between the pyridine
nitrogen (N3) and the hydrazone NH proton. It is also evident
that in 2-E the naphthalene bay area is oriented toward the NH
hydrogen. The measured torsional angles30 indicate that the
systems are almost planar.31 The comparison of the bond
lengths32 with those of similar intramolecularly H-bonded hy-
drazones (phenylhydrazone derivatives condensed with
diesters),33 leads us to conclude that the aryl rings are in
conjugation with the pyridine and/or ester moieties. The in-
tramolecular H-bond, the presence of a pseudo-six membered
aromatic ring, and the π-delocalization all contribute to the
planarity of the molecules.
Computational Studies. Both isomers of 1 and 2 were
subjected to DFT calculations36 at the M06/cc-pVTZ level.37,38
The optimized geometries of both compounds predicted the E
isomer to be more stable than Z in both acetonitrile and toluene
as solvents. In the case of compound 1, isomer E is ΔGo = 1.2 kcal
molꢀ1 (1.5 kcal molꢀ1 exp.) more stable than Z, whereas in 2,
isomer E is ΔGo = 1.5 kcal molꢀ1 (2.0 kcal molꢀ1 exp.) more
stable in acetonitrile as solvent. These data are in close agreement
1
with the H NMR spectroscopy results. The calculations com-
plement the X-ray structural analysis as they enable probing of
the Z isomers, which could not be crystallized. The Z isomers of 1
and 2 deviate from planarity because of repulsion between the
electron lone pairs on the imine and pyridine nitrogens (Figures
Calc-S1 and Calc-S2 in the SI). A large torsion angle is present
between the NpyrꢀCpyrꢀCimindN atoms (37.3ꢀ for 1-Z and
37.6ꢀ for 2-Z) as compared to the planar E isomers (1.6ꢀ for 1-E
and 2.2ꢀ for 2-E). The fact that the Z isomers are equally
nonplanar supports the conclusion that the additional electronic
stabilization of 2-E (relative 1-E) is the reason behind the slight
difference in isomer ratio (ΔGo) observed in CD3CN.
The fully characterized 1H NMR spectrum of 1-E in CD3CN
(Figure 2, and Figure S1 in the SI) shows the presence of a distinct
intramolecular H-bond between the pyridine nitrogen and the
hydrazone proton. This H-bond manifests itself by a highly
deshielded NH signal at 14.56 ppm. A small signal is also observed
at 11.48 ppm stemming from the minor isomer, i.e., 1-Z. The E:Z
isomer ratio in CD3CN is 93:7 for 1 based on the integration ofthe
NHsignals. One-dimensionalNOESY(FigureS2 intheSI) clearly
shows that the pyridine is H-bonded with the hydrazone NH
proton. The assigned structure in the solution phase agrees well
with the solid-state structure. The H-bond between the pyridine
and the hydrazone NH in 2-E (Figure S3 in the SI) is evident by
the signal resonating at 15.80 ppm. This proton is clearly affected
by the naphthalene’s ring current effect, as it resonates further
We also analyzed the NMR chemical shifts of 1 and 2 by
comparing their calculated relative isotropic nuclear shieldings
using the M06-L (no HartreeꢀFock exchange) functional.39
A
comparison of the isotropic nuclear magnetic shielding constants
for the hydrazone NH proton in a pair of rotamers of 2-E
(calculated 16.7 and 16.3 ppm, see SI for details) showed small
9814
dx.doi.org/10.1021/ja200699v |J. Am. Chem. Soc. 2011, 133, 9812–9823

Journal of the American Chemical Society
ARTICLE
Figure 2. 1H NMR spectra (500 MHz) in CD3CN of (a) 1-E with peak assignments; (b) 1-Z-Hþ with peak assignments, recorded after the addition of
2.2 equiv of TFA to 1-E; (c) a mixture of 1-E and 1-Z, taken after passing 1-Z-Hþ over a plug of K2CO3; and (d) 1-E after the mixture was left to
equilibrate for 3 h at rt.
1
This process was monitored by H NMR spectroscopy; the
addition of acid shifted the NH proton signal from 14.50 to 12.90
ppm (Figure 2b). This high-field shift indicates that the ester
group is now H-bonded with the NH proton (1-Z-Hþ).20d As
expected the protonation also causes the pyridine protons to shift
downfield.42 Only proton H1 seems to move upfield (8.76 to
8.63 ppm) as it is no longer influenced by the H-bond and
RAHB.43 The phenyl protons also shift downfield, the most
prevalent change being for proton H5 (from 7.38 to 7.62 ppm),
indicating again that the pyridine ring is protonated. The 1D
NOESY experiments (Figure S6 in the SI) show that the
protonated pyridine nitrogen orients itself such that it is directed
away from the phenyl ring. The color of the solution goes back to
light-yellow when passed through a plug of potassium carbonate
(K2CO3),44 and a mixture of isomers is observed in the 1H NMR
spectrum (Figure 2c). Distinct NH signals are observed at 11.48
and 14.56 ppm for the Z and E configurations, respectively. The
“metastable” Z isomer thermally equilibrates with time to give
back the more stable E isomer, showing the complete reversibility
of the switching process (Figure 2d).
Scheme 4. Acid/Base Controlled E T Z Isomerization of 1
differences of only ∼0.4 ppm between them. Our results suggest
that the difference in the chemical shifts between 1 and 2 (exp.
14.6 and 15.8 ppm; calc. 15.6 and 16.7 ppm for the E isomers of 1
and 2, respectively)40 cannot be explained solely on the premise
of ring current effect differences originating from both the
relative orientation and the different aryl groups. RAHB34 is
definitely participating in determining the chemical shift of the
NH protons.
A similar process is observed when 2-E is protonated.28a The
addition of 2.0 equiv of TFA to 2-E is accompanied by an upfield
shift of the hydrazone NH proton from 15.8 to 13.9 ppm (Figure
S7 in the SI). This shift indicates that the ester moiety is
H-bonded with the NH proton (2-Z-Hþ). The protonation
causes the pyridine signals to shift downfield as expected.
NOESY experiments indicate that the position of the naphthyl
AcidꢀBase Switching Monitored by 1H NMR Spectroscopy.
The addition of 2.2 equiv of trifluoroacetic acid (TFA)41 to a
CD3CN solution 1-E changes its color from light- to dark-yellow.
It is expected that the protonation of the pyridine group in 1-E
will lead to breaking of the H-bond and rotation around the
CdN double bond to form protonated 1-Z-Hþ (Scheme 4).
9815
dx.doi.org/10.1021/ja200699v |J. Am. Chem. Soc. 2011, 133, 9812–9823

Journal of the American Chemical Society
ARTICLE
Figure 3. ORTEP drawing (50% probability ellipsoids) of 1-Z-Hþ and 2-Z-Hþ. The protons are placed in calculated positions, except for the hydrazone
N(2)H and N(3)H protons.
Figure 4. UV/vis spectra of the E, Z and Z-Hþ isomers of the phenyl-
based molecular switch, 1. All data were recorded in MeCN at 298 K.
Figure 5. Changes in the UV/vis spectra during the acid/base switching
of 1-E. All data were recorded in MeCN at 298 K. To a 2.68 ꢁ 10ꢀ5 M
solution of 1-E (a), was added TFA (b), until there was no observable
ring did not change as result of protonation (Figure S9 in the SI)
change (c). NEt3 was then added (d) to switch the system back (e). The
and that the protonated pyridine nitrogen atom is oriented away
from the naphthyl core (Figure S10 in the SI). After the solution
is passed through a plug of K2CO3, two NH signals can be
observed, one for the E isomer at 15.80 ppm and another for the
Z isomer at 12.80 ppm. The Z isomer thermally equilibrates after
a period of time to restore the original isomer ratio (97:3). This
switching cycle with acid and base was repetitively demonstrated
over five runs for both the naphthyl and phenyl derivatives with
TFA as acid and triethylamine (NEt3) as base.
signals of (a) and (e) are overlapping.
AcidꢀBase Switching Monitored by UV/Vis Spectroscopy.
The complete switching cycles of compounds 1 and 2 were also fully
characterized by UV/vis spectroscopy (Figures 4, 5, and S11ꢀS16 in
the SI). The E isomer of compounds 1 (Figure 4) and 2 (Figure S12
in the SI) is characterized by an intense and broad absorption band
centered at λmax = 366 nm (ε366 = 2.14 ꢁ 104 Mꢀ1 cmꢀ1) and
392 nm (ε392 = 1.96 ꢁ 104 Mꢀ1 cmꢀ1), respectively. These bands
result from the πꢀπ* transitions of the aromatic units and are in
agreement with the X-ray data that showed that the systems are fully
conjugated. The extended aromatic unit in 2 leads to a bathochromic
shift of 26 nm as compared to 1. Both 1-E and 2-E possess weaker
and almost isoabsorptive absorptions at higher energies as well (1-E:
The protonation of the pyridine ring was also confirmed by
X-ray crystallography analysis (Figure 3). The structures of the
protonated Z isomers, 1-Z-Hþ and 2-Z-Hþ, clearly show that the
intramolecular hydrogen bond (NꢀH O: 1.905(1) Å,
3 3 3
133.00(10)ꢀ in 1-Z-Hþ and 1.968(1) Å, 135.50(10)ꢀ in 2-Z-
Hþ) is now between the ester carbonyl and the NH proton, in
accordance with the NMR studies. Moreover, the naphthalene
ring in 2-Z-Hþ adopts the same rotameric form in the solid-state
as observed in solution. The bond lengths and torsional angles45
show that these systems are also fully conjugated and planar.
ε
280 =6.33ꢁ 103 Mꢀ1 cmꢀ1 and 2-E:ε279 =6.67ꢁ 103 Mꢀ1 cmꢀ1).
The protonation of the pyridine subunit in these compounds
(Figures 4 and S12 in SI), and the subsequent isomerization
to the Z-Hþ isomers is supported by the bathochromic shift
9816
dx.doi.org/10.1021/ja200699v |J. Am. Chem. Soc. 2011, 133, 9812–9823

Journal of the American Chemical Society
ARTICLE
Table 1. Rate Constants and Gibbs Activation Energy Para-
meters for the Z T E Isomerization of 1 and 2 in Different
Deuterated Solvents
concentration of TFA is required (∼18 vs 2 equiv) to protonate
the system in acetone, which can change the ionic strength of the
solvent. Moreover, it has been reported in the literature that
acetone can form hydrogen bonds with the acidic NH proton,25e
which can affect the isomerization rate. The differences in rate
constants when changing the polarity of the solvent are not
substantial, and when compared with reported results25 one
might conclude that the isomerization process goes through the
lateral inversion mechanism. However, both derivatives show a
linear relationship between log k and the Dimroth parameter
(ET)51 (Figure S18 in the SI), which is a clear indication for the
involvement of a polarized transition state in the process.
Mechanism Calculations. In order to gain further insight into
the isomerizationmechanismof 1 and 2weperformed calculations
on the possible pathways including inversion and different flavors
of rotation. The inversion pathway happens through a change in
the hybridization of the N atom; from sp2 to sp at the linear
transition state. The sp2 lone pair changes to a transitory filled p
orbital to allow the inversion to occur. Using the M06 functional,37
we calculated a barrier to inversion of 67.0 kcal/mol for 1. This
process exhibits a predicted barrier for the Z to E transformation
that is not consistent with the experimental observations.
The rotation mechanism can occur via several pathways,
depending on the reorganization of the electrons participating
in the CdN π-bond: heterolysis, homolysis, and tautomeriza-
tion. Heterolysis of the CdN bond produces a zwitterionic
transition state, while the homolysis produces a triplet inter-
mediate. We performed calculations using the M06 functional on
the singlet state rotation that leads to a charge separated
transition state. Our computational methodology (M06/cc-
pVTZ)37,38 predicts a barrier of 55.6 kcal/mol for the heterolytic
rotation around the CdN bond of 1 in acetonitrile, also not
consistent with the experimental observations.
Homolysis leads to a stable triplet intermediate where the
dihedral CpyrꢀCdNꢀN is 90ꢀ. Therefore, in principle, the
potential energy surface for the rotation process follows the
singlet energy valley at low dihedral angles and the triplet energy
valley at high dihedral angles. We estimate the crossover point
around ∼60ꢀ from relaxed coordinate scans for both the singlet
state and the triplet state. We calculated that the barrier for the
homolytic rotation is at least 42.7 kcal/mol for 1 in acetonitrile,
which is not consistent with our experiments.
We also considered a mechanism where a hydrazoneꢀazo
tautomerization reorganizes the π-electronic framework along
the extended delocalized system (Figure 6), thus enabling rotation
around a CꢀN single bond. Several reports have proposed such an
isomerization mechanism in TCAHs when using acid/base
catalysts;20d however, such an isomerization has not been con-
sidered thus far as a viable pathway in the absence of a catalyst.52
We believe that the pyridyl group (significantly more basic than
the carbonyl) greatly favors proton exchange and acts as an
internal “catalyst”. We envision a mechanism where the isomer-
ization proceeds via a rotational transition structure that has
an azo form with a dihedral angle CpyrꢀCiminꢀNꢀN (and
CesterꢀCiminꢀNꢀN) at ∼90ꢀ. This is in contrast to a triplet state
(from homolysis) that would exhibit pyramidalization (from
rehybridization) of the Cimin atom. We calculated the barrier for
such a tautomerization and found it to be 23.8 and 24.3 kcal/mol
for 1 and 2, respectively, in acetonitrile. The magnitude of these
barriers is consistent with our experimental data (Table 1).53
We further characterized 1-E and 2-E, as well as 1TS and 2TS,
using natural bond orbital (NBO) formalism and software54
1
2
k (ꢁ 104 sꢀ1) / ΔGq
k (ꢁ 104 sꢀ1) / ΔGq
solvent (ET)
(kcal molꢀ1
)
(kcal molꢀ1
)
toluene-d8 (33.9)
CDCl3 (39.1)
0.82 ( 0.23/22.70 ( 0.16 0.57 ( 0.21/22.90 ( 0.21
1.76 ( 0.11/22.30 ( 0.03 1.04 ( 0.02/22.60 ( 0.01
3.34 ( 0.26/21.90 ( 0.04 9.60 ( 0.23/21.30 ( 0.01
2.58 ( 0.14/22.10 ( 0.03 3.54 ( 0.04/21.80 ( 0.01
acetone-d6 (42.2)
CD3CN (46)
of ∼32ꢀ34 nm in the main absorption band (1-Z-Hþ:ε398 =2.35ꢁ
104 Mꢀ1 cmꢀ1 and 2-Z-Hþ: ε426 = 1.94 ꢁ 104 Mꢀ1 cmꢀ1). The
deprotonation of the Z-Hþ isomers using NEt3 leads to the
“metastable” Z isomers, which is accompanied by a strong hypso-
chromic shift of ∼46 nm of the main absorption band (1-Z: ε352
=
1.94 ꢁ 104 Mꢀ1 cmꢀ1 and 2-Z: ε380 = 1.75 ꢁ 104 Mꢀ1 cmꢀ1).46
The reversibility of the switching process was demonstrated by
UV/vis spectroscopy as well (Figures 5 and S11 in SI). The
absorption maximum of 1-E at λmax = 366 nm (Figure 5a) shifts
bathochromicly to λmax = 392 nm (Figure 5c) when titrated with
TFA (25 equiv). New peaks emerge at λ = 307 (ε307 = 6.57 ꢁ 103
M
ꢀ1 cmꢀ1) and 257 nm (ε257 = 1.04 ꢁ 104 Mꢀ1 cmꢀ1) in the
protonated form. When 1-Z-Hþ is titrated with NEt3 (Figure 5d)
the original absorption is regenerated (Figure 5d). Four isosbestic
points were visible during the entire switching cycle at λ = 274,
292, 320, and 379 nm(FigureS11 in the SI) showingthatonly two
species are involved in the process.47 A similar behavior is also
observed for the naphthyl derivative (Figures S12ꢀ14 in the SI);
the absorption of 2-E at λmax = 392 nm, is shifted to λmax = 426 nm
upon protonation with TFA (Figure S13c in SI). Four isosbestic
points are observed at λ = 241, 264, 339, and 408 nm during this
process. Treating 2-Z-Hþ with excess of NEt3 restores the original
spectrum (Figure S13e in SI), attesting to the reversibility of the
switching process. These acidꢀbase switching cycles can be
reversibly repeated multiple times for both 1 and 2.
Mechanistic Studies: Computation, Kinetics and Thermo-
dynamics. NMR Analysis. In order to gain insight into the
switching mechanism of the H-bonded arylhydrazone switches,
we first performed a thorough study of the Z f E thermal
equilibration using 1H NMR spectroscopy. The rate of equilibra-
tion was studied by following the changes in the intensity of the
NH signal of the Z configuration as a function of time. The rate
constants (k) were then calculated from nonlinear least-squares
fittings and the Gibbs energy of activation (ΔGq) was calculated
using the Eyring equation.48 The initial equilibrium ratio was
taken into account when performing these calculations and only
the contribution of kZfE (referred to as k from now on) to the
measured rates was used in determining the different kinetic and
thermodynamic parameters (see SI for more details).49
The rate measurements were conducted multiple times at
room temperature (Figure S17 in SI) in different solvents having
various polarities (Table 1): CD3CN, acetone-d6, CDCl3 and
toluene-d8. It was observed that the isomerization rate increases
and the ΔG298qvalue decreases when going from the nonpolar to
the more polar solvents. The difference in the rate between the
extremities, toluene-d8 and CD3CN, is 6.2 times for 2 and 3.1
times for 1.50 In acetone-d6 the rates are slightly higher than in
CD3CN and this might be explained by the fact that a higher
9817
dx.doi.org/10.1021/ja200699v |J. Am. Chem. Soc. 2011, 133, 9812–9823

Journal of the American Chemical Society
ARTICLE
Figure 6. Potential free energy (M06/cc-pVTZ) landscape for the proposed intermediates and transition structure for the rotational isomerization
process of switches 1 and 2 in acetonitrile. Selected natural bond orders and charges from natural localized molecular orbitals and natural populations for
structures 1-E, 2-E, 1TS, and 2TS.
Table 2. Activation Parameters for the Z T E Isomerization of 1 and 2 in Toluene-d8 and CD3CN
molecular switch (solvent)
Ea (kcal molꢀ1
)
ΔHq (kcal molꢀ1
)
TΔSq (kcal molꢀ1
)
ΔSq (cal molꢀ1 Kꢀ1
)
1 (toluene-d8)
2 (toluene-d8)
1 (CD3CN)
2 (CD3CN)
6.7 ( 1.1
8.3 ( 1.3
9.6 ( 0.4
11.2 ( 0.7
6.1 ( 1.1
7.7 ( 1.3
9.0 ( 0.4
10.6 ( 0.7
ꢀ16.7 ( 0.2
ꢀ15.2 ( 0.2
ꢀ13.1 ( 0.1
ꢀ11.2 ( 0.1
ꢀ56.6 ( 3.7
ꢀ51.9 ( 4.5
ꢀ44.5 ( 1.4
ꢀ38.0 ( 2.3
(Figure 6). Compounds 1-E and 2-E exhibit a highly delocalized
π-electron system. In contrast, in compounds 1TS and 2TS, the
π-system of the N atoms and terminal aryl (phenyl in 1 or
naphthyl in 2) lie orthogonal to the pyridyl-ester π-orbitals. This
renders the electronics in both transition structures almost
identical. However, the different aryl groups—phenyl or
naphthyl—have marked effects on the electronics of 1-E and
2-E. This is exhibited by the change in the bond order between Ca
and Nb from 1-E to 1TS and from 2-E to 2TS (Figure 6). For the
case of phenyl (1), the CaꢀNb natural bond order decreases from
0.93 to 0.89, while the same bond order increases from 0.82 to
0.90 for naphthyl (2). These differences are also reflected (albeit
to a minor degree) in the neighboring bonds, which are more
relevant to the rotational barrier. These data suggest that the
difference in the barriers to rotation in 1 and 2 is associated with
the partial localization of the π-electron system from the highly
delocalized ground state to the less delocalized transition state.
Furthermore, we calculated the dipole moments for the wave
functions of structrues 1-E, 1TS, 2-E and 2TS in a dielectric
continuum for acetonitrile. We found that the ground states
have a much smaller dipole moment: 2.7 and 3.0 D for 1-E and
2-E, respectively, while the transition states exhibit a much
higher dipole moment: 6.8 and 6.7 D for 1TS and 2TS,
respectively (Figures Calc-S3 and Calc-S4 in the SI). This is in
agreement with our experimental determination of the depen-
dence of the isomerization rates on solvent polarity, which
suggests that the rate-determining step involves a more polar
transition structure.
Thermodynamic Parameters. Relying solely on the changes in
rate constant (and subsequently the Gibbs energy of activation)27
as a function of solvent polarity cannot be used as the only basis to
support the isomerization mechanism in these H-bonded arylhy-
drazones. In order to reach a better conclusion about the
mechanism, one has to look at the full picture, which entails a full
analysis of not only the rates of the process but also the contribu-
tion of the entropy and enthalpy parameters. To accomplish this
we extended the rate constant measurement studies in toluene-d8
and CD3CN to a range of temperatures: 294.5ꢀ338.5 K for
toluene-d8 and 268.5ꢀ308.5 K for CD3CN. In toluene-d8 (Table
S1 in the SI) the rates are almost the same for 1 and 2 over the
range of temperatures, whereas in CD3CN the rates are consis-
tently higher for 2 (Table S2 in the SI). These data were applied in
calculating (Table 2) the enthalpy of activation (ΔHq), entropy of
activation (ΔSq), and energy of activation (Ea) using the Eyring
and Arrhenius equations (Figures S19ꢀ20 in the SI). The large
negative ΔSq value of both systems regardless of the solvent can be
interpreted as further evidence for the involvement of a polarized
transition state in the isomerization process, given that at the rate-
determining step, solvent molecules may associate to solvate the
more polar transition structure. This parameter is less negative in
more polar CD3CN, which requires less reorganization to stabilize
the polar transition state.
9818
dx.doi.org/10.1021/ja200699v |J. Am. Chem. Soc. 2011, 133, 9812–9823

Journal of the American Chemical Society
ARTICLE
Table 3. Kinetic Data and Activation Parameters for the E to Z-Hþ Isomerization of 1 and 2 in CH3CN
molecular switch
kEfZa ( ꢁ 10ꢀ2 sꢀ1
)
Ea (kcal molꢀ1
)
ΔHq (kcal molꢀ1
)
TΔSq (kcal molꢀ1
)
ΔSq (cal molꢀ1 Kꢀ1
)
1
2
2.7 ( 0.3
5.2 ( 0.3
15.6 ( 0.5
16.3 ( 0.7
15.1 ( 0.5
15.8 ( 0.7
ꢀ4.70 ( 0.05
ꢀ3.6 ( 0.5
ꢀ15.8 ( 1.9
ꢀ11.5 ( 1.7
a Determined at 25.0(2) ꢀC.
When comparing 1 and 2, it is evident that the electronics of the
aryl group affect the transition state.The naphthyl group is a better
EWG and can better stabilize the polarized transition state, thus
requiring less solvent reorganization and resulting in a less negative
ΔSq value. On the other hand the naphthyl group makes the NH
proton more acidic and hence the H-bond with the ester carbonyl is
stronger as manifested in the larger ΔHq and Ea values. The H-bond
has to be broken first in order for the isomerization to proceed, and
this explains the positive enthalpy values for both switches. These
two opposing effects cause the isomerization rates of 1 and 2 to be
almost the same in toluene-d8. In CD3CN, the ΔHq and Ea values
are larger than in toluene-d8 because of the solvent’s promotion of
H-bonding; however, the change for both systems is almost the
same (ΔΔHq is 2.9 and 3.0 kcal molꢀ1, for 1 and 2, respectively).
The same holds true for the Ea values. Conversely, the stabilization
of the transition state is larger in 2 compared to 1 (ΔΔSq is 13.9 and
pre-equilibrium affording the E-Hþ kinetic intermediate. The
rate-limiting step therefore corresponds to the EHþ f ZHþ
isomerization that can be described as follows:
KH
E a E-Hþ/ fast pre-equilibrium
kE=Z
E-Hþ/ f Z-Hþ rate-limiting step
Nonlinear least-squares fittings of the variations of kobs versus
[TFA]tot allowed us to determine the monomolecular rate
constant kEHþf ZHþ (referred to as kEfZ from now on for the
sake of simplicity) of the E f Z-Hþ isomerization process
(Table 3). Interestingly, the first order isomerization rate con-
stant kEfZ calculated for the naphthyl derivative 2 (kEfZ
=
=
5.2(3) ꢁ 10ꢀ2 sꢀ1) is about 2 times higher than that of 1 (kEfZ
2.7(3) ꢁ 10ꢀ2
s
ꢀ1). As previously discussed, this feature sup-
12.1 cal molꢀ1 ꢀ1, respectively), which results in the 0.3 kcal
K
ports the involvement of a polarized transition state in the
isomerization process. The more extended naphthyl π-system
stabilizes the transition state of 2 better than in 1, and hence, the
reaction goes faster.
molꢀ1 difference in the ΔGq value (Table 1) between 1 and 2 in
CD3CN, and the rate differences between them when going from
the less polar to the more polar solvent. Overall, the opposing
contributions of enthalpy and entropy result in the observed small
dependence of isomerization rate on solvent polarity.
The activation parameters for the E f Z-Hþ isomerization
process (Table 3) in CH3CN have been determined as well using
the kEfZ rates (Figures S23 and S24 in the SI). As stated earlier,
the rather large negative ΔSq value for both systems is evidence
for the rotation mechanism.24 The lesser negative values mea-
sured for the E f Z-Hþ isomerization process with respect to the
thermal Z f E one indicate that there are fewer reorganization
constraints for the stabilization of the polar transition state. This
may be explained by the fact that the H-bond within the E-Hþ
kinetic intermediates has been already disrupted because of the
protonation of the pyridine moiety. One also has to take into
account the fact that the polarity and ionic strength of CH3CN
might have been altered as a result of the addition of more than
100 equiv of TFA to the solution (See experimental details in the
UV/Vis Analysis. TheEfZ-Hþ isomerizationsofcompounds1
and 2 were probed in CH3CN using a stopped flow spectro-
photometer (Figures S21 and S22 in SI). The isomerization was
easily monitored because of significant spectral variations induced
by the protonation of the pyridine unit and by the subsequent
switching of the corresponding molecules. Kinetic measurements
were carried out under pseudo-first-order conditions ([TFA]tot g
10 ꢁ [1]tot or [2]tot). In all cases, a single-exponential signal versus
time (Figures S21 and S22 in the SI) was recorded when
monitoring the main πꢀπ* absorption band at 409 and 450 nm
for 1 and 2, respectively. These two wavelengths were chosen
because they correspond to the maximum spectral amplitude
difference between the reactants and the products. The mono-
exponential nature of the recorded kinetic signal indicates a first-
order reaction kinetics with respect to 1 or 2. The absorbance
values measured at the beginning and at the end of the experimental
kinetic traces were found to be in agreement with the values
expected from the measurements performed at equilibrium. These
results show that there are no additional fast or slow steps that
might belost while mixingthe reagents (∼3ms). The [TFA]tot was
SI). When comparing the entropies of activation (ΔSq
=
ꢀ15.8(1.9) cal molꢀ1 Kꢀ1 for 1 and ΔSq = ꢀ11.5(1.7) cal
molꢀ1 Kꢀ1 for 2) it becomes clear that 2 has a more stabilized
transition state as a result of its extended π-system. The ΔHq and
Ea values measured for 1 and 2 (Table 3) were found to be within
the error range, which is in agreement with the fact that in the E-
Hþ kinetic intermediates, the H-bonds have been already
dissociated.
On the basis of the kinetic and thermodynamic studies we can
conclude that both the E-Hþ T Z-Hþ and the Z T E isomeriza-
tion processes in these H-bonded arylhydrazone switches go
through a polar transition state. When we couple these results
with our computational studies we can conclude that the isomer-
ization mechanism involves hydrazoneꢀazo tautomerization fol-
lowed by rotation around a CꢀN single bond (Figure 6), and not
the standard rotation around a CdN double bond (Scheme 2). If
we were to only consider the rate data, we would have come to the
wrong conclusion, as it happened with others before us.27
The Catalyzing Effect of NEt3 on the Z-HþfE Isomeriza-
tion Process. We also examined the kinetics of the Z-Hþ f E
isomerization process using UV/vis spectroscopy (Figures S25
varied, and the average pseudo-first-order rate constants (kobs
)
were determined (see SI for more details).55 The variations of the
corresponding kobs with [TFA]tot are given in Figures S21 and S22
in the SI and obey the following relationship:56
a½TFAꢂtot
kobs
¼
þ c
1 þ b½TFAꢂtot
where a = kEHþfZHþ ꢁ KH, b = KH and c = kZHþfEHþ
The E f Z-Hþ reaction is triggered only when the pyridine is
protonated. This acidꢀbase reaction is extremely fast and
reaches equilibrium within the mixing time of the stopped-flow
apparatus (∼3 ms). It can be therefore considered as a fast
9819
dx.doi.org/10.1021/ja200699v |J. Am. Chem. Soc. 2011, 133, 9812–9823

Journal of the American Chemical Society
ARTICLE
and S26 in the SI). Pseudo-first-order conditions ([NEt3]tot g 10
ꢁ [1-Z-Hþ]tot or [2-Z-Hþ]tot) were used for these measure-
ments as well. The “metastable” Z isomer of both compounds 1
and 2, obtained from the deprotonation of the Z-Hþ isomer
could be characterized spectrophotometrically (Figure 4 and
Figure S11 in the SI). A strong hypsochromic shift of about
46 nm was observed upon deprotonation of the pyridinium unit,
and a single rate-limiting step subsequent to this protolytic
equilibrium was measured that was assigned to the Z f E
isomerization reaction. The latter process was accompanied by
bathochromic (Δλ ≈ 12ꢀ14 nm) and hyperchromic shifts that
eventually afforded absorption spectra that entirely matched
those of 1-E or 2-E. The [NEt3]tot were varied, and the average
pseudo-first-order rate constants (kobs) were determined.55 The
corresponding kobs linearly varies with [NEt3]tot with a signifi-
cant intercept at the origin. As we discussed earlier, base can be
used to catalyze the equilibration process in TCAHs
(Scheme 1).20 We can therefore rationalize our kinetic data by
assuming a similar catalyzing influence of the NEt3 base on the Z
f E isomerization process. This actually can explain why during
our NMR studies we barely saw the Z isomer when we
deprotonated the Z-Hþ one by adding NEt3 to the NMR
tube.44 The statistical processing of the kinetic data allowed for
the determination of the second-order rate constants kNEt3 (Mꢀ1
isomer gets protonated in a fast pre-equilibrium to afford the
corresponding E-Hþ kinetic intermediate, which then isomerizes
into the more stable Z-Hþ protonated isomer in a single rate
limiting step.
The corresponding kinetic data, and activation parameters, of
both the forward and backward isomerization processes demon-
strate that the isomerization mechanism involves a polar transi-
tion state. Consequently, the isomerization around the CdN
double bond cannot take place via the lateral inversion mechan-
ism. Thus far, the existence of a polar transition state in these
systems was automatically linked to a rotation mechanism
around a polar CdN double bond.24 However, our calculations
have shown that hydrazoneꢀazo tautomerization followed by
rotation around a CꢀN single bond is the most likely isomeriza-
tion mechanism in these H-bonded systems. This mechanism
may apply generally to other hydrazones as well.
This study shows that for an accurate determination of the
isomerization mechanism in hydrazones, and subsequently imi-
nes, it is important to study the contribution of enthalpy and
entropy to the Gibbs energy of activation, and not rely solely on
isomerization rates. We are currently working on taking advan-
tage of these insights to better tune the performance of these
hydrazone-based molecular switches. Now that the nature of the
isomerization mechanism has been established, we can use these
rotary switches10 as the basis of molecular rotors and motors that
can be incorporated and used in functional materials.57
s
ꢀ1) that correspond to the bimolecular Z f E isomerization
pathway with catalysis by NEt3 for 1 (kNEt3 = 1.0 ( 0.1 Mꢀ1 sꢀ1
)
and 2 (kNEt3 = 7.2 ( 0.8 Mꢀ1 sꢀ1). The intercept at the origin is
assumed to correspond to the first-order rate constant for the
uncatalyzed pathway and was found to be equal to kZfE = 2.1 (
0.9 ꢁ 10ꢀ3 sꢀ1 for 1 and kZfE = 3.7 ( 0.3 ꢁ 10ꢀ3 sꢀ1 for 2 under
our experimental conditions. It is noteworthy that these values
are about 1 order of magnitude higher than those obtained by 1H
NMR spectroscopy studies (Table 1). This discrepancy can be
easily explained if one takes into account the fact that in the UV/
vis spectroscopy studies the solutions contained large excess of
’ ASSOCIATED CONTENT
S
Supporting Information. Experimental procedures,
b
NMR spectra of key compounds, UV/vis absorption spectra,
kinetic and thermodynamic measurement data, and computa-
tional details. These materials are available free of charge via the
Internet at http://pubs.acs.org.
the salt Et3HNþ CF3CO2ꢀ (a large excess (100 equiv) of TFA
’ AUTHOR INFORMATION
3
was used to generate the Z-Hþ isomers, and as a result, at least
the same amount of NEt3 was added to neutralize the solution).
These large amounts of acid and base can significantly change the
ionic strength of the solvent and, hence, influence the isomeriza-
tion rates. These additional kinetic data once again show that,
even in the presence of a catalyzing base, the rate measured for 2
(kNEt3) is significantly higher than that of 1, thus lending further
support to the involvement of a polar transition state in the
isomerization process.
Corresponding Author
ivan.aprahamian@dartmouth.edu
’ ACKNOWLEDGMENT
This work was supported by Dartmouth College, the Burke
Research Initiation Award, the Centre National de la Recherche
Scientifique (CNRS) and the University of Strasbourg (UMR
7509 CNRS-UdS) in France. We thank Wayne T. Casey for the
help with NMR spectroscopy and Dr. Richard Staples (Michigan
State University) for X-ray analysis. Computational facilities were
funded by grants from ARO-DURIP and ONR-DURIP.
’ CONCLUSIONS
We have synthesized and characterized a novel hydrazone-
based molecular switch and thoroughly studied the isomerization
mechanism of this phenyl-containing system and of a previously
reported naphthyl-based one. These unique switches represent
the initial steps toward chemically activated configurational
rotary switches. The switching process takes place with a change
in pH; the E-isomer undergoes protonation with TFA that leads
to isomerization around the CdN bond. When the protonated
compound Z-Hþ is passed through a plug of K2CO3 a mixture of
isomers (E and Z) is obtained that equilibrates with time to
reinstate the initial isomer ratio. On the other hand NEt3
catalyzes the isomerization process and the Z isomer cannot be
observed by 1H NMR spectroscopy. The E-Hþ T Z-Hþ process
was also studied using spectrophotometric methods. The E
’ REFERENCES
(1) (a) Feringa, B. L., Ed. Molecular Switches; Wiley-VCH:
Weinheim, Germany, 2001. (b) Kelly, T. R., Ed. Top. Curr. Chem.
2005, 262, volume on Molecular Machines. (c) Khuong, T V.; Nunez,
J. E.; Godinez, C. E.; Garcia-Garibay, M. A. Acc. Chem. Res. 2006,
39, 413–422. (d) Kay, E. R.; Leigh, D. A.; Zerbetto, F. Angew. Chem.,
Int. Ed. 2007, 46, 72–191. (e) Champin, B.; Mobian, P.; Sauvage, J.-P.
Chem. Soc. Rev. 2007, 36, 358–366. (f) Astumian, R. D. Phys. Chem.
Chem. Phys. 2007, 9, 5067–5083.(g) Balzani, V.; Credi, A.; Venturi,
M. Molecular Devices and Machines: Concepts and Perspectives for the
Nanoworld; Wiley-VCH: Weinheim, Germany, 2008. (h) Vives, G.;
Tour, J. M. Acc. Chem. Res. 2009, 42, 473–487. (i) Stoddart, J. F.
Chem. Soc. Rev. 2009, 38, 1802–1820.
9820
dx.doi.org/10.1021/ja200699v |J. Am. Chem. Soc. 2011, 133, 9812–9823

Journal of the American Chemical Society
ARTICLE
(2) Molecular Motors; Schilwa, M., Ed.; Wiley-VCH: Weinheim,
Germany, 2003.
(10) Kottas, G. S.; Clarke, L. I.; Horinek, D.; Michl, J. Chem. Rev.
2005, 105, 1281–1376.
(3) (a) Al-Atar, U.; Fernandes, R.; Johnsen, B.; Baillie, D.; Branda,
N. R. J. Am. Chem. Soc. 2009, 131, 15966–15967. (b) Elbaz, J.; Wang,
Z.-G.; Orbach, R.; Willner, I. Nano Lett. 2009, 9, 4510–4514. (c) Silvi, S.;
Constable, E. C.; Housecroft, C. E.; Beves, J. E.; Dunphy, E. L.;
Tomasulo, M.; Raymo, F. M.; Credi, A. Chem.—Eur. J. 2009,
15, 178–185. (d) Wang, X.; Zhu, J.; Smithrud, D. B. J. Org. Chem.
2010, 75, 3358–3370. (e) Gassensmith, J. J.; Matthys, S.; Lee, J.-J.;
Wojcik, A.; Kamat, P.; Smith, B. D. Chem.—Eur. J. 2010, 16, 2916–2921.
(f) Davis, J. J.; Orlowski, A. G.; Rahman, H.; Beer, P. D. Chem. Commun
2010, 54–63. (g) Russew, M.-M.; Hecht, S. Adv. Mater. 2010,
22, 3348–3360.
(4) (a) Amabilino, D. B.; Ashton, P. R.; Balzani, V.; Brown, C. L.;
Credi, A.; Frechet, J. M. J.; Leon, J. W.; Raymo, F. M.; Spencer, N.;
Stoddart, J. F.; Venturi, M. J. Am. Chem. Soc. 1996, 118, 12012–12020.
(b) Collin, J.; Gavina, P.; Sauvage, J.-P. New J. Chem. 1997, 21, 525–528.
(c) Lane, A. S.; Leigh, D. A.; Murphy, A. J. Am. Chem. Soc. 1997,
119, 11092–11093. (d) Loeb, S. J.; Wisner, J. A. Chem. Commun
2000, 1939–1940. (e) Brouwer, A. M.; Frochot, C.; Gatti, F. G.; Leigh,
D. A.; Mottier, L.; Paolucci, F.; Roffia, S.; Wurpel, G. W. Science 2001,
291, 2124–2128. (f) Jeppesen, J. O.; Nielsen, K. A.; Perkins, J.; Vignon,
S. A.; Di Alberto, F.; Ballardini, R.; Gandolfi, M. T.; Venturi, M.; Balzani,
V.; Becher, J.; Stoddart, J. F. Chem.—Eur. J. 2003, 9, 2982–3007.
(g) Badjiꢀc, J. D.; Balzani, V.; Credi, A.; Silvi, S.; Stoddart, J. F. Science
2004, 303, 1845–1849. (h) Kang, S.; Vignon, S. A.; Tseng, H.; Stoddart,
J. F. Chem.—Eur. J. 2004, 10, 2555–2564. (i) Aprahamian, I.; Miljanic,
O. S.; Dichtel, W. R.; Isoda, K.; Yasuda, T.; Kato, T.; Stoddart, J. F. Bull.
Chem. Soc. Jpn. 2007, 80, 1856–1869. (j) Durola, F.; Lux, J.; Sauvage, J.-
P. Chem.—Eur. J. 2009, 15, 4124–4134.
(5) (a) Jimenez, M. C.; Dietrich-Buchecker, C.; Sauvage, J.-P. Angew.
Chem., Int. Ed. 2000, 39, 3284–3287. (b) Chiu, S.; Rowan, S. J.; Cantrill,
S. J.; Stoddart, J. F.; White, A. J. P.; Williams, D. J. Chem.—Eur. J. 2002,
8, 5170–5183. (c) Berna, J.; Leigh, D. A.; Lubomska, M.; Mendoza,
S. M.; Perez, E. M.; Rudolf, P.; Teobaldi, G.; Francesco, Z. Nat. Mater.
2005, 4, 704–710. (d) Ebron, V. H.; Yang, Z. W.; Seyer, D. J.; Kozlov,
M. E.; Oh, J. Y.; Xie, H.; Razal, J.; Hall, L. J.; Ferraris, J. P.; MacDiarmid,
A. G.; Baughman, R. H. Science 2006, 311, 1580–1583. (e) Stadler, A.;
Kyritsakas, N.; Graff, R.; Lehn, J.-M. Chem.—Eur. J. 2006,
12, 4503–4522. (f) Song, C.; Swager, T. M. Org. Lett. 2008,
10, 3575–3578. (g) Clark, P. G.; Day, M. W.; Grubbs, R. H. J. Am.
Chem. Soc. 2009, 131, 13631–13633. (h) Juluri, B. K.; Kumar, A. S.; Liu,
Y.; Ye, T.; Tang, Y.-W.; Flood, A. H.; Fang, L.; Stoddart, J. F.; Weiss,
P. S.; Huang, T. J. ACS Nano 2009, 3, 291–300. (i) Miwa, K.; Furusho,
Y.; Yashima, E. Nat. Chem. 2010, 2, 444–449.
(6) (a) Chatterjee, M. N.; Kay, E. R.; Leigh, D. A. J. Am. Chem. Soc.
2006, 128, 4058–4073. (b) Serreli, V.; Lee, C.; Kay, E. R.; Leigh, D. A.
Nature 2007, 445, 523–527. (c) Alvarez-Pꢀerez, M.; Goldup, S. M.; Leigh,
D. A.; Slawin, A. M. Z. J. Am. Chem. Soc. 2008, 130, 1836–1838.
(7) (a) Shin, J.; Pierce, N. A. J. Am. Chem. Soc. 2004,
126, 10834–10835. (b) Tian, Y.; He, Y.; Chen, Y.; Yin, P.; Mao, C.
Angew. Chem., Int. Ed. 2005, 44, 4355–4358. (c) Yin, P.; Choi, H. M. T.;
Calvert, C. R.; Pierce, N. A. Nature 2008, 451, 318–322. (d) Green, S. J.;
Bath, J.; Turberfield, A. J. Phys. Rev. Lett. 2008, 101, 238101.
(e) Omabegho, T.; Sha, R.; Seeman, N. C. Science 2009, 324, 67–71.
(f) von Delius, M.; Geertsema, E. M.; Leigh, D. A. Nat. Chem. 2010,
2, 96–101.
(11) (a) Kelly, T. R.; De Silva, S. H.; Silva, R. A. Nature 1999,
401, 150–152. (b) Koumura, N.; Zijlstra, R .W.; van Delden, R. A.;
Harada, N.; Feringa, B. L. Nature 1999, 401, 152–155. (c) Leigh, D. A.;
Wong, J. K. Y.; Dehez, F.; Zerbetto, F. Nature 2003, 424, 174–179.
(d) Hernꢀandez, J. V.; Kay, E. R.; Leigh, D. A. Science 2004,
306, 1532–1537. (e) Fletcher, S. P.; Dumur, F.; Pollard, M. M.; Feringa,
B. L. Science 2005, 310, 80–82. (f) Lin, Y; Dahl, B. J.; Branchaud, B. P.
Tetrahedron Lett. 2005, 46, 8359–8362. (g) Dahl, B. J.; Branchaud, B. P.
Org. Lett. 2006, 8, 5841–5844. (h) Feringa, B. L. J. Org. Chem. 2007,
72, 6635–6652. (i) Kelly, T. R.; Cai, X. L.; Damkaci, F.; Panicker, S. B.;
Tu, B.; Bushell, S. M.; Cornella, I.; Piggott, M. J.; Salives, R.; Cavero, M.;
Zhao, Y.; Jasmin, S. J. Am. Chem. Soc. 2007, 129, 376–386.
(12) By configuration we mean cisꢀtrans isomerism around double
bonds as suggested by convention. We are distinguishing here between
configurational and conformational rotation for the simple fact that
energy barriers associated with these two processes are quite different.
These differences in isomerization rates can be very important in future
applications.
(13) Chaur, M. N.; Collado, D.; Lehn, J.-M. Chem.—Eur. J. 2011,
17, 248–258.
(14) (a) Vicini, P.; Zani, F.; Cozzini, P.; Doytchinova, I. Eur. J. Med.
Chem. 2002, 37, 553–564. (b) Loncle, C.; Brunel, J.; Vidal, N.;
Dherbomez, M.; Letourneux, Y. Eur. J. Med. Chem. 2004,
39, 1067–1071. (c) Savini, L.; Chiasserini, L.; Travagli, V.; Pellerano,
C.; Novellino, E.; Cosentino, S.; Pisano, M. B. Eur. J. Med. Chem. 2004,
39, 113–122. (d) Cocco, M. T.; Congiu, C.; Lilliu, V.; Onnis, V. Bioorg.
Med. Chem. 2005, 14, 366–372. (e) Masunari, A.; Tavares, L. C. Bioorg.
Med. Chem. 2007, 15, 4229–4236. (f) Vicini, P.; Incerti, M.; La Colla, P.;
Loddo, R. Eur. J. Med. Chem. 2009, 44, 1801–1807.
(15) (a) Rowan, S. J.; Cantrill, S. J.; Cousins, G. R. L.; Sanders,
J. K. M.; Stoddart, J. F. Angew. Chem., Int. Ed. 2002, 41, 898–952.
(b) Corbett, P. T.; Leclaire, J.; Vial, L.; West, K. R.; Wietor, J.-L.;
Sanders, J. K. M.; Otto, S. Chem. Rev. 2006, 106, 3652–3711. (c) Lehn,
J.-M. Chem Soc. Rev. 2007, 36, 151–160.(d) Miller, B. L. Dynamic
Combinatorial Chemistry. In Drug Discovery, Bioorganic Chemistry, and
Materials Science; John Wiley & Sons, Inc.: Hoboken, NJ, 2010.
(16) (a) Barboiu, M.; Lehn, J.-M. Proc. Natl. Acad. Sci. U.S.A. 2002,
99, 5201–5206. (b) Stadler, A.-M.; Kyritsakas, N.; Lehn, J.-M. Chem.
Commun 2004, 2024–2025. (c) Hecht, S.; Huc, I., Eds. Foldamers;
Wiley-VCH: Weinheim, Germany, 2007. (d) Stadler, A.-M.; Ramírez, J.;
Lehn, J.-M. Chem.—Eur. J. 2010, 16, 5369–5378.
(17) (a) Petitjean, A.; Khoury, R. G.; Kyritsakas, N.; Lehn, J.-M. J.
Am. Chem. Soc. 2004, 126, 6637–6647. (b) Ulrich, S.; Petitjean, A.; Lehn,
J.-M. Eur. J. Inorg. Chem. 2010, 1913–1928.
(18) McCarthy, C. G. In The Chemistry of the Carbon-Nitrogen Bond;
Patai, S., Ed.; John Wiley and Sons: New York, 1970; pp 392ꢀ399.
(19) Parmerter, S. M. In Organic Reactions; Adam, R., Ed.; Wiley-
VCH: New York, 1959; Vol. 10, p 1.
(20) (a) Yao, H. C. J. Org. Chem. 1964, 29, 2959–2963. (b) Bose,
A. K.; Kugajevsky, I. Tetrahedron 1967, 23, 1489–1497. (c) Karabatsos,
G. J.; Taller, R. A. Tetrahedron 1968, 24, 3923–3937. (d) Mitchell, A. D.;
Nonhebel, D. C. Tetrahedron Lett. 1975, 16, 3859–3862. (e) Mitchell,
A. D.; Nonhebel, D. C. Tetrahedron 1979, 35, 2013–2019. (f) Bertolasi,
V.; Ferretti, V.; Gilli, P.; Issa, Y. M.; Sherif, O. E. J. Chem. Soc., Perkin
Trans. 2 1993, 2223–2228.
(8) (a) Gu, H.; Chao, J.; Xiao, S.-J.; Seeman, N. C. Nature 2010,
465, 202–205. (b) Lund, K.; Manzo, A. J.; Dabby, N.; Michelotti, N.;
Johnson-Buck, A.; Nangreave, J.; Taylor, S.; Pei, R.; Stojanovic, M. N.;
Walter, N. G.; Winfree, E.; Yan, H. Nature 2010, 465, 206–210.
(9) (a) Collier, C.; Mattersteig, G.; Wong, E.; Luo, Y.; Beverly, K.;
Sampaio, J.; Raymo, F.; Stoddart, J.; Heath, J. R. Science 2000,
289, 1172–1175. (b) Flood, A.; Stoddart, J.; Steuerman, D.; Heath,
J. R. Science 2004, 306, 2055–2056. (c) Green, J. E.; Choi, J. W.; Boukai,
A.; Bunimovich, Y.; Johnston-Halperin, E.; DeIonno, E.; Luo, Y.; Sheriff,
B. A.; Xu, K.; Shin, Y. S.; Tseng, H.; Stoddart, J. F.; Heath, J. R. Nature
2007, 445, 414–417.
(21) (a) Zollinger, H. Colour Chemistry. Synthesis, Properties and
Application of Organic Dyes and Pigments; VCH: Weinheim, 1991; pp
45ꢀ68. (b) Kelemen, J.; Moss, S.; Sauter, H.; Winkler, T. Dyes Pigm.
1982, 3, 27–47. (c) Hihara, T.; Okada, Y.; Morita, Z. Dyes Pigm. 2006,
69, 151–176. (d) Lee, H. Y.; Song, X.; Park., H.; Baik, M.-H.; Lee, D. J.
Am. Chem. Soc. 2010, 132, 12133–12144.
(22) (a) Kessler, H. Tetrahedron 1974, 30, 1861–1870. (b) Padwa,
A.; Albrecht, F. J. Am. Chem. Soc. 1974, 96, 4849–4857. (c) Padwa, A.
Chem. Rev. 1977, 77, 37–68.
(23) (a) Curtin, D. Y.; Grubbs, E. J.; McCarty, C. G. J. Am. Chem. Soc.
1966, 88, 2775–2786. (b) Kerek, F.; Ostrogovich, G.; Simon, Z. J. Chem.
9821
dx.doi.org/10.1021/ja200699v |J. Am. Chem. Soc. 2011, 133, 9812–9823

Journal of the American Chemical Society
ARTICLE
Soc. B 1971, 541–544. (c) Raban, M.; Carlson, E. J. Am. Chem. Soc. 1971,
93, 685–691. (d) Hall, G. E.; Middleton, W. J.; Roberts, J. D. J. Am.
Chem. Soc. 1971, 93, 4778–4781. (e) Galvez, J.; Guirado, A. J. Comput.
Chem. 2010, 31, 520–531 and references therein.
(24) For examples of the rotation mechanism in hydrazones,
please see: (a) Tobin, J. C.; Hegarty, A. F.; Scott, F. L. J. Chem. Soc. B
1971, 2198–2202. (b) Wong, J. L.; Zady, M. F. J. Org. Chem. 1975,
40, 2512–2516. (c) Pichon, R.; Saint, J.; Le; Courtot, P. Tetrahedron
1981, 37, 1517–1524.
Å for 2-E. The length of the imino double bond (CdN) is 1.305(1) Å for
1-E and 1.313(1) Å for 2-E.
(33) (a) Kaberia, F.; Vickery, B.; Willey, G. R.; Drew, M. G. B. J.
Chem. Soc., Perkin Trans. 2 1980, 1622–1626. (b) Vickery, B.; Willey,
G. R.; Drew, M. G. B. J. Chem. Soc., Perkin Trans. 2 1981, 155–160.
(c) Vickery, B.; Willey, G. R.; Drew, M. G. B. J. Chem. Soc., Perkin Trans. 2
1981, 1454–1458. (d) Drew, M. G. B.; Vickery, B.; Willey, G. R. J. Chem.
Soc., Perkin Trans. 2 1982, 1297–1303. (e) Pavloviꢀc, G.; Racanꢀe, L.;
ꢁ
Ciꢁcak, H.; Traliꢀc-Kulenoviꢀc, V. Dyes Pigm. 2009, 83, 354–362.
(25) For examples of the inversion mechanism in hydrazones,
please see: (a) Shvo, Y.; Nahlieli, A. Tetrahedron Lett. 1970,
49, 4273–4274. (b) Kessler, H.; Leibfritz, L. Liebigs Ann. Chem. 1970,
737, 53–60. (c) Bensai, R.; Taddei, F. J. Chem. Soc., Perkin Trans. 2
1985, 1629–1632. (d) Timpe, H. J.; Mueller, U.; Schuetz, R. J. Prakt.
Chem. 1986, 328, 181–189. (e) Singh, D. K.; Balasubrahmanyam, S. N.;
Prasad, N. Indian J. Chem., Sect. B 1990, 29B, 804–810. (f) Pihlaja, K.;
Simeonov, M. F.; Fueloep, F. J. Org. Chem. 1997, 62, 5080–5088.
(g) Srivastava, A.; Srivastava, V.; Verma, S. M. Indian J. Chem., Sect. B
1997, 36B, 236–241. (h) Ramanathan, S.; Lemal, D. M. J. Org. Chem.
2007, 72, 1566–1569.
(26) For some examples of conflicting reports on imine isomeriza-
tion mechanism, please see: (a) Marullo, N. P.; Wagener, E. H. J. Am.
Chem. Soc. 1966, 88, 5034–5035. (b) V€ogtle, F.; Mannschr, A.; Staab,
H. A. Justus Liebigs Ann. Chem. 1967, 708, 51–56. (c) Marullo, N. P.;
Wagener, E. H. Tetrahedron Lett. 1969, 10, 2555–2558. (d) Kessler, H.;
Leibfritz, L. Tetrahedron Lett. 1970, 1423–1426. (e) Leibfritz, L.; Kessler,
H. Chem. Commun. 1970, 655–657.
(27) For some examples of conflicting reports on hydrazone iso-
merization mechanism, please see ref 25a and: (a) Kalinowski, H.-O.;
Kessler, H.; Leibfritz, D.; Pfeffer, A. Chem. Ber 1973, 106, 1023–1032.
(b) Stassinopoulou, C. I.; Zioudrou, C.; Karabatsost, G. J. Tetrahedron
Lett. 1972, 13, 3671–3674. (c) Stassinopoulou, C. I.; Zioudrou, C.;
Karabatsost, G. J. Tetrahedron 1976, 32, 1147–1151. (d) Lu, S.-I.
J. Comput. Chem. 2009, 30, 2176–2180.
(28) (a) Landge, S. M.; Aprahamian, I. J. Am. Chem. Soc. 2009,
131, 18269–18271. (b) Su, X.; Aprahamian, I. Org. Lett. 2011, 13, 30–33.
(c) Su, X.; Robbins, T. F.; Aprahamian, I. Angew. Chem., Int. Ed. 2011,
50, 1841–1844.
(29) For some examples of pH activated molecular switches,
please see:(a) Dolain, C.; Maurizot, V.; Huc, I. Angew. Chem., Int. Ed.
2003, 42, 2738–2740. (b) Keaveney, C. M.; Leigh, D. A. Angew. Chem.,
Int. Ed. 2004, 43, 1222–1224. (c) Cheng, K.-W.; Lai, C.-C.; Chiang, P.-
T.; Chiu, S.-H. Chem. Commun. 2006, 2854–2856. (d) Silvi, S.; Arduini,
A.; Pochini, A.; Secchi, A.; Tomasulo, M.; Raymo, F. M.; Baroncini, M.;
Credi, A. J. Am. Chem. Soc. 2007, 129, 13378–13379. (e) Vella, S. J.;
Tiburcio, J.; Loeb, S. J. Chem. Commun. 2007, 4752–4754. (f) Silvi, S.;
Arduini, A.; Pochini, A.; Secchi, A.; Tomasulo, M.; Raymo, F. M.;
Baroncini, M.; Credi, A. J. Am. Chem. Soc. 2007, 129, 13378–13379.
(g) Uchiyama, S.; Iwai, K.; de Silva, A. P. Angew. Chem., Int. Ed. 2008,
47, 4667–4669. (h) Fang, L.; Hmadeh, M.; Wu, J.; Olson, M. A.; Spruell,
J. M.; Trabolsi, A.; Yang, Y.-Y.; Elhabiri, M.; Albrecht-Gary, A.-M.;
Stoddart, J. F. J. Am. Chem. Soc. 2009, 131, 7126–7134. (i) Elbaz, J.;
Shimron, S.; Willner, I. Chem. Commun. 2010, 1209–1211. (j) Leblond,
J.; Gao, H.; Petitjean, A.; Leroux, J.-C. J. Am. Chem. Soc. 2010,
132, 8544–8545.
(30) When viewed along the central NꢀN bond, the torsional angle
is ꢀ176.73(10)ꢀ for 1-E and ꢀ179.81(10)ꢀ for 2-E, and between the
NpyrꢀCpyrꢀCimindN atoms it is ꢀ3.16(10)ꢀ for 1-E and ꢀ8.23(10)ꢀ
for 2-E,
(31) These types of hydrazones usually adopt planar structures. For
other examples, please see: (a) Vickery, B.; Willey, G. R.; Drew, M. G. B.
J. Chem. Soc., Perkin Trans. 2 1981, 155–160. (b) Drew, M. G. B.;
Vickery, B.; Willey, G. R. Acta Crystallogr., Sect. B 1982, 38, 1530–1535.
(c) Frohberg, P.; Drutkowski, G.; Wagner, C. Eur. J. Org. Chem.
2002, 1654–1663.
(f) Sethukumar, A.; Arul Prakasam, B. J. Mol. Struct. 2010, 963, 250–257.
(34) (a) Bertolasi, V.; Gilli, P.; Ferretti, V.; Gilli, G. J. Am. Chem. Soc.
1991, 113, 4917–4925. (b) Bertolasi, V.; Gilli, P.; Ferretti, V.;ꢃGilli, G.;
ꢁ
Vaughan, K. New J. Chem. 1999, 23, 1261–1267. (c) Simunek, P.;
Svobodovꢀa, M.; Bertolasi, V.; Pretto, L.; Lyꢁcka, A.; Machꢀaꢁcek, V. New J.
Chem. 2007, 31, 429–438.
(35) The reason for this is most probably the unfavorable peri
interaction in the other conformation (rotamer).
(36) Jaguar 7.6 User Manual, release 110; Schrodinger: New York,
2009.
(37) (a) Zhao, Y.; Truhlar, D. G. J. Chem. Theory Comput 2008,
4, 1849–1868. (b) Zhao, Y.; Truhlar, D. G. Theo. Chem. Acc. 2008,
120, 215–241.
(38) Dunning, T .H., Jr. J. Chem. Phys. 1989, 90, 1007–1023.
(39) Zhao, Y.; Truhlar, D. G. J. Phys. Chem. A 2008, 112, 6794–6799.
(40) The error in the calculated absolute nuclear magnetic shifts was
∼1 ppm. However, the error in the relative nuclear magnetic shifts was
observed to be ∼0.1 ppm.
(41) Only 1.0 equiv of TFA is needed to fully protonate the starting
material, ethyl-2-pyridylacetate. Subsequently, when CH3SO3H is used
as the acid, only 1.0 equiv is needed to fully protonate the pyridine
subunit. It is clear from these data that the H-bond with the hydrazone
NH proton is decreasing the basicity of the pyridine nitrogen. We used
the kinetic studies (see the equations provided in the UV/vis analysis
section) to determine the protonation constant (log KH) of the pyridyl
unit in these hydrazone-based systems and found them to be 3.1 ( 0.2
and 2.84 ( 0.2 for 1 and 2, respectively. These values are comparable,
suggesting a weak influence of the aromatic substituents; however, they
are significantly lower than the protonation constant of pyridine in
CH3CN (pKa = 12.33): Kaljurand, I.; Rodima, T.; Leito, I.; Koppel, I. A.;
Schwesinger, R. J. Org. Chem. 2000, 65, 6202–6208.
(42) Tables of Spectral Data for Structure Determination of Organic
Compounds; Pretsch, E., Ed.; Springer-Verlag: Berlin, Germany, 1989;
Vol. 2.
(43) In ethyl-2-pyridylacetate this same proton resonates at 8.50
ppm and shifts to 8.70 ppm upon protonation with TFA. This indicates
that the shift to 8.76 ppm in 1-E has to do with the H-bonding and
RAHB. In 1-Z the pyridine is not H-bonded, and proton H1 resonates at
8.55 ppm, clearly showing the effect of the H-bond on this proton.
(44) When NEt3 is used as the base, the fully equilibrated spectrum
is generated instantaneously, and this process is not observed.
(45) The bond length between (i) the aromatic carbon C(1) and
nitrogen N(2) is 1.396(1) Å for 1-Z-Hþand 1.395(1) Å for 2-Z-Hþ);
(ii) nitrogens N(2) and N(1) is 1.317(1) Å for 1-Z-Hþand 1.319(1) Å
for 2-Z-Hþ; and (iii) the CdN double bond is 1.305(1) Å for 1-Z-
Hþand 1.301(1) Å for 2-Z-Hþ, whereas the torsional angle between the
N(1)dCiminꢀCdO atoms is 1.49(10)ꢀ for 1-Z-Hþand 9.61(10)ꢀ for 2-
Z-Hþ.
(46) Regardless of the isomerization or the protonation states of the
hydrazone-based switches 1 and 2, the main absorption of 2 is constantly
10ꢀ20% less intense than that of 1.
(47) It is noteworthy that, at the beginning of the spectrophotometric
titrations of 1-E and 2-E by TFA (Figures S15 and S16 in the SI),
unexpected spectral variations were observed. These changes were
attributed to the initial protonation of the Z isomer, which has more
basic pyridyl nitrogen, as it is not involved in H-bonding as is the case in
the E isomer. This observation supports the presence ofa mixture ofZ and
E isomers in solution, in agreement with the NMR data. The E isomers
remain by far the predominant species in solution (estimated >95%).
(32) The bond length between the aromatic carbon and the
hydrazone nitrogen (N2) is 1.401(1) Å for 1-E and 1.400(1) Å for 2-
E, and the N(1)ꢀN(2) bond length is 1.321(1) Å for 1-E and 1.326(1)
9822
dx.doi.org/10.1021/ja200699v |J. Am. Chem. Soc. 2011, 133, 9812–9823

Journal of the American Chemical Society
ARTICLE
(48) (a) Frost, A. A.; Pearson, R. G. Kinetics and Mechanism; John
Wiley and Sons, Inc.: New York, NY, 1953. (b) Connors, K. A. Chemical
Kinetics: The Study of Reaction Rates in Solution; John Wiley and Sons,
Inc.: New York, 1990; pp 60ꢀ61.
(49) The NMR spectroscopy measurements give kobs which consists
of two contributions: kEfZ þ kZfE. The former is faster by 2 orders of
magnitude from the latter, and hence, its contribution to kobs is minimal.
Therefore, we can assume that kobs ≈ kZfE. Nevertheless, the kZfE
values were used for the different calculations.
(50) The isomerization rate previously reported for 2 in CD3CN
(see ref 28a) was overestimated because base catalysis was not taken into
account.
(51) When studying mechanisms that involve polar transition states,
it is not enough to only use the dielectric constant values of the solvents
in assessing the dependence of rates on solvent polarity. The Dim-
rothꢀReichardt parameter ET is an empirical measure of solvent
polarity, which is based on the solvatochromism of a pyridinium
phenoxide. This parameter mainly reflects the electrophilicity and the
H-bond donor activity of the solvent and is generally used when
intramolecular bonded systems are studied. (a) Dimroth, K.; Reichardt,
C.; Siepmann, T.; Bohlmann, F. Liebigs Ann. Chem. 1963, 661, 1–37.
(b) Reichardt, C. Angew. Chem., Int. Ed. 1965, 4, 29–40. (c) Reichardt,
C.; Dimroth, K. Fortschr. Chem. Forsch. 1968, 11, l–73.
(52) The proposed tautomerization is similar to the one that takes
place in 2-pyridone: Wong, M. W.; Wiberg, K. B.; Frisch, M. J. J. Am.
Chem. Soc. 1992, 114, 1645–1652.
(53) Given that the accuracy of our computational method is likely
to be around ∼1 kcal/mol, we believe that our calculations are not able
to distinguish the difference between the barriers of 1 and 2.
(54) Reed, A. E.; Curtiss, L. A.; Weinhold, F. Chem. Rev. 1988,
88, 899–926.Glendening, E. D.; Badenhoop, J, K.; Reed, A. E.; Carpen-
ter, J. E.; Bohmann, J. A.; Morales, C. M.; Weinhold, F. NBO 5.0;
Theoretical Chemistry Institute, University of Wisconsin: Madison,
2001.
(55) Biokine, User’s Manual; Bio-Logic Company: Echirolles, France,
1991.
(56) The ordinate at the origin of the fitted curves in Figures S21 and
S22 in SI is very close to zero, which means that the backward reaction
Z-Hþ f E-Hþ is a very minor process. This makes sense as E-Hþ is
most likely an unstable species. Thus, kZHþfEHþ , kEHþfZHþ and so
we can assume that kobs ≈ kEHþfZHþ. For this reason we have used c = 0
in our calculations.
(57) Michl, J.; Sykes, E. C. H. ACS Nano 2009, 1042–1048.
9823
dx.doi.org/10.1021/ja200699v |J. Am. Chem. Soc. 2011, 133, 9812–9823