Symmetric halogen bonding is preferred in solution

Article abstract of DOI:10.1021/ja301341h

Halogen bonding is a recently rediscovered secondary interaction that shows potential to become a complementary molecular tool to hydrogen bonding in rational drug design and in material sciences. Whereas hydrogen bond symmetry has been the subject of sys

Full text of DOI:10.1021/ja301341h

Article
pubs.acs.org/JACS
Symmetric Halogen Bonding Is Preferred in Solution
Anna-Carin C. Carlsson,† Jurgen Grafenstein,† Adnan Budnjo,†,∥ Jesse L. Laurila,†,§ Jonas Bergquist,‡
̈
̈
Alavi Karim,† Roland Kleinmaier,† Ulrika Brath,† and Mate Erdelyi*,†,#
́
́
́
†Department of Chemistry and Molecular Biology, University of Gothenburg, SE-412 96 Gothenburg, Sweden
‡Analytical Chemistry, Department of Chemistry - Biomedical Center, Uppsala University, SE-751 24, Uppsala, Sweden
#The Swedish NMR Centre, Medicinaregatan 5c, SE-413 90 Gothenburg, Sweden
S
* Supporting Information
ABSTRACT: Halogen bonding is a recently rediscovered secondary
interaction that shows potential to become a complementary molecular tool
to hydrogen bonding in rational drug design and in material sciences. Whereas
hydrogen bond symmetry has been the subject of systematic studies for
decades, the understanding of the analogous three-center halogen bonds is yet
in its infancy. The isotopic perturbation of equilibrium (IPE) technique with
13C NMR detection was applied to regioselectively deuterated pyridine
complexes to investigate the symmetry of [N−I−N]+ and [N−Br−N]+
halogen bonding in solution. Preference for a symmetric arrangement was observed for both a freely adjustable and for a
conformationally restricted [N−X−N]+ model system, as also confirmed by computation on the DFT level. A closely attached
counterion is shown to be compatible with the preferred symmetric arrangement. The experimental observations and
computational predictions reveal a high energetic gain upon formation of symmetric, three-center four-electron halogen bonding.
Whereas hydrogen bonds are generally asymmetric in solution and symmetric in the crystalline state, the analogous bromine and
iodine centered halogen bonds prefer symmetric arrangement in solution.
development, for example. In the assessment of the possible use
of HB and XB as complementary molecular tools, under-
standing their similarities and differences is a prerequisite.
A classic concern regarding HB is its symmetry.37,38
Originating from its proposed relevance for enzyme cataly-
sis,39,40 HB symmetry has been studied intensely for decades.41
Whereas asymmetric in solution ([N−H···N]+ ⇄ [N···H−
N]+),20 hydrogen bonds were shown to be symmetric
([N···H···N]) in crystals.42,43 In contrast, our understanding
of XB symmetry is still in its infancy.44 Its dependency on the
local environment, such as steric and electronic factors, solvent
polarity, and the influence of counterions, has not yet been
addressed.
Herein, we present an investigation of the symmetry of
systems comprised of an electropositive bromine(I)or iodine(I)
bound to two nitrogenous electron donors. In such
[N···X···N]+-type systems, a halogen (X) may either be
centered between the two electron donors [N···X···N]+, or it
may be closer to one of the nitrogens [N−X···N]+. The former,
symmetric system corresponds to two N···X halogen bonds of
equal distance and strength and is characterized by a single-well
energy potential (Figure 1a), whereas the second corresponds
to a covalent N−X bond and a weaker, longer N···X halogen
bond. On the basis of the observations made for closely related
HBs,41 the latter system is expected to exist as a rapidly
interconverting mixture of the nonsymmetric tautomers [N−
INTRODUCTION
■
Halogen bonding (XB) refers to the moderately strong,
directional, noncovalent interaction of an electropositive
halogen (X = Cl, Br, or I) and an electron donating species
(D) that can be neutral or anionic in character.1,2 In the general
definition Y−X···D,1 X is covalently attached to atom Y (Y = C,
N, X, etc.) and serves as Lewis acid that accepts electron
density from the Lewis base D. The Y−X···D angle is
approximately 180°, and the distance between the halogen
bonding atoms (X···D) is shorter than the sum of their van der
Waals radii. XB interactions were observed first in the 19th
century by Guthrie,3 and were investigated a hundred years
later in the solid state by Hassel.4,5 This phenomenon had been
neglected for decades, and has only after its recent rediscovery
received immense interest,6−8 with two decades of studies
resulting in applications in a variety of research fields, for
example, crystal engineering,1,9−12 supramolecular chemistry,13
polymer sciences,14 liquid crystals,8,15,16 conductive materi-
als,9,17,18 and medicinal chemistry.10−12,19,20 Halogen bonds
have so far mostly been studied in the solid and gaseous
phases,30,31 and in silico.21−25 Lately, the exploration of their
applicability in biological systems has started,11,19,20,26−28 yet so
far only a handful of studies of XB have been carried out in
solution.29−36
The strong resemblance of XB to hydrogen bonding (HB)
has been frequently emphasized.1 As HB is by far the most
commonly used secondary interaction for directing molecular
recognition processes, reaching an improved understanding of
XB is expected to open up exciting opportunities for drug
Received: February 9, 2012
Published: March 5, 2012
© 2012 American Chemical Society
5706
dx.doi.org/10.1021/ja301341h | J. Am. Chem. Soc. 2012, 134, 5706−5715

Journal of the American Chemical Society
Article
optimal N−N distance for N−X−N interactions, and 2a,b is
thereby expected to possibly promote nonsymmetric geo-
metries. Symmetric complexes of 2 may be achieved by
energetically penalized adjustment of its covalent backbone or
by alteration of the N−X distances, which in turn prevent
optimal orbital overlap. On the other hand, 1a,b permits an N−
X−N axial rotation of the pyridine rings,44 whereas the
aromatic rings of 2a,b are forced to maintain coplanarity, which
may facilitate conjugation of the pyridine π-system to a p-
orbital of the coordinating halogen.48 Here, the investigation of
the effect of geometric restraints on halogen bond symmetry is
presented. Data for the corresponding bis(pyridinium) triflate
HB complexes 1c and 2c (Figure 2), which are known to be
asymmetric,38,44 [N−H···N]+ ⇄ [N···H−N]+, are included for
comparison.
Figure 1. Energy potentials for the halogen motion in an N−X−N
system: (a) single-well, and (b) double-well. The potential energy
variation is shown as a function of the nitrogen−halogen distance.
Here, N−X refers to a covalent, N···X to a secondary bond.
X···N]+ and [N···X−N]+, which correspond to the two energy
minima of a double-well potential (Figure 1b). Originating
from the shallow energy barrier between these geometries and
the rapidity of the chemical exchange process, most standard
spectroscopic methods give coalesced signals for such
equilibrating tautomers and are thereby incapable of differ-
entiation between a static symmetric structure, [N···X···N]+,
and a rapidly equilibrating pair of asymmetric geometries, [N−
X···N]+ ⇄ [N···X−N]+. Isotopic substitution causes a slight
perturbation of equilibrium processes, but it does not affect
static geometries and it can therefore be utilized for
differentiation of such systems using NMR spectroscopic
detection.38 Model systems 1a,b and 2a,b (Figure 2) comprised
RESULTS AND DISCUSSION
■
For symmetry investigations in solution applying the NMR
technique isotopic perturbation of equilibrium (IPE), mixtures
of selectively deuterated and nondeuterated compounds are
required,38,49 and were prepared following the procedures
summarized below.
Synthesis. For the preparation of nondeuterated and
deuterated analogues of bis(pyridine)halonium triflates 1a,b
and 1a,b-d the synthetic protocol of Brown et al. was
followed.50 Pyridine and pyridine-2-d in dry dichloromethane
were reacted with Br2 or I2 in the presence of silver triflate,
affording mixtures of [N−Br−N]+ complexes 1a/1a-d or [N−
I−N]+ complexes 1b/1b-d in high yields. Nondeuterated 1,2-
bis(2-pyridylethynyl)benzene halonium triflates 2a,b were
synthesized following the route depicted in Scheme 1.
Microwave-assisted Sonogashira coupling of diiodobenzene
(4) and 2 equiv of 2-ethynylpyridine (5) afforded compound 3
in high yield.51 Formation of the [N−X−N]+ triflate complex
with Br2 or I2 in the presence of silver triflate provided the
bromine(I) and the iodine(I) complexes 2a and 2b,
respectively, in moderate to high yields.50 For the generation
of the deuterated complexes 2a,b-d, first regioselective
deuteration of 2-chloropyridine (6) at C6 was achieved using
BuLi-Li-DMAE.52 Subsequent rapid Sonogashira coupling with
TMS-acetylene, followed by KF-mediated deprotection53 of the
alkyne resulted in 6-deutero-2-ethynylpyridine (5-d). Sonoga-
shira coupling of 5-d and 8 furnished 3-d in moderate yield.
The selectively deuterium-labeled [N−X−N]+ complexes 2a,b-
d were obtained using the conditions given above for 2a,b.
NMR Spectroscopic Studies. Isotopic perturbation of
equilibrium (IPE) is a sensitive NMR spectroscopic method
that relies on the observation of secondary isotope effects for
distinguishing a single symmetric molecule from rapidly
interconverting asymmetric tautomers.49 Its major advantage
is that it succeeds even when the rapid exchange causes the
NMR signals from the individual tautomers to coalesce into
one set of signals, which in turn are virtually identical to those
of a single static structure. IPE has been applied successfully, for
example, to symmetry determinations of carbocations,54 O−
H···O55,56 and N−H···N57,58 HB molecular systems, thiapente-
lenes,59 and metal chelating complexes.60 It requires sub-
stitution of a proton with deuterium (or, more generally, one
isotope with another, e.g.,16O with 18O)61 close to the
interaction site, causing changes in the vibration spectrum of
the molecule, which eventually affect the observed vibrationally
averaged NMR parameters.62 IPE relies most commonly on 13C
NMR spectroscopic detection and on the analysis of mixtures
Figure 2. Whereas 1 permits free adjustment of N−X distances for
most favorable XB or HB interactions, the steric requirements of 2
enforce nonoptimal geometries for a symmetric N−X−N interaction.
For 1a and 2a, X = Br; for 1b and 2b, X = I; for 1c and 2c, X = H.
of two pyridine-type Lewis bases binding an electropositive
iodine or bromine were studied. They permit linear N−X−N
geometry45,46 necessary for XB,1 but differ in the ease of
availability of N···X distances optimal for two symmetric XB
interactions. Importantly, 1a,b and 2a,b should not be confused
with iodonium or bromonium salts as they consist of a halogen
and two pyridines contributing to the N−X bond by a single
electron each.47,48 Hence, the question addressed here is
whether the interactions of Br(I) and I(I) to the two
coordinating nitrogens are identical, interpretable as two
N···X halogen bonds (Figure 1a), or if one of the N−X
bonds is stronger than the other, yielding an asymmetric
complex having one classical covalent, N−X, and one classical
N···X halogen bond (Figure 1b). Whereas 1a,b may freely
adjust to allow for the energetically most favorable interaction,
symmetric or asymmetric, the geometrical restriction of the 1,2-
diethynylbenzene moiety of 2a,b offers a slightly longer than
5707
dx.doi.org/10.1021/ja301341h | J. Am. Chem. Soc. 2012, 134, 5706−5715

Journal of the American Chemical Society
Article
a
Scheme 1. Outline of the Synthesis of 2 and 2-d (for a, X = Br; for b, X = I)
a
Reagents and conditions: (a) 2.3 equiv 5, Pd(PPh3)2Cl2, CuI, Et2NH, DMF, MW, 120 °C, 9 min; (b) Br2 or I2, AgOTf, dry CH2Cl2, rt, N2 or Ar,
10-15 min; (c) 0,7 equiv 5, Pd(PPh3)2Cl2, CuI, Et2NH, DMF, MW, 120 °C, 4 min; (d) Pd(PPh3)2Cl2, CuI, Et2NH, DMF, MW, 120 °C, 13 min; (e)
(1) n-BuLi, Li-DMAE, dry n-hexane, −78 °C, N2, 1 h; (2) MeOD, −78 − 0 °C, 30 min; (f) TMS-acetylene, Pd(PPh3)2Cl2, CuI, PPh3, Et2NH, DMF,
MW, 120 °C, 27 min; (g) 7-d, KF, MeOH, rt, 16 h.
n
of a nondeuterated molecule and its deuterated analogue. In
such a mixture, each observed NMR signal turns into two, one
originating from the deuterated molecule and one from the
corresponding nondeuterated one, with a small shift difference
nΔobs, between them, where n denotes the number of
intervening bonds between the site of the 1H-to-2H substitution
and the investigated carbon (eq 1).
second component, Δeq, is present only for rapidly
tautomerizing systems. Isotope substitution slightly changes
the vibrational zero-point energies (ZPE) of the two tautomers,
and according to Boltzmann’s law, the equilibrium constant K
n
between the tautomers. This in turn gives a contribution Δeq
to the isotopic shift that is significant and determined by eq 3:
n
Δ
= D(K − 1)/[2(K + 1)]
eq
(3)
n
Δ
= δ
C(D)
− δ
C(H)
obs
(1)
where D equals the chemical shift difference between the
signals of the tautomeric forms (halogenated N+−X and
nonhalogenated N, in this particular case). It is noteworthy that
sizable nΔeq need not be restricted to small n. According to the
van’t Hoff equation,63 K is temperature dependent, and so is
The observed shift difference consists of two independent
n
isotope shift contributions: the intrinsic isotope shift Δ0 and
n
the equilibrium isotope shift Δeq (eq 2).
n
n
n
n
therefore Δeq. To induce large isotope effects, selective
Δ
= Δ + Δ
eq
obs
0
(2)
deuterium substitution was performed as close as possible to
the [N−X−N]+ interaction site, at the C2 position of one of
the pyridines of 1a−c and 2a−c (see Figure 2). Here, the
monodeuterated analogues of 1 and 2 are referred to as 1-d and
2-d, respectively.
In both static and rapidly equilibrating systems, the isotope
substitution slightly changes the vibrationally averaged
equilibrium geometry and thus causes changes to the intrinsic
n
shift, Δ0, on all nearby 13C nuclei. It is usually small (<0.4
Isotope shifts nΔobs were obtained by acquisition of 13C {1H,
2H}NMR spectra of the isotopologues of 1 and 2 dissolved in
CD2Cl2. Similar to previous, related41 measurements of
ppm) and negative as the 13C NMR signal corresponding to the
heavier, deuterated molecule occurs at a lower chemical shift.
n
The magnitude of Δ0 attenuates rapidly as n increases. The
5708
dx.doi.org/10.1021/ja301341h | J. Am. Chem. Soc. 2012, 134, 5706−5715

Journal of the American Chemical Society
Article
intrinsic isotope shifts, mixtures of pyridine and pyridine-2-d,
and of 1,2-bis(2′-pyridylethynyl)benzene (3) and its mono-
deuterated analogue 3-d, respectively, were used as these
Table 2. Temperature Coefficients (ppm K) of the Isotope
Shifts Reported in Table 1, Observed for CD2Cl2 Solutions
1
2
3
4
3
structure
C2 Δobs
C3 Δobs
C4 Δobs
C5 Δobs
C6 Δobs
n
represent static structures (nΔobs = Δ0) most similar to the
Pyridine
−5
−3
−5
−6
−7
−7
−10
−5
−5
−6
−6
−10
−9
−7
0
0
+2
0
−3
−3
−3
−7
-
investigated XB systems 1 and 2. Isotopic shifts for the halogen
bonded complexes and for the static reference systems and the
most similar analogous equilibrating systems (N−H−N
complexes,57,58 1c and 2c) are reported in Table 1. Small
1a
1b
1c
2a
2b
2c
3
1
0
−5
−3
−2
−3
−3
−6
0
0
−3
+15
−2
Table 1. Experimental 13C NMR Isotope Shifts (ppb) at 298
K Measured for CD2Cl2 Solutions
−11
−7
0
+2
1
2
3
4
3
structure
C2 Δobs
C3 Δobs
C4 Δobs
C5 Δobs
C6 Δobs
altered polarity of the environment affects the electron density
of the nitrogen lone pair by a dipolar mechanism. Similar
through-space polarization by dipolar interaction has been
reported previously, for example, for ethers.67 As the overlap of
the nitrogen lone pair with the vicinal C−H/C−D bond has a
significant influence on the magnitude of the IE,68 its
temperature dependence indirectly reflects the influence of
temperature on the polarity of the environment.
Pyridine
−341
−307
−336
−333
−315
−334
−299
−333
−140
−139
−145
−126
−138
−145
−133
−135
0
+17
+20
+45
+23
+24
+21
+13
+14
0
−15
−29
−30
−52
−25
−28
−31
−17
1a
1b
1c
2a
2b
2c
3
0
+20
0
0
0
+15
The importance of the local environment on symmetry has
been the subject of debates concerned with HB.37,69 For
example, it was proposed that the symmetry of weakly binding
charged complexes might be affected by the disorder of the
local environment generated by a strongly binding asymmetri-
cally positioned counterion.69,70 Such an effect is expected to
have the strongest influence in nonpolar solvents. To assess the
interaction strength of the [N−X−N]+ XB center with the
triflate counterion, translational diffusion coefficients were
acquired for CD2Cl2 solutions of the investigated molecules.
Diffusion rates for the [N···X···N]+ interaction sites were
equilibrium shifts are expected originating from (a) the very
small difference in the Lewis basicity of the two nitrogens that
is reflected by an equilibrium constant close to unity for the
potential tautomerization process [N−X···N]+ ⇄ [N···X−
N]+,44 and (b) from the comparably small chemical shift
difference (D, eq 3) of the halogenated and nonhalogenated, or
protonated and nonprotonated (ca. 8 ppm at C2),64 tautomeric
states. Such small equilibrium isotope effects do not permit
straightforward, direct differentiation between a static structure
and a tautomerizing system.44 However, the similarity of the
observed isotope effects (IEs) at defined positions for the
corresponding XB complexes (1a−2a, 1b−2b), for the
reference compounds for the static (pyridine, 3) and the
equilibrating systems (1c, 2c) indicate the comparability of the
two types of model systems. Originating from their large
1
observed by H NMR, and that of triflate anion by 19F NMR.
Comparable diffusion coefficients (Table 3) of the correspond-
1
Table 3. 15N NMR Chemical Shifts, and H and 19F NMR
Translational Diffusion Coefficients
1
2
magnitudes, the Δobs and Δobs values suffer least from
measurement errors, and their comparison among the various
compounds is therefore expected to provide the most accurate
conclusions. Yet, the magnitude pattern of the 3−4 bond IEs
may also be informative. In CD2Cl2 solutions, comparable 1Δobs
structure
δ(15N) ppm
D(1H) × 10−10 m2/s
D(19F) × 10−10 m2/s
Pyridine
−67.0
−142.9
−175.1
−134.1
−141.2
−165.0
−137.9
−64.5
30.6
13.9
14.0
20.3
10.7
10.7
11.7
14.5
-
1a
1b
1c
2a
2b
2c
3
14.2
15.0
14.4
10.4
11.7
10.7
-
2
and Δobs were observed for the static reference compounds
pyridine and 3 (Table 1), and for the iodine(I) complexes 1b
and 2b. The bromine(I) complexes 1a and 2a show
consequently the lowest IEs, which are comparable to or
lower than those observed for the equilibrating references 1c
and 2c. For distinguishing between a static geometry and the
corresponding tautomeric mixture (Figure 1), the temperature
dependence of the isotope shifts, expressed as the slopes of
nΔobs versus reciprocal temperature,65 was studied (Table 2).
The small temperature coefficients for the CD2Cl2 solutions of
1a,b are indicative of single symmetric structures, whereas the
higher coefficients of 1c confirm its expected asymmetry. In
similarity to 1a,b, the temperature coefficients of the isotope
shifts of 2a,b are comparable and are closer in magnitude to
those of the static 3, than to those of the tautomeric reference
2c. Thus, IPE analysis indicates single, symmetric structures for
1a,b and 2a,b in CD2Cl2 solution, whereas it confirms the
asymmetry for 1c and 2c.57,58
ing positively and negatively charged species indicate tight
complexes for each investigated system. Hence, despite strong
binding, the triflate ion does not perturb the symmetry of the
[N−X−N]+ XB in apolar CD2Cl2 solutions. This fact is
explained either by a symmetric arrangement of the triflate,
with equal distances to the pyridine nitrogens, or by a close,
asymmetric orientation of the counterion, which is incapable of
breaking the strongly stabilized symmetric [N···X···N]+ halogen
bond. The generally higher diffusion rates of the bipyridyl
complexes, 1a−c, as compared to the 1,2-bis(2′-
pyridylethynyl)benzene complexes, 2a−c, originate from their
smaller size, that is, smaller solvent accessible surface area.
Strong interaction with the counterion without loss of
symmetry suggests a significant energetic gain upon formation
of the static, symmetric 3-center-4-electron XB,48,71 which
The observed temperature dependence of the intrinsic
isotope effects of the static pyridine and 3 is well-explained
by the temperature dependence of solvent polarity.66 Hence,
5709
dx.doi.org/10.1021/ja301341h | J. Am. Chem. Soc. 2012, 134, 5706−5715

Journal of the American Chemical Society
Article
observation is in excellent agreement with the previously
reported exceptionally high strength of analogous halogen and
hydrogen bonds, for example, [I−I···I]− and [F−H···F]− of
160−180 kJ/mol.1,72
As a sensitive indicator for the changes of the nitrogen bond
order and electron density,73 the 15N chemical shifts were
measured for free pyridine and for compounds 1−3 using
1H,15N-HMBC experiments.74,75 For each compound, a single
nitrogen signal was detected (Table 3).
the 3-center-4-electron bonds make a DFT description of 1a−c
and 2a−c challenging, owing on the one hand to the
incomplete description of nondynamic electron correlation in
these bonds,91 and on the other to the self- interaction error
inherent to DFT.92 However, reference calculations44 with
second-order Møller−Plesset perturbation theory (MP2)93 for
1a−c confirm that DFT provides a reasonable description of
these complexes.
Table 4 shows computed N−X and N−N bond distances for
1a,b and 2a,b, along with reference distances calculated for
Upon complexation, alteration of the nitrogen chemical shift
was observed for 1−2, but to a different extent for the
protonated (pyridine−1c, −67 ppm) and the bromine
(pyridine−1a, −76 ppm) and iodine centered (pyridine−1b,
−108 ppm) halogen bonded systems (Table 3). A larger
chemical shift change may be interpretable as a stronger
interaction and may imply an increasing covalent character in
the order H < Br < I. The 15N chemical shift change upon
formation of the bromine centered halogen bond of 1a is
similar to that upon protonation (1c, Table 3), whereas the
iodine centered XB of 1b caused a 15N chemical shift alteration
comparable to the shift change caused by N-alkylation (N-
methylpyridinium iodide, δN15 = −180.5 ppm). This latter
observation may be rationalized in several ways. (a) It may be a
result of stabilization of the [N−I−N]+ interaction by σ-overlap
of an empty iodine p-orbital and the filled nonbonding orbital
of the nitrogens simultaneously with an efficient overlap of a
filled p-orbital of I(I) and the nitrogen π*-orbitals, which are
involved in the aromatic system of the complexing pyridines.48
Electron donation from the pyridine π-orbitals to the bromine
or iodine d-orbitals may also contribute. This explanation is
supported by the coplanarity of bis(pyridine)halonium
complexes in crystals.46,48,50 (b) It may be due to an
exceptionally strong electrostatic Nδ‑···I+ interaction; however,
in light of the noncomparable smaller ΔδN upon protonation of
pyridine, this explanation appears less likely. (c) A large
chemical shift change may also be rationalized as indication of a
stronger halogen bond by resonance stabilization41 or (d) by an
increased covalent character.76 Geometry optimization, dis-
cussed in detail below, predicts dN−I = 2.3 Å for 2b, which is
close to the sum of the covalent radii of N and I (2.1 Å) and
significantly shorter than the sum of their van der Waals radii
(3.5 Å). The ca. 10 ppm higher 15N shift of 2b compared to 1b
may reflect its weaker N−I bonds, interpretable here as the
consequence of the steric restraint introduced by the
diethynylbenzene backbone that strives for longer than optimal
N−N distance for the symmetric [N···I···N]+ complex.
Notably, the XB symmetry of 2b is retained. The 15N chemical
shift changes of 1a and 2a reveal comparable, yet weaker
interactions. A less favorable interaction of the bromine(I)
complex 2a compared to 1a is further indicated by its
apparently lower stability in solutions.
Table 4. Computationally Predicted N−X and N−N
Distances [Å] for the Equilibrium Geometries of 1a−c, 2a−
a
c, and 3
a
Calculations were done at the B3LYP level with the basis set and
ECPs described in the computational section, employing PCM with
CD2Cl2 as solvent. Values in italics refer to unstable symmetric
structures. Those for 1a−c are taken from reference 44.
protonated (1c, 2c) and halogenated pyridine as well as for 3.
DFT calculations predict symmetric geometries for 1a,b and
2a,b (Figure 3, Supporting Information) and asymmetric ones
for 1c and 2c, confirming the experimental findings. The N−Br
and N−I bond distances in 1a,b and 2a,b are less than 0.3 Å
longer than their counterparts in the hypothetic brominated or
iodinated pyridine, which indicates that the N−X bonds in 1a,b
and 2a,b show a significant covalent character. For the stable
asymmetric structures of 1c and 2c, the elongation of the
shorter N−H bond compared to protonated pyridine is less
than 0.07 Å, and thus closely resembles that in protonated
pyridine. A longer N−N distance was predicted for 3 (4.68 Å)
than for the complexes 1a (4.28 Å), 1b (4.60 Å), and, most
notably, 1c (2.58 and 2.80 Å for the symmetric and asymmetric
structure, respectively). As a consequence of the strain provided
by the backbone of 3, longer N−X distances and lower stability
for 2a−c as compared to 1a−c are expected. This conjecture is
only partially confirmed by computation (Table 4). The largest
difference between the complexes of the freely adjustable 1 and
of the rigidified 2 is predicted for the protonated complex. For
Computation. For theoretical confirmation of the exper-
imental results, the equilibrium geometries and energies of 1a−
c,44 2a−c and 3 were calculated using density functional theory
(DFT), employing the B3LYP exchange and correlation
functional.77−80 The LANL08 basis set81 in conjunction with
the LANL2DZ effective core potential82−84 were used for Br
and I, Pople’s 6-311++G(d,p) basis set85−87 for N and the
central H atoms in 2c and 3c, and Pople’s 6-311G(d,p) basis
set86,87 for the remaining atoms. Solvent effects were accounted
for by the Polarizable Continuum Model (PCM),88,89 with
CH2Cl2 as solvent.44 All calculations were done with the
Gaussian09 program package.90 The delocalized electrons in
5710
dx.doi.org/10.1021/ja301341h | J. Am. Chem. Soc. 2012, 134, 5706−5715

Journal of the American Chemical Society
Article
Table 5. Computed Frequencies [cm−1] for the N−X and
a
N−N Stretching Vibrations for 1a−c, 2a−c, and 3
Figure 3. DFT geometry optimization (DFT, B3LYP LANL08)
predicts a static, symmetric [N···X···N]+ 2a complex with dN−Br = 2.1
Å. This distance is between the sum of the covalent (1.9 Å) and the
sum of the van der Waals radii (3.4 Å) of N and Br. Optimized
structures of 2b and 2c are shown in the Supporting Information.
the stable asymmetric structure, the longer N−H bond is
elongated by 0.35 Å in 2c as compared to 1c, and the N−N
distance increases by 0.2 Å. For the symmetric structure of 2c,
the N−H distance increases by 0.03 Å, and the N−N distance
adjusts by 0.045 Å. It should further be noted that, in 2c, the
N−H−N bonds become bent with a bond angle of 165°.
For 2a and 2b, the increase in the N−X and N−N distances
as compared to 1a and 1b is below 0.001 Å. These findings
suggest that 3, in spite of its conjugated covalent nature, is
easily distorted. This proposal is supported by the X-ray
crystallographic studies of related substances.94−96 The
equilibrium geometry of 3 is influenced by Coulomb repulsion
between the lone pairs of the nitrogen atoms, which repulsion
is neutralized upon complexation of a cation in 2a−c. The
strain introduced by rearrangement of 3 from its equilibrium
geometry to the geometries found in 2a,b is slightly negative
(ca. −2 kJ mol−1). Remarkably, the comparably large geometric
rearrangement of 3 for forming 2c yields an estimated strain
energy of only 6 kJ mol−1 (for a detailed discussion of the strain
energies see Supporting Information). Consequently, the N−
X−N three-centered bond of 2a,b shows comparable geo-
metries to those of 1a,b with the primary difference that a
coplanar geometry is enforced for the pyridine rings of 2a,b,
whereas free rotation around the N−X−N axis is allowed for
the pyridines of 1a,b.
Informative measures for the stability of the studied
complexes are provided by the computed frequencies of their
N−X stretching vibrations (Table 5) and their estimated
relative stabilities (Table 6). The signature of the asymmetric
N−X−N stretching vibrations is consistent with symmetric
1a,b and 2a,b and asymmetric 1c and 2c. Comparable
vibrational frequencies, with less than 10 cm−1 difference,
were predicted for the Br and I complexes (Table 5). As a
general trend, all frequencies of 2a,b are approximately 20 cm−1
(symmetric) to 30 cm−1 (antisymmetric vibration modes)
lower than their counterparts in 1a,b. However, this difference
does not necessarily indicate a lower stability of 2a,b as
compared to 1a,b, but may rather be the consequence of the
larger reduced masses for 2a,b. For 1c and 2c, the N−H and
N···H stretching vibrations show markedly different frequencies
(ratio >20), reflecting the large difference in their bond
strengths. This difference is more distinct for 2c than for 1c, in
line with the calculated bond length alternations discussed
above. The estimated relative stabilities of the complexes 1a−c,
shown in Table 6, indicate higher binding energies for the
a
Calculations done at B3LYP level with the basis set and ECPs
described in the computational section, employing PCM with CD2Cl2
as solvent. Values in italics refer to unstable symmetric structures.
Values for 1a−c are taken from the literature.44 Symmetric N−X−N
b
c
stretching vibration unless otherwise stated. Antisymmetric N−X−N
d
stretching vibration unless otherwise stated. N−X stretching
e
vibration. N···X stretching vibration.
halogen(I) as compared to the protonated complex (difference
60−70 kJ mol−1). It should be noted that the bond energies of
1a,b and 2a,b are comparable in magnitude to that of a covalent
bond. The difference between protonated and halogenated
complexes may be partly ascribed to the higher stability of
HOTf compared to BrOTf and IOTf. The computed bond
energies of 2a−c (ΔE) are approximately 20 kJ mol−1 weaker
than their counterparts in 1a−c. At first glance, this result might
be ascribed to the reorganization energy of 3. However, this
energy is close to zero for the formation of 2a,b and about 6 kJ
mol−1 for the formation of 2c, as shown above. Also, the
observed energy difference is nearly constant from 2a to 2c,
whereas the reorganization energy increases from 2a to 2c.
Thus, the different formation energies are likely to originate
from electronic effects, for example, from differences in the
degree of delocalization of the π-systems in the complexes of 1
versus 2. The ΔG values are equal (c) or slightly larger (a,b) for
2 as compared to 1, reflecting a smaller entropy loss upon
formation of 2a,c than of 1a,c, depending on the number of
entities associating throughout complexation.
A comparison of the energies for planar and twisted
conformers of 1a−c (Table 6) confirm the hypothesis that
the pyridine rings of 1a,b freely rotate around the N−X−N axis.
For 1c, the twisted conformation appears to be favored;
however, this prediction should not be overly trusted since an
incorrect treatment of the hindered rotation around the N−X−
N axis as soft vibration cannot be ruled out,97 and hence, the
entropy balance between the planar and the twisted forms may
be described incorrectly.
5711
dx.doi.org/10.1021/ja301341h | J. Am. Chem. Soc. 2012, 134, 5706−5715

Journal of the American Chemical Society
Article
are often asymmetric,44,50,98 whereas [N−I−N]+ are symmet-
ric99 in the solid state. In contrast, [N−H−N] hydrogen bonds
are asymmetric in solution and symmetric in crystals.38 This
fact reveals that despite the often emphasized similarities of HB
and XB,1 there are distinct differences as well that may become
of vast importance in the development of XB to a
complementary molecular tool to HB in rational drug design
strategies. The higher stability of the symmetric arrangement of
3-center-4-electron halogen bonds compared to hydrogen
bonds in solutions may originate from a more efficient overlap
of the nitrogen lone pairs with the p-orbital48 of I+ or Br+ as
compared to the possible overlap with the empty s orbital of
H+. The advantageous orbital overlap yields a covalent
character for the investigated XB systems, revealed by 15N
chemical shift data analysis as well as by computation.
Experimental findings were supported by DFT analysis
predicting symmetric [N···X···N]+ arrangements for 1a,b and
2a,b.
a
Table 6. Predicted Stabilities [kJ mol−1] of 1a−c and 2a−c
An important aspect of this work is that the systems studied
provide simple models for the investigation of 3-center-4-
electron halogen bonds. A well-known example for such a
−
halogen bonded complex is the I3 ion, which has been the
target of intense debate in the halogen bonding commun-
ity.100,101 The triiodide ion was previously reported both to be
symmetric (dI−I = 2.90 Å)102,103 and asymmetric (dI−I = 2.83
and 3.03 Å)46,104−106 in the solid state, depending on its
environment. It is often cited as the strongest halogen bonded
complex (180 kJ mol−1).1 The suggestion of the exceptional
halogen bond donor strength of Xδ+ is further confirmed by this
a
Calculations were done at B3LYP level of theory with the basis set
and ECPs described in the computational section, employing the PCM
with CD2Cl2 as solvent. ΔE denotes the electronic energy, and ΔG the
Gibbs free energy at 298 K. Values for 1a−c are taken from reference
44. Underlined values are potentially flawed by the improper treatment
of a hindered rotation, such as soft vibration in the entropy term.97
investigation. Related trihalide ions, such as Br3 , I2Br−, and
−
IBr2 , were typically found to be asymmetric in crystals.104 The
−
assessment of the symmetry of trihalide ions in solutions is
considerably more difficult than the investigation of the related
model compounds 1 and 2, and could therefore so far only be
predicted by computation.107−111
b
Refers to the asymmetric structure.
The inversion barrier for 2c is 16 kJ mol−1 higher in energy
than that for 1c (Table 6), in agreement with the fact that the
steric strain in 3 hinders inversion (the N−N distance is shorter
in the symmetric form of 2c than in the asymmetric one).
However, ZPE and entropy effects largely equalize this
difference between 1c and 2c. Further computational
confirmation for the symmetric XB geometries of 2a,b were
provided by relaxed potential energy surface (PES) calculations
at the B3LYP/LACVP* level using CH2Cl2 continuum solvent
A further impact of the present investigation is that [N−X−
N]+ complexes are common reagents used in electrophilic
halogenations.50,112−121 Gaining an improved understanding of
the structure of such halogenation agents, for example, of the
Barluenga’s reagent,113−115 is therefore of relevance in synthetic
organic chemistry. In these reactions, X+ is transferred from an
[N−X−N]+ bond to a CC double bond through a [C−X−
C]+ intermediate. Whereas the symmetry of the [C−X−C]+
intermediate has been extensively studied,122−124 the symmetry
of the halonium transfer reagent has so far received little
attention.44,125,126
The presented NMR and computational study of 1c agrees
well with the previous conclusions on hydrogen bond
asymmetry based on computational (DFT),127 solution37,38
and solid state127 NMR spectroscopic investigations. However,
the proposal of a general asymmetry of HB in solutions has
been questioned by the Limbach group, who based on solution
and solid state NMR studies argued that HB symmetry depends
on the local environment (counterion, neighboring molecules),
and that strong hydrogen bonds may be symmetric or exhibit
only small deviations from symmetry.70,128−130 Moreover, for
hydrogen bonds exhibiting tautomeric behavior, they proposed
that isotopic perturbation observed by NMR may not originate
from equilibrium isotope effects caused by zero point energy
changes, but from equilibrium averaged intrinsic or geometric
isotope effects arising from anharmonic ground state
vibrations.131 It should be noted that recent UV−vis
spectroscopic studies of Limbach129 support the findings of
model, as implemented in Jaguar (Schrodinger, Inc.), and
̈
scanned for geometries with varying N−X distances in 0.05 Å
steps. The PES scans predict a single energy minimum with
d
N−Br = 2.19 Å for 2a and dN−I = 2.30 Å for 2b, in contrast with
a double well potential for 2c with optimal dN−H distances of
1.05 Å and 1.96 Å.
CONCLUSIONS
■
Combined NMR spectroscopic and computational investiga-
tion indicates the strong preference for the static, symmetric
arrangement of bromine and iodine centered [N···X···N]+-type
XB in solutions. Translational diffusion measurements revealed
a tight interaction of the positively charged halogen bonded
complex with the triflate counterion. Such close interactions
were previously concluded to be responsible for asymmetries
observed in crystals; however, they do not cause any observable
destabilization of the symmetric arrangements of 1a,b and 2a,b
in CH2Cl2 solutions. Interestingly, related [N−Br−N]+ systems
5712
dx.doi.org/10.1021/ja301341h | J. Am. Chem. Soc. 2012, 134, 5706−5715

Journal of the American Chemical Society
Article
Perrin.56 The conclusion of hydrogen bond asymmetry based
on IPE studies has recently been further criticized by Singleton
et al.,132 who by computational methods, yet ignoring the
related computational work of Ohta,123 proposed that IPE data
attributed to isotopic perturbation of equilibria may in certain
cases be better interpreted as the consequence of isotope
induced desymmetrization of symmetrical potential energy
surfaces. Spectroscopic observation (IPE) and computational
(DFT) prediction of symmetric geometries for 1a,b and 2a,b
whereas of asymmetric ones for the closely related 1c and 2c do
2013) under grant agreement no. 259638. We thank the
Swedish Research Council (2007:4407; 621-2008-3562) and
the Carl Tryggers Foundation for financial support.
REFERENCES
■
(1) Metrangolo, P.; Neukirch, H.; Pilati, T.; Resnati, G. Acc. Chem.
Res. 2005, 38, 386.
(2) Politzer, P.; Lane, P.; Concha, M. C.; Ma, Y.; Murray, J. S. J. Mol.
Model. 2007, 13, 305.
(3) Guthrie, F. J. Chem. Soc. 1863, 16, 239.
(4) Hassel, O.; Hvoslef, J. Acta Chem. Scand. 1954, 8, 873.
(5) Hassel, O. Science 1970, 170, 497.
(6) Metrangolo, P.; Meyer, F.; Pilati, T.; Resnati, G.; Terraneo, G.
Angew. Chem., Int. Ed. 2008, 47, 6114.
(7) Cavallo, G.; Metrangolo, P.; Pilati, T.; Resnati, G.; Sansotera, M.;
Terraneo, G. Chem. Soc. Rev. 2010, 39, 3772.
(8) Roper, L. C.; Prasang, C.; Kozhevnikov, V. N.; Whitwood, A. C.;
Karadakov, P. B.; Bruce, D. W. Cryst. Growth Des. 2010, 10, 3710.
(9) Fourmigue, M.; Auban-Senzier, P. Inorg. Chem. 2008, 47, 9979.
(10) Bissantz, C.; Kuhn, B.; Stahl, M. J. Med. Chem. 2010, 53, 5061.
(11) Lu, Y.; Shi, T.; Wang, Y.; Yang, H.; Yan, X.; Luo, X.; Jiang, H.;
Zhu, W. J. Med. Chem. 2009, 52, 2854.
not support this proposal. Furthermore, the magnitude of the
2h
J
scalar coupling determined through the N−H···N
NN
intermolecular hydrogen bond (10.2
0.4 Hz) of bis-
(pyridinium) tetrachlorogallate [(Py-H···Py)(GaCl4)],127
closely related to compound 1c, by solid state NMR best fits
to an asymmetric arrangement (dN1−N2 = 2.707 Å, dN1−H
=
0.946 Å, dN2−H = 1.761 Å). This observation agrees well with
the asymmetric geometry of 1c (Table 4). The X-ray
crystallographic analysis of bis(2-methylpyridinium) tetraphe-
nylborate revealed an electron density pattern best compatible
with a single, symmetric, strongly hydrogen bonded geometry;
yet the possibility that the N−H−N hydrogen would occupy a
double well rather than a single well potential in crystal could
not be excluded.99 This may be rationalized by (a) the
inhibition of dynamic characteristics in crystals,127 or by an
inherent weakness of X-ray crystallography (b) to recognize
dynamic processes and (c) to accurately refine hydrogen
positions.133 The alternative views on hydrogen bond
symmetry indicate that the analysis of weakly binding
complexes remains a challenge. Unequivocal determination of
HB and XB symmetry necessitates the careful, simultaneous use
of alternative spectroscopic and computational methods.
(12) Parisini, E.; Metrangolo, P.; Pilati, T.; Resnati, G.; Terraneo, G.
Chem. Soc. Rev. 2011, 40, 2267.
(13) Resnati, G.; Metrangolo, P. Chem.Eur. J. 2001, 7, 2511.
(14) Metrangolo, P.; Resnati, G.; Pilati, T.; Liantonio, R.; Meyer, F. J.
Polym. Sci., Part A; Polym. Chem. 2007, 45, 1.
(15) Bruce, D. W.; Metrangolo, P.; Meyer, F.; Prasang, C.; Resnati,
G.; Terraneo, G.; Whitwood, A. C. New J. Chem. 2008, 32, 477.
(16) Bruce, D. W.; Metrangolo, P.; Meyer, F.; Pilati, T.; Prasang, C.;
Resnati, G.; Terraneo, G.; Wainwright, S. G.; Whitwood, A. C.
Chemistry 2010, 16, 9511.
(17) Fourmigue, M.; Batail, P. Chem. Rev. 2004, 104, 5379.
(18) Brezgunova, M.; Shin, K. S.; Auban-Senzier, P.; Jeannin, O.;
Fourmigue, M. Chem. Commun. 2010, 46, 3926.
(19) Lu, Y. X.; Wang, Y.; Zhu, W. L. Phys. Chem. Chem. Phys. 2010,
12, 4543.
(20) Hardegger, L. A.; Kuhn, B.; Spinnler, B.; Anselm, L.; Ecabert, R.;
Stihle, M.; Gsell, B.; Thoma, R.; Diez, J.; Benz, J.; Plancher, J. M.;
Hartmann, G.; Banner, D. W.; Haap, W.; Diederich, F. Angew. Chem.,
Int. Ed. 2011, 50, 314.
ASSOCIATED CONTENT
* Supporting Information
■
S
Details on the synthesis depicted in Scheme 1, spectroscopic
data for compound identification, and details on the NMR and
computational symmetry investigations. This material is
available free of charge via the Internet at http://pubs.acs.org.
(21) Clark, T.; Hennemann, M.; Murray, J. S.; Politzer, P. J. Mol.
Model. 2007, 13, 291.
(22) Politzer, P.; Murray, J. S.; Clark, T. Phys. Chem. Chem. Phys.
2010, 12, 7748.
(23) Riley, K. E.; Hobza, P. J. Phys. Chem. B 2008, 112, 3157.
(24) Tawarada, R.; Seio, K.; Sekine, M. J. Org. Chem. 2008, 73, 383.
(25) Eskandari, K.; Zariny, H. Chem. Phys. Lett. 2010, 492, 9.
(26) Auffinger, P.; Hays, F. A.; Westhof, E.; Ho, P. S. Proc. Natl. Acad.
Sci. U.S.A. 2004, 101, 16789.
(27) Voth, A. R.; Hays, F. A.; Ho, P. S. Proc. Natl. Acad. Sci. U.S.A.
2007, 104, 6188.
(28) Voth, A. R.; Khuu, P.; Oishi, K.; Ho, P. S. Nat. Chem. 2009, 1,
74.
(29) Metrangolo, P.; Panzeri, W.; Recupero, F.; Resnati, G. J. Fluorine
Chem. 2002, 114, 27.
(30) Mele, A.; Metrangolo, P.; Neukirch, H.; Pilati, T.; Resnati, G. J.
Am. Chem. Soc. 2005, 127, 14972.
(31) Sarwar, M. G.; Dragisic, B.; Sagoo, S.; Taylor, M. S. Angew.
Chem., Int. Ed. 2010, 49, 1674.
(32) Dimitrijevic, E.; Kvak, O.; Taylor, M. S. Chem. Commun. 2010,
46, 9025.
(33) Sarwar, M. G.; Dragisic, B.; Salsberg, L. J.; Gouliaras, C.; Taylor,
M. S. J. Am. Chem. Soc. 2010, 132, 1646.
(34) Serpell, C. J.; Kilah, N. L.; Costa, P. J.; Felix, V.; Beer, P. D.
Angew. Chem., Int. Ed. 2010, 49, 5322.
(35) Bruckmann, A.; Pena, M. A.; Bolm, C. Synlett 2008, 900.
(36) Erdelyi, M. Chem Soc Rev 2012. DOI: 10.1039/C2CS15292D.
AUTHOR INFORMATION
Corresponding Author
mate@chem.gu.se
■
Present Addresses
∥Department of Chemistry, University of Bergen, NO-5020
Bergen, Norway.
§AstraZeneca R&D Molndal, Medicinal Chemistry, S-431 83
̈
Molndal, Sweden.
̈
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We are grateful to Alexei A. Neverov (Department of
Chemistry, Queen’s University, Canada) for helpful advice
regarding the synthesis of halonium complexes,50 and to
Charles L. Perrin (Department of Chemistry and Biochemistry,
University of California San Diego, CA) for instructive
discussions. We thank Per-Ola Norrby (Department of
Chemistry and Molecular Biology, University of Gothenburg)
for valuable advice regarding the relaxed scan calculations. The
research leading to these results has received funding from the
European Union Seventh Framework Programme (FP7/2007-
5713
dx.doi.org/10.1021/ja301341h | J. Am. Chem. Soc. 2012, 134, 5706−5715

Journal of the American Chemical Society
Article
(37) Perrin, C. L. Science 1994, 266, 1665.
(85) Clark, T.; Chandrasekhar, J.; Spitznagel, G. W.; Schleyer, P. v. R.
J. Comput. Chem. 1983, 4, 294.
(86) Francl, M. M.; Pietro, W. J.; Hehre, W. J.; Binkley, J. S.; Gordon,
M. S.; Defrees, D. J.; Pople, J. A. J. Chem. Phys. 1982, 77, 3654.
(87) Raghavachari, K.; Binkley, J. S.; Seeger, R.; Pople, J. A. J. Chem.
Phys. 1980, 72, 650.
(88) Mennucci, B.; Tomasi, J. J. Chem. Phys. 1997, 106, 5151.
(89) Cossi, M.; Scalmani, G.; Rega, N.; Barone, V. J. Chem. Phys.
2002, 117, 43.
(38) Perrin, C. L. Pure Appl. Chem. 2009, 81, 571.
(39) Gerlt, J. A.; Kreevoy, M. M.; Cleland, W.; Frey, P. A. Chem. Biol.
1997, 4, 259.
(40) Cleland, W. W.; Kreevoy, M. M. Science 1994, 264, 1887.
(41) Perrin, C. L. Acc. Chem. Res. 2010, 43, 1550.
(42) Borisov, E. V.; Skorodumov, E. V.; Pachevskaya, V. M.; Hansen,
P. E. Magn. Reson. Chem. 2005, 43, 992.
(43) Ellison, R. D.; Levy, H. A. Acta Crystallogr. 1965, 19, 260.
(44) Carlsson, A. C. C.; Grafenstein, J.; Laurila, J. L.; Bergquist, J.;
Erdelyi, M. Chem. Commun. 2012, 17, 1458.
(45) Schuster, I. I.; Roberts, J. D. J. Org. Chem. 1979, 44, 2658.
(46) Hassel, O.; Hope, H. Acta Chem. Scand. 1961, 15, 407.
(47) Lin, G. H. Y.; Hope, H. Acta Crystallogr., Sect. B: Struct. Sci.
1972, B 28, 643.
̈
(90) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci,
B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H.
P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.;
Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima,
T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J., J. A.;
Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin,
K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.;
Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega,
N.; Millam, N. J.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.;
Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.;
Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.;
Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.;
Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, O.;
Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian09;
Gaussian, Inc.: Wallingford, CT, 2010.
(91) Becke, A. D. J. Chem. Phys. 2003, 119, 2972.
(92) Grafenstein, J.; Cremer, D. Theor. Chem. Acc. 2009, 123, 171.
(93) Møller, C.; Plesset, M. S. Phys. Rev. 1934, 46, 0618.
(94) Wandel, H.; Wiest, O. J. Org. Chem. 2002, 67, 388.
(95) Baldwin, K. P.; Matzger, A. J.; Scheiman, D. A.; Tessier, C. A.;
Vollhardt, K. P. C; Youngs, W. J. Synlett 1995, 1215.
(96) Yamakawa, J.; Ohkoshi, M.; Takahashi, F.; Nishiuchi, T.;
Kuwatani, Y.; Nishinaga, T.; Yoshida, M.; Iyoda, M. Chem. Lett. 2008,
37, 784.
(48) Alcock, N. W.; Robertson, G. B. J. Chem. Soc. Dalton Trans.
1975, 2483.
(49) Siehl, H. U. Adv. Phys. Org. Chem. 1987, 23, 63.
(50) Neverov, A. A.; Feng, H. X.; Hamilton, K.; Brown, R. S. J. Org.
Chem. 2003, 68, 3802.
(51) Erdelyi, M.; Gogoll, A. J. Org. Chem. 2001, 66, 4165.
(52) Gros, P.; Choppin, S.; Mathieu, J.; Fort, Y. J. Org. Chem. 2002,
67, 234.
(53) Erdelyi, M.; Langer, V.; Karlen, A.; Gogoll, A. New J. Chem.
2002, 26, 834.
(54) Saunders, M.; Telkowski, L.; Kates, M. R. J. Am. Chem. Soc.
1977, 99, 8070.
(55) Perrin, C. L.; Lau, J. S.; Kim, Y. J.; Karri, P.; Moore, C.;
Rheingold, A. L. J. Am. Chem. Soc. 2009, 131, 13548.
(56) Perrin, C. L.; Lau, J. S. J. Am. Chem. Soc. 2006, 128, 11820.
(57) Perrin, C. L.; Ohta, B. K. J. Am. Chem. Soc. 2001, 123, 6520.
(58) Perrin, C. L.; Ohta, B. K. Bioorg. Chem. 2002, 30, 3.
(59) Perrin, C. L.; Kim, Y. J.; Kuperman, J. J. Phys. Chem. A 2001,
105, 11383.
̈
(60) Perrin, C. L.; Kim, Y. J. Inorg. Chem. 2000, 39, 3902.
(61) Perrin, C. L.; Karri, P. Chem. Commun. 2010, 46, 481.
(62) Perrin, C. L.; Flach, A. Angew. Chem., Int. Ed. 2011, 50, 7674.
(63) van’t Hoff, J. H. Z. Phys. Chem. 1887, 1, 483.
(64) Perrin, C. L.; Karri, P. J. Am. Chem. Soc. 2010, 132, 12145.
(65) Perrin, C. L.; Kim, Y. J. J. Am. Chem. Soc. 1998, 120, 12641.
(66) Strauss, C. R.; Trainor, R. W. Aust. J. Chem. 1995, 48, 1665.
(67) Perrin, C. L.; Erdelyi, M. J. Am. Chem. Soc. 2005, 127, 6168.
(68) Perrin, C. L.; Ohta, B. K.; Kuperman, J.; Liberman, J.; Erdelyi,
M. J. Am. Chem. Soc. 2005, 127, 9641.
(97) Ayala, P. Y.; Schlegel, H. B. J. Chem. Phys. 1998, 108, 2314.
(98) Blair, L. K.; Parris, K. D.; Hii, P. S.; Brock, C. P. J. Am. Chem.
Soc. 1983, 105, 3649.
(99) Glidewell, C.; Holden, H. D. Acta Crystallogr., B 1982, 38, 667.
(100) In Categorizing Halogen Bonding and Other Noncovalent
Interactions Involving Halogen Atoms, IUCr 2011 Satellite Workshop:
Siguenza, Spain, 2011.
̈
(101) Abate, A.; Brischetto, M.; Cavallo, G.; Lahtinen, M.;
Metrangolo, P.; Pilati, T.; Radice, S.; Resnati, G.; Rissanen, K.;
Terraneo, G. Chem Commun 2010, 46, 2724.
(69) Perrin, C. L. Acc. Chem. Res. 2010, 43, 1550.
(70) Pietrzak, M.; Wehling, J. P.; Kong, S.; Tolstoy, P. M.;
Shenderovich, I. G.; Lopez, C.; Claramunt, R. M.; Elguero, J.;
Denisov, G. S.; Limbach, H. H. Chemistry 2010, 16, 1679.
(71) Hope, H.; Lin, G. H. Y. Chem. Commun. 1970, 169.
(72) Emsley, J. Chem. Soc. Rev. 1980, 9, 91.
(73) Duthaler, R. O.; Roberts, J. D. J. Am. Chem. Soc. 1978, 100,
4969.
(74) Meier, S.; Benie, A. J.; Duus, J. O.; Sorensen, O. W.
ChemPhysChem 2009, 10, 893.
(102) Slater, R. C. L. Acta Crystallogr. 1959, 12, 187.
(103) Bent, H. A. Chem. Rev. 1968, 68, 587.
(104) Davies, J. E.; Nunn, E. K. Chem. Commun. 1969, 1374.
(105) Mooney, R. C. L. Z. Kristallogr. 1935, 90, 143.
(106) Tasman, H. A.; Boswijk, K. H. Acta Crystallogr. 1955, 8, 59.
(107) Zhang, F. S.; Lynden-Bell, R. M. Phys. Rev. Lett. 2003, 90.
(108) Zhang, F. S.; Lynden-Bell, R. M. Eur. Phys. J. D 2005, 34, 129.
(109) Margulis, C. J.; Coker, D. F.; Lynden-Bell, R. M. Chem. Phys.
Lett. 2001, 341, 557.
(110) Margulis, C. J.; Coker, D. F.; Lynden-Bell, R. M. J. Chem. Phys.
(75) Parella, T. Magn. Reson. Chem. 1998, 36, 467.
(76) Chen, J. G.; McAllister, M. A.; Lee, J. K.; Houk, K. N. J. Org.
Chem. 1998, 63, 4611.
2001, 114, 367.
(111) Sato, H.; Hirata, F.; Myers, A. B. J. Phys. Chem. A 1998, 102,
2065.
(77) Becke, A. D. J. Chem. Phys. 1993, 98, 5648.
(78) Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B: Condens. Matter
1988, 37, 785.
(79) Vosko, S. H.; Wilk, L.; Nusair, M. Can. J. Phys. 1980, 58, 1200.
(80) Stephens, P. J.; Devlin, F. J.; Chabalowski, C. F.; Frisch, M. J. J.
Phys. Chem. 1994, 98, 11623.
(112) Arsequell, G.; Espuna, G.; Valencia, G.; Barluenga, J.; Carlon,
R. P.; Gonzalez, J. M. Tetrahedron Lett. 1999, 40, 7279.
(113) Barluenga, J. Pure Appl. Chem. 1999, 71, 431.
(114) Barluenga, J.; Rodriguez, M. A.; Campos, P. J. J. Org. Chem.
1990, 55, 3104.
(115) Barluenga, J.; Rodriguez, M. A.; Gonzalez, J. M.; Campos, P. J.
(81) Roy, L. E.; Hay, P. J.; Martin, R. L. J. Chem. Theory Comput.
Tetrahedron Lett. 1990, 31, 4207.
(116) Barluenga, J.; Rodriguez, M. A.; Campos, P. J. Tetrahedron Lett.
1990, 31, 2751.
(117) Barluenga, J.; Campos, P. J.; Lopez, F.; Llorente, I.; Rodriguez,
2008, 4, 1029.
(82) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 270.
(83) Wadt, W. R.; Hay, P. J. J. Chem. Phys. 1985, 82, 284.
(84) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 299.
M. A. Tetrahedron Lett. 1990, 31, 7375.
5714
dx.doi.org/10.1021/ja301341h | J. Am. Chem. Soc. 2012, 134, 5706−5715

Journal of the American Chemical Society
Article
(118) Barluenga, J.; Gonzalez-Bobes, F.; Murguia, M. C.; Ananthoju,
S. R.; Gonzalez, J. M. Chemistry 2004, 10, 4206.
(119) Lopez, J. C.; Bernal-Albert, P.; Uriel, C.; Valverde, S.; Gomez,
A. M. J. Org. Chem. 2007, 72, 10268.
(120) Huang, K. T.; Winssinger, N. Eur. J. Org. Chem. 2007, 1887.
(121) Barluenga, J.; Gonzalez, J. M.; Garciamartin, M. A.; Campos, P.
J. Tetrahedron Lett. 1993, 34, 3893.
(122) Ohta, B. K.; Hough, R. E.; Schubert, J. W. Org. Lett. 2007, 9,
2317.
(123) Ohta, B. K.; Scupp, T. M.; Dudley, T. J. J. Org. Chem. 2008, 73,
7052.
(124) Olah, G. A.; Bollinge., J.; Brinich, J. J. Am. Chem. Soc. 1968, 90,
2587.
(125) Sabin, J. R. J. Mol. Struct. 1971, 7, 407.
(126) Haque, I.; Wood, J. L. J. Mol. Struct. 1968, 2, 217.
(127) Claramunt, R. M.; Perez-Torralba, M.; Maria, D. S.; Sanz, D.;
Elena, B.; Alkorta, I.; Elguero, J. J. Magn. Reson. 2010, 206, 274.
(128) Benedict, H.; Limbach, H. H.; Wehlan, M.; Fehlhammer, W.
P.; Golubev, N. S.; Janoschek, R. J. Am. Chem. Soc. 1998, 120, 2939.
(129) Koeppe, B.; Tolstoy, P. M.; Limbach, H. H. J. Am. Chem. Soc.
2011, 133, 7897.
(130) Lesnichin, S. B.; Tolstoy, P. M.; Limbach, H. H.; Shenderovich,
I. G. Phys. Chem. Chem. Phys. 2010, 12, 10373.
(131) Sharif, S.; Denisov, G. S.; Toney, M. D.; Limbach, H. H. J. Am.
Chem. Soc. 2006, 128, 3375.
(132) Bogle, X. S.; Singleton, D. A. J. Am. Chem. Soc. 2011, 133,
17172.
(133) Almlof, J.; Ottersen, T. Acta Crystallogr., A 1979, 35, 137.
5715
dx.doi.org/10.1021/ja301341h | J. Am. Chem. Soc. 2012, 134, 5706−5715