The Origin of Chalcogen-Bonding Interactions

Article abstract of DOI:10.1021/jacs.7b08511

Favorable molecular interactions between group 16 elements have been implicated in catalysis, biological processes, and materials and medicinal chemistry. Such interactions have since become known as chalcogen bonds by analogy to hydrogen and halogen bonds. Although the prevalence and applications of chalcogen-bonding interactions continues to develop, debate still surrounds the energetic significance and physicochemical origins of this class of σ-hole interaction. Here, synthetic molecular balances were used to perform a quantitative experimental investigation of chalcogen-bonding interactions. Over 160 experimental conformational free energies were measured in 13 different solvents to examine the energetics of O···S, O···Se, S···S, O···HC, and S···HC contacts and the associated substituent and solvent effects. The strongest chalcogen-bonding interactions were found to be at least as strong as conventional H-bonds, but unlike H-bonds, surprisingly independent of the solvent. The independence of the conformational free energies on solvent polarity, polarizability, and H-bonding characteristics showed that electrostatic, solvophobic, and van der Waals dispersion forces did not account for the observed experimental trends. Instead, a quantitative relationship between the experimental conformational free energies and computed molecular orbital energies was consistent with the chalcogen-bonding interactions being dominated by n → σ? orbital delocalization between a lone pair (n) of a (thio)amide donor and the antibonding σ? orbital of an acceptor thiophene or selenophene. Interestingly, stabilization was manifested through the same acceptor molecular orbital irrespective of whether a direct chalcogen···chalcogen or chalcogen···H-C contact was made. Our results underline the importance of often-overlooked orbital delocalization effects in conformational control and molecular recognition phenomena.

Full text of DOI:10.1021/jacs.7b08511

Article
Cite This: J. Am. Chem. Soc. XXXX, XXX, XXX-XXX
pubs.acs.org/JACS
The Origin of Chalcogen-Bonding Interactions
†
‡
,†
Dominic J. Pascoe, Kenneth B. Ling, and Scott L. Cockroft*
†EaStCHEM School of Chemistry, University of Edinburgh, Joseph Black Building, David Brewster Road, Edinburgh EH9 3FJ, U.K.
‡Syngenta, Jealott’s Hill International Research Centre, Bracknell, Berkshire RG42 6EY, U.K.
*
S Supporting Information
ABSTRACT: Favorable molecular interactions between group
6 elements have been implicated in catalysis, biological
1
processes, and materials and medicinal chemistry. Such
interactions have since become known as chalcogen bonds
by analogy to hydrogen and halogen bonds. Although the
prevalence and applications of chalcogen-bonding interactions
continues to develop, debate still surrounds the energetic
significance and physicochemical origins of this class of σ-hole
interaction. Here, synthetic molecular balances were used to
perform a quantitative experimental investigation of chalcogen-bonding interactions. Over 160 experimental conformational free
energies were measured in 13 different solvents to examine the energetics of O···S, O···Se, S···S, O···HC, and S···HC contacts
and the associated substituent and solvent effects. The strongest chalcogen-bonding interactions were found to be at least as
strong as conventional H-bonds, but unlike H-bonds, surprisingly independent of the solvent. The independence of the
conformational free energies on solvent polarity, polarizability, and H-bonding characteristics showed that electrostatic,
solvophobic, and van der Waals dispersion forces did not account for the observed experimental trends. Instead, a quantitative
relationship between the experimental conformational free energies and computed molecular orbital energies was consistent with
the chalcogen-bonding interactions being dominated by n → σ* orbital delocalization between a lone pair (n) of a (thio)amide
donor and the antibonding σ* orbital of an acceptor thiophene or selenophene. Interestingly, stabilization was manifested
through the same acceptor molecular orbital irrespective of whether a direct chalcogen···chalcogen or chalcogen···H−C contact
was made. Our results underline the importance of often-overlooked orbital delocalization effects in conformational control and
molecular recognition phenomena.
INTRODUCTION
studies have suggested that dispersion and orbital delocalization
■
(
17,24,29,33−37
effects may also make important contributions.
For
It is reasonable to expect that electron-rich group 16
chalcogen) elements such as oxygen, sulfur, and selenium
may not form particularly favorable contacts with each other.
However, chalcogen−chalcogen contacts are so commonly
observed in X-ray crystal structures that they have become
example, X-ray crystallographic data have revealed the striking
directional dependency of some σ-hole interactions, which is
38−41
consistent with geometry dependent orbital effects.
Due to the difficulty associated with the measurement of
weak interactions in solution, there remains a paucity of
1
−3
known as chalcogen-bonding interactions.
Chalcogen-
bonding interactions have been invoked in such diverse areas
quantitative experimental investigations of chalcogen-bonding
4
,5
6,7
8,9
10
31,36
as catalytic, synthetic, materials, biological, medici-
interactions.
Furthermore, developing a quantitative under-
1
,11
12−14
nal, and supramolecular chemistry.
Chalcogen-bonding
standing of the nature of these interactions is further
complicated by the challenges associated with dissecting
multiple competing influences and solvent effects, which are
both hard to predict, and may dominate the experimental
interactions are themselves considered to be a subclass of “σ-
1
5
hole interactions”, which are most well-known for their
16−18
association with halogen-bonding interactions (group 17).
Alongside the halogens and chalcogens, tetrel elements (group
42−45
behavior.
1
9
20,21
1
1
4), pnictogens (group 15),
and even aerogens (group
Here we have used synthetic molecular balances (Figure 1)
to perform a quantitative experimental investigation of
chalcogen-bonding interactions. Experimental conformational
free energies were compared with theory to examine the
empirical significance of solvent-mediated electrostatic and
solvophobic effects (Figure 2), van der Waals dispersion
(Figures 2 and 3), and orbital delocalization (Figures 4−7).
2
2
8) have been identified as being able to engage in σ-hole
interactions. Despite the undoubted prevalence of σ-hole
interactions, their energetic significance in solution, and the
underlying physicochemical origins are the subject of
2
3−30
debate.
σ-Holes were originally defined as being associated
with a region of positive electrostatic potential that projects
15
along the Z-axis opposite to a σ bond. In line with the original
definition, some experimental characteristics of σ-hole inter-
actions can be qualitatively, and sometimes quantitatively,
Received: August 10, 2017
Published: October 6, 2017
30−32
correlated with electrostatic potentials.
However, other
©
XXXX American Chemical Society
A
DOI: 10.1021/jacs.7b08511
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
comparable to those of OH to OC H-bonds measured in
68
structurally related molecular balances. Varying the thiophene
substituent had a substantial influence on the preference for
O···S contacts, following the trend Me < H < Cl < COOMe <
COMe < CHO (Figure 2, left). Interestingly, the O···Se contact
in compound 1c was slightly more favorable than the O···S
contact in compound 1a-CHO, despite the increased steric
bulk and the lack of an electron-withdrawing group on the
selenophene ring. β-Thiophene compounds 1b-H and 1b-
CHO, which could not form O···S contacts, had a weaker
preference for the closed conformer compared to the
corresponding α-thiophenes 1a-H and 1a-CHO that could
form direct O···S contacts. Thioformamide balances 2a-Me, 2a-
H, and 2a-Cl that could potentially host S···S contacts had ∼1.5
kJ mol− decreased preference for the closed conformer
compared to the equivalently substituted 1a balances that
hosted O···S contacts. Indeed, while balances in series 1a and
1
1c had minimized structures containing planar O···S or O···Se
Figure 1. Molecular balances used in the present investigation to
investigate chalcogen-bonding interactions.
contacts, such a planar structure and corresponding S···S
contact was only seen in balance 2a-Cl. Similarly, β-thiophenes
in the 1b series were calculated to have planar structures,
hosting CO···HC contacts, while the equivalent β-
substituted thioformamide 2b did not, and instead adopted a
EXPERIMENTAL EVALUATION OF
■
CHALCOGEN-BONDING INTERACTIONS
1
46
propeller-like conformation. Consistent with previous studies,
We used molecular balances for our quantitative experimental
investigation of chalcogen-bonding interactions (Figure 1).
Molecular balances provide useful tools for the quantification of
interactions, since the position of a conformational equilibrium
depends on the magnitude of intramolecular interactions and
there was little difference in the energies of secondary
conformers in which X/Y-carbonyl substituents were flipped,
suggesting that no significant secondary chalcogen···chalcogen
interactions were present in the X/Y-carbonyl substituted
69
42,43,47,48
compounds.
the competing solvent effects (Figure 1A).
Accord-
ingly, molecular balances have been used to measure a wide
range of interaction classes including those involving
EVALUATION OF SOLVENT-MEDIATED
ELECTROSTATIC AND SOLVOPHOBIC
CONTRIBUTIONS
■
4
9−51
52−61
62−67
fluorine,
arenes,
and carbonyl groups.
More
specifically, the molecular balances shown in Figure 1 are
derived from previous investigations of solvent effects and
Solvents are known to exert both electrostatic (including
H-bonding interactions) and solvophobic influences
on the conformational preferences of molecular
42,47,68
hydrogen bonding interactions.
The new designs in
Figure 1 host chalcogen-bonding interactions in the closed
conformers (Figure 1A,D, right) that are absent in the open
conformation (Figure 1A,D, left). Since rotation about the
42−45,47,48,60,70
balances.
The conformational free energy differ-
ences in Figure 2 show striking solvent independence for
balances that preferred the closed conformation. For example,
conformational free energies across compound series 1 were
similar in solvophobic H-bonding solvents such as methanol-d4
(
thio)formamide is slow on the NMR time scale at room
1
9
temperature, integration of the discrete F NMR resonances
corresponding to each conformer provides direct access to the
conformational equilibrium constant K and, therefore, the
and dimethyl sulfoxide-d compared to very apolar solvents,
6
conformational free energy difference, ΔG = −RT ln K.
such as carbon disulfide and benzene-d . The only significant
EXP
6
The compounds shown in Figure 1 containing a range of
potential O, S, and Se contacts were synthesized (see the SI).
An X-ray crystal structure of balance 1a-Cl (CSD deposition
no. 1563020) and density functional theory (DFT) calculations
confirmed that most of the α-substituted 1a and 2a series of
molecular balances accommodated chalcogen···chalcogen con-
tacts in the closed conformation (Figure 2 and Figures S10−
S12). In addition, balances containing β-substituted thiophenes
that were incapable of forming direct chalcogen···chalcogen
contacts in the closed conformer were synthesized with the
intention of serving as controls (1b series and 2b, Figure 1B,E).
Conformers were assigned using HMBC/NOESY NMR
spectroscopy and by the comparison of experimental and
computed conformational ratios (see the SI and below). The
conformational free energy differences between the open and
closed conformers were measured for each balance in 13
different solvents (Figure 2).
changes in conformational free energies were seen when the
very strong H-bond donor perfluoro-tert-butyl alcohol was used
as the solvent. Conformational free energies in this solvent were
−1
found to be driven toward the open conformer by ∼2 kJ mol
compared to the other solvents due to its ability to form strong
competitive H-bonding interactions with formyl carbonyl
groups (Figure 2, bottom). There have been previous reports
of very weak solvent effects on some other σ-hole
2
9,33,34,71,72
interactions,
but such observations are not univer-
3
1
sal. The lack of solvent dependence in the present
investigation is particularly surprising considering that the
conformational free energies of similar formamide molecular
balances hosting H-bonding and aromatic interactions were
found to be strongly dependent on the H-bond donor and
42,68
acceptor abilities of the solvent.
These findings indicate that
the chalcogen-bonding interactions in the present investigation
do not have a substantial solvophobic, electrostatic or dipolar
origin (Table S18). Although, the balances in the present
investigation were not soluble in water, given the apparent
universality of the observed solvent independence, it might be
All of the compounds in series 1a and 1c preferred the closed
conformers in which O···S, or O···Se contacts were formed
−1
(
<−7.4 to −1 kJ mol ). Such conformational preferences are
B
DOI: 10.1021/jacs.7b08511
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 2. Experimental conformational free energies (ΔGEXP) measured in 13 different solvents at 298 K. Corresponding minimized structures of
each of the molecular balances calculated in the gas phase using B3LYP/6-311G* are shown. Colors correspond to those used in structures shown in
−1
Figure 1. Where the conformational equilibrium constant, K > 20, conformational energies are plotted at −7.4 kJ mol with error bars truncated
−1
beyond −10.0 kJ mol . All data and errors are tabulated in the SI.
reasonable to expect similar conformational preferences in
aqueous solution.
highly polarizable solvent carbon disulfide, compared to
43
methanol-d , which has a low bulk polarizability. The solvent
4
with the lowest bulk polarizability in our investigation is
perfluoro-tert-butyl alcohol, which should favor the closed
conformer if contributions from dispersion forces in the
chalcogen···chalcogen contacts are significant. Instead, the
conformational free energies in perfluoro-tert-butyl alcohol are
driven toward the open conformer compared to all of the other
solvents. This indicates that solvation of the formyl oxygen
atoms by hydrogen bonding is more energetically significant
than any contribution from residual differences in dispersion
forces in the solution phase. Furthermore, the experimental
conformational free energies were compared with those
calculated in the gas-phase using DFT methods that both did,
and did not, include dispersion corrections (M06-2X and
ωB97X-D vs B3LYP). The strongest correlation was found
against conformational energies (ΔECALC) calculated using the
EVALUATION OF VAN DER WAALS DISPERSION
CONTRIBUTIONS
■
Having ruled out substantial solvophobic and electrostatic
contributions to chalcogen-bonding interactions in our
investigation, we then set out to consider van der Waals
dispersion forces. Bulk solvent polarizability has been shown to
describe the extent to which the solvent competes with, and
43,73
attenuates dispersion forces between functional groups.
Solvents with low bulk polarizability would be expected to favor
closed conformers that accommodate chalcogen···chalcogen
interactions involving polarizable S and Se atoms, while highly
polarizable solvents would be expected to favor the open
conformer to expose polarizable groups to the solvent.
However, Figure 2 shows that there is a negligible difference
between the conformational free energies measured in the
2
non-dispersion corrected B3LYP method (R = 0.94, Figure
C
DOI: 10.1021/jacs.7b08511
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 4. (A) Correlation of calculated orbital energies in the open vs
closed molecular balance conformers. Data points that fall below the
line formed by the gray points are stabilized in the closed conformer
due to (B) resonance delocalization modulated by structural
planarization (orange), and n → σ* orbital delocalization (teal)
associated with either (C) direct chalcogen···chalcogen contacts, or
(D) chalcogen···H−C contacts. Solid filled points are orbital energies
for α-thiophene series 1a-X and α-selenophene balance 1c. Points with
black outlines are the β-thiophene balances in the 1b-Y series. Open
circles correspond to the only thioformamide balance hosting a
favorable S−S contact, 2a-Cl. Alternative correlations using M06-2X/
6
-311G* and ωB97X-D/6-311G*, plus a comparison of full vs
simplified molecular balance data are provided in the SI.
of orbital delocalization are commonly overlooked in the
context of molecular interactions, which are often considered to
be “non-bonding” or “non-covalent”. We point out that the
terminology used to describe electron delocalization effects in
“
non-bonded” interactions is often inconsistent: polarization,
Figure 3. Correlations of experimental conformational free energies
donor−acceptor interactions, charge transfer, partial covalency,
orbital mixing, and orbital interactions, among others, have all
been used to describe a broadly similar ground-state
measured in CDCl at 298 K (ΔGEXP) vs those predicted at the
3
indicated levels of theory in the gas phase (ΔECALC). An additional
correlation using the SM8 implicit solvent model for chloroform
showed no improvement in the correlation coefficient (Figure S23).
11,32,74−78
phenomenon.
Such inconsistencies may arise, at
least in part, from the challenge of obtaining systematic, direct
experimental measurements of weak interactions in solution,
and further establishing causal association with quantum
3
A). In contrast, conformational energies calculated using
77
dispersion-corrected (DFT-D) methods formed substantially
poorer correlations (R = 0.88 and 0.84, Figures 3B,C). Thus,
these correlations, combined with the very limited solvent
dependence of the conformational free energies indicate that
differences in dispersion forces make negligible contributions to
the chalcogen-bonding interactions that govern the observed
conformational free energies.
mechanical descriptors.
2
Nonetheless, n → π* orbital delocalization from a lone pair
(
n) into the carbonyl antibonding orbital (π*) has been
62−66
proposed to stabilize carbonyl−carbonyl interactions,
6
7
alongside competing dipolar electrostatic explanations.
Similarly, n → σ* delocalization from a lone pair orbital (n)
into the antibonding orbital of a σ-bond (σ*) has been
suggested by theory to stabilize interactions involving
EVALUATION OF ORBITAL DELOCALIZATION
CONTRIBUTIONS
24,29,33−41,79
■
chalcogens.
Thus, we set about performing a
comprehensive orbital analysis of our molecular balances.
Our orbital analysis began by performing geometry
minimizations on the open and closed conformations of
molecular balances bearing a range of substituents (all of the
compounds shown in Figure 1 and more, see the SI) using both
DFT and DFT-D methods. We hypothesized that the energies
of particular orbitals in the open and closed conformers could
be compared to reveal orbital interactions that specifically
stabilized one conformer over the other. To avoid the splitting
of the orbitals arising from the canonical resonance forms of the
aromatic electrons that were not involved in the chalcogen
So far, we have discounted electrostatic, solvophobic, and
dispersion forces as the primary determinants of the chalcogen-
bonding interactions in our molecular balances. Others have
proposed that orbital delocalization effects may play a role in
various classes of σ-hole interactions based on spectroscopic,
17,24,29,33−41
structural, and computational analyses.
Delocaliza-
tion effects are long recognized aspects of bond theory; most
chemists are familiar with the concepts of inductive polarization
along σ-bonds, resonance involving π-bonds, and hyper-
conjugation between σ- and π-bonds. However, similar forms
D
DOI: 10.1021/jacs.7b08511
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 5. Orbital decomposition analysis illustrating the hypothetical combination of molecular fragments A + B in the three orientations shown. (A)
HOMO containing the formyl oxygen lone pair (n) is stabilized in (C) and (D) by the same set of antibonding σ* orbitals of the (seleno/thio)phene
fragment, irrespective of the orientation of the connected ring and the specific intramolecular contacts present. The preferred conformers of the
compounds investigated are shown in Figure 2. All orbitals and minimized geometries were calculated using B3LYP/6-311G*.
interactions, the fluorophenyl moiety was replaced with a
proton, and a single-point energy calculation was performed on
each structure (retaining the geometry of the complete
balance). The use of such fragments greatly simplified the
task of assigning pairs of open/closed orbitals (see the SI for
validation). The resulting comparison of orbital energies for all
of the balances with planar structures from Figure 2 is
presented in Figure 4. The line formed by the gray points in
Figure 4A corresponds to the vast majority of orbitals in which
there is little difference in energy between the open and closed
conformers. Data points that fall below the gray background
line correspond to orbitals that are more stable in the closed
than the open conformer. Two sets of data sit below the
background line (orange and teal, Figure 4A). Upon inspection
of the molecular orbitals, the orange data were found to
correspond to through-bond, resonance delocalization of the
lone pair orbital that lies above and below the plane of the
amide into the coplanar aromatic system (orange, Figure 4B).
Such delocalized orbitals were, accordingly, only present in
molecular balances that had planar closed conformations. The
teal series corresponded to orbitals in which the other,
orthogonal lone pair orbital of the amide was delocalized into
the S−C (or Se−C) σ-bond of the adjacent thiophene (or
selenophene) (Figures 4C,D and 5C,D). Thus, these orbitals
were consistent with the occurrence of stabilizing n → σ*
orbital interactions.
even higher energy, unoccupied, antibonding molecular orbital
of the thiophene (or selenophene) (Figure 5B). Interestingly,
this decomposition analysis showed that the same molecular
orbitals (Figures 5A,B) combine to stabilize the formamide
lone pair, irrespective of the α/β-connectivity, or the
orientation of the thiophene ring (Figure 5C,D). Furthermore,
we used Natural Bond Orbital (NBO) analysis to examine the
occurrence and stabilizing character of specific n → σ* orbital
interactions. In simplistic terms, NBOs are theoretical
constructs that are intermediate between molecular orbitals
76
(
such as those shown in Figure 5) and the constituent atomic
80
orbitals. NBOs reveal orbital delocalization that includes both
covalent bonds and orbital interactions that can be considered
as having “partial” covalent character. Indeed, NBO analysis has
previously been used to analyze putative n → σ* and n → π*
29,33−36,65,66
interactions.
NBO analysis of our balances revealed
the potential for stabilizing n → σS*−C and n → σS*e−C
delocalization where direct chalcogen···chalcogen contacts
occurred, while weaker n → σ* , n → σ* , and n → σ*
C−C
C−S
H−C
NBOs were present in the β-connected thiophene balances
(
Figure S23 and Table S45).
The occurrence of such orbital interactions should be
indicated by lengthening of the accepting bond in the closed
conformer relative to the open conformer of each molecular
balance. Computational geometry minimizations revealed
lengthening of the bonds aligned with the (thio)amide contact
in the closed conformation (blue bonds, Figure 6). The extent
of bond lengthening did not correlate with the experimental
conformational free energies measured in the molecular
balances, since changes in electron density were also modulated
by the adjacent X and Y substituents (Figure 1). Consistent
with this suggestion, bond lengthening also occurred at
electron-accepting substituents (purple bonds, Figure 6).
We confirmed the identity of these delocalized n → σ*
orbitals by further decomposition of the molecular balance
fragments into the constituent (thio)formamide (e.g., Figure
5
A) and thiophene (or selenophene) components (e.g., Figure
5B). This hypothetical decomposition analysis indicated that
the stabilized, delocalized orbitals of the type shown in Figures
C,D and 5C,D did indeed result from the hybridization of a
high-energy, but occupied, lone-pair orbital (Figure 5A) with an
4
E
DOI: 10.1021/jacs.7b08511
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 6. Calculated bond lengthening (B3LYP/6-311G*) in the
closed vs open conformers for molecular balances hosting chalcogen−
chalcogen contacts. Further bond length differences are provided in
Figure S13.
Having confirmed the identity and possible stabilizing nature
of n → σ* orbital contributions to chalcogen-bonding
interactions, we sought a quantitative energetic relationship
between experiment and theory. Unfortunately, we found no
correlation between the experimental conformational free
energy differences measured in the molecular balances and
the n → σ* orbital delocalization energies output from the
NBO calculations (Figure S22 and Table S45). Indeed, one
limitation of NBO analysis is that it can be challenging to
ascribe an easily understood physical meaning to NBOs.
Instead, we compared the computed energies of physically
relevant molecular orbitals with our experimental conforma-
tional free energies. A striking correlation was found between
the energies of the molecular orbitals identified in Figure 5C,D
2
that contained n → σ* orbital delocalization (R = 0.99, Figure
7A). The energies relating to balances containing both direct
chalcogen···chalcogen and chalcogen···H−C contacts (teal and
black outlined points, respectively) were found to fit on the
same correlation. This finding was consistent with the
involvement of the same σ* acceptor orbital (Figure 5C,D),
irrespective of the orientation or connectivity of the thiophene
79
ring. Contrasting with previous suggestions, the β-connected
thiophenes (black outlines in Figure 7A) were weaker lone pair
acceptors than the equivalently substituted α-connected
variants (filled circles in Figure 7A). However, it is important
to note that the relative acceptor abilities may not be general, as
they are likely to be influenced by the geometric constraints
imposed by our intramolecular system. The single point
associated with selenophene balance 1c was an outlier (Figure
S19) indicating the increased favorability of this interaction
compared to the O···S and S···S interactions. In comparison,
the energies of the resonance delocalized orbitals (Figure 7B),
along with other molecular orbitals (Figure S21) did not form
good correlations with the same experimental data. Thus, the
strong correlation in Figure 7A establishes a quantitative link
between the experimentally determined conformational free
energies and the theoretically determined energies of n → σ*
delocalized orbitals involved in stabilizing the chalcogen-
bonding interactions.
Figure 7. Correlations of the calculated energies of orbitals stabilized
by (A) n → σ* orbital delocalization (teal) and (B) resonance
delocalization modulated by planarization (orange). Solid filled points
correspond to α-substituted thiophenes, while β-substituted thio-
phenes are indicated with black outlines. Calculations were performed
on structures of the type shown inset using B3LYP/6-311G*. X and Y
=
substituents as shown in Figure 1. Alternative correlations using
DFT-D methods are provided in the SI.
1). The conformational free energies of balances hosting
chalcogen-bonding interactions were found to be surprisingly
solvent independent, ruling out substantial contributions from
81
electrostatic and solvophobic effects (Figure 2). This solvent
independence combined with comparison against dispersion-
corrected calculated conformational energies further indicated
that van der Waals dispersion forces did not account for the
observed interaction trends (Figure 3). The latter finding was
consistent with previous studies that have found substantial
attenuation of dispersion forces between functional groups due
CONCLUSION
■
We have performed a quantitative, experimental investigation of
chalcogen-bonding interactions. Synthetic molecular balances
were used to examine solvent and substituent effects on a range
of chalcogen···chalcogen and chalcogen···HC contacts (Figure
F
DOI: 10.1021/jacs.7b08511
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
to competitive dispersion interactions with the surrounding
(11) Reid, R. C.; Yau, M.-K.; Singh, R.; Lim, J.; Fairlie, D. P. J. Am.
Chem. Soc. 2014, 136, 11914−11917.
4
3,61,70,73
solvent.
Instead, computed changes in bond lengths
(12) Lim, J. Y. C.; Marques, I.; Thompson, A. L.; Christensen, K. E.;
and NBO analysis pointed toward the involvement of
stabilizing contributions from n → σ* orbital delocalization
between the lone pair on a (thio)amide donor and the
antibonding σ* orbitals of the adjacent thiophene (or
selenophene) acceptor. A quantitative relationship between
the energy of the orbital hosting n → σ* orbital delocalization
and the experimental data was seen. Interestingly, thiophene
rings were found to accept electrons into the same antibonding
molecular orbital in both α- and β-connected thiophenes, either
via direct chalcogen···chalcogen or chalcogen···HC contacts.
Intriguingly, our quantitative comparison of experimental and
computational data reveals empirical behavior most consistent
with a dominant contribution from orbital delocaliza-
Fel
13) Gleiter, R.; Werz, D. B.; Rausch, B. J. Chem. - Eur. J. 2003, 9,
676−2683.
14) Robinson, S. W.; Mustoe, C. L.; White, N. G.; Brown, A.;
Thompson, A. L.; Kennepohl, P.; Beer, P. D. J. Am. Chem. Soc. 2015,
37, 499−507.
15) Murray, J. S.; Lane, P.; Politzer, P. J. Mol. Model. 2009, 15, 723−
29.
16) Beale, T. M.; Chudzinski, M. G.; Sarwar, M. G.; Taylor, M. S.
́
ix, V.; Beer, P. D. J. Am. Chem. Soc. 2017, 139, 3122−3133.
(
2
(
1
(
7
(
Chem. Soc. Rev. 2013, 42, 1667−1680.
(17) Cavallo, G.; Metrangolo, P.; Milani, R.; Pilati, T.; Priimagi, A.;
Resnati, G.; Terraneo, G. Chem. Rev. 2016, 116, 2478−2601.
(18) Clark, T.; Hennemann, M.; Murray, J. S.; Politzer, P. J. Mol.
Model. 2007, 13, 291−296.
2
9,33−36
tion.
Our results highlight the energetic significance of
(
2
(
1
19) Bauza,
013, 52, 12317−12321.
20) Murray, J. S.; Lane, P.; Politzer, P. Int. J. Quantum Chem. 2007,
07, 2286−2292.
21) Zahn, S.; Frank, R.; Hey-Hawkins, E.; Kirchner, B. Chem. - Eur.
J. 2011, 17, 6034−6038.
22) Bauza, A.; Frontera, A. Angew. Chem., Int. Ed. 2015, 54, 7340−
343.
23) Burling, F. T.; Goldstein, B. M. J. Am. Chem. Soc. 1992, 114,
313−2320.
24) Bleiholder, C.; Gleiter, R.; Werz, D. B.; Ko
007, 46, 2249−2260.
25) Cozzolino, A. F.; Vargas-Baca, I.; Mansour, S.; Mahmoudkhani,
A. H. J. Am. Chem. Soc. 2005, 127, 3184−3190.
26) Esrafili, M. D.; Mohammadian-Sabet, F. Chem. Phys. Lett. 2015,
34, 210−215.
27) Murray, J. S.; Lane, P.; Clark, T.; Riley, K. E.; Politzer, P. J. Mol.
́
A.; Mooibroek, T. J.; Frontera, A. Angew. Chem., Int. Ed.
orbital delocalization in molecular interactions.
ASSOCIATED CONTENT
Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/jacs.7b08511.
■
*
S
(
(
7
(
2
(
2
(
́
Experimental, characterization, and computational data
(
PDF)
Crystallographic data for compound 1a-Cl (CIF)
̈
ppel, H. Inorg. Chem.
AUTHOR INFORMATION
Corresponding Author
scott.cockroft@ed.ac.uk
■
(
6
(
*
ORCID
Scott L. Cockroft: 0000-0001-9321-8997
Author Contributions
The manuscript was written through contributions of all
authors.
Notes
Model. 2012, 18, 541−548.
(28) Duarte, D. J. R.; Sosa, G. L.; Peruchena, N. M. J. Mol. Model.
2013, 19, 2035−2041.
(29) Iwaoka, M.; Komatsu, H.; Katsuda, T.; Tomoda, S. J. Am. Chem.
Soc. 2004, 126, 5309−5317.
(
(
30) Stone, A. J. J. Am. Chem. Soc. 2013, 135, 7005−7009.
The authors declare no competing financial interest.
31) Garrett, G. E.; Gibson, G. L.; Straus, R. N.; Seferos, D. S.;
Taylor, M. S. J. Am. Chem. Soc. 2015, 137, 4126−4133.
(32) Politzer, P.; Murray, J. S.; Clark, T. Phys. Chem. Chem. Phys.
2013, 15, 11178−11189.
ACKNOWLEDGMENTS
We thank Syngenta and the EPSRC for funding. We thank Dr.
Stefan Borsley for helpful feedback.
■
(33) Iwaoka, M.; Komatsu, H.; Katsuda, T.; Tomoda, S. J. Am. Chem.
Soc. 2002, 124, 1902−1909.
(34) Barton, D. H. R.; Hall, M. B.; Lin, Z.; Parekh, S. I.; Reibenspies,
REFERENCES
■
J. J. Am. Chem. Soc. 1993, 115, 5056−5059.
35) Komatsu, H.; Iwaoka, M.; Tomoda, S. Chem. Commun. 1999,
05−206.
36) Iwaoka, M.; Tomoda, S. J. Am. Chem. Soc. 1996, 118, 8077−
084.
37) Bleiholder, C.; Werz, D. B.; Ko
Soc. 2006, 128, 2666−2674.
38) Bauza, A.; Quinonero, D.; Deya, P. M.; Frontera, A.
CrystEngComm 2013, 15, 3137−3144.
39) Iwaoka, M.; Takemoto, S.; Tomoda, S. J. Am. Chem. Soc. 2002,
24, 10613−10620.
40) Rittner, R.; Ducati, L. C.; Tormena, C. F.; Fiorin, B. C.; Braga,
C. B. Spectrochim. Acta, Part A 2011, 79, 1071−1076.
41) Hudson, B.; Nguyen, E.; Tantillo, D. J. Org. Biomol. Chem. 2016,
(
1) Beno, B. R.; Yeung, K.-S.; Bartberger, M. D.; Pennington, L. D.;
Meanwell, N. A. J. Med. Chem. 2015, 58, 4383−4438.
(
2
(
8
(
1
(
1
2) Murray, J. S.; Lane, P.; Politzer, P. Int. J. Quantum Chem. 2008,
08, 2770−2781.
3) Murray, J. S.; Lane, P.; Clark, T.; Politzer, P. J. Mol. Model. 2007,
3, 1033−1038.
4) Benz, S.; Lop
Angew. Chem., Int. Ed. 2017, 56, 812−815.
5) Robinson, E. R. T.; Walden, D. M.; Fallan, C.; Greenhalgh, M.
D.; Cheong, P. H.-Y.; Smith, A. D. Chem. Sci. 2016, 7, 6919−6927.
6) Shiina, I.; Nakata, K.; Ono, K.; Onda, Y.-s.; Itagaki, M. J. Am.
Chem. Soc. 2010, 132, 11629−11641.
7) Cox, P. A.; Leach, A. G.; Campbell, A. D.; Lloyd-Jones, G. C. J.
Am. Chem. Soc. 2016, 138, 9145−9157.
8) Mahmudov, K. T.; Kopylovich, M. N.; Guedes da Silva, M. F. C.;
Pombeiro, A. J. L. Dalton Trans. 2017, 46, 10121−10138.
9) Werz, D. B.; Gleiter, R.; Rominger, F. J. Am. Chem. Soc. 2002,
24, 10638−10639.
10) Fick, R. J.; Kroner, G. M.; Nepal, B.; Magnani, R.; Horowitz, S.;
Houtz, R. L.; Scheiner, S.; Trievel, R. C. ACS Chem. Biol. 2016, 11,
48−754.
(
̈
ppel, H.; Gleiter, R. J. Am. Chem.
(
́
ez-Andarias, J.; Mareda, J.; Sakai, N.; Matile, S.
(
(
(
1
(
(
(
(
(
14, 3975−3980.
(42) Mati, I. K.; Adam, C.; Cockroft, S. L. Chem. Sci. 2013, 4, 3965−
(
1
(
3965.
(43) Adam, C.; Yang, L.; Cockroft, S. L. Angew. Chem., Int. Ed. 2015,
54, 1164−1167.
(44) Cockroft, S. L.; Hunter, C. A. Chem. Commun. 2009, 3961−
7
3963.
G
DOI: 10.1021/jacs.7b08511
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
(
45) Cockroft, S. L.; Hunter, C. A. Chem. Commun. 2006, 3806−
(80) Weinhold, F.; Landis, C. R. Chem. Educ. Res. Pract. 2001, 2, 91−
104.
3
808.
(
(
46) Mati, I. K.; Cockroft, S. L. Chem. Soc. Rev. 2010, 39, 4195−4205.
47) Muchowska, K. B.; Adam, C.; Mati, I. K.; Cockroft, S. L. J. Am.
(81) Our findings rule out a purely electrostatic model for chalcogen-
bonding interactions, though they may be, at least qualitatively,
compatible with an electrostatic plus polarization model. Murray, J. S.;
Politzer, P. Wiley Interdiscipl. Rev. Comp. Mol. Sci. 2017, e1326.
Chem. Soc. 2013, 135, 9976−9979.
48) Yang, L.; Adam, C.; Cockroft, S. L. J. Am. Chem. Soc. 2015, 137,
0084−10087.
49) Hof, F.; Scofield, D. M.; Schweizer, W. B.; Diederich, F. Angew.
Chem., Int. Ed. 2004, 43, 5056−5059.
50) Li, P.; Maier, J. M.; Vik, E. C.; Yehl, C. J.; Dial, B. E.; Rickher, A.
E.; Smith, M. D.; Pellechia, P. J.; Shimizu, K. D. Angew. Chem., Int. Ed.
017, 56, 7209−7212.
51) Ams, M. R.; Fields, M.; Grabnic, T.; Janesko, B. G.; Zeller, M.;
Sheridan, R.; Shay, A. J. Org. Chem. 2015, 80, 7764−7769.
52) Maier, J. M.; Li, P.; Hwang, J.; Smith, M. D.; Shimizu, K. D. J.
Am. Chem. Soc. 2015, 137, 8014−8017.
53) Tatko, C. D.; Waters, M. L. J. Am. Chem. Soc. 2002, 124, 9372−
373.
54) Gung, B. W.; Wekesa, F.; Barnes, C. L. J. Org. Chem. 2008, 73,
803−1808.
55) Newcomb, L. F.; Gellman, S. H. J. Am. Chem. Soc. 1994, 116,
993−4994.
56) Kim, E.-i.; Paliwal, S.; Wilcox, C. S. J. Am. Chem. Soc. 1998, 120,
1192−11193.
57) Paliwal, S.; Geib, S.; Wilcox, C. S. J. Am. Chem. Soc. 1994, 116,
497−4498.
58) Gardarsson, H.; Schweizer, W. B.; Trapp, N.; Diederich, F.
Chem. - Eur. J. 2014, 20, 4608−4616.
59) Motherwell, W. B.; Moïse, J.; Aliev, A. E.; Nic,
(
1
(
(
2
(
(
(
9
(
1
(
4
(
1
(
4
(
(
̌
M.; Coles, S. J.;
Horton, P. N.; Hursthouse, M. B.; Chessari, G.; Hunter, C. A.; Vinter,
J. G. Angew. Chem., Int. Ed. 2007, 46, 7823−7826.
(
60) Emenike, B. U.; Bey, S. N.; Bigelow, B. C.; Chakravartula, S. V.
S. Chem. Sci. 2016, 7, 1401−1407.
61) Hwang, J.; Dial, B. E.; Li, P.; Kozik, M. E.; Smith, M. D.;
Shimizu, K. D. Chem. Sci. 2015, 6, 4358−4364.
62) Bartlett, G. J.; Newberry, R. W.; VanVeller, B.; Raines, R. T.;
Woolfson, D. N. J. Am. Chem. Soc. 2013, 135, 18682−18688.
63) Newberry, R. W.; Orke, S. J.; Raines, R. T. Org. Lett. 2016, 18,
614−3617.
64) Bartlett, G. J.; Choudhary, A.; Raines, R. T.; Woolfson, D. N.
Nat. Chem. Biol. 2010, 6, 615−620.
65) Newberry, R. W.; VanVeller, B.; Guzei, I. A.; Raines, R. T. J. Am.
Chem. Soc. 2013, 135, 7843−7846.
66) Choudhary, A.; Gandla, D.; Krow, G. R.; Raines, R. T. J. Am.
Chem. Soc. 2009, 131, 7244−7246.
67) Fischer, F. R.; Wood, P. A.; Allen, F. H.; Diederich, F. Proc. Natl.
Acad. Sci. U. S. A. 2008, 105, 17290−17294.
68) Dominelli-Whiteley, N.; Brown, J. J.; Muchowska, K. B.; Mati, I.
(
(
(
3
(
(
(
(
(
K.; Adam, C.; Hubbard, T. A.; Elmi, A.; Brown, A. J.; Bell, I. A. W.;
Cockroft, S. L. Angew. Chem., Int. Ed. 2017, 56, 7658−7662.
(
69) In addition, NBO calculations revealed no such secondary
interactions.
70) Yang, L.; Adam, C.; Nichol, G. S.; Cockroft, S. L. Nat. Chem.
013, 5, 1006−1010.
71) Sarwar, M. G.; Dragisic, B.; Salsberg, L. J.; Gouliaras, C.; Taylor,
M. S. J. Am. Chem. Soc. 2010, 132, 1646−1653.
72) Robertson, C. C.; Perutz, R. N.; Brammer, L.; Hunter, C. A.
Chem. Sci. 2014, 5, 4179−4183.
73) Yang, L.; Brazier, J. B.; Hubbard, T. A.; Rogers, D. M.; Cockroft,
(
2
(
(
(
S. L. Angew. Chem., Int. Ed. 2016, 55, 912−916.
(
(
(
74) Stone, A. J.; Price, S. L. J. Phys. Chem. 1988, 92, 3325−3335.
75) Bent, H. A. Chem. Rev. 1968, 68, 587−648.
76) Reed, A. E.; Curtiss, L. A.; Weinhold, F. Chem. Rev. 1988, 88,
8
(
99−926.
77) Gonthier, J. F.; Steinmann, S. N.; Wodrich, M. D.;
Corminboeuf, C. Chem. Soc. Rev. 2012, 41, 4671−4687.
(78) Grabowski, S. J. Chem. Rev. 2011, 111, 2597−2625.
(79) Alabugin, I. V.; Zeidan, T. A. J. Am. Chem. Soc. 2002, 124,
3
175−3185.
H
DOI: 10.1021/jacs.7b08511
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX