Subtle "supramolecular buttressing effects" in Cucurbit[7]uril/guest assemblies

Article abstract of DOI:10.1039/c3ob40250a

Biphenyl derivatives bearing a dimethylsulfonium group at position 3 and three different substituents at position 4 (H, F and CH₃) have been prepared as probes to test the validity of the "supramolecular buttressing" concept. We define the latter as the alteration, by a neighboring unit, of a substituent effect on intermolecular recognition. In this case, the 4-substituents exert some pressure on the 3-dimethylsulfonium groups and control the ratio of their syn and anti conformations. As free species, biphenyls bearing 4-H and 4-F substituents are present as approximately equimolar mixtures of syn and anti-conformers, while the biphenyl scaffold with a 4-CH₃ group adopts the anti-conformation exclusively. The 3-dimethylsulfonium substituents then interact with one of the carbonylated portals of Cucurbit[7]uril (CB[7]), and their conformations affect the position of the guests inside the cavity of the macrocycle, thereby validating our "supramolecular buttressing" model. Surprisingly however, binding affinities towards CB[7] are barely affected by the nature of the 4-substituents and the conformations of the neighboring sulfonium groups, despite very different electronic densities presented to the CB[7] portal in their syn or anti conformations. Solvation was found to dramatically smoothen host-guest Columbic interactions, although the latter remain important in the recognition process. Replacing the positively charged 3-dimethylsulfonium unit with an isopropyl substituent decreases the affinity of the biphenyl guest by 1000-fold. The Royal Society of Chemistry 2013.

Full text of DOI:10.1039/c3ob40250a

Organic&
BiomolecularChemistry
www.rsc.org/obc
Volume 11 | Number 19 | 21 May 2013 | Pages 3075–3262
ISSN 1477-0520
PAPER
Roymon Joseph and Eric Masson
Subtle“supramolecular buttressing effects” in Cucurbit[7]uril/guest assemblies

Organic &
Biomolecular Chemistry
View Article Online
View Journal | View Issue
PAPER
Subtle “supramolecular buttressing effects” in
Cucurbit[7]uril/guest assemblies†
Cite this: Org. Biomol. Chem., 2013, 11,
3116
Roymon Joseph and Eric Masson*
Biphenyl derivatives bearing a dimethylsulfonium group at position 3 and three different substituents at
position 4 (H, F and CH3) have been prepared as probes to test the validity of the “supramolecular but-
tressing” concept. We define the latter as the alteration, by a neighboring unit, of a substituent effect on
intermolecular recognition. In this case, the 4-substituents exert some pressure on the 3-dimethylsulfo-
nium groups and control the ratio of their syn and anti conformations. As free species, biphenyls bearing
4-H and 4-F substituents are present as approximately equimolar mixtures of syn and anti-conformers,
while the biphenyl scaffold with a 4-CH3 group adopts the anti-conformation exclusively. The 3-dimethyl-
sulfonium substituents then interact with one of the carbonylated portals of Cucurbit[7]uril (CB[7]), and
their conformations affect the position of the guests inside the cavity of the macrocycle, thereby validat-
ing our “supramolecular buttressing” model. Surprisingly however, binding affinities towards CB[7] are
barely affected by the nature of the 4-substituents and the conformations of the neighboring sulfonium
groups, despite very different electronic densities presented to the CB[7] portal in their syn or anti confor-
mations. Solvation was found to dramatically smoothen host–guest Columbic interactions, although the
latter remain important in the recognition process. Replacing the positively charged 3-dimethylsulfonium
unit with an isopropyl substituent decreases the affinity of the biphenyl guest by 1000-fold.
Received 3rd February 2013,
Accepted 26th March 2013
DOI: 10.1039/c3ob40250a
www.rsc.org/obc
the attacking base; metalation actually takes place para to the
triethylsilyl substituent (see the green arrow in Fig. 1b).
Introduction
The term “buttressing effect” was proposed by Westheimer in
To the best of our knowledge, the concept of chemical but-
1950 to describe the impact of meta-substituents on the tor- tressing has never been extended to supramolecular systems.
sional barriers of substituted biphenyls along their aryl–aryl We propose to define “supramolecular buttressing” as the
axis.1 meta-Substituents were found to buttress (i.e. to support alteration, by a neighboring unit, of a substituent effect on
or strengthen) ortho-substituents, by limiting their ability to intermolecular recognition (see Fig. 1c; remote group R2 exerts
undergo distortion during the torsional isomerization process; pressure on substituent R1, which is then forced to alter the
the consequence was an increase of the torsional barrier (see interaction between units A and B). In this study, we present a
Fig. 1a). A particularly spectacular example is the 6.8 kcal
mol−1 penalty imposed by 3- and 3′-iodine atoms on the acti-
vation barrier of 2,2′,3,3′-tetraiodo-5,5′-dicarboxybiphenyl com-
pared to its 2,2′-diiodo analog (30.2 vs. 23.4 kcal mol−1).1 Since
then, buttressing effects have been invoked on numerous
occasions to describe the alteration, by a neighboring or a
more remote group, of a substituent effect on the mobility,2
reactivity3 and physical properties4 of various structures. For
example, Schlosser and co-workers showed that the metalation
of arenes is blocked by meta-triethylsilyl groups (see substitu-
ent R2 in Fig. 1b), which buttress ortho-substituents R1 (Cl,3a,b
Br3c and CF3)3d and force them to shield the ipso position from
Department of Chemistry and Biochemistry, Ohio University, Athens, Ohio 45701,
Fig. 1 (a) Barriers of torsional isomerisation of biphenyl scaffolds affected by
buttressing. (b) Regioselectivity of arene metalation controlled by buttressing
effects. (c) “Supramolecular buttressing”.
USA. E-mail: masson@ohio.edu; Fax: +1-740-593-0148; Tel: +1-740-593-9992
†Electronic supplementary information (ESI)
available. See DOI:
10.1039/c3ob40250a
3116 | Org. Biomol. Chem., 2013, 11, 3116–3127
This journal is © The Royal Society of Chemistry 2013

View Article Online
Organic & Biomolecular Chemistry
Paper
case of supramolecular buttressing, and assess how the
remote substituent (R2 in Fig. 1c) affects the binding affinity
between the two molecular partners A and B, and the structure
of assembly A·B.
Host–guest assemblies formed upon interaction between
Cucurbit[7]uril (CB[7])5,6 and positively charged guests are
ideal A·B systems, since (1) binding affinities can be exception-
ally high in aqueous medium (up to 5 × 1015 M−1!),7 (2) the
structures of the assemblies are often very well defined, and
(3) the position of the guest inside the cavity of the macrocycle
can be readily assessed by 1H NMR spectroscopy: hydrogen
nuclei residing at the core of the hollow macrocycle are shifted
upfield, and hydrogens outside the cavity undergo a downfield
shift that weakens as the distance between the hydrogen atom
and the carbonylated portal increases.8 In order to test supra-
molecular buttressing, CB[7] guests (unit A in Fig. 1c) must
bear (1) a hydrophobic moiety that is encapsulated by the
macrocycle, (2) a positively charged substituent R1 that inter-
acts with the carbonyl portal of CB[7], and (3) a neighboring
group R2, which affects the geometry or the rigidity of substitu-
ent R1. Biphenyls 1–3 (see Fig. 2a) satisfy these guidelines,
since the nature of buttressing substituents R2 is expected to
control the conformation of the dimethylsulfonium group R1
(syn or anti, see Fig. 2a), and consequently affect its interaction
with CB[7].
Biphenyl 1 was prepared by diazotisation of 3-(methylthio)-
aniline, followed by nucleophilic aromatic substitution with
potassium iodide, Suzuki coupling with 4-tolylboronic acid
and methylation of the thioether with trimethyloxonium tetra-
fluoroborate (see Schemes 1a and 1d). Biphenyl 2 was obtained
by first converting 5-bromo-2-fluorophenol to the correspond-
ing benzenethiol 2c via carbamation, Newman–Kwart
rearrangement9 and decarbamation (see Scheme 1b); thiol 2c
was then methylated, and the resulting sulfide 2d was coupled
with 4-tolylboronic acid and methylated again to afford our
target (Scheme 1d). 4-Bromo-2-nitrotoluene was converted to
the corresponding benzenethiol 3c via reduction to aniline 3a,
diazotisation, substitution with potassium ethyl xanthate and
saponification. Thiol 3c was then submitted to the same
sequence of reactions as analog 2c to afford biphenyl 3 (see
Schemes 1c and 1d).
Scheme 1 (a) Preparation of sulfide 1a; preparation of key thiol intermediates
(b) 2c and (c) 3c; (d) preparation of biphenyls 1–3 from the corresponding
thiols.
The very low rotation barriers along the Caryl–S bond of
derivatives 1–3 (2.1, 3.8 and 6.9 kcal mol−1, respectively)10
render the accurate determination of the syn/anti ratio by NMR
experiments at low temperature impossible.11 Therefore, we
decided to extract the information by using density functional
theory (DFT) as well as Hartree–Fock and second order Møller–
Plesset perturbation (MP2). The geometries of the three pairs
Fig. 2 (a) Equilibrium between syn and anti conformations of biphenyls 1–3.
(b) Electrostatic potential map superimposed on the isodensity surface of biphe-
nyl anti-3 (isovalue 0.004); rainbow color coding, with dark blue indicating par-
ticularly positive regions and yellow neutral regions.
This journal is © The Royal Society of Chemistry 2013
Org. Biomol. Chem., 2013, 11, 3116–3127 | 3117

View Article Online
Paper
Organic & Biomolecular Chemistry
Table 1 Calculated stabilities ΔGsyn→anti of anti-conformers relative to syn con-
formers in aqueous medium and, between parentheses, in the gas phasea
B97-D3-
(BJ)
PW6B95-D3-
(BJ)
B3LYP-D
B97-D
MP2
1
2
3
−0.2 (−0.1) −0.2 (−0.1) −0.2 (−0.1) −0.2 (−0.1)
+0.8 (+2.3) +0.2 (+1.7) −0.3 (+1.2) +0.4 (+1.9)
−2.4 (−2.0) −2.5 (−2.1) −3.2 (−2.7) −3.2 (−2.8)
−0.2 (−0.1)
0.0 (+1.5)
−3.5 (−3.1)
a In kcal mol−1; negative values indicate anti-preference.
of syn and anti-conformers were optimized using the
B3LYP-D12,13 hybrid and triple-ζ, doubly polarized def2-TZVPP
basis sets.14 Single-point calculations were then carried out
using the B97-D,13,15 B97-D3(BJ)15,16 and PW6B95-D3(BJ)16,17
functionals to refine the electronic contributions to the syn/
anti preferences, and vibrational analysis was performed with
the B3LYP-D hybrid to extract Gibbs free energies at 25 °C (see
Table 1). Energies were finally corrected with a solvation term,
calculated using the polarizable continuum solvation model
(IEFPCM)18 at the B97-D level. Since the interaction between
Fig. 3 1H NMR spectra of (a) biphenyl 1 and (b) its inclusion complex with
CB[7]; (c) biphenyl 2, (d) assembly 2·CB[7], (e) biphenyl 3 and (f ) assembly 3·
CB[7]. Measurements carried out in deuterium oxide. See Fig. 2 for numbering.
substituents R2 and the methyl groups of the dimethylsulfo- conformers, respectively.10 The destabilization is most likely
nium unit (or its opposite lone pair of electrons) operates due to steric congestion when the fluoro and dimethylsulfo-
essentially through space, with interatomic distances ranging nium groups are vicinal.
from 2 to 5 Å, all tested functionals have been corrected for
In aqueous solution however, the syn preference of biphenyl
medium and long-range dispersive interactions using 2 is almost entirely counterbalanced by stronger solvation of
Grimme’s DFT-D13 or DFT-D3(BJ)16 methods (the latter being the anti-conformer (−44.8 vs. −43.3 kcal mol−1). This effect is
the most recent version). These corrections are especially likely due to the concomitant shielding of a large surface sur-
important since dispersive interactions are maximal at rounding the positively charged sulfonium unit and of one
approximately 3 Å.19 The choice of the four functionals was side of the fluorine substituent from solvent interactions. A
based in part on one of our recent studies, where we showed similar preference is observed with biphenyl 3, albeit weaker.
that the B3LYP-D and B97-D hybrids afforded highly accurate According to calculations, syn and anti-conformers of biphe-
barriers of torsional isomerisation for a 46-member set of sub- nyls 1 and 2 co-exist in 42 : 58 and 60 : 40 ratios, respectively
stituted biphenyls,20 and on Grimme’s recent thorough bench- (ΔGsyn→anti −0.18 0.03 and +0.2 0.4 kcal mol−1), and biphe-
mark,21 which recommends the use of the B97-D3(BJ) GGA nyl 3 is present exclusively in its anti-conformation (ΔGsyn→anti
and the PW6B95-D3(BJ) hybrid. Some six years ago, the B97-D −3.0 0.5 kcal mol−1; syn/anti ratio > 1 : 99). Experimentally,
functional had also been shown to accurately evaluate inter- we can confirm the trends in syn/anti selectivities by NOESY
action energies in supramolecular assemblies.19 Finally, we experiments. In the case of biphenyl 1, volume integrals of
carried out optimizations and vibrational analysis at the MP2/ NOESY cross-peaks between H(7)/H(2) and H(7)/H(4) (see
def2-TZVPP level to compare this widely used method with Fig. 2 for numbering) are of similar magnitude, and confirm
DFT-D.
the presence of an approximately equimolar ratio of syn and
In the gas phase, all functionals as well as MP2 calculations anti-conformers. An intense H(7)/H(2) cross-peak and the
indicate that (1) biphenyl 1 is present as a 1 : 1 mixture of syn absence of H(7)/4-CH3 cross-peaks indicate that biphenyl 3
and anti-conformers (ΔGsyn→anti −0.08 kcal mol−1 on average), adopts exclusively the anti-conformation.
A
moderate
(2) the syn conformation of fluorinated biphenyl 2 is clearly H(7)/H(2) cross-peak is also observed with biphenyl 2, approxi-
favored (ΔGsyn→anti +1.7
0.4 kcal mol−1 on average, corre- mately half as intense as the H(7)/H(2) signal of biphenyl 3
sponding to a 95 : 5 syn/anti ratio), and (3) biphenyl 3 adopts (the volume integrals of H(3′)/4′-CH3 and H(5′)/4′-CH3 cross-
almost exclusively the anti-conformation (ΔGsyn→anti −2.5
peaks are used as reference in both cases). This again suggests
0.5 kcal mol−1; syn/anti ratio 1 : 99; see Table 1). We should the presence of approximately equimolar amounts of syn and
note that although favorable Columbic and van der Waals anti-conformers in biphenyl 2.
interactions between the 4-fluoro-substituent of biphenyl 2
Upon interaction with CB[7], the terminal 4′-CH3 group, as
and the two methyl groups of the sulfonium unit are certainly well as hydrogens H(2′), H(3′), H(5′) and H(6′) from the neigh-
responsible for the observed syn preference, the presence of boring aryl unit, undergo strong upfield shifts (0.53, 1.02 and
the fluorine atom is overall destabilizing, regardless of the con- 0.79 ppm on average in the case of 4′-CH3 hydrogens, meta’-
formation of the sulfonium group: in a virtual equilibrium, and ortho’-hydrogens, respectively); to the contrary, dimethyl-
biphenyl 1 and fluorobenzene are favored over biphenyl 2 and sulfonium hydrogens H(7) are shifted upfield (0.21 ppm on
benzene, by 3.1 and 5.1 kcal mol−1 with syn and anti average; see Fig. 3). These shifts indicate that the 4′-tolyl
3118 | Org. Biomol. Chem., 2013, 11, 3116–3127
This journal is © The Royal Society of Chemistry 2013

View Article Online
Organic & Biomolecular Chemistry
Paper
Table 2 Chemical shifts measured upon CB[7] encapsulationa
H(4′)
H(3′), H(5′)
H(2′), H(6′)
H(2)
H(6)
H(7)
1
2
3
−0.52
−0.51
−0.57
−1.01
−1.00
−1.06
−0.80
−0.81
−0.77
+0.09
+0.09
+0.23
+0.03
−0.02
+0.11
+0.18
+0.18
+0.27
a In ppm; negative values indicate upfield shifts. See Fig. 2 for
numbering.
Table 3 Thermodynamic parameters describing the interaction between
guests 1–5 and CB[7], as determined by ITC
ΔHa
ΔSb
TΔSc
ΔGd
Ke
1
2
3
4
5
−8.7 ( 0.1) −1.2 ( 0.1) −0.4 ( 0.1) −8.4 ( 0.1) 1.3 ( 0.1) × 106
−8.3 ( 0.1) 0.7 ( 0.2) 0.2 ( 0.1) −8.6 ( 0.1) 1.8 ( 0.1) × 106
−8.9 ( 0.1) −2.2 ( 0.1) −0.7 ( 0.1) −8.2 ( 0.1) 1.1 ( 0.1) × 106
−6.2 ( 0.1) −6.6 ( 0.3) −2.0 ( 0.1) −4.2 ( 0.1) 1.2 ( 0.1) × 103
−8.8 ( 0.3) −6.9 ( 1.0) −2.1 ( 0.3) −6.8 ( 0.3) 8.9 ( 0.5) × 104
Fig. 4 Optimized structures of assemblies syn-1·CB[7] and anti-1·CB[7] in the
“gas phase” and in “aqueous medium”, calculated at the B97-D/SVP level
without and with the COSMO solvation model, respectively.
a Binding enthalpy [kcal mol−1]. b Binding entropies [cal mol K−1].
c Entropic component to the interaction [kcal mol−1]. d Free energy of
binding [kcal mol−1]. e Binding affinity [M−1]. Errors between
parentheses.
moieties of biphenyls 1–3 are located inside the cavity of the
macrocycle, and the dimethylsulfonium groups are dominat-
ing one of the portals.
The modelling of the interaction between CB[7] and biphe-
nyls 1–3 is challenging. All tested optimization methods (force to −8.9 kcal mol−1 with methylated biphenyl 3 (see Table 3);
fields, semi-empirical and DFT-D) propose that CB[7] is sitting this enthalpic improvement is counterbalanced by an increas-
between the two aryl units, with 4′-CH3 substituents located ingly high entropic penalty along the same sequence (TΔS
outside the macrocycle (see complexes anti-1·CB[7]gas and syn- ranging from +0.20 kcal mol−1 with biphenyl 2 to −0.67 kcal/
1·CB[7]gas, Fig. 4), in stark contrast with experimental results. mol for biphenyl 3), thereby reversing the trend in affinity. We
Such an arrangement may well be valid in the gas phase to should note however, that this possible enthalpy–entropy com-
maximize dispersive interactions between the 4′-CH3 substitu- pensation spans a very narrow range of enthalpies and entro-
ent (as well as hydrogens H(3′) and H(5′)) and the carbonylated pies, compared to the uniquely wide distributions of these two
portal of CB[7]. Only when DFT-D optimizations (at the B97- parameters observed with CB[7]/guest complexes; with binding
D/SVP level) are carried out with corrections for solvation affinities of approximately 106
M
−1, enthalpy contributions
(either by using IEFPCM or the conductor-like screening have been measured between −14 and −2 kcal mol−1, and
model COSMO),22 the 4′-CH3 group translates further inside entropy between −7 and +6 kcal mol−1, respectively!6e There-
the cavity of CB[7] to allow proper solvation of the carbonylated fore, we will not attempt to justify the minor variations
rim (see assemblies anti-1·CB[7]aq and syn-1·CB[7]aq, Fig. 4). described above.
Although improved compared to “gas phase” optimizations,
Relative binding affinities between pairs of biphenyls
1
the “solvated” assemblies are still not satisfactory, since hydro- obtained by competitive H NMR titrations in the presence of
gens H(2) and 4′-CH3 are located at the level of the CB[7] CB[7] afford similar results; a minor discrepancy was obtained
portal, while experimentally, they undergo contrasted chemical for the binding affinity of biphenyl 2 relative to biphenyl 1 (a
shifts (+0.09 and −0.52 ppm, respectively, in the case of biphe- 2.3-fold difference was determined by NMR experiments, com-
nyl 1; see Table 2). We suspect that solvation effects are still pared to a 1.4-fold preference by ITC).
underestimated during optimizations, and thus the 4′-CH3
These unexpectedly similar binding affinities let us envi-
substituent should sit further inside the cavity of CB[7]. sion two borderline scenarios when biphenyls 1 and 2 interact
However, one should note that the exchange rates between with CB[7]: (1) anti conformers may undergo torsional isomeri-
CB[7] and biphenyls 1–3 are fast on the NMR time scale, and sation along their Caryl–S axis to afford complexes syn-1·CB[7]
thus energy profiles describing the binding processes are and syn-2·CB[7] exclusively. However, if the latter option were
inevitably shallow and difficult to model accurately.
favorable, biphenyl 3, which is forced to remain in its anti-con-
Binding affinities between CB[7] and biphenyls 1–3 were formation, would display a much weaker affinity for CB[7]
subsequently determined by isothermal titration calorimetry compared to biphenyls 1 and 2 (the 2.8 kcal mol−1 penalty for
(ITC). Surprisingly, they were very similar, with binding con- anti → syn isomerisation would translate into an approximately
stants equal to 1.3 × 106, 1.8 × 106 and 1.1 × 106 M−1, respect- 600-fold decrease in binding affinity). (2) On the opposite side
ively. A barely significant increase in binding enthalpies is of the continuum, syn-conformers may isomerize upon encap-
measured with increasingly electron-donating 4-substituents, sulation with CB[7], thereby affording assemblies anti-1·CB[7]
ranging from −8.3 kcal mol−1 in the case of fluorobiphenyl 2 and anti-2·CB[7] exclusively. This option is plausible since
This journal is © The Royal Society of Chemistry 2013
Org. Biomol. Chem., 2013, 11, 3116–3127 | 3119

View Article Online
Paper
Organic & Biomolecular Chemistry
electrostatic potential maps of the biphenyls in their anti con-
One can thus conclude that in this study, supramolecular
formation show an electron-deficient area surrounding the two buttressing only affects the position of the guests inside CB[7]
methyl groups of the sulfonium unit exposed to the CB[7] (or more generally the geometry of our A·B assemblies; see
portal (see Fig. 2b). However, if this scenario were correct, the introduction), but not the binding affinities of the two units.
dimethylsulfonium hydrogens H(7) of biphenyls 1 and 3 When the biphenyl derivatives adopt an anti-conformation,
would undergo similar downfield shifts of approximately 0.27 the two methyl groups of the dimethylsulfonium substituent
upon encapsulation (since biphenyl 3 must remain in its anti- are in direct contact with the “upper” carbonylated portal of
conformation while interacting with CB[7], see Fig. 3, spectra CB[7], and tend to lift the encapsulated 4-tolyl unit further
e and f). This is clearly not the case; hydrogens H(7) of biphenyl inside the cavity. Chemical shifts observed for H(4′) and H(3′)
1 are deshielded by only 0.18 ppm (see Fig. 3, spectra a and b, upon binding are −0.57 and −1.06 ppm for biphenyl anti-3, vs.
and Table 2). This effect is particularly visible in a competitive −0.52 and −1.01 ppm with approximately 1 : 1 mixtures of syn-
binding experiment between biphenyls 1 and 3 and CB[7] (see and anti-biphenyls 1 and 2; during the “lifting process” of anti-
Fig. 5), with hydrogens H(7) of biphenyl 1 resonating at higher conformers, H(2′) and H(6′) hydrogens are moving closer to
frequencies than biphenyl 3 in the absence of CB[7] (3.32 vs. the upper rim (the chemical shift upon CB[7] binding is
3.29 ppm, see spectrum a), and at lower frequencies when −0.77 ppm in anti-biphenyl 3 vs. −0.81 ppm in mixtures of syn-
both guests are encapsulated inside the macrocycle (3.50 vs. and anti-biphenyls 1 and 2). In their syn-conformations, biphe-
3.57 ppm; spectrum o). 4′-Methyl hydrogen H(4′) and nyls 1 and 2 lack proper contact between the dimethylsulfo-
especially hydrogens H(2) of biphenyls 1 and 3 can also been nium substituent and the upper CB[7] rim, and thus the
used as probes, and show very different chemical shifts upon encapsulated 4-tolyl moiety is located closer to the opposite
interaction with CB[7]. H(4′) hydrogens are shifted upfield by “lower” portal compared to anti-conformers.
0.52 and 0.57 ppm, respectively, and H(2) nuclei are shifted
downfield by 0.09 ppm in the case of biphenyl 1 and up to blies is very surprising, considering that Columbic interactions
0.23 ppm in biphenyl 3 (see Table 2). between the upper carbonylated portal of CB[7] and the region
The lack of clear anti preference in CB[7]/biphenyl assem-
The dismissal of the two borderline scenarios described of low electron density surrounding the two methyl groups of
above, reinforced by uniform CB[7]/guest binding affinities, the sulfonium substituent should favor this conformation (see
strongly suggests that CB[7] encapsulation has little, if any Fig. 2b). Moreover, repulsion between the rim of CB[7] and the
effect on the syn/anti ratios of biphenyls 1–3. The identical sulfur lone pair in a syn conformation should have further
chemical shifts of hydrogens H(2), H(4′) and H(7) measured destabilized this arrangement. To assess the impact of the
upon encapsulation of biphenyls 1 and 2 (+0.09, −0.52 and positively charged substituent on the binding affinity, we pre-
+0.18 ppm, respectively; see Table 2) indicate that the two pared biphenyl 4 (see Scheme 2), which bears a neutral isopro-
guests are not only similarly affected by CB[7], but have indeed pyl substituent instead of a dimethylsulfonium group at
similar syn/anti ratios in both their free and bound forms (as position 3, and a carboxylate group at the 4-position to
long as the effects of the macrocycle and of the 4-substituents enhance water solubility. Similar chemical shift patterns are
are cumulative, which is likely).
observed upon interaction of biphenyls 3 and 4 with CB[7] (see
Fig. 6; the lack of chemical shift of the 2-isopropyl hydrogen
nucleus H(3) and the moderate downfield shifts of the two
Fig. 5 1H NMR spectra of (a) free biphenyl 1; (b) free biphenyl 3 and (c) biphe-
nyls 1 and 3. (d)–(o) Competitive 1H NMR titration of biphenyls 1 and 3 (2.0 mM
each) in the presence of increasing amounts of CB[7] (up to 6.0 mM; 0.50 mM
increments) in deuterium oxide. Red arrows indicate the trend of hydrogens H
(7) and H(4’) in biphenyl 1, and blue arrows in biphenyl 3.
Scheme 2 Preparation of carboxylate 4, an isosteric, yet negatively charged
analog of biphenyls 1–3.
3120 | Org. Biomol. Chem., 2013, 11, 3116–3127
This journal is © The Royal Society of Chemistry 2013

View Article Online
Organic & Biomolecular Chemistry
Paper
while leaving the lower carbonylated portal unshielded and
fully available for interactions with the solvent; in this case,
the sulfonium group would be fully detached from the upper
CB[7] rim, regardless of its conformation. Water-mediated
Columbic interactions would be consistent with terphenyl 5
binding to CB[7] with a stronger affinity than biphenyl 4 (i.e.
greater than 1.2 × 103 M−1). (2) The terphenyl axle may pierce
through both CB[7] portals to allow direct contact between the
sulfonium unit and the upper carbonylated rim of the macro-
cycle; in this case, an affinity ranging from 1.2 × 103 M−1
(biphenyl 4) to 1.3 × 106 M−1 (biphenyl 1) is expected. The ratio
of the affinity of biphenyl 1 and terphenyl 5 would then corre-
spond to the energy penalty imposed to the assembly for dis-
rupting the solvation of the lower CB[7] portal.
Fig. 6 1H NMR spectra of (a) biphenyl 4 and (b) its inclusion complex with CB[7].
See Scheme 2 for numbering.
neighboring methyl hydrogens H(7) clearly indicate anti-
preference).
The affinity of biphenyl 4 towards CB[7] reached only 1.2 ×
103 M−1, approximately 1000 times weaker than biphenyl 1,
thereby indicating that Columbic forces do play a role in the
interaction between biphenyls 1–3 and CB[7], even when the
latter adopt a syn conformation. To reconcile this dilemma, we
suggest that CB[7]-bound biphenyls 1 and 2 leave more water-
accessible surface area in their syn conformation, and allow
for a better solvation of the partially shielded upper carbony-
lated portal. Since the 4-tolyl group does not pierce through
the lower CB[7] portal, Columbic interactions between the sulf-
onium unit of syn-biphenyls 1 and 2 and the upper CB[7] rim
are likely mediated by water. We also note that the 1000-fold
decrease in binding affinity, when the positive sulfonium
anchor is replaced with an approximately isosteric isopropyl
group, is reminiscent of the difference in CB[n] affinity
between organic ammonium salts and their conjugate base
(ratios between 16 and 3.2 × 104 are common).6e,23
1H NMR spectroscopy validated the second scenario, since
H(4′′) hydrogens were barely shifted upfield upon addition of
CB[7], contrary to hydrogens connected to the central phenyl
unit at positions 2′, 6′, 3′ and 5′ which were shifted upfield by
approximately 0.5 ppm. More peculiar is the weak or moderate
upfield shift of all signals, including those from the dimethyl-
sulfonium group H(7) (see Fig. 7, spectra a and b), which indi-
cates that CB[7] shuttles along the terphenyl axle and
sometimes even surrounds the positively charged group. The
concomitant formation at higher CB[7] concentrations
(>2.5 mM) of what is likely the loose ternary complex 5·(CB[7])2
was another surprise. Under these conditions, hydrogens H(7)
and those at the opposite terminal phenyl ring were further
shifted upfield, while hydrogens from the central aryl unit
underwent downfield shifts (see Fig. 7, spectrum c). ITC experi-
ments afforded a binding affinity of 8.9 × 104 M−1 24
15 times
,
weaker than biphenyl 1. This difference represents a 1.6 kcal
mol−1 penalty for the disruption of the solvation shell sur-
rounding the lower CB[7] portal, and is surprisingly entirely
entropic in nature (binding enthalpies of biphenyls 1 and ter-
phenyl 5 are −8.7 and –8.8 kcal mol−1, while their binding
entropies are −0.4 and −2.1 kcal mol−1, respectively). A similar
13-fold difference was measured when comparing the affinities
of pentyl- and hexylammonium towards CB[6] in a sodium
chloride solution;25 in this case, the aliphatic tail of the hexyl-
ammonium cation disrupted the interaction between sodium
and the CB[6] portal.
Finally, we wanted to assess whether the dimethylsulfo-
nium group could still interact with the carbonylated rim of
CB[7], possibly via water mediation, if the separation between
the two units became longer. To that aim, we designed terphe-
nyl 5 (see Scheme 3), and envisioned two modes of interaction
with CB[7]: (1) similarly to biphenyl 1, the terminal 4′′-methyl
may behave as a pivot inside the cavity of the macrocycle,
Fig. 7 1H NMR spectra of (a) terphenyl 5 (2.0 mM), (b) assembly 5·CB[7] as the
major component in the presence of 2.5 mM CB[7], and (c) a likely mixture of
[2]pseudorotaxane 5·CB[7] and [3]pseudorotaxane 5·(CB[7])2 in the presence of
a higher concentration of CB[7] (7.0 mM). See Scheme 3 for numbering.
Scheme 3 Preparation of terphenyl 5 by two consecutive Suzuki couplings.
This journal is © The Royal Society of Chemistry 2013
Org. Biomol. Chem., 2013, 11, 3116–3127 | 3121

View Article Online
Paper
Organic & Biomolecular Chemistry
Solvation effects are not only key to the understanding of
CB[n] recognition, but are also the main contributors to the
Conclusions
We have shown that the buttressing of 3-dimethylsulfonium subtlety of supramolecular buttressing.
groups by vicinal 4-substituents mildly affects the geometry of
CB[7]-bound biphenyls 1–3, and in particular the extent of
guest penetration inside the cavity of the macrocycle, thereby
Experimental section
illustrating our concept of “supramolecular buttressing”.
Generalities
When the biphenyl derivatives adopt an anti-conformation,
Starting materials were purchased from Sigma-Aldrich
the two methyl groups of the dimethylsulfonium substituent
are in direct contact with the upper carbonylated portal of CB[7],
and tend to lift the encapsulated 4-tolyl unit further inside
the cavity. In their syn-conformations, biphenyls 1 and 2 lack
proper support from the dimethylsulfonium substituent at the
upper CB[7] rim, and therefore the trapped 4-tolyl unit is
sitting closer to the opposite lower portal. However, the 4-tolyl
group does not pierce through the lower portal to maximize
Coulombic interactions between the sulfonium unit and the
opposite rim of CB[7]; to the contrary, the lower portal remains
fully solvated, and Coulombic interactions between the sulfo-
nium unit and the upper CB[7] rim are likely mediated by
water. The predominant role of solvation in the overall binding
processes makes modelling particularly challenging. The
polarizable continuum solvation model or the conductor-like
screening model improve the accuracy of host–guest structure
optimization, but are still not satisfactory. Whenever guests
are not anchored to both CB[7] portals via dipole–dipole inter-
actions, and their encapsulated moieties are not tightly nested
inside the cavity of the macrocycle, we recommend using mod-
eling with extra caution.
(St. Louis, MO), TCI America (Portland, OR) and Cambridge
Isotope Laboratories (Andover, MA). 1H NMR spectra were
recorded at 300 MHz using a Bruker 300 spectrometer (Billerica,
MA) and 1H-decoupled 13C NMR spectra were obtained at
75.5 MHz using the same Bruker 300 spectrometer at 25 °C.
Chemical shifts
δ refer to the residual HDO signal (δ
4.80 ppm) when the solvent is D2O or to the residual CH3CN
signal (δ 1.96 ppm) when the solvent is acetonitrile-d3. Pro-
ducts were also characterized by high-resolution mass spectro-
metry (HRMS) performed at the COSMIC facility of the Old
Dominion University (Norfolk, VA) using a Bruker Daltonics 12
Tesla APEX-Qe FTICR mass spectrometer with an Apollo II Ion
Funnel. CB[7] binding assays, including competitive binding
experiments in the presence of two guests, were carried out
using previously described methods.26 ITC experiments were
performed using an iTC200 calorimeter (Microcal Inc., GE
Healthcare, Piscataway, NJ). Computational work was carried
out using the Glenn cluster (IBM 1350) at the Ohio Super-
computer Center (Columbus, OH).
Intermediates and final products
While supramolecular buttressing affects geometries, it
does not influence binding affinities between CB[7] and the
3-Iodothioanisole (1a). To a solution of 3-(methylthio)-
biphenyl guests. Regardless of the nature of the 4-substituent aniline (1.0 g, 7.2 mmol) in water (10 mL) at 0 °C was added
(H, F and CH3), all binding constants range from 1.1 to 1.8 × conc. HCl (3.7 mL) followed by an aqueous solution of sodium
106 M−1. Interactions between CB[7] and biphenyls 1–3 have nitrite (1.0 g, 15 mmol) in water (5.0 mL) over 30 min. The
thus no significant effect on syn/anti ratios, although CB[7] solution was kept at 0 °C for 1 h until the dropwise addition of
binding to anti-conformers had been expected to be much a potassium iodide (2.4 g, 15 mmol) solution in water (10 mL).
more favorable compared to syn-conformers. Here again, sol- The reaction mixture was then kept at 25 °C for 6 h. The
vation dominates the recognition process and smoothes the product was extracted with dichloromethane (3 × 50 mL),
influence of Coulombic interactions. The latter still play a role washed with water (0.10 L) and brine (0.10 L), dried with
though; when the 3-dimethylsulfonium unit is replaced by the Na2SO4 and concentrated. The product was purified by chro-
approximately isosteric isopropyl substituent, and a negative matography (silica gel; eluent: petroleum ether–CH2Cl2, 19 : 1)
carboxylate group is attached to position 4 to enhance water to afford a light yellow liquid (1.2 g, 67%). 1H NMR (CD3CN): δ
solubility, the affinity of the guest decreases by 1000-fold. Also, 7.63 (s, Ar-H, 1H), 7.53 (d, J = 9.0 Hz, Ar-H, 1H), 7.28 (d, J = 9.0
when biphenyl scaffolds are replaced by a terphenyl unit, the Hz, Ar-H, 1H), 7.08 (t, J = 6.0 Hz, Ar-H, 1H), 2.15 (s, S(CH3), 3H)
longer axle pierces through both CB[7] portals to allow direct ppm. 13C NMR: δ 141.6, 135.0, 134.9, 131.4, 126.4, 95.4 (ArC),
contact between the 3-dimethylsulfonium substituent and the 16.1 (S(CH3)) ppm. HRMS (ESI) m/z calcd for C7H7IS ([M + H]+)
upper CB[7] rim. Solvation of the lower rim is thus reminiscent 250.938590, found 250.938563.
of a thin membrane capping a container: interactions between
Methyl(4′-methylbiphenyl-3-yl)sulfane (1b). A solution of
sulfonium groups and the upper rim of CB[7], even when potassium carbonate (0.82 g, 6.0 mmol) in H2O (10 mL) was
mediated by a thin water layer, can prevent perforation of the added to a solution of sulfide 1a (0.75 g, 3.0 mmol), 4-tolyl-
lower rim; to the contrary, the guest pierces the lower rim boronic acid (0.61 g, 4.5 mmol) and palladium(0) tetrakis-
when the separation between the upper rim and the positive (triphenylphosphine) (0.32 g, 0.30 mmol) in N,N-dimethyl-
surface of the guest would have been too long otherwise.
formamide (50 mL) under inert atmosphere. The resulting
Overall, supramolecular buttressing is a valid concept, mixture was heated at 120 °C for 12 h. After cooling to 25 °C,
although it operates in a milder manner than initially envi- the reaction was filtered through a pad of celite. The filtrate
sioned, at least with these CB[7]-containing assemblies. was then poured into ice-cold water (30 mL), acidified with
3122 | Org. Biomol. Chem., 2013, 11, 3116–3127
This journal is © The Royal Society of Chemistry 2013

View Article Online
Organic & Biomolecular Chemistry
Paper
1.0 M HCl (30 mL), and extracted with dichloromethane (3 × NMR (CD3CN): δ 7.71–7.62 (m, Ar-H, 2H), 7.18 (t, J = 8.7 Hz,
30 mL). The organic fractions were washed with water (60 mL) Ar-H, 1H), 3.10 (s, NCH3, 3H), 3.00 (s, NCH3, 3H) ppm.
and brine (60 mL), dried with anhydrous sodium sulfate, then 13C NMR: δ 163.2 (d, J = 248 Hz, ArCF), 164.3 (ArSCO), 140.9
concentrated in vacuo. The product was purified by chromato- (ArC), 135.8 (d, J = 8.3 Hz, ArC), 120.2 (d, J = 20.3 Hz, ArC),
graphy (silica gel; eluent: hexane–dichloromethane, 49 : 1) to 118.7 (ArC), 116.8 (d, J = 3.7 Hz, ArC), 37.4 (N(CH3)2) ppm.
afford a colourless oil (0.42 g, 65%). 1H NMR (CD3CN): δ HRMS (ESI) m/z calcd for C9H9BrFNOS ([M + Na]+) 299.946446,
7.54–7.52 (m, Ar-H, 3H), 7.38–7.37 (m, Ar-H, 2H), 7.28–7.26 (m, found 299.946317.
Ar-H, 3H), 2.53 (s, S(CH3), 3H), 2.39 (s, Ar-CH3, 3H) ppm. 13C
5-Bromo-2-fluorobenzenethiol (2c). To a solution of thio-
NMR: δ 142.6 (ArC), 140.3 (ArC), 138.6 (ArC), 138.5 (ArC), 130.6 carbamate 2b (3.2 g, 12 mmol) in ethylene glycol (60 mL) was
(ArC × 2), 130.4 (ArC), 127.9 (ArC × 2), 125.8 (ArC), 125.4 (ArC), added potassium hydroxide (0.97 g, 17 mmol) in water (17 mL)
124.5 (ArC), 21.3 (ArC), 15.8 (S(CH3)) ppm. HRMS (ESI) m/z and the mixture was heated to 150 °C for 6 h. The reaction was
calcd for C14H14S ([M + H]+) 215.088898, found 215.088971.
Dimethyl(4′-methylbiphenyl-3-yl)sulfonium
cooled to 0 °C, diluted with water and acidified with 1.0 M HCl
tetrafluoro- (0.10 L) before extraction with ethyl acetate (3 × 0.10 L). The
borate (1). Trimethyloxonium tetrafluoroborate (76 mg, combined organic layers were washed with water (0.15 L),
0.51 mmol) was added to a solution of biphenyl 1b (0.10 g, brine (0.15 L) and dried with Na2SO4. Evaporation of the
0.47 mmol) in nitromethane (3.0 mL) under nitrogen atmos- solvent afforded the title compound as a colorless oil (1.6 g,
phere. The reaction mixture was heated to 80 °C for 12 h. After 67%). 1H NMR (CD3CN): δ 7.58 (d, J = 8.7 Hz, Ar-H, 1H),
cooling to 25 °C, methanol (10 mL) was added, and the solvent 7.37–7.33 (m, Ar-H, 1H), 7.08 (t, J = 9.1 Hz, Ar-H, 1H), 4.14 (s,
was evaporated under vacuum. Addition of diethyl ether SH, 1H) ppm. 13C NMR: δ 159.2 (d, J = 242 Hz, ArCF), 133.6
(15 mL) resulted in the formation of the title compound as a (ArC), 130.8 (d, J = 7.8 Hz, ArC), 123.0 (d, J = 21.4 Hz), 118.1
1
white solid (0.14 g, 95%); m.p. 113–114 °C. H NMR (CD3CN): (ArC), 117.5 (d, J = 3.9 Hz, ArC) ppm. HRMS (ESI) m/z calcd for
δ 8.16 (s, Ar-H, 1H), 8.03 (d, J = 7.7 Hz, Ar-H, 1H), 7.89 (d, J = C6H3BrFS ([M]2+) 409.824025, found 409.824090.
7.9 Hz, Ar-H, 1H), 7.78 (t, J = 8.0, Ar-H, 1H), 7.65 (d, J = 8.1 Hz,
5-Bromo-2-fluorothioanisole (2d). A mixture of thiol 2c
Ar-H, 2H), 7.36 (d, J = 7.9 Hz, Ar-H, 2H), 3.23 (s, S(CH3)2, 6H), (1.5 g, 7.2 mmol), methyl iodide (5.1 g, 36 mmol) and dry
2.42 (s, Ar-CH3, 3H) ppm. 13C NMR: δ 144.6 (ArC), 140.1 (ArC), potassium carbonate (3.0 g, 22 mmol) in acetone (60 mL) was
136.4 (ArC), 133.6 (ArC), 132.4 (ArC), 130.9 (ArC × 2), 129.1 heated to 40 °C for 15 h, before being concentrated. The
(ArC), 128.8 (ArC), 128.2 (ArC × 2), 127.0 (ArC), 29.3 (S(CH3)2), residue was then dissolved in ethyl acetate (0.10 L) and washed
21.3 (ArCH3). HRMS (ESI) m/z calcd for C15H17S ([M]+) with water (0.10 L) and brine (75 mL), and dried with Na2SO4.
229.104548, found 229.104452.
O-(5-Bromo-2-fluorophenyl)dimethylthiocarbamate
The evaporation of the solvent afforded the product as a light
(2a). yellow liquid (1.4 g, 88%). 1H NMR (CD3CN): δ 7.42 (dd, J = 9.1,
Potassium carbonate (3.6 g, 26 mmol) was added to a solution 2.3 Hz, Ar-H, 1H), 7.35–7.30 (m, Ar-H, 1H), 7.02 (t, J = 9.7 Hz,
of 5-bromo-2-fluorophenol (5.0 g, 26 mmol) in water (30 mL), Ar-H), 2.49 (s, S(CH3), 3H) ppm. 13C NMR: δ 159.7 (d, J =
and the resulting solution was kept at 25 °C for 15 min, then 243 Hz, ArCF), 130.5 (d, J = 3.3 Hz, ArC), 130.1 (d, J = 7.5 Hz,
cooled to 10 °C.
A solution of N,N-dimethylthiocarb- ArC), 129.8 (d, J = 18.8 Hz, ArC), 117.9 (d, J = 3.3 Hz, ArC),
amoylchloride (4.2 g, 34 mmol) in tetrahydrofuran (10 mL) 117.7 (d, J = 23.3 Hz, ArC), 14.9 (S(CH3)) ppm.
was added subsequently. The reaction mixture was stirred at
(4-Fluoro-4′-methylbiphenyl-3-yl)(methyl)sulfane (2e). Prepared
25 °C for 6 h and extracted with ethyl acetate (3 × 75 mL). The similarly to biphenyl 1b, with 5-bromo-2-fluorothioanisole 2d
combined organic layers were washed with water (0.15 L) and (0.50 g, 2.3 mmol) instead of 3-iodothioanisole. The product
brine (0.15 L) and concentrated in vacuo. The product was puri- was purified by chromatography (silica gel; eluent: hexane–
fied by chromatography (silica gel; eluent: hexane–dichloro- dichloromethane, 19 : 1) to afford a colorless oil (0.27 g, 51%).
methane, 9 : 1) to afford a white solid (3.6 g, 50%); m. 1H NMR (CD3CN): δ 7.56–7.52 (m, Ar-H, 3H), 7.46–7.41 (m,
p. 100–102 °C. 1H NMR (CD3CN): δ 7.48–7.41 (m, Ar-H, 1H), Ar-H, 1H), 7.32 (s, Ar-H, 1H), 7.29 (s, Ar-H, 1H), 7.17 (t, J =
7.37–7.34 (m, Ar-H, 1H), 7.16 (t, J = 9.7 Hz, Ar-H, 1H), 3.38 (s, 8.8 Hz, 1H), 2.56 (s, S(CH3), 3H), 2.40 (s, Ar-CH3, 3H) ppm.
NCH3, 3H), 3.31 (s, NCH3, 3H) ppm. 13C NMR: δ 186.9 (s, 13C NMR: δ 160.3 (d, J = 242 Hz, ArCF), 139.0 (d, J = 3.4 Hz,
ArOCS), 155.1 (d, J = 248 Hz, ArCF), 143.1 (d, J = 13.1 Hz, ArC), 138.6 (ArC), 137.8 (ArC), 130.6 (ArC × 2), 127.9 (ArC × 2),
ArCO), 131.2 (d, J = 7.8 Hz), 129.3 (ArC), 119.1 (d, J = 20.5 Hz), 127 (d, J = 17.8 Hz, ArC), 127 (d, J = 2.8 Hz, ArC), 126.1 (d, J =
116.4 (d, J = 3.4 Hz), 41.7 (N(CH3)2) ppm. HRMS (ESI) m/z 8.3 Hz, ArC), 116.3 (d, J = 22.0 Hz, ArC), 21.2 (ArCH3), 15.2 (d,
calcd for C9H9BrFNOS ([M
299.946435.
+
Na]+) 299.946446, found J = 2.3 Hz, S(CH3)2) ppm. HRMS (ESI) m/z calcd for C14H12FS
([M]+) 232.071705, found 232.071651.
S-(5-Bromo-2-fluorophenyl)dimethylthiocarbamate
(2b).
(4-Fluoro-4′-methylbiphenyl-3-yl)dimethylsulfonium tetrafluoro-
Dimethylthiocarbamate 2a (3.6 g, 13 mmol) was dissolved in borate (2). Obtained similarly to sulfonium 1, using biphenyl
diphenyl ether (0.10 L) and heated to 260 °C for 10 h. After 2e (0.10 g, 0.43 mmol) instead of biphenyl 1b. White solid
1
cooling, the reaction mixture was loaded onto a silica gel (0.12 g, 86%); m.p. 144–145 °C. H NMR (CD3CN): δ 8.11–8.04
column and eluted with hexane. After the elution of diphenyl (m, Ar-H, 2H), 7.62 (d, J = 8.0 Hz, Ar-H, 2H), 7.56 (t, J = 9.1 Hz,
ether, the product was isolated (eluent: hexane–ethyl acetate Ar-H, 1H), 7.36 (d, J = 7.9 Hz, Ar-H, 2H), 3.28 (s, S(CH3)2, 3H),
1 : 1) as a light brown solid (3.2 g, 89%); m.p. 80–81 °C. 1H 2.42 (s, Ar-CH3, 3H). 13C NMR: δ 161.7 (d, J = 255 Hz, ArCF),
This journal is © The Royal Society of Chemistry 2013
Org. Biomol. Chem., 2013, 11, 3116–3127 | 3123

View Article Online
Paper
Organic & Biomolecular Chemistry
140.9 (d, J = 3.2 Hz, ArC), 140.0 (ArC), 136.4 (d, J = 8.7 Hz, ArC), concentrated. The product was then dissolved in ethyl acetate
135.7 (ArC), 131.0 (2 × ArC), 130.0 (ArC), 128.1 (2 × ArC), 119.2 (0.10 L), washed with water (2 × 75 mL) and brine (75 mL), and
(d, J = 21 Hz, ArC), 113.5 (d, J = 14.7 Hz, ArC), 28.6 (d, J = dried with Na2SO4. Evaporation of the solvent afforded the
1
2.1 Hz, S(CH3)2), 21.2 (ArCH3) ppm. HRMS (ESI) m/z calcd for title compound as a light yellow liquid (1.8 g, 97%). H NMR
C15H16S ([M]+) 247.095126, found 247.094982.
(CD3CN): δ 7.31 (s, Ar-H, 1H), 7.22 (d, J = 8.1 Hz, Ar-H, 1H),
5-Bromo-2-methylaniline (3a).27 Tin chloride dihydrate 7.09 (d, J = 8.0 Hz, Ar-H, 1H), 2.49 (s, S(CH3), 3H), 2.25 (s,
(21 g, 93 mmol) was added portion wise to a solution of Ar-CH3, 3H) ppm. 13C NMR: δ 142.0, 135.5, 132.5, 128.4, 127.4,
4-bromo-2-nitrotoluene (5.0 g, 23 mmol) in ethyl acetate 121.1 (ArC), 20.0 (ArCH3), 15.5 (S(CH3)) ppm.
(0.10 L) at 0 °C. The reaction mixture was heated to reflux for
(4,4′-Dimethylbiphenyl-3-yl)methylsulfane
(3e). Prepared
4 h and neutralized with 1.0 M NaOH (0.10 L) after cooling to similarly to biphenyl 1b, with 5-bromo-2-methylthioanisole 3d
0 °C. The reaction mixture was filtered and the precipitate (0.50 g, 2.0 mmol) instead of 3-iodothioanisole. The product
washed with ethyl acetate (0.20 L). The combined organic was purified by chromatography (silica gel; eluent: hexane–
layers were washed with water (2 × 0.10 L) and brine (0.10 L), dichloromethane 99 : 1) to afford a colorless oil, which solidi-
dried with anhydrous Na2SO4 and concentrated in vacuo. The fies upon standing (0.32 g, 61%); m.p. 30–31 °C. 1H NMR
product was purified by chromatography (silica gel; eluent: (CD3CN): δ 7.56 (d, J = 7.9 Hz, Ar-H, 2H), 7.42 (s, Ar-H, 1H),
hexane–ethyl acetate, 9 : 1) to afford a dark brown liquid (4.2 g, 7.34–7.23 (m, 4H, Ar-H), 2.55 (s, S(CH3), 3H), 2.39 (s, Ar-CH3,
1
97%). H NMR (CD3CN): δ 6.89 (d, J = 7.9 Hz, Ar-H, 1H), 6.82 3H), 2.34 (s, Ar-CH3, 3H) ppm. 13C NMR: δ 140.8, 139.7, 139.2,
(s, Ar-H, 1H), 6.72 (d, J = 7.9 Hz, Ar-H, 1H), 4.20 (br, NH2, 2H), 138.6, 135.6, 132.0 (ArC), 131.0 (ArC × 2), 128.2 (ArC × 2), 124.4
2.07 (Ar-CH3, 3H) ppm. 13C NMR: 148.6, 132.7, 122.0, 120.7, (ArC), 124.1 (ArC), 21.7 (ArCH3), 20.2 (ArCH3), 15.8 (S(CH3))
120.4, 117.4 (ArC), 17.2 (ArCH3) ppm. HRMS (ESI) m/z calcd for ppm. HRMS (ESI) m/z calcd for C15H16S ([M + H]+) 229.104548,
C7H8BrN ([M + H]+) 185.991288, found 185.991305.
found 229.104679.
(4,4′-Dimethylbiphenyl-3-yl)dimethylsulfonium tetrafluoro-
S-5-Bromo-2-methylphenyl O-ethyl carbonodithioate (3b).27
Concentrated HCl (8.0 mL) was added to a cooled suspension borate (3). Obtained similarly to sulfonium 1, using biphenyl
of aniline 3a (8.5 g, 46 mmol) in water (25 mL) followed by a 3e (0.10 g, 0.44 mmol) instead of biphenyl 1b. White solid
solution of sodium nitrite (3.2 g, 46 mmol) in water (10 mL) (0.12 g, 83%); m.p. 140–141 °C. 1H NMR (CD3CN): δ 8.01 (s,
over a 30 min period. The solution was kept at 0 °C for 2 h, Ar-H, 1H), 7.91 (d, J = 8.0 Hz, Ar-H, 1H), 7.65 (d, J = 8.0 Hz,
and a solution of potassium ethyl xanthate (15 g, 91 mmol) in Ar-H, 2H), 7.56 (d, J = 8.1 Hz, Ar-H, 1H), 7.36 (d, J = 7.8 Hz,
water (15 mL) was added dropwise. The reaction mixture was Ar-H, 2H), 3.20 (s, S(CH3)2, 6H), 2.65 (s, Ar-CH3, 3H), 2.42 (s,
then heated to 55 °C for 1 h. The product was extracted with Ar-CH3, 3H) ppm. 13C NMR: δ 143.0, 140.1, 139.8, 136.5, 133.7,
diethyl ether (3 × 0.10 L), washed with water (2 × 0.10 L) and 133.5 (ArC), 130.9 (ArC × 2), 128.1 (ArC × 2), 126.7, 126.1 (ArC),
brine (0.10 L), dried with Na2SO4 and concentrated. The 29.0 (S(CH3)2), 21.3 (ArCH3), 19.6 (ArCH3) ppm. HRMS (ESI)
product was purified by chromatography (silica gel; eluent: m/z calcd for C16H19S ([M]+) 243.120198, found 243.120065.
hexane); further purification by recrystallization in hexane
4-Bromo-2-isopropylaniline
(4a). N-Bromosuccinimide
afforded pure white needles (2.1 g, 16%); m.p. 62–63 °C. 1H (7.2 g, 41 mmol) was added to a mixture of 2-isopropylaniline
NMR (CD3CN): δ 7.68 (s, Ar-H, 1H), 7.57 (d, J = 8.2 Hz, Ar-H, (5.0 g, 37 mmol) and ammonium acetate (0.29 g, 3.7 mmol) in
1H), 7.31 (d, J = 8.2 Hz, Ar-H, 1H), 4.60 (q, J = 7.1 Hz, OCH2, acetonitrile (0.15 L), and the resulting solution was kept at
2H), 2.36 (s, Ar-CH3, 3H), 1.30 (t, J = 7.08 Hz, CH3, 3H) ppm. 25 °C for 1 h. The reaction mixture was concentrated, diluted
13C NMR: δ 213.0 (CS), 143.3, 139.5, 135.1, 134.0, 132.9, 120.3 with water (0.15 L) and extracted with ethyl acetate (3 ×
(ArC), 72.4 (ArCH3), 20.82 (OCH2), 14.3 (ArCH3) ppm. HRMS 75 mL). The combined organic layers were combined and
(ESI) m/z calcd for C10H12BrOS2 ([M + H]+) 290.950745, found washed with brine (0.10 L), dried with Na2SO4 and concen-
290.950881.
trated in vacuo. The product was purified by chromatography
5-Bromo-2-methylbenzenethiol (3c).27 Xanthate 3b (3.0 g, (silica gel; eluent: hexane–ethyl acetate, 9 : 1) to afford a color-
1
10 mmol) was heated to reflux with potassium hydroxide less liquid (4.2 g, 53%). H NMR (CD3CN): δ 7.21 (s, Ar-H, 1H),
(1.7 g, 31 mmol) in ethanol (30 mL) for 5 h. The solution was 7.09 (d, J = 8.6 Hz, Ar-H, 1H), 6.61 (d, J = 8.5 Hz, Ar-H, 1H),
neutralized with HCl (10% in water), extracted with diethyl 4.10 (br, Ar-NH2, 2H), 2.90 (hept., J = 6.8 Hz, Ar-CH, 1H), 1.21
ether (3 × 75 mL), dried with Na2SO4 and evaporated to afford (d, J = 6.8 Hz, CH-(CH3)2, 6H) ppm. 13C NMR: δ 145.0, 135.6,
1
a light yellow liquid (1.9 g, 89%). H NMR (CD3CN): δ 7.82 (s, 129.9, 128.9, 117.9, 110.0 (ArC), 28.3 (CH(CH3)2), 22.5 ((CH3)2)
Ar-H, 1H), 7.24 (d, J = 8.1 Hz, Ar-H, 1H), 7.11 (d, J = 8.1 Hz, Ar- ppm. HRMS (ESI) m/z calcd for C9H12BrN ([M
H, 1H), 3.98 (s, Ar-SH, 1H), 2.25 (s, Ar-CH3, 3H) ppm. 13C NMR: 214.022588, found 214.022652.
+
H]+)
δ 136.1, 135.6, 133.1, 132.5, 129.7, 120.3 (ArC), 20.82 (ArCH3)
ppm. HRMS (ESI) m/z calcd for C7H6BrS ([M]2+) 401.874168, aniline 4a (4.0 g, 19 mmol) in water (20 mL) was added conc.
found 401.874474. HCl (9.3 mL) dropwise, before cooling the mixture to 0 °C. An
4-Bromo-1-iodo-2-isopropylbenzene (4b). To a suspension of
5-Bromo-2-methylthioanisole (3d). A mixture of thiol 3c ice-cold solution of sodium nitrite (2.6 g, 37 mmol; 20 mL) was
(1.7 g, 8.6 mmol), methyl iodide (7.3 g, 51 mmol) and dry added dropwise and the reaction mixture was allowed to stand
potassium carbonate (3.6 g, 26 mmol) in acetone (60 mL) were for 1 h at 0 °C until the addition of a solution of potassium
heated to 40 °C for 15 h. The reaction mixture was iodide (6.2 g, 37 mmol) in water (25 mL). The reaction mixture
3124 | Org. Biomol. Chem., 2013, 11, 3116–3127
This journal is © The Royal Society of Chemistry 2013

View Article Online
Organic & Biomolecular Chemistry
Paper
was then kept at 25 °C for 6 h. The product was extracted with 4-bromophenylboronic acid (0.80 g, 4.0 mmol) in methanol
ethyl acetate (2 × 0.10 L), and the combined organic layers (5 mL) were added to a mixture of 3-iodothioanisole (1a; 1.0 g,
were washed with brine (0.10 L), dried with Na2SO4 and evapo- 4.0 mmol) and palladium(0) tetrakis(triphenylphosphine)
rated. The product was purified by chromatography (silica gel; (0.21 g, 0.20 mmol) in degassed toluene (20 mL). The reaction
eluent: hexane–dichloromethane, 19 : 1) to afford a light brown mixture was heated to 80 °C for 24 h under an inert atmos-
liquid (2.8 g, 46%). 1H NMR (CD3CN): δ 7.72 (d, J = 8.4 Hz, phere. After evaporating the solvent, the precipitate was treated
Ar-H, 1H), 7.45 (s, Ar-H, 1H), 7.07 (d, J = 8.4 Hz, Ar-H, 1H), 3.14 with a 2.0 M sodium carbonate solution (25 mL) and with con-
(hept., J = 6.8 Hz, CH(CH3)2, 1H), 1.21 (d, J = 6.5 Hz, CH(CH3)2, centrated ammonia (5.0 mL), then it was extracted with ethyl
6H) ppm. 13C NMR: δ 153.8, 141.9, 131.8, 130.2, 123.7, 99.6 acetate (2 × 0.20 L). The combined organic layers were passed
(ArC), 39.1 (CH(CH3)2), 23.1 ((CH3)2) ppm.
through a pad of celite, and the filtrate was washed with water
4-Bromo-2-isopropylbenzoic acid (4c). A solution of butyl- (0.10 L), dried with Na2SO4 and evaporated. The product was
lithium (2.0 M) in hexane (3.7 mL, 9.3 mmol) was added to a purified by chromatography (silica gel; eluent: hexane–dichloro-
solution of bromoarene 4b (2.8 g, 8.4 mmol) in dry tetrahydro- methane (19 : 1) to afford a light yellow liquid (0.66 g, 59%). 1H
furan (50 mL), and the reaction mixture was kept at −75 °C for NMR (CD3CN): δ 7.66–7.57 (m, Ar-H, 4H), 7.51 (s, Ar-H, 1H),
2 h before being poured onto an excess of freshly crushed dry 7.41–7.40 (m, Ar-H, 2H), 7.30 (m, Ar-H, 1H), 2.55 (s, S(CH3),
ice. The resulting slurry was subsequently acidified with a 3H) ppm. 13C NMR: δ 141.7, 141.0, 140.9 (ArC), 133.2 (ArC × 2),
solution of hydrogen chloride in diethyl ether (10 mL). After 130.9 (ArC), 130.3 (ArC × 2), 126.7, 125.7, 124.8, 122.8 (ArC),
evaporation of the solvent, the precipitate was diluted with 16.0 (S(CH3)) ppm. HRMS (ESI) m/z calcd for C13H11BrS ([M]+)
water (50 mL) and the product extracted with ethyl acetate (3 × 277.975935, found 277.976001.
75 mL). The residue crystallized as white needles in hexane
Methyl(4′′-methyltriphenyl-3-yl)sulfane (5b). Prepared simi-
(1.2 g, 58%); m.p. 105–106 °C. 1H NMR (CD3CN): δ 9.65 (br, larly to biphenyl 1b, with biphenyl 5a (0.50 g, 1.8 mmol)
Ar-COOH, 1H), 7.66 (m, Ar-H, 2H), 7.45 (d, J = 8.3 Hz, Ar-H, instead of 3-iodothioanisole (1a). The product was purified by
1H), 3.78 (hept., J = 6.8 Hz, CH(CH3)2, 1H), 1.25 (d, J = 6.8 Hz, chromatography (silica gel; eluent: hexane–ethyl acetate, 9 : 1
CH(CH3)2, 6H) ppm. 13C NMR: δ 169.2 (ArCOOH), 153.2, 132.7, to afford a white solid (0.35 g, 67%); m.p. 112–113 °C. 1H NMR
130.6, 130.0, 129.8, 127.2 (ArC), 30.4 (CH(CH3)2), 24.0 ((CH3)2) (CD3CN): δ 7.74 (s, Ar-H, 4H), 7.61 (t, J = 6.0 Hz, Ar-H, 3H),
ppm. HRMS (ESI) m/z calcd for C10H10BrO2 ([M − H]−) 7.45 (dt, J = 15.0, 9.0 Hz, Ar-H, 2H), 7.33–7.28 (m, Ar-H, 3 H),
240.985869, found 240.986963.
2.57 (s, S(CH3), 3H), 2.41 (s, Ar-CH3, 3H) ppm. 13C NMR:
3-Isopropyl-4′-methylbiphenyl-4-carboxylic acid (4d). Pre- δ 142.2, 141.2, 140.5, 140.2, 138.6, 138.4 (ArC), 130.7 (ArC × 2),
pared similarly to biphenyl 1b, with 4-bromo-2-isopropylben- 130.5 (ArC), 128.5 (ArC × 2), 128.2 (ArC × 2), 127.78 (ArC × 2),
zoic acid 4c (0.40 g, 1.6 mmol) instead of 3-iodothioanisole. 126.2, 125.4, 124.6, 21.2 (ArCH3), 15.7 (S(CH3)) ppm. HRMS
The title compound crystallized as white prisms in acetonitrile (ESI) m/z calcd for C20H18S ([M + H]+) 291.120198, found
1
(0.26 g, 62%); m.p. 168–169 °C. H NMR (CD3CN): δ 9.45 (br, 291.120379.
ArCOOH, 1H), 7.83 (d, J = 8.1 Hz, Ar-H, 1H), 7.70 (s, Ar-H, 1H),
Dimethyl(4′′-methyltriphenyl-3-yl)sulfonium trifluoromethane-
7.61 (d, J = 8.0 Hz, Ar-H, 2H), 7.51 (d, J = 8.0 Hz, Ar-H, 1H), sulfonate (5). Iodomethane (24 mg, 0.17 mmol) and silver tri-
7.31 (d, J = 7.8 Hz, Ar-H, 2H), 3.87 (hept., J = 6.8 Hz, CH(CH3)2, flate (23 mg, 90 μmol) were added to a solution of sulfide 5b
1H), 2.41 (s, Ar-CH3, 3H), 1.30 (d, J = 6.8 Hz, CH(CH3)2, 6H) (25 mg, 80 μmol) in acetone and the reaction mixture was kept
ppm. 13C NMR: δ 169.6 (ArCOOH), 151.5, 145.3, 139.2, 138.2 at 25 °C for 12 h. The precipitate was filtered, and evaporation
131.7 (ArC), 130.7 (ArC × 2), 129.3 (ArC), 128.1 (ArC × 2), 125.7, of the filtrate afforded the title compound as a white solid
125.0 (ArC), 30.7 (CH(CH3)2), 24.2 ((CH3)2), 21.2 (ArCH3) ppm. (32 mg, 82%); m.p. 137–138 °C. 1H NMR (CD3CN): δ 8.21 (s,
HRMS (ESI) m/z calcd for C17H17O2 ([M − H]−) 253.12230, Ar-H, 1H), 8.12 (d, J = 9.0 Hz, Ar-H, 1H), 7.91–7.79 (m, Ar-H,
found 253.123337.
6H), 7.64 (d, J = 8.1 Hz, Ar-H, 2H), 7.34 (d, J = 7.8 Hz, 2H), 3.21
Sodium 3-isopropyl-4′-methylbiphenyl-4-carboxylate (4). (s, S(CH3)2, 6H), 2.42 (s, Ar-CH3, 3H) ppm. 13C NMR: δ 144.3,
NaOH (97% pellets; 8.5 mg, 0.21 mmol) was added to a sus- 142.3, 139.0, 138.0, 137.9, 133.8, 132.4 (ArC), 130.7 (ArC × 2),
pension of carboxylic acid 4d (50 mg, 0.19 mmol) in water 129.5, 128.9 (ArC), 128.8 (ArC × 2), 128.5 (ArC × 2), 127.9
(5.0 mL), and the solution was kept at 25 °C for 1 h, before (ArC × 2), 126.9 (ArC), 29.4 (S(CH3)2), 21.2 (ArCH3) ppm. HRMS
being concentrated. The residue was dried under high (ESI) m/z calcd for C21H21OS ([M]+) 305.135848, found
vacuum; white solid (54 mg, 99%); m.p. >300 °C. 1H NMR 305.135464.
(D2O): δ 7.61–7.58 (m, Ar-H, 3H), 7.47–7.44 (m, Ar-H, 1H),
7.34–7.32 (m, Ar-H, 3H), 3.26 (hept., J = 6.7 Hz, CH(CH3)2, 1H),
Computational work
2.36 (s, Ar-CH3, 3H), 1.26 (d, J = 6.8 Hz, 6H) ppm. 13C NMR: δ Calculations were carried out with the TURBOMOLE suite of
179.5 (Ar-COONa), 145.3, 140.5, 138.7, 138.1, 137.6 (ArC), 129.8 programs (version 6.3.1),28 and with the Gaussian 09
(ArC × 2), 126.9 (ArC × 2), 126.4, 124.0, 123.8 (ArC), 30.5 (CH package.29 For TURBOMOLE jobs, input files were first gene-
(CH3)2), 23.4 ((CH3)2), 20.2 (ArCH3) ppm. HRMS (ESI) m/z calcd rated with the graphical interface TmoleX (version 3.3)30 with
for C17H17O2 ([M + 2Na]+) 299.101846, found 299.101804.
default parameters, and corrected manually or with TURBO-
(4′-Bromobiphenyl-3-yl)(methyl)sulfane (5a). An aqueous MOLE 6.3.1 to satisfy calculation needs; the m4 grid31 was
solution of sodium carbonate (2.0 M; 5.0 mL) and used throughout this study. Gaussian 09 jobs were generated
This journal is © The Royal Society of Chemistry 2013
Org. Biomol. Chem., 2013, 11, 3116–3127 | 3125

View Article Online
Paper
Organic & Biomolecular Chemistry
using GaussView 5.0,32 and corrected manually; the ultrafine
grid was employed. In all cases, convergence criteria were 10−7
Hartree and 10−4 atomic units as the maximum norm of the
Cartesian gradient. Def2-TZVPP basis sets14 were used for the
evaluation of free guests, and def2-SVP basis sets for CB[7]-
containing assemblies. All DFT-D and MP2 calculations were
carried out with TURBOMOLE within the resolution of the
identity approximation. IEFPCM corrections to the electronic
energies were obtained using Gaussian 09 by calculating the
energy difference between single point calculations carried out
with and without the solvent on B3LYP-D-optimized structures.
The B97-D functional as implemented in Gaussian 09 was
used for these single point calculations. COSMO was applied
with default parameters using TURBOMOLE.
( j) A. J. Prinsen, A. Koeberg-Telder and H. Cerfontain,
Tetrahedron, 1970, 26, 1953; (k) H. Tomioka, K. Kimoto,
H. Murata and Y. Izawa, J. Chem. Soc., Perkin Trans. 1, 1991,
471; (l) C. K. Cheung, R. S. Wedinger and W. J. le Noble,
J. Org. Chem., 1989, 54, 570.
4 (a) R. N. Armstrong, H. L. Ammon and J. N. Darnow, J. Am.
Chem. Soc., 1987, 109, 2077; (b) W. F. Forbes and
W. A. Mueller, J. Am. Chem. Soc., 1957, 79, 6495;
(c) H. A. Karnes, M. L. Rose, J. W. Collat and
M. S. Newman, J. Am. Chem. Soc., 1968, 90, 458;
(d) H. A. Karnes, B. D. Kybett, M. H. Wilson, J. L. Margrave
and M. S. Newman, J. Am. Chem. Soc., 1965, 87, 5554;
(e) Y. H. Lai and P. Chen, J. Chem. Soc., Perkin Trans. 2,
1989, 1665.
5 J. Kim, I. S. Jung, S. Y. Kim, E. Lee, J. K. Kang, S. Sakamoto,
K. Yamaguchi and K. Kim, J. Am. Chem. Soc., 2000, 122,
540.
Acknowledgements
6 For recent reviews, see: (a) J. Lagona, P. Mukhopadhyay,
S. Chakrabarti and L. Isaacs, Angew. Chem., Int. Ed., 2005,
44, 4844; (b) K. Kim, N. Selvapalam, Y. H. Ko, K. M. Park,
D. Kim and J. Kim, Chem. Soc. Rev., 2007, 36, 267;
(c) W. H. Huang, S. Liu and L. Isaacs, Cucurbit[n]urils, in
Modern Supramolecular Chemistry: Strategies for Macrocycle
Synthesis, ed. F. Diederich, P. J. Stang and R. R. Tykwinski,
Wiley-VCH Verlag GmbH & Co., Weinheim, 2008, p. 113;
(d) L. Isaacs, Chem. Commun., 2009, 619; (e) E. Masson,
X. Ling, R. Joseph, L. Kyeremeh-Mensah and X. Lu, RSC
Adv., 2012, 2, 1213.
7 (a) S. Moghaddam, C. Yang, M. Rekharsky, Y. H. Ko,
K. Kim, Y. Inoue and M. K. Gilson, J. Am. Chem. Soc., 2011,
133, 3570; (b) M. V. Rekharsky, T. Mori, C. Yang, Y. H. Ko,
N. Selvapalam, H. Kim, D. Sobransingh, A. E. Kaifer, S. Liu,
L. Isaacs, W. Chen, S. Moghaddam, M. K. Gilson, K. Kim
and Y. Inoue, Proc. Natl. Acad. Sci. U. S. A., 2007, 104,
20737.
This work was supported by the American Chemical Society
Petroleum Research Fund (PRF No. 51053-ND4), the Depart-
ment of Chemistry and Biochemistry, the College of Arts and
Sciences and the Vice President for Research at Ohio Univer-
sity. We are extremely grateful to the Ohio Supercomputer
Center (OSC) in Columbus for its generous allocation of com-
puting time.
Notes and references
1 M. Rieger and F. H. Westheimer, J. Am. Chem. Soc., 1950,
72, 19.
2 (a) G. Yamamoto, M. Suzuki and M. Ōki, Bull. Chem. Soc.
Jpn., 1983, 56, 809; (b) I. Eventova, E. B. Nadler, E. Rochlin,
J. Frey and Z. Rappoport, J. Am. Chem. Soc., 1993, 115,
1290; (c) K. Maeda, Y. Okamoto, O. Toledano, D. Becker,
S. E. Biali and Z. Rappoport, J. Org. Chem., 1994, 59, 5473;
(d) F. Imashlro, K. Takegoshl, T. Terao and A. Salka, J. Phys.
Chem., 1982, 86, 3306; (e) S. L. Chien and R. Adams, J. Am.
Chem. Soc., 1934, 56, 1787; (f) L. A. Paquette and
J. M. Gardlik, J. Am. Chem. Soc., 1980, 102, 5033;
(g) L. A. Paquette, J. M. Gardlik, L. K. Johnson and
K. J. McCullough, J. Am. Chem. Soc., 1980, 102, 5026.
8 (a) K. Moon and A. E. Kaifer, Org. Lett., 2004, 6, 185;
(b) W. L. Mock and N. Y. Shih, J. Org. Chem., 1986, 51,
4440; (c) H. J. Buschmann, A. Wego, A. Zielesny and
E. Schollmeyer, J. Inclusion Phenom. Macrocyclic Chem.,
2006, 54, 241.
9 G. C. Lloyd-Jones, J. D. Moseley and J. S. Renny, Synthesis,
2008, 661.
3 (a) C. Heiss, E. Marzi and M. Schlosser, Eur. J. Org. Chem., 10 Calculated at the PW6B95-D3(BJ)/def2-TZVPP//B3LYP-D/
2003, 4625; (b) J. Gorecka, C. Heiss, R. Scopelliti and
M. Schlosser, Org. Lett., 2004, 6, 4591; (c) C. Heiss, F. Cottet
def2-TZVPP level; see p. 3 and experimental section for
details.
and M. Schlosser, Eur. J. Org. Chem., 2005, 5236; 11 A. Mazzanti, L. Lunazzi, R. Ruzziconi, S. Spizzichino and
(d) M. Schlosser, C. Heiss and F. Leroux, Eur. J. Org. Chem., M. Schlosser, Chem.–Eur. J., 2010, 16, 9186.
2006, 735; (e) M. Schlosser, F. Cottet, C. Heiss, O. Lefebvre, 12 (a) A. D. Becke, J. Chem. Phys., 1993, 98, 1372;
M. Marull, E. Masson and R. Scopelliti, Eur. J. Org. Chem.,
2006, 729; (f) F. Mongin, E. Marzi and M. Schlosser,
(b) P. J. Stephens, F. J. Devlin, C. F. Chabalowski and
M. J. Frisch, J. Phys. Chem., 1994, 98, 11623.
Eur. J. Org. Chem., 2001, 2771; (g) R. W. Alder, Chem. Rev., 13 S. Grimme, J. Comput. Chem., 2006, 27, 1787.
1989, 89, 1215; (h) M. Decouzon, P. Ertl, O. Exner, J. F. Gal 14 F. Weigend and R. Ahlrichs, Phys. Chem. Chem. Phys., 2005,
and P. C. Maria, J. Am. Chem. Soc., 1993, 115, 12071;
(i) A. F. Pozharskii, O. V. Ryabtsova, V. A. Ozeryanskii,
A. V. Degtyarev, O. N. Kazheva, G. G. Alexandrov and
7, 3297; basis sets available at http://bases.turbo-forum.
com/tbl/tbl.html and http://www.cosmologic.de/basis-sets/
basissets.php.
O. A. Dyachenko, J. Org. Chem., 2003, 68, 10109; 15 A. D. Becke, J. Chem. Phys., 1997, 107, 8554.
3126 | Org. Biomol. Chem., 2013, 11, 3116–3127
This journal is © The Royal Society of Chemistry 2013

View Article Online
Organic & Biomolecular Chemistry
Paper
16 (a) S. Grimme, S. Ehrlich and L. Goerigk, J. Comput. Chem.,
2011, 32, 1456; (b) S. Grimme, J. Antony, S. Ehrlich and
H. Krieg, J. Chem. Phys., 2010, 132, 154104.
17 Y. Zhao and D. G. Truhlar, J. Phys. Chem. A, 2005, 109,
5656.
18 (a) R. Improta, G. Scalmani, M. J. Frisch and V. Barone,
J. Chem. Phys., 2007, 127, 074504; (b) R. Improta, V. Barone,
G. Scalmani and M. J. Frisch, J. Chem. Phys., 2006, 125,
GmbH since 2007; available from http://www.turbomole.
com and distributed by COSMOlogic GmbH & Co. KG,
D-51381 Leverkusen (http://www.cosmologic.de); see also:
(a) R. Ahlrichs, M. Bär, M. Häser, H. Horn and C. Kölmel,
Chem. Phys. Lett., 1989, 162, 165; (b) M. Sierka,
A. Hogekamp and R. Ahlrichs, J. Chem. Phys., 2003, 118,
9136; (c) P. Deglmann, K. May, F. Furche and R. Ahlrichs,
Chem. Phys. Lett., 2004, 384, 103.
054103; (c) J. Tomasi, B. Mennucci and R. Cammi, Chem. 29 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
Rev., 2005, 105, 2999.
19 S. Grimme, J. Antony, T. Schwabe and C. Mück-Lichtenfeld,
Org. Biomol. Chem., 2007, 5, 741.
20 E. Masson, Org. Biomol. Chem., 2013, DOI: 10.1039/
c3ob26704k, Advance Article.
21 L. Goerigk and S. Grimme, Phys. Chem. Chem. Phys., 2011,
13, 6670.
22 (a) A. Klamt and G. Schüürmann, J. Chem. Soc., Perkin
Trans. 2, 1993, 5, 799; (b) A. Klamt and V. Jonas, J. Chem.
Phys., 1996, 105, 9972; the COSMO model uses a hypotheti-
cal solvent of infinite dielectric constant; the error caused
by this approximation is insignificant for highly polar
solvent like water.
23 A. Praetorius, D. M. Bailey, T. Schwarzlose and W. M. Nau,
Org. Lett., 2008, 10, 4089.
24 A 1 : 1 binding model was applied.
M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone,
B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato,
X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng,
J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda,
J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao,
H. Nakai, T. Vreven, J. A. Montgomery Jr., J. E. Peralta,
F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers,
K. N. Kudin, V. N. Staroverov, R. Kobayashi, J. Normand,
K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar,
J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene,
J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo,
R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin,
R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin,
K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador,
J. J. Dannenberg, S. Dapprich, A. D. Daniels, Ö. Farkas,
J. B. Foresman, J. V. Ortiz, J. Cioslowski and D. J. Fox,
Gaussian, Inc., Wallingford, CT, 2009.
25 M. V. Rekharsky, Y. H. Ko, N. Selvapalam, K. Kim and
Y. Inoue, Supramol. Chem., 2007, 19, 39.
26 X. Ling, E. L. Samuel, D. L. Patchell and E. Masson, Org.
Lett., 2010, 12, 2730.
30 COSMOlogic GmbH & Co. KG, D-51381 Leverkusen (http://
www.cosmologic.de).
31 K. Eichkorn, F. Weigend, O. Treutler and R. Ahlrichs,
Theor. Chem. Acc., 1997, 97, 119.
27 A. Gopalsamy, US Pat, 0276030 A1, 2007.
28 A development of the University of Karlsruhe and For- 32 Gaussian, Inc., Wallingford, CT 06492 (http://www.
schungszentrum Karlsruhe GmbH, 1989–2007; TURBOMOLE
gaussian.com/g_prod/gv5.htm).
This journal is © The Royal Society of Chemistry 2013
Org. Biomol. Chem., 2013, 11, 3116–3127 | 3127