Atropisomerization in confined space; Cucurbiturils as tools to determine the torsional barrier of substituted biphenyls

Article abstract of DOI:10.1002/ejoc.201301460

The torsional barrier of biphenyls bearing a prochiral dimethylsulfonium group at their 3-position could be determined by variable-temperature ¹H NMR spectroscopy only after encapsulation into cucurbit[7]-or-[8]uril, which triggered the splitting of the two methyl signals. Confinement of the biphenyl units into the macrocycles amplifies the dissymmetry caused by the various ortho-and ortho′-substituents and represents a new tool that can be used to access atropisomerization barriers of bi(hetero)aryl derivatives. Cucurbiturils were used for the first time as analytical accessories to access coveted kinetic parameters, namely the torsional barriers of substituted biphenyls. Copyright

Full text of DOI:10.1002/ejoc.201301460

FULL PAPER
DOI: 10.1002/ejoc.201301460
Atropisomerization in Confined Space; Cucurbiturils as Tools to Determine the
Torsional Barrier of Substituted Biphenyls
Roymon Joseph[a] and Eric Masson*[a]
Keywords: Atropisomerism / Host-guest systems / Supramolecular chemistry / Biaryls / Conformation analysis
The torsional barrier of biphenyls bearing a prochiral di-
methylsulfonium group at their 3-position could be deter-
mined by variable-temperature 1H NMR spectroscopy only
after encapsulation into cucurbit[7]- or -[8]uril, which trig-
gered the splitting of the two methyl signals. Confinement of
the biphenyl units into the macrocycles amplifies the dissym-
metry caused by the various ortho- and orthoЈ-substituents
and represents a new tool that can be used to access atropiso-
merization barriers of bi(hetero)aryl derivatives.
Introduction
Barriers to torsional isomerization along the Caryl–Caryl
axis of bi(hetero)aryl derivatives can be determined by
using a limited number of analytical methods, each covering
a rather specific range of activation energies (see Fig-
ure 1).[1] For example, dynamic nuclear magnetic resonance
spectroscopy (DNMR) may allow the determination of bar-
riers ranging from 5 to 20 kcalmol–1;[2] measuring the op-
tical activity of enantiomerically enriched or pure atropiso-
mers as a function of time at a given temperature affords
Figure 1. Analytical methods allowing the determination of tor-
sional barriers of bi(hetero)aryl derivatives.
activation energies ranging from 23 to 33 kcalmol–1; finally, is 1.8 kcalmol–1, i.e., 95% of calculated barriers for any new
dynamic separation techniques such as dynamic high- series of biphenyl scaffolds are 99% likely to be within
performance liquid chromatography (DHPLC), gas 1.8 kcalmol–1 of experimental data. Barriers up to
chromatography (DGC), stopped-flow DGC and capillary 45 kcalmol–1 could also be calculated by using B3LYP–
zone electrophoresis (CZE, see Figure 1) cover a wide range D,[4,5] B97–D[5,6] or TPSS–D3[7] functionals with slightly
of activation barriers, from approximately 15 to lower accuracy (see Figure 1).[1]
45 kcalmol–1. We note, however, that DNMR is only suc-
Of course, calculations suffer from the inescapable weak-
cessful when prochiral groups connected to the biaryl scaf- ness that predictions may yet fall outside the margin of tol-
folds resonate at different frequencies at low temperatures, erance. New experimental tools to determine these acti-
and appear equivalent at higher temperatures, which is by vation barriers are thus always desirable.
no means guaranteed.[3]
In this study, we show that encapsulation of selected bi-
We have recently shown[1] that torsional barriers of sub- phenyl derivatives bearing prochiral groups into cucurbit[7]-
stituted biphenyls ranging from 6 to 30 kcalmol–1 could be and cucurbit[8]uril (CB[7] and CB[8])[8] allows the straight-
calculated with very high precision and accuracy by density forward determination of their torsional barriers by 1H
functional theory (DFT) by using the dispersion-corrected DNMR; such determination would have been impossible in
B3LYP–D hybrid,[4,5] and triple-ζ doubly-polarized def2- the absence of the macrocycle, due to the indistinguishable
TZVPP basis sets, with enthalpic and entropic corrections resonance of the prochiral groups, regardless of tempera-
at nonzero temperatures; only four barriers out of 39 devi- ture.[9]
ated by more than 1.0 kcalmol–1 from experimental data,
with negligible mean deviation and a mean absolute devia-
tion as low as 0.47 kcalmol–1; the tolerance of this method
Results and Discussion
Guests 1–3 were chosen as model structures to illustrate
this new method (see Scheme 1), because their biphenyl
scaffold was expected to sit inside the cavity of CB[7] and
CB[8], and because the positively charged sulfonium sub-
stituent should enhance binding affinities by interacting
[a] Department of Chemistry and Biochemistry, Ohio University,
181 Clippinger Hall, Athens, Ohio 45701, USA
E-mail: masson@ohio.edu
http://www.ohio.edu/chemistry/faculty/masson.php
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201301460.
Eur. J. Org. Chem. 2014, 105–110
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
105

R. Joseph, E. Masson
FULL PAPER
with the carbonylated portal of the macrocycles. In a recent 0.93 ppm, for the CH2 and CH3 signals, respectively (num-
study, we showed that the affinity of CB[7] for a series of bered 8 and 9 in Figure 2, spectra d and e). Exchanges are
biphenyl derivatives bearing a sulfonium unit at the 3-posi- fast on the NMR time-scale in all cases. Whereas guests 1
tion could reach 106 m–1.[10] The two methyl groups of the and 3 are encapsulated by CB[7] with fast and intermediate
sulfonium unit are intended to be used as diastereotopic exchange rates, respectively, the 2Ј-ethyl substituent of guest
probes during DNMR experiments, but merge into one sin- 2 prevents the formation of any complex with CB[7] (see
1
glet in H NMR spectra carried out in deuterium oxide, the Supporting Information for titrations of guests 1–3 with
[D4]MeOH and [D3]acetonitrile (with the exception of bi- CB[7] and CB[8]). Binding affinities were measured by iso-
phenyl 3 in the latter solvent), or overlap with the signal of thermal titration calorimetry (ITC), and range from
residual water in [D6]DMSO.
1.3ϫ105 to 8.3ϫ106 m–1 (Table 1).
Scheme 1. Preparation of biphenyls 1–3 and structure of biphenyls
4.
Figure 2. 1H NMR spectra of (a) biphenyl 1, (b) assembly 1·CB[7],
(c) assembly 1·CB[8], (d) biphenyl 2, (e) assembly 2·CB[8], (f) bi-
phenyl 3, and (g) assembly 3·CB[8]. Measurements carried out in
D2O. See Scheme 1 for numbering; signals 8 and 9 refer to the 2Ј-
ethyl substituent of biphenyl 2.
Iodoarene 1a was obtained after metalation of 2,4-di-
fluorothioanisole with sec-butyllithium and interception of
the aryllithium intermediate with iodine, and was sub-
sequently coupled to (2-fluorophenyl)boronic acid to afford
biphenyl 1b. Methylation of the latter with trimethylox-
Biphenyls 1–3 are likely very mobile inside the macro-
onium tetrafluoroborate afforded guest 1. A similar pro- cycles, and a series of plausible complexes with similar sta-
cedure was used to obtain guests 2 and 3 (Scheme 1). bilities (differences in electronic contributions lower than
In all three cases, CB[8] encapsulated the phenyl ring 0.8 kcalmol–1) could be obtained from DFT calculations at
bearing substituent R, in deuterium oxide, as indicated by the TPSS-D3(BJ)/def2-TZVP level[7,11] (this functional has
significant upfield shifts of hydrogen nuclei located at the been successfully tested in supramolecular systems,[12] see
3Ј- and 5Ј-positions (0.42 and 0.39 ppm on average) and experimental section for details). Figure 3 shows the opti-
especially the 6Ј-position (δ = 0.66 ppm on average). mized structure of assembly 2·CB[8], with the ethyl substit-
Furthermore, the ethyl sidechain of guest 2 is situated deep uent R located deep inside the cavity of the macrocycle, in
1
inside the cavity of CB[8], with upfield shifts of 0.72 and accordance with H NMR experiments.
Table 1. Binding affinities of guests 1–3 towards CB[n] compounds; activation parameters for the torsional isomerization of biphenyls 1–
3 in CB[n] compounds, and comparison to computational data.
[a]
[e]
[f]
Ka
Tc[b]
[°C]
ΔH‡[c]
TΔS‡[d]
ΔG‡
ΔG‡
ΔΔG‡[g]
ΔG‡Ј[h]
exp
calc
[M–1]
[kcalmol–1]
[kcalmol–1]
[kcalmol–1]
[kcalmol–1]
[kcalmol–1]
[kcalmol–1]
1·CB[7]
1·CB[8]
2·CB[8]
3·CB[8]
3·CB[8][i]
4.4 (Ϯ0.2)ϫ106
1.3 (Ϯ0.1)ϫ105
8.3 (Ϯ0.4)ϫ106
51
54
54
18.2 (Ϯ 0.5)
15.1 (Ϯ 0.3)
14.7 (Ϯ 0.3)
14.2 (Ϯ 0.4)
1.0 (Ϯ 0.5)
–1.2 (Ϯ 0.3)
–0.8 (Ϯ 0.3)
–4.5 (Ϯ 0.5)
17.2
16.3
15.5
18.7
17.0
15.8 (16.3)
15.8 (16.3)
14.4 (15.4)
18.5 (17.4)
18.1 (17.0)
1.4 (0.9)
0.5 (0.0)
1.1 (0.1)
0.2 (1.3)
–1.1 (0.0)
13.9
13.9
14.8
16.0
15.4
6.5 (Ϯ0.4)ϫ106 114
56
[a] Binding affinity. [b] Coalescence temperature. [c] Activation enthalpy. [d] Entropic contribution at coalescence temperature. [e] Free
energy of activation as determined by DNMR experiments. [f] Free energy of activation calculated for the free guest at the B3LYP-D/
def2-TZVPP level, and, in parentheses, after correction for solvation using the IEFPCM model. [g] Deviation of the experimental torsional
barrier inside CB[n] from the calculated barrier of the free guest in the gas phase, and, in parentheses, in aqueous solution. [h] Calculated
torsional barrier of the free guest in the gas phase, after replacement of the dimethylsulfonium unit with a hydrogen atom. [i] In the
presence of CB[8] (0.18 equiv.). All standard errors on experimental free energies of isomerization are 0.1 kcalmol–1.
106
www.eurjoc.org
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2014, 105–110

Torsional Barriers of Substituted Biphenyls
conformations of these CB[n]/guest assemblies is very fast
on the NMR time-scale, and much faster than rotation
along the Caryl–Caryl axis (sharp aromatic signals are ob-
served in all four complexes).[13]
Free energies of activation were determined by using the
Eyring equation and by plotting ln(k/T) vs. 1/T, where k is
the rate of isomerization at temperature T. Enthalpies of
activation ΔH‡ could be obtained from the slope of the best
straight line, and ΔS‡ from its intersection with the y-axis
(see Table 1 and Figure 4, c). Coalescence temperatures
were 51, 54, and 54 °C in the case of assemblies 1·CB[7],
1·CB[8] and 2·CB[8], respectively, and extrapolation af-
forded a coalescence temperature of 114 °C for guest 3 in
CB[8].
With the exception of assembly 1·CB[7], activation en-
thalpies for the three guests were similar (14.2–
15.1 kcalmol–1), and activation entropies (TΔS‡ ranging
from –4.5 to –0.8 kcalmol–1) were responsible for the differ-
ences in free energies of activation. The latter in CB[8] were
16.3, 15.5 and 18.7 kcalmol–1, respectively. The barrier of
torsional isomerization of guest 1 in CB[7] was slightly
higher than in CB[8] (17.2 vs. 16.3 kcalmol–1 at 51 °C). It
is tempting to attribute the difference to the tighter confine-
ment of guest 1 inside CB[7] compared with CB[8]; how-
ever, the difference merely means that the affinity of the
transition state of guest 1 towards CB[7] is only 9.1 times
greater than towards CB[8], compared to 35 times for the
ground state.
Figure 3. Optimized structure of complex 2·CB[8].
The key feature of the spectra of complexes 1·CB[7] and
1·CB[8] to 3·CB[8] is the splitting of the dimethylsulfonium
signals into two well-defined singlets, with remarkably wide
differences in resonance frequencies in some cases (0.02,
0.10, 0.36 and 0.28 ppm, respectively, see Figure 2; no split
is observed in assembly 3·CB[7]). We attribute the splitting
to the loose confinement of the guest inside the macro-
cycles, which still creates a different chemical environment
in the immediate surroundings of the two methyl groups,
and “amplifies” the dissymmetry caused by ortho-substitu-
ents R (see Figure 2).
DNMR experiments were then carried out, and rates of
torsional isomerization for guests 1–3 in CB[8] and biphenyl
1 in CB[7] were determined by line-shape analysis by using
a two-site exchange model (see Figure 4, a and b). This
model is valid because the exchange rate between various
The inconveniently high coalescence temperature of com-
plex 3·CB[8] is caused (1) by the higher activation barrier
of the torsional pathway, and (2) by a significant difference
in resonance frequencies between the two diastereotopic
methylsulfonium groups (δ = 0.28 ppm, see Figure 5). To
narrow the latter, we carried out DNMR experiments by
using a substoichiometric amount of CB[8] (0.18 equiv.,
which translates into a 0.05 ppm difference between the two
methylsulfonium groups, see Figure 5, b). Coalescence was
then observed at 56 °C.
Figure 5. 1H NMR spectra of biphenyl 3, (a) in the absence of
CB[8], and in the presence of (b) CB[8] (0.18 equiv.) and (c) CB[8]
(1.0 equiv.).
Figure 4. (a) Variable-temperature 1H NMR experiments; reso-
nance of the dimethylsulfonium groups in assembly 1·CB[8].
(b) NMR simulation. (c) Eyring plot depicting the rates of tor-
sional isomerizations of guest 1 inside CB[7] (empty blue circles),
guest 1 in CB[8] (filled blue circles), guest 2 in CB[8] (red circles)
and guest 3 in CB[8] (green circles) at various temperatures.
Although the exchange between free guest 3, free CB[8]
and the corresponding assembly is fast on the NMR time-
scale at 25 °C (i.e., its activation barrier is lower than ap-
Eur. J. Org. Chem. 2014, 105–110
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
107

R. Joseph, E. Masson
FULL PAPER
proximately 14 kcalmol–1), it must be greater than the gain
As far as biphenyls 1–3 are concerned, syn conforma-
in free energy upon CB[8] binding (an affinity of tions (with the two methylsulfonium groups pointing
6.5ϫ106 m–1 corresponds to a 8.9 kcalmol–1 stabilization). towards the fluorine atom at the 2-position, see Figure 3)
This barrier is thus not negligible, and the DNMR simula- are preferred in both ground and transition states by 2.5
tions had to be carried out with a four-site exchange model. and 2.0 kcalmol–1 on average, respectively, compared to
Activation energies of 17.0 and 11.7 (Ϯ0.2) kcalmol–1 were anti conformations [electronic energies obtained with the
obtained for the torsional and guest egression pathways, PW6B95-D3(BJ)[11,17] functional after optimization and vi-
respectively, at coalescence temperature. The torsional bar- brational analysis at the B3LYP–D level]. As we have re-
rier of free guest 3 at 56 °C was then readily extrapolated ported in a previous study,[10] solvation smoothens the syn
to 16.9 kcalmol–1 (see the Supporting Information for de- preference, which reaches only 0.6 and 0.5 kcalmol–1 at the
tails). A similar method could be applied to guests 1 and 2 ground and transition states, respectively. In all three bi-
with fractions of bound guests greater than approximately phenyl compounds, the buttressing effect of the dimeth-
75%, should coalescence temperatures between 25 and ylsulfonium unit at the 3-position on the neighboring fluor-
50 °C be desired (see the Supporting Information for ti- ine substituent decreases isomerization rates significantly.
trations of guests 1–3 with CB[n] macrocycles).
Torsional barriers of biphenyls bearing a hydrogen atom at
A question that needs to be addressed is whether acti- the 3-position instead of the sulfonium group were calcu-
vation barriers inside CB[n] compounds are similar to those lated (Table 1), and found to be 0.4 to 2.5 kcalmol–1 lower.
in the gas phase and in organic solvents. The slight differ- Whereas in guests 2 and 3, the ethyl or trifluoromethyl R
ences measured between assemblies 1·CB[7] and 1·CB[8] groups at the 2Ј-position pass over the hydrogen atom at
(0.9 kcalmol–1), as well as between free guest 3 and its the 6-position in the transition state, calculations suggest
CB[8] complex (1.0 kcalmol–1 at 56 °C)[14] suggest so. that the 2Ј-fluoro substituent of guest 1 passes preferentially
Whereas solvent effects on torsional barriers have only been over the buttressed 2-fluoro group, a 2.0 kcalmol–1 advan-
assessed systematically on two occasions by using deriva- tage compared to passage over the fluorine atom at the 6-
tives 4a and 4b (see Scheme 1), they were found to barely position. Whereas the sulfonium unit exerts steric but-
affect Gibbs energies of activation (by less than tressing on its ortho-positioned neighbor, its positive charge
1.2 kcalmol–1 among 33 solvents). Because activation ener- slightly alleviates the electronic density around the fluorine
gies are determined by DNMR with various solvents to co- atom, thereby making it more favorable to pass over the
ver adequate temperature ranges[15] (and therefore, an error 2Ј-fluoro group. When solvation is taken into account, the
in the order of 1.2 kcal/mol is tacitly accepted by neglecting preference for the 2-F/2Ј-F torsional pathway is reduced to
solvent effects), torsional barriers measured inside CB[n] 0.6 kcalmol–1.
macrocycles should be no less valid than those measured
Finally, the scope and limitations of this method was
with various solvents. Furthermore, our previously pub- assessed. When deuterium oxide is used without the ad-
lished DFT calculations were very accurate, even when ne- dition of cosolvents, coalescence or extrapolated
glecting solvation effects unless charged substituents were coalescence temperatures should range from approximately
connected to the ortho and orthoЈ positions of the biphenyl –10 to 110 °C; the corresponding activation barriers are ap-
scaffolds.[1] The cavity of CB[n] compounds, as a chemical proximately 12.5 to 20.5 kcalmol–1. The addition of cosol-
environment, is thus expected to follow the same trend, and vents, resulting in a decrease in the melting point of the
not to cause any major alteration of the free activation ener- mixture, may allow the determination of barriers ranging
gies, compared to gas or solution phase measurements. To from 11 to 12.5 kcalmol–1, although concerns about the sol-
support this argument, we calculated the barriers of bi- ubility of the CB[n] assemblies under these conditions
phenyl compounds 1–3 at the B3LYP–D/def2-TZVPP level should be raised. We expect that this method will be ex-
by using our previously benchmarked method.[1] Deviations tended to various bi(hetero)aryl derivatives, and that the
from experimental data inside CB[7] or CB[8] (less than latter will not be limited to positively charged scaffolds be-
1.4 kcalmol–1, see Table 1) are all within the tolerance limit cause CB[n] also interact with selected neutral[18] and even
of the method (1.8 kcalmol–1). Because a fraction of the negatively charged guests.[10] As shown in this study and in
guest surface is exposed to the aqueous environment even others,[10] binding affinities towards CB[n] are not sensitive
when surrounded by CB[n], we also compared the torsional to minor conformational changes of the guest, therefore the
barriers measured inside CB[7] or CB[8] with calculated macrocycles are not expected to impact on the torsional
barriers obtained after correction with solvation energies barriers more than changes in solvent composition.
(determined with the polarizable continuum solvation
model, IEFPCM).[16] In all three cases, again, deviations
were all within the tolerance interval (less than
Conclusions
1.3 kcalmol–1, see Table 1). One can thus conclude that
CB[n] encapsulation of biphenyl derivatives is a valid tool
We have shown that encapsulation of selected biphenyl
with which to determine their torsional barriers as free spe- derivatives into CB[7] and CB[8] “amplifies” the dissym-
cies. If necessary, DNMR experiments can be carried out metry caused by the 2-, 2Ј-, 6-, and 6Ј-substituents, and
by using various concentrations of CB[n], and the activation causes remarkable differences in resonance frequencies of
barrier in neat solvent can even be extrapolated.
a prochiral dimethylsulfonium group connected to their 3-
108
www.eurjoc.org
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2014, 105–110

Torsional Barriers of Substituted Biphenyls
1
pound (55 mg, 80%) as a light-yellow powder; m.p. 96–97 °C. H
position. In the absence of the macrocycle, both methyl
groups resonate at the same frequencies, regardless of tem-
perature. We also determined that CB[n] encapsulation has
a very limited impact on the torsional barriers, and we com-
pared the latter with calculated values obtained by using a
highly accurate benchmarked density functional method.
To the best of our knowledge, this study is the first case of
CB[n] use as an analytical accessory to access coveted ki-
netic parameters, and is anticipated to be applicable to a
wide range of bi(hetero)aryl derivatives.
NMR (D2O): δ = 8.13 (q, J = 8.4 Hz, 1 H, Ar-H), 7.66 (q, J =
6.9 Hz, 1 H, Ar-H), 7.60–7.49 (m, 2 H, Ar-H), 7.46–7.38 (m, 2 H,
Ar-H), 3.40 [s, 6 H, S(CH3)2] ppm. 13C NMR (CD3CN): δ = 163.8
(dd, J = 258.4, 7.5 Hz, ArCF), 159.4 (dd, J = 264.7, 7.5 Hz, ArCF),
159.4 (d, J = 248.1 Hz, ArCF), 133.7 (d, J = 11.6 Hz, ArC), 133.2
(d, J = 8.4 Hz, ArC), 133.2 (ArC), 125.9 (d, J = 3.2 Hz, ArC),
117.0 (d, J = 21.5 Hz, ArC), 116.1 (t, J = 19.9 Hz, ArC), 115.7 (dd,
J = 24.8, 3.8 Hz, ArC), 115.4 (ArC), 109.9 (dd, J = 15.5, 3.8 Hz,
ArCS), 28.7 [d, J = 23.7 Hz, S(CH3)2] ppm. HRMS (ESI): m/z
calcd. for C14H12F3S [M]+ 269.060632; found 269.059939.
2-Fluoro-3-iodothioanisole (2a): A solution of sec-butyllithium (1.4
m in hexane, 5.0 mL, 7.0 mmol) was added to a solution of penta-
methyldiethylenetriamine (1.3 g, 7.7 mmol) in anhydrous THF
(50 mL) at –75 °C, and the reaction mixture was kept at this tem-
perature for 10 min. A solution of 2-fluorothioanizole (1.0 g,
7.0 mmol) in THF (5.0 mL) was added dropwise and the reaction
mixture was kept at –75 °C for 2 h, followed by the addition of
iodine (1.8 g, 7.0 mmol) in one portion. The resulting mixture was
warmed to 25 °C before the addition of aq. Na2S2O3 (2.0 mL,
56 mg, 0.35 mmol). The solution was diluted with water (0.10 L)
and extracted with diethyl ether (3ϫ 40 mL); the organic layers
were dried with Na2SO4 and the solvents evaporated. The product
after column chromatography (silica gel; hexane/ethyl acetate, 19:1)
was a mixture of starting material and product with similar Rf val-
ues, and was used in the next step without further purification.
HRMS (ESI): m/z calcd. for C7H6FIS [M]+ 267.921344; found
267.921328.
Experimental Section
2,4-Difluoro-3-iodothioanisole (1a): A solution of sec-butyllithium
(1.4 m in hexane, 4.5 mL, 6.2 mmol) was added to a solution of
2,4-difluorothioanisole (1.0 g, 6.2 mmol) in anhydrous THF
(60 mL), and the reaction mixture was kept at –75 °C for 2 h before
the addition of iodine (1.6 g, 6.2 mmol). The reaction mixture was
then warmed to 25 °C, aq. Na2S2O3 (2.0 mL, 50 mg, 0.32 mmol)
was added, and the mixture was diluted with water (40 mL) and
extracted with diethyl ether (2ϫ 50 mL); the organic layers were
dried with Na2SO4 and the solvents evaporated. The product was
purified by chromatography (silica gel; hexane/ethyl acetate, 19:1)
to afford 1a (1.3 g, 73%) as a white solid; m.p. 42–43 °C. 1H NMR
(CD3CN): δ = 7.40 (m, 1 H, Ar-H), 7.03 (t, J = 8.8 Hz, 1 H, Ar-
H), 2.48 (s, 3 H, S-CH3) ppm. 13C NMR: δ = 161.8 (dd, J = 243.3,
5.3 Hz, ArCF), 160.3 (dd, J = 240.8, 6.0 Hz, ArCF), 130.8 (dd, J
= 8.3, 3.8 Hz, ArC), 122.7 (dd, J = 20.3, 3.8 Hz, ArC), 112.6 (dd,
J = 24.0, 3.8 Hz, ArC), 72.2 (t, J = 30.0 Hz, ArC), 16.1 (d, J =
2.3 Hz, CH3) ppm. HRMS (ESI): m/z calcd. for C7H5F2IS [M]+
285.911922; found 285.912114.
[(2Ј-Ethyl-2-fluorobiphenyl-3-yl)][(methyl)]sulfane (2b): Prepared in
a similar manner to biphenyl 1b, with 2-fluoro-3-iodothioanisole
(2a; 0.50 g, 1.9 mmol) and (2-ethylphenyl)boronic acid (0.42 g,
2.8 mmol) instead of 2,4-difluoro-3-iodothioanisole and (2-
fluorophenyl)boronic acid. The product was purified by
chromatography (silica gel; hexane/dichloromethane, 19:1) to af-
ford 2b (0.12 g, 26% over two steps) as a colorless oil. 1H NMR
(CD3CN): δ = 7.38–7.24 (m, 5 H, Ar-H), 7.18 (t, J = 9.6 Hz, 1 H,
Ar-H), 4.82 (t, J = 7.0 Hz, 1 H, Ar-H), 2.51 [s, 3 H, S(CH3)], 2.46
(q, J = 7.6 Hz, 2 H, Ar-CH2), 1.03 (t, J = 7.6 Hz, 3 H, CH2-
CH3) ppm. 13C NMR: δ = 157.3 (d, J = 238.5 Hz, ArCF), 143.6,
135.7, 131.1 (ArC), 129.9 (d, J = 17.3 Hz, ArC), 129.5 (ArC), 129.4
(d, J = 3.0 Hz, ArC), 127.7 (d, J = 2.3 Hz, ArC), 127.2 (d, J =
18.0 Hz, ArC), 126.8 (ArC), 125.6 (d, J = 4.5 Hz, ArC), 27.1
(ArCH2), 15.7 (CH2CH3), 15.2 (d, J = 2.3 Hz, SCH3) ppm. HRMS
(ESI): m/z calcd. for C15H15FS [M]+ 246.873010; found 246.087235.
Methyl(2,2Ј,6-trifluorobiphenyl-3-yl)sulfane (1b): A solution of po-
tassium carbonate (0.34 g, 2.5 mmol) in H2O (5.0 mL) was added
to a solution of sulfide 1a (0.35 g, 1.2 mmol), (2-fluorophenyl)bo-
ronic acid (0.26 g, 1.8 mmol) and tetrakis(triphenylphosphine)pal-
ladium(0) (0.13 g, 0.12 mmol) in N,N-dimethylformamide (20 mL)
under an inert atmosphere. The resulting mixture was heated to
120 °C for 12 h. After cooling to 25 °C, the reaction mixture was
filtered through a pad of Celite and the filtrate was then poured
into ice-cold water (70 mL), acidified with 1.0 m HCl (5.0 mL), and
extracted with dichloromethane (3ϫ 25 mL). The organic fractions
were washed with water (60 mL) and brine (60 mL), dried with
Na2SO4, then concentrated in vacuo. The product was purified by
column chromatography (silica gel; hexane/ethyl acetate, 9:1) to af-
ford 1b (0.28 g, 90%) as a colorless oil. 1H NMR (CD3CN): δ =
7.57–7.42 (m, 3 H, Ar-H), 7.36–7.27 (m, 2 H, Ar-H), 7.15 (t, J =
8.9 Hz, 1 H, Ar-H), 2.51 (s, 3 H, S-CH3) ppm. 13C NMR: δ = 161.0
(d, J = 246.0 Hz, ArCF), 159.5 (dd, J = 245.3, 6 Hz, ArCF), 158.2
(dd, J = 243.8, 6.8 Hz, ArCF), 133.4 (ArC), 131.8 (d, J = 8.3 Hz,
ArC), 130.4 (dd, J = 9.8, 4.5 Hz, ArC), 125.5 (d, J = 3.9 Hz, ArC),
122.7 (dd, J = 18.8, 3.8 Hz, ArC), 117.6 (d, J = 15.6 Hz, ArC),
116.7 (d, J = 21.8 Hz, ArC), 113.5 (t, J = 21.0 Hz, ArC), 112.9 (dd,
J = 23.2, 3.8 Hz, ArC), 16.1 (d, J = 2.3 Hz, CH3) ppm. HRMS
(ESI): m/z calcd. for C13H9F3S [M]+ 254.037157; found 254.036825.
(2Ј-Ethyl-2-fluorobiphenyl-3-yl)dimethylsulfonium (2): Trimethylox-
onium tetrafluoroborate (60 mg, 0.42 mmol) was added to a solu-
tion of biphenyl 2b (80 mg, 0.33 mmol) in nitromethane (4.0 mL)
under a nitrogen atmosphere. The reaction mixture was heated to
80 °C for 12 h. After cooling to 25 °C, methanol (10 mL) was
added and the solvent was evaporated under vacuum. Addition of
diethyl ether (10 mL) resulted in the formation of 2 (70 mg, 83%)
1
as a colorless oil that solidified upon standing; m.p. 62–63 °C. H
NMR (D2O): δ = 7.96 (t, J = 7.2 Hz, 1 H, Ar-H), 7.77 (t, J =
7.2 Hz, 1 H, Ar-H), 7.59 (t, J = 7.8 Hz, 1 H, Ar-H), 7.49 (m, 2 H,
Ar-H), 7.14–7.35 (m, 1 H, Ar-H), 7.30 (d, J = 7.5 Hz, 1 H, Ar-H),
3.33 [s, 6 H, S(CH3)2], 2.48 (q, J = 7.8 Hz, 2 H, Ar-CH2), 1.02 (t,
J = 7.5 Hz, 3 H, CH3) ppm. 13C NMR (CD3CN): δ = 159.3 (d, J
Dimethyl(2,2Ј,6-trifluorobiphenyl-3-yl)sulfonium Tetrafluoroborate
(1): Trimethyloxonium tetrafluoroborate (36 mg, 0.25 mmol) was = 250.5 Hz, ArCF), 143.7 (ArC), 139.7 (d, J = 4.5 Hz, ArC), 133.5
added to a solution of biphenyl 1b (50 mg, 0.19 mmol) in nitro-
methane (2.0 mL) under a nitrogen atmosphere. The reaction mix-
ture was stirred at 25 °C for 12 h. Methanol (5.0 mL) was then
added and the solvents were evaporated under vacuum. Addition
of diethyl ether (10 mL) resulted in the formation of the title com-
(ArC), 132.5 (d, J = 16.5 Hz, ArC), 131.2, 131.2, 130.5, 129.9
(ArC), 127.5 (d, J = 4.5 Hz, ArC), 127.1 (ArC), 113.8 (d, J =
15.8 Hz, ArC), 28.6 (ArCH2), 26.9 [S(CH3)2], 15.7 (CH2CH3) ppm.
HRMS (ESI): m/z calcd. for C16H18FS [M]+ 261.110776; found
261.110079.
Eur. J. Org. Chem. 2014, 105–110
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
109

R. Joseph, E. Masson
FULL PAPER
[2-Fluoro-2Ј-(trifluoromethyl)biphenyl-3-yl](methyl)sulfane (3b): Pre-
pared in a similar manner to biphenyl 2b, with (2-trifluoromethyl-
phenyl)boronic acid (0.74 g, 3.9 mmol) instead of (2-ethylphenyl)-
boronic acid. The product was purified by chromatography (silica
gel; hexane/ethyl acetate, 99:1) to afford 3b (0.10 g, 13%) as a color-
thank Dr. Venkat Reddy Dudipala (University of Akron, OH) for
his technical support with DNMR experiments, Dr. Giuseppe Bal-
acco (Nucleomatica; Molfetta, Italy) for his suggestions regarding
DNMR simulations, as well as the Ohio Supercomputer Center
(OSC) in Columbus for its generous allocation of computing time.
1
less oil. H NMR (CD3CN): δ = 7.85 (d, J = 9.0 Hz, 1 H, Ar-H),
7.71 (t, J = 7.5 Hz, 1 H, Ar-H), 7.63 (t, J = 7.8 Hz, 1 H, Ar-H),
7.43–7.38 (m, 2 H, Ar-H), 7.25 (t, J = 7.8 Hz, 1 H, Ar-H), 7.12 (t,
J = 7.2 Hz, 1 H, Ar-H), 2.52 [s, 3 H, S(CH3)] ppm. 13C NMR: δ =
157.7 (d, J = 160.0 Hz, ArCF), 135.4, 133.6, 133.4, 130.1 (ArC),
129.8 (d, J = 30.0 Hz, ArC), 129.4 (ArC), 128.8 (d, J = 2.6 Hz,
ArC), 128.1 (d, J = 17.3 Hz), 127.6–127.3 (m, ArC ϫ2), 125.4 (d,
J = 4.3 Hz, ArC), 123.7 (ArC), 15.4 [S(CH3)] ppm. HRMS (ESI):
m/z calcd. for C14H10F4S [M]+ 286.043385; found 286.043445.
[1] For references pertaining to each analytical method, see: E.
Masson, Org. Biomol. Chem. 2013, 11, 2859–2871.
[2] For recent examples, see: a) A. Mazzanti, L. Lunazzi, M. Min-
zoni, J. E. Anderson, J. Org. Chem. 2006, 71, 5474–5481; b) L.
Lunazzi, A. Mazzanti, M. Minzoni, J. Org. Chem. 2006, 71,
9297–9301; c) R. Ruzziconi, S. Spizzichino, A. Mazzanti, L.
Lunazzi, M. Schlosser, Org. Biomol. Chem. 2010, 8, 4463–4471;
d) R. Ruzziconi, S. Spizzichino, L. Lunazzi, A. Mazzanti, M.
Schlosser, Chem. Eur. J. 2009, 15, 2645–2652; e) L. Lunazzi,
M. Mancinelli, A. Mazzanti, S. Lepri, R. Ruzziconi, M.
Schlosser, Org. Biomol. Chem. 2012, 10, 1847–1855.
[3] F. Leroux, M. Maurin, N. Nicod, R. Scopelliti, Tetrahedron
Lett. 2004, 45, 1899–1902.
[2-Fluoro-2Ј-(trifluoromethyl)biphenyl-3-yl]dimethylsulfonium Tetra-
fluoroborate (3): Obtained in a similar manner to sulfonium 2, by
using biphenyl 3b (43 mg, 0.15 mmol) instead of biphenyl 2b, yield
1
50 mg (86%); light-yellow powder. H NMR (D2O): δ = 8.00 (t, J
[4] a) A. D. Becke, J. Chem. Phys. 1993, 98, 1372–1377; b) P. J.
Stephens, F. J. Devlin, C. F. Chabalowski, M. J. Frisch, J. Phys.
Chem. 1994, 98, 11623–11627.
= 6.6 Hz, 1 H, Ar-H), 7.92 (d, J = 7.2 Hz, 1 H, Ar-H), 7.81 (t, J
= 8.1 Hz, 1 H, Ar-H), 7.75–7.66 (m, 2 H, Ar-H), 7.59 (t, J = 7.8 Hz,
1 H, Ar-H), 7.47 (d, J = 7.0 Hz, 1 H, Ar-H), 3.33 [S(CH3)2, 6
H] ppm. 13C NMR (CD3CN): δ = 159.5 (d, J = 251.6 Hz, ArC),
13936, 135.8, 133.4, 132.5, 131.1 (ArC), 130.6 (d, J = 16.7 Hz,
ArC), 129.8 (d, J = 30.0 Hz, ArC), 127.8 (q, J = 5.1 Hz, ArC),
127.4 (d, J = 4.2 Hz, ArC), 127.1, 123.5 (ArC), 113.8 (d, J =
15.1 Hz, ArC), 28.8 [d, J = 40.0 Hz, S(CH3)2] ppm. HRMS (ESI):
m/z calcd. for C15H13F4S [M]+ 301.066860; found 301.066219.
[5] S. Grimme, J. Comput. Chem. 2006, 27, 1787–1799.
[6] A. D. Becke, J. Chem. Phys. 1997, 107, 8554–8560.
[7] a) J. Tao, J. P. Perdew, V. N. Staroverov, G. E. Scuseria, Phys.
Rev. Lett. 2003, 91, 146401–146404; b) S. Grimme, J. Antony,
S. Ehrlich, H. Krieg, J. Chem. Phys. 2010, 132, 154104–154119.
[8] a) J. Lagona, P. Mukhopadhyay, S. Chakrabarti, S. L. Isaacs,
Angew. Chem. 2005, 117, 4922; Angew. Chem. Int. Ed. 2005, 44,
4844–4870; b) E. Masson, X. Ling, R. Joseph, L. Kyeremeh-
Mensah, X. Lu, RSC Adv. 2012, 2, 1213–1247.
Computational Details: Conformations of CB[n]/guest assemblies
were screened at the semi-empirical level (PM7) by using MOPAC
2012,[19] with built-in parameters. The five most stable conforma-
tions were further optimized by using TURBOMOLE 6.3.1
(COSMOlogic GmbH & Co. KG, 51381 Leverkusen) at the TPSS-
D3(BJ)/def2-SVP level with corrections for solvation using the
COSMO model.[20] Convergence criteria were 10–6 Hartree and
10–3 atomic units as the maximum norm of the Cartesian gradient.
Electronic energies were then refined in a single-point calculation
with def2-TZVP basis sets (see the Supporting Information for co-
ordinates of the most stable structure of complex 2·CB[8]). Kinetic
parameters for the torsional isomerization of biphenyls 1–3 and
their analogues bearing a hydrogen atom at the 3-position were
calculated by using a known procedure.[1]
[9] This observation is valid even when using an 800 MHz spec-
trometer.
[10] R. Joseph, E. Masson, Org. Biomol. Chem. 2013, 11, 3116–
3127.
[11] S. Grimme, S. Ehrlich, L. Goerigk, J. Comput. Chem. 2011, 32,
1456–1465.
[12] S. Grimme, Chem. Eur. J. 2012, 18, 9955–9964.
[13] This is not the case in assembly 2·CB[7]; broad aromatic signals
are observed even when the guest is bound quantitatively.
[14] The torsion barrier of guest 3 inside CB[8] is 17.9 kcalmol–1 at
56 °C.
[15] A combination of solvents is sometimes used to determine the
torsion barrier of one single biphenyl derivative, see for exam-
ple: G. Bott, L. D. Field, S. Sternhell, J. Am. Chem. Soc. 1980,
102, 5618–5626.
[16] G. Scalmani, M. J. Frisch, J. Chem. Phys. 2010, 132, 114110–
Supporting Information (see footnote on the first page of this arti-
cle): Titration of biphenyl 1 with CB[7] and biphenyl compounds
1–3 with CB[8], and characterization of the assemblies; DNMR
experiments; ITC results; model for the extrapolation of activation
energies to 0% CB[n]; Cartesian coordinates of the optimized struc-
ture of complex 2·CB[8]; optimized geometries and activation pa-
rameters for torsional isomerization of all biphenyl compounds
presented in this study.
114124.
[17] Y. Zhao, D. G. Truhlar, J. Phys. Chem. A 2005, 109, 5656–5667.
[18] a) M. V. Rekharsky, T. Mori, C. Yang, Y. H. Ko, N. Selvapa-
lam, H. Kim, D. Sobransingh, A. E. Kaifer, S. Liu, L. Isaacs,
W. Chen, S. Moghaddam, M. K. Gilson, K. Kim, Y. Inoue,
Proc. Natl. Acad. Sci. USA 2007, 104, 20737–20742; b) S.
Moghaddam, C. Yang, M. Rekharsky, Y. H. Ko, K. Kim, Y.
Inoue, M. K. Gilson, J. Am. Chem. Soc. 2011, 133, 3570–3581;
c) M. Florea, W. M. Nau, Angew. Chem. 2011, 123, 9510; An-
gew. Chem. Int. Ed. 2011, 50, 9338–9342.
[19] James J. P. Stewart, Stewart Computational Chemistry, Colo-
rado Springs, CO, USA, http://OpenMOPAC.net.
[20] a) A. Klamt, G. Schüürmann, J. Chem. Soc. Perkin Trans. 2
1993, 5, 799–805; b) A. Klamt, V. Jonas, J. Chem. Phys. 1996,
105, 9972–9981.
Acknowledgments
This work was supported by the American Chemical Society Petro-
leum Research Fund (PRF No. 51053-ND4), the Department of
Chemistry and Biochemistry, the College of Arts and Sciences and
the Vice President for Research at Ohio University. The authors
Received: September 26, 2013
Published Online: November 13, 2013
110
www.eurjoc.org
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2014, 105–110