Synthesis and structure of m-terphenyl thio-, seleno-, and telluroethers

Article abstract of DOI:10.1021/jo101299x

Several routes for the synthesis of m-terphenyl thio-, seleno-, and telluroethers were investigated. m-Terphenyl iodides react with diphenyl diselenides or ditellurides (CsOH·H₂O, DMSO, 110 °C) to give the desired compounds in 19-84% yield whic

Full text of DOI:10.1021/jo101299x

pubs.acs.org/joc
Synthesis and Structure of m-Terphenyl Thio-, Seleno-, and Telluroethers
Uzma I. Zakai,† Anna Bzoch-Mechkour,‡ Neil E. Jacobsen,† Leif Abrell,† Guangxin Lin,†
Gary S. Nichol,† Thomas Bally,*,‡ and Richard S. Glass*,†
†Department of Chemistry and Biochemistry, The University of Arizona, Tucson, Arizona 85721,
United States, and ‡Department of Chemistry, University of Fribourg, CH-1700 Fribourg, Switzerland
rglass@u.arizona.edu; thomas.bally@unifr.ch
Received July 1, 2010
Several routes for the synthesis of m-terphenyl thio-, seleno-, and telluroethers were investigated.
m-Terphenyl iodides react with diphenyl diselenides or ditellurides (CsOH H2O, DMSO, 110 °C) to
3
give the desired compounds in 19-84% yield which significantly extends the previously reported
such reactions because o-benzyne cannot be an intermediate as previously suggested. However, the
most general synthetic route was that involving reaction of 2,6-diaryl Grignard reagents with sulfur,
selenium, or tellurium electrophiles. The m-terphenyl thio-, seleno-, and telluroethers were char-
acterized spectroscopically and, in one case, by single-crystal X-ray analysis. Certain of these
compounds showed atropisomerism and barriers for interconversion of isomers were determined
by variable-temperature NMR spectroscopy. The barriers for interconverting the syn and anti
atropisomers increase on going from the analogous S to Se to Te compounds. Calculations on this
isomerization revealed that the barriers are due to rotation about the aryl-aryl bond and that the
barriers for rotation about the aryl-chalcogen bond are much lower.
Introduction
is well-known. It has been synthesized4 via a Newman-
Kwart rearrangement starting from 2,6-diphenylphenol.
The preparation of 2,4,6-Ph3C6H2SH, 1l,5 involves treat-
ment of 2,4,6-Ph3C6H2Li(OEt2)2, obtained from 2,4,6-
Ph3C6H2Br in turn made by bromination of 2,4,6-Ph3C6H3,
with elemental sulfur. Similarly reaction of 2,6-di(p-tolyl)-
phenyl magnesium bromide with elemental sulfur provided
thiophenol 1m.6 However, it should be noted that the
Grignard reagent was obtained by sequential treatment of
1,3-dichlorobenzene with n-BuLi followed by p-tolyl-mag-
nesium bromide, which is an advantageous method reported
by Saednya and Hart.7
Little is known about the chemical consequences of
through-space interaction of chalcogenoethers and aromatic
π-systems, in particular, whether chalcogen lone pair-
aromatic π-interactions would result in lowered oxidation
potentials. To investigate this issue, m-terphenyl chalcogen-
oethers 1-3 were synthesized and characterized. Only two
phenyl-chalcogenoethers with two flanking o-aryl groups
have been previously reported: 2,4,6- Ph3C6H2SMe, 1i,1
and 2,6 (o-MeO-p-t-BuC6H3)-4-Me-C6H2SMe, 1j (Table 1).2
However, analogous thiols, selenols, and metal complexes3
as well as a ditelluride have been made. m-Terphenylthiol 1k
(4) (a) Bishop, P. T.; Dilworth, J. R.; Zubieta, J. A. Chem. Commun. 1985,
257–259. (b) Bishop, P. T.; Dilworth, J. R.; Nicholson, T.; Zubieta, J. Dalton
ꢀ
Trans. 1991, 385–392. (c) Lawson Daku, L. M.; Pecaut, J.; Lenormand-
Foucaut, A.; Vieux-Melchior, B.; Iveson, P.; Jordanov, J. Inorg. Chem. 2003,
42, 6824–6850. (d) Moseley, J. D.; Lenden, P. Tetrahedron 2007, 63, 4120–
4125.
(5) Ruhlandt-Senge, K.; Power, P. P. Bull. Soc. Chim. Fr. 1992, 129, 594–
(1) Suzuki, Y.; Toda, T.; Mukai, T. Heterocycles 1976, 4, 739–742.
(2) Bryant, J. A.; Helgeson, R. C.; Knobler, C. B.; de Grandpre, M. P.;
Cram, D. J. J. Org. Chem. 1990, 55, 4622–4634.
(3) (a) Twamley, B.; Haubrich, S. T.; Power, P. P. Adv. Organomet. Chem.
1999, 44, 1–65. (b) Clyburne, J.; McMullen, A. C. Coord. Chem. Rev. 2000,
210, 73–99.
598.
DOI: 10.1021/jo101299x
r
Published on Web 11/11/2010
J. Org. Chem. 2010, 75, 8363–8371 8363
2010 American Chemical Society

Zakai et al.
SCHEME 1. Synthesis of m-Terphenyl- and Substituted
m-Terphenylchalcogenoethers
JOCArticle
TABLE 1. Compounds Synthesized in this Study and Reported in the
Literaturea
R
R0
Ar
R
Ar
1a
1b
1c
1d
Me
Me
Me
Me
H
o-MeOC6H4
m-MeOC6H4
p-MeOC6H4
Ph
2a
2b
2c
2d
Me o-MeOC6H4
Me p-MeOC6H4
Me 2,5-(MeO)2C6H3
Ph Ph
H
H
H
3a-f, respectively. In another paper15 cyclic voltammetric
studies reveal that the irreversible oxidation potentials for
thioethers 1a,c,d,f, and g and selenoethers 2a and 2b but not
telluroethers 3d and 3e are less positive than expected on the
basis of models. This suggests possible through-space S and
1e
1f
1g
Me Me Ph
2e
2f
Ph o-MeOC6H4
Ph p-MeOC6H4
Ph 2,5-(MeO)2C6H3
Me
Me
Ph
H
H
H
2,5-(MeO)2C6H3
2-naphthyl
o-MeOC6H4
2g
1h
2h10
2i12
H
H
H
2,4,6-Ph3C6H2
2,4,6-Me3C6H2
2,6-(i-Pr)2C6H3
1i1
1j2
1k4
1l5
1m6
1n8
1o9
Me Ph Ph
Me Me o-MeO-p-t-BuC6H3 2j13
Se π interaction. In addition, analysis of the photoelec-
H
H
H
H
H
H
Ph Ph
Ph
3a
3b
3c
3d
3e
3f
Me o-MeOC6H4
Me 2,5-(MeO)2C6H3
Ph Ph
Ph o-MeOC6H4
Ph 1-naphthyl
Ph p-MeC6H4
3 3 3
tron spectrum of 1d shows that its HOMO is a combination
of sulfur lone pair and π-MO.
H
H
H
p-MeC6H4
2,4,6-Me3C6H2
2,4,6-(i-Pr)3C6H2
aFor the known compounds the literature reference is given as a
superscript.
Similarly, 2,6-(20,40,60-Me3C6H2)2C6H3SH, 1n,8 and 2,6-
(20,40,60-i-Pr3C6H2)3C6H3SH, 1o,9 were obtained by reaction
of the corresponding aryl lithium with elemental sulfur.
Synthesis of 2,4,6-Ph3C6H2SeH, 2h, was achieved by sequen-
tial bromination, lithiation with n-BuLi, and reaction with
elemental selenium of 1,3,5-Ph3C6H3.10 The corresponding
diselenide was also reported by the reaction of 2,4,6-Ph3-
C6H2MgBr with elemental selenium.11 The preparation of
2,6-(20,40,60-Me3C6H2)C6H3SeH,12 2i, and 2,6-(20,60-i-Pr2C6-
H3)C6H3SeH, 2j,13 paralleled that of their sulfur analogues
thiophenols 1n and 1o, respectively, except that selenium was
used instead of sulfur. Finally, [2,4,6-Ph3C6H2Te]2 was made
by reaction of the corresponding Grignard reagent with
elemental tellurium.14
It is anticipated that the two flanking phenyl groups in the
m-terphenyl chalcogenoethers will adopt a perpendicular
conformation such as in structure A below, which allows
for chalcogen lone pair-arene π through-space interaction
in the radical cations. This paper reports the synthesis and
structural characterization of m-terphenylthioethers 1a-i,
as well as the related seleno- and telluroethers 2a-g and
Results and Discussion
Syntheses. m-Terphenylmethylthioether 1d was synthe-
sized as shown in eq 1 of Scheme 1 in 42% yield after puri-
fication. The alternative sequence, that is, introduction of the
thioether moiety followed by Suzuki coupling, as shown in
eq 2, produced thioether 1e. Although these routes provided
the best syntheses of thioethers 1d and 1e, an alternative
method based on the m-terphenyl synthesis reported by
Saednya and Hart7 proved to be the most generally useful
for our purposes. Reaction of 1,3-dichlorobenzene 8 with
n-butyllithium followed by an aryl Grignard reagent affords
m-terphenyl and substituted m-terphenyl Grignard reagents 9
as shown in Table 2. The mechanism suggested7 for forma-
tion of Grignard 9 from 1,3-dichlorobenzene 8 is shown in
Scheme 2. This intermediate was shown to react with Hþ
or I2 to give the corresponding m-terphenyl or substituted
m-terphenyl or substituted m-terphenyliodide, respectively.
Reaction of Grignard 9 with MeSSO2Me, PhSSPh, MeSe-
SeMe, PhSeSePh, MeTeTeMe, or PhTeTePh afforded the
corresponding chalcogenoether in the yield listed in Table 2.
Since it had been reported16 that reaction of bromo- or
iodobenzene with diphenyl diselenide or diphenyl disulfide
(6) Ohta, S.; Ohki, Y.; Ikagawa, Y.; Suizu, R.; Tatsumi, K. J. Organomet.
Chem. 2007, 692, 4792–4799.
(7) Saednya, A.; Hart, H. Synthesis 1996, 1455–1458.
(8) Ellison, J. J.; Ruhlandt-Senge, K.; Power, P. P. Angew. Chem., Int. Ed.
1994, 33, 1178–1180.
(9) Niemeyer, M.; Power, P. P. Inorg. Chem. 1996, 35, 7264–7272.
(10) Hauptmann, R.; Kliss, R.; Schneider, J.; Henkel, G. Z. Anorg. Allg.
Chem. 1998, 624, 1927–1936.
(11) Kataeva, L. M.; Kataev, E. G. Zh. Obshch. Khim. 1962, 32, 2710–
2713.
with CsOH H2O in DMSO afforded diphenyl selenide or
3
diphenyl sulfide, respectively, in 64-70% yield, the reaction
of substituted m-terphenyl iodides with diphenyl dichalco-
genides was studied. Remarkably, m-terphenyl and substi-
tuted m-terphenylchalcogenides were formed in 19-84%
(12) Ellison, J. J.; Ruhlandt-Senge, K.; Hope, H. H.; Power, P. P. Inorg.
Chem. 1995, 34, 49–54.
(13) Niemeyer, M.; Power, P. P. Inorg. Chim. Acta 1997, 263, 201–207.
€
(14) Schulz Lang, E.; Maichle-Mossmer, C.; Strahle, J. Z. Anorg. Allg.
€
Chem. 1994, 620, 1678–1685.
(15) Ammam, M.; Zakai, U. I.; Wilson, G. S.; Glass, R. S. Pure Appl.
Chem. 2010, 82, 555–563.
(16) Varala, R.; Ramu, E.; Adapa, S. R. Bull. Chem. Soc. Jpn. 2006, 79,
140–141.
8364 J. Org. Chem. Vol. 75, No. 24, 2010

Zakai et al.
JOCArticle
TABLE 2. Synthesis of Compounds 1-3 by Reaction of Grignard
Reagent 9 with MeSSO2Me or RXXR (R = Me, Ph; X = S, Se, Te)
TABLE 3. Reaction of m-Terphenyl Iodide and Substituted m-Terphenyl
Iodides, 2,6-Ar2C6H3I, with Diphenyldichalcogenides, PhXXPh, and
CsOH H2O in DMSO (110°, 24-27 h)
3
Ar
electrophile
product
yield (%)
o-MeOC6H4
m-MeOC6H4
p-MeOC6H4
2,5-(MeO)2C6H3
2-naphthyl
o-MeOC6H4
o-MeOC6H4
p-MeOC6H4
2,5-(MeO)2C6H3
C6H5
o-MeOC6H4
p-MeOC6H4
2,5-(MeO)2C6H3
o-MeOC6H4
3,5-(MeO)2C6H3
C6H5
o-MeOC6H4
1-naphthyl
p-MeC6H4
MeSSO2Me
MeSSO2Me
MeSSO2Me
MeSSO2Me
MeSSO2Me
PhSSPh
MeSeSeMe
MeSeSeMe
MeSeSeMe
PhSeSePh
PhSeSePh
PhSeSePh
PhSeSePh
MeTeTeMe
MeTeTeMe
PhTeTePh
PhTeTePh
PhTeTePh
PhTeTePh
1a
1b
1c
1f
1g
1h
2a
2b
2c
2d
2e
2f
2g
3a
3b
3c
3d
3e
3f
67
74
16
26
59
54
30
34
64
46
62
45
29
69
37
67
21
24
41
Ar
Ph
X
product
yield (%)
Se
Te
Se
Se
Te
2d
3c
2f
2e
3d
58
84
19
4-MeO
2-MeO
25
stituted m-terphenyl radical appears feasible; although a
nucleophilic addition-elimination pathway is also possible,
especially in view of the successful reaction with 4-methyl-
thiophenol.
To explore the generality of this reaction, reaction of the
substituted m-terphenyl iodides listed in Table 3 with a
variety of secondary amines (aniline, indoline, indole, car-
bazole, morpholine, piperidine, and N-phenylpiperazine)
and CsOH H2O in DMSO was monitored by GC-MS.
3
Little or no desired product was obtained. Similarly, reaction
of 2,6-dimethylphenyliodide, -bromide, or -chloride with
Ph2X2 (X = S, Se, Te) or 4-XC6H4SH (X = H, Me, NH2,
OMe, NO2) or 1-butanethiol or 2-mercaptoethanol or sec-
SCHEME 2. Mechanism for Formation of Grignard 9 from 1,3-
Dichlorobenzene 8
ondary amines and CsOH H2O in DMSO gave little or no
3
desired product with one exception: reaction of 2,6-di-
methylphenyliodide, PhTeTePh, and CsOH H2O in DMSO
3
produced phenyl-2,6-dimethylphenyltelluride in 41% yield.
Similarly, reactions of 2,6-dimethoxyphenyliodide, iodophe-
nyl-2,4,6-tricarboxylic acid, or bromo or chloro phenyl-
2,6-dicarboxylic acid with Ph2X2 (X=S, Se, Te) or4-XC6H4SH
(X=H, Me, NH2, OMe, NO2) or 1-butanethiol or 2-mer-
captoethanol or secondary amines and CsOH H2O in
3
DMSO were unsuccessful.
NMR Spectroscopy. The structures of the substituted
m-terphenyl methyl and phenylchalcogenides synthesized
yield as seen in Table 3. The reaction was unsuccessful with
diphenyl disulfide, di(4-aminophenyl) disulfide, and di-
(4-nitrophenyl) disulfide with any of the substituted m-terphenyl
iodides listed in Table 3. However, in our hands, in another
process that had literature precedents with simpler aryl
halides,174-methylthiophenol reacted with m-terphenyl
iodide to give the corresponding sulfide in 33% yield. This
reaction was unsuccessful or gave trace amounts of the
desired products with the other substituted m-terphenyl
iodides listed in Table 3, as well as with other thiols
(4-amino-, 4-methoxy-, and 4-nitrophenylthiol, 1-butanethiol,
and 2-mercaptoethanol). Since it had been suggested18 that
this reaction occurs via a benzyne intermediate, our results
show that another mechanism must be operative with sub-
stituted m-terphenyl iodides, which cannot eliminate to form
o-benzyne. Since a frequently encountered product in our
reactions is the substituted m-terphenyl in which iodine is
replaced by hydrogen, a free radical mechanism via the sub-
were determined by spectroscopic analysis (IR, 1H and 13
C
NMR), mass spectrometry including high resolution MS,
and in one case, by single crystal X-ray diffraction analysis.
An interesting feature of the 1H NMR spectrum of thioether
1a in CD2Cl2 is that there are two methoxy peaks at room
temperature as seen in Figure 1.
This is expected if the flanking phenyl rings are orthogonal to
the central one (such as in geometry A), owing to atropisomer-
ism. In this situation, there are syn and anti isomers, as
illustrated in Figure 2. Indeed one of these conformations is
adopted in the solid state, as seen in the single crystal X-ray
crystal structure analysis of thioether 1a shown in Figure 3, in
which the anti isomer is evident. The angles between the central
benzene ring and the mean planes through the o-methoxyphe-
nyl rings are 75.49(4)° and 80.41(5)°. The angle between the
central benzene ring and the plane of the methyl C, S, and
aromatic carbon to which the sulfur is bonded is 69.37(9)°.
Variable-temperature 1H NMR spectroscopic studies pro-
vide more insight into this atropisomerism. The low boiling
point of CD2Cl2 limits high-temperature NMR studies for
thioether 1a, but in DMSO-d6 the OMe signals coalesce near
(17) Varala, R.; Ramu, E.; Alam, M. M.; Adapa, S. R. Chem. Lett. 2004,
33, 1614–1615.
(18) Varala, R.; Ramu, E.; Alam, M. M.; Adapa, S. R. Synlett 2004,
1747–1750.
J. Org. Chem. Vol. 75, No. 24, 2010 8365

Zakai et al.
JOCArticle
FIGURE 1. 300 MHz 1H NMR spectrum of 1a in CD2Cl2 at
ambient temperature (inset showing expanded OMe region).
FIGURE 4. Variable-temperature 300 MHz 1H NMR spectro-
scopic study of thioether 1a in CD2Cl2 (OMe region on left, SMe
region on right).
FIGURE 2. Syn and anti atropisomers of methoxy-substituted
m-terphenylchalcogenoethers.
FIGURE 5. Variable-temperature 300 MHz 1H NMR spectro-
scopic study of selenoether 2a in DMSO-d6 (OMe region).
on increasing the temperature.19 Similarly the 1H NMR signal
for the -SMe moiety in thioether 1a is a broad singlet at
ambient temperature, but the SeMe (2a) and TeMe (3a) signals
at ambient temperature are two peaks that coalesce on warm-
ing (see Figures 7 and 8, respectively).
The o-methoxy Ph-chalcogen compounds 1h, 2e, and 3d
also show two methoxy peaks that collapse into one on
warming (see Supporting Information for stacked plots). The
13C NMR spectra for all of the o-methoxy compounds 1a, 1h,
2a, 2e, 3a, and 3d show two peaks for the MeO carbons at
ambient temperature. In addition, the 77Se NMR spectra of
selenoethers 2a and 2e show two Se signals at room temperature
at δ 374.7 and 375.0 for 2a and 371.7 and 374.6 ppm for 2e.
FIGURE 3. Molecular structure of thioether 1a from X-ray crystal-
lographic structure analysis.
30° (as seen in the stacked plot in the Supporting Information).
On cooling a solution of thioether 1a in CD2Cl2, the SMe
signal split into two signals as shown in Figure 4. The barrier
for interconversion of the m-methoxy analogue thioether 1b is
much lower than that for thioether 1a. Consequently, there is
one peak for the methoxy groups at ambient temperature,
which broadens and separates into two peaks at low tempera-
ture in CD2Cl2 (see Supporting Information). The p-methoxy
analogue is incapable of atropisomerism and shows one peak
for the methoxy groups at all temperatures.
The selenium and tellurium analogues with o-methoxy
groups, 2a and 3a, respectively, show even higher coales-
cence temperatures in DMSO-d6 than for thioether 1a, and
stacked plots for these compounds are shown in Figures 5
and 6, respectively, illustrating their collapse to a single peak
(19) Although the peak shapes observed for 1c and 3a can be fitted well
using two atropisomers, those observed for 2a are not as well-fitted. A better
fit is obtained using three atropisomers but the calculations outlined later in
the paper do not support another conformer for 2a.
8366 J. Org. Chem. Vol. 75, No. 24, 2010

Zakai et al.
JOCArticle
FIGURE 8. Variable-temperature 300 MHz 1H NMR spectro-
scopic study of telluroether 3a in DMSO-d6 (TeMe region).
FIGURE 6. Variable-temperature 300 MHz 1H NMR spectro-
scopic study of telluroether 3a in DMSO-d6 (OMe region).
FIGURE 9. Experimental (solid lines) and simulated (broken lines)
1H NMR peaks for selenoether 2a at 45 C [(a) SeMe resonance, (b)
the OMe resonance] and for telluroether 3a [(c) TeMe resonance at
60 C, (d) OMe resonance at 75 C] in DMSO-d6 near the coalescence
temperatures (see Table 4 for parameters).
FIGURE 7. Variable-temperature 300 MHz 1H NMR spectro-
scopic study of selenoether 2a in DMSO-d6 (SeMe region).
obtained for MeO coalescence for thioether 1a in DMSO-
d6 cannot be directly compared with that for SMe coales-
cence in CD2Cl2.
Analysis of the coalescence data provided excellent fits of
the observed spectra and calculated spectra. The fits around
the critical coalescence temperatures for the OMe and SeMe
signals for selenoether 2a and the OMe and TeMe signals for
telluroether 3a are shown in Figure 9. The fitting parameters
as well as the calculated barriers are listed in Table 4.
A detailed analysis of the barriers was required because the
variable-temperature NMR spectra show that the chemical
shifts vary significantly with temperature. This is dramati-
cally illustrated for the Me-chalcogen signals in the spectra of
selenoether 2a and telluroether 3a as seen in Figures 7 and 8.
The WinDNMR program (which predicts this variation)
allowed fitting the parameters at each temperature and thus
to determine the activation barriers for interconversions of
the syn and anti atropisomers. Another important feature of
these interconversions is that they are solvent-dependent,
which is seen by comparing the ambient temperature spectra
of thioether 1a in the OMe region in CD2Cl2 (Figure 1 insert)
and DMSO-d6 (see Figure 9). Thus the energy barriers
It is interesting that the barriers for interconversion of
the atropisomers increase on going from sulfur to selenium
to tellurium. Thus, the increasing size of the chalcogen (S <
Se < Te)20 and not the increasing aryl-chalcogen bond
length (S -C < Se-C < Te-C), which would tend to reduce
steric interaction, is key in determining the magnitude of this
barrier. In addition, comparison of the variable-temperature
1H NMR data for Ph-chalcogen compounds 1h, 2e, and 3d
(see Supporting Information) with that of the corresponding
Me-chalcogen compounds (1a, 2a, and 3a, respectively)
show comparable coalescence temperatures. Thus increasing
the size of the substituent attached to the chalcogen (Ph >
Me) appears to have little or no effect. Perusal of Table 4 also
(20) The van der Waals radius for S, Se, and Te is 1.80, 1.90, and 2.06,
respectively: Mantina, M.; Chamberlin, A. C.; Valero, R.; Cramer, C. J.;
Truhlar, D. G. J. Phys. Chem. A 2009, 113, 5806–5812.
J. Org. Chem. Vol. 75, No. 24, 2010 8367

Zakai et al.
JOCArticle
TABLE 4. Fitting Parameters for Variable-Temperature 1H NMR Spectra of Thioether 1a, Selenoether 2a, and Telluroether 3a near Coalescence
compound
group
temp (°C)
solvent
line width (Hz)
vA - vB (Hz)
kAB þ kBA (s-1
)
% A
ΔGq (AfB) (kJ/mol)
1a
2a
2a
3a
3a
SMe
15
45
45
60
75
CD2Cl2
1.25
2.00
2.00
1.25
1.60
3.47
5.01
-6.68
3.12
9.20
16.8
20.3
10.2
25.1
52.1
51.1
50.3
52.5
51.6
66.9
72.5
71.9
77.5
78.5
SeMe
OMe
TeMe
OMe
DMSO-d6
DMSO-d6
DMSO-d6
DMSO-d6
-7.30
FIGURE 10. Interconversion of atropisomers of thioether 1a, selenoether 2a, and telluroether 3a (structures from B3LYP/6-31G*
calculations for thioether 1a; note that the two anti structures are identical).
TABLE 5. Calculated 1H NMR Spectroscopic Chemical Shifts for Me
reveals that the barriers for interconversion of the atrop-
isomers of selenoether 2a and telluroether 3a, determined
using either the coalescence of the chalcogen-Me or o-OMe
signals, are very similar. This suggests that both coalescence
phenomena are measuring the same process: rotation about
the aryl-aryl bond.
Groups in Thioether 1aa
isomer
δ SMe (ppm)
δ OMe(1) (ppm)
δ OMe(2) (ppm)
syn-a, X = S
syn-s, X = S
anti, X = S
1.56
1.86
1.90
3.69
3.70
3.64
3.69
3.66
3.61
aGIAO/WP04/cc-pVDZ in DMSO as a continuum solvent, scaled
using the optimized factors given in ref 21.
It is surprising that only two signals are observed for the
Me-X (X = chalcogen) and o-methoxy groups in thioether
1a, selenoether 2a, and telluroether 3a because there are in
fact three possible atropisomers: the anti isomer and two syn
isomers (syn-a where the X-Me group is facing away from the
methoxy groups; syn-s where the X-Me group is facing
toward the methoxy groups); see Figure 10 where the three
transition states for interconversion of the three atrop-
isomers are also shown. The observation of only two signals
for the O-Me and X-Me groups in the 1H NMR spectra may
be due to any of three reasons: (a) two of the signals in the
three species have identical or nearly identical chemical
shifts, (b) one of the conformers is present in amounts too
small to be observed, or (c) some of the conformers inter-
convert rapidly on the NMR time scale, at the temperatures
that are accessible in our experiments.
but that syn-s lacks this symmetry element (the X-Me group
is tilted toward one side and the dihedral angles to the two
phenyl groups are not the same). The calculated structures
are depicted in the Supporting Information along with their
geometric coordinates.
1
Based on these geometries, the H NMR chemical shifts
for the S-Me and O-Me moieties in the atropisomers of
thioether 1a (see Table 5) were calculated following the
protocol outlined in ref 21. The results of these calculations
suggest that the chemical shifts for the SMe groups in
the three conformers of thioether 1a should not overlap,
although there may be overlap of the signals for the OMe
groups. This excludes hypothesis (a), at least with regard to the
X-Me signals.
To decide which of the above three hypotheses is correct,
we sought help from quantum chemical calculations. First
we optimized the structures of all three atropisomers (and the
four transition states between them) by the B3LYP/6-31G*
method (see Figure 10). Thereby we found that conformer
syn-a has a plane of symmetry for both X = S and X = Se,
Next the relative enthalpies and free energies of the three
conformers and the transition states between them were
(21) Jain, R.; Bally, T.; Rablen, P. R. J. Org. Chem. 2009, 74, 4017–4023.
€
(22) Gunther, H. NMR Spectroscopy, 2nd; Wiley: New York, 1992;
pp 343.
8368 J. Org. Chem. Vol. 75, No. 24, 2010

Zakai et al.
JOCArticle
TABLE 6. Relative Energies in kJ/mola of Different Atropisomers of
Thioether 1a and Selenoether 2a and Barriers for Their Interconversion
(See Also Figure 10)
phenyl group, are much higher and in good accord with the
experimental free energies of activation reported in Table 4.
In contrast to the low barriers discussed above, these barriers
increase on going from X = S to X = Se, in accord with
experiment. The height of the barriers for rotation of a lateral
phenyl group are dictated by the repulsion between the o-H
atom of the phenyl ring and the central heteroatom. As the
van der Waals radius of Se is larger than that of S, that
repulsion is larger and the barrier therefore higher for Se.23
Thus it is to be assumed that the one of the X-Me re-
sonances is the average of those for conformers syn-a and
syn-s, while the other is that of the anti conformer. The same
applies to the O-Me resonances, except in this case there is
averaging between two resonances for the anti conformer.
Atropisomerism in the oxygen analogues of compounds
1-3 (m-terphenyl ethers) has never been studied, but House
and coworkers24 have measured the barrier for interconvert-
ing the syn and anti isomers of 1,8-diarylanthracenes 10.
Surprisingly, they found that replacing the OH by an OAc
group had little effect on that barrier.25,26 This was ascribed
to splaying out of the 1,8-aryl groups that is evident in the
X-ray crystal structure of acetate 10b.24
gas phase
SCRF @DMSO
compound
conformer
Hrel
1.03
(0)
0.35
Grel
2.04
(0)
0.14
Hrel
Grel
1a (X = S) syn-s
syn-a
anti
1.26
(0)
2.27
(0)
1.21
16.20
15.71
59.55
60.22
3.32
1.00
21.76
20.17
68.32
70.40
3.32
TS(syn-a f syn-s) 19.43 24.99
TS(anti f anti)
TS(syn-s f anti)
TS(anti f syn-a)
19.03 23.49
58.51 67.28
59.45 69.62
3.01
0.00
1.12
2a (X = Se) syn-s
syn-a
anti
3.01
0.00
1.15
0.00
2.44
0.00
2.48
TS(syn-a f syn-s) 16.48 18.64
13.51
11.64
65.50
64.80
15.67
13.35
73.44
71.25
TS(anti f anti)
TS(syn-s f anti)
TS(anti f syn-a)
15.27 16.98
64.04 71.98
64.83 71.27
aCalculated by the B2PLYP-D/cc-pVDZ//B2PLYP-D/cc-pVDZ
method, using B3LYP/6-31G* thermal corrections and entropies.
calculated, using the recently introduced double hybrid
B2PYLP method, corrected for dispersion energy contribu-
tions, in the gas phase and in DMSO as a continuum solvent
(see Table 6). First we note that for X=S the three con-
formers are within ca. 3 kJ/mol, i.e., one would expect all of
them to be present in solution above room temperature (where
RT g2.5 kJ/mol). For X = Se the situation is similar in DMSO
(in the gas phase syn-s is more strongly disfavored). Thus,
hypothesis (b) must also be discarded.
Finally, the barriers for interconverting the three atrop-
isomers were examined (see Table 6). These calculations show
that the barriers for rotation around the Ph-X(Me) bond
are too low to be detected by dynamic NMR in the tempera-
ture range that is accessible to us, both for X=S and X=
Se, i.e., that rapid equilibration between conformers syn-a
and syn-s is to be expected (and similar rotation in the anti
conformer is also rapid on the NMR time scale). Interest-
ingly these barriers decrease on going from S to Se, which can
be understood easily because the increase of the C-X bond
In conclusion, the effective syntheses and structural stud-
ies reported here and the electrochemical studies reported
(25) Interestingly, House et al.23 report a minor component in acetate 10b
that rapidly interconverts with the syn isomer. They assign the structure of
this minor component to the syn isomer with the OAc cis to the aryl methyl
groups which equilibrates with the major syn isomer with the OAc trans to the
aryl methyl groups. If this assignment for the minor component is correct,
then the rotation barrier about the aryl-O bond in 10b is greater than the
rotation barriers about the aryl-X bond (X = S, Se, Te) in the substituted and
unsubstituted m-terphenyl chalcogenoethers 1-3.
(26) (a) Lunazzi, L.; Mancinelli, M.; Mazzanti, A. J. Org. Chem. 2007, 72,
5391–5394. (b) Casarini, D.; Lunazzi, L.; Mazzanti, A. Eur. J. Org. Chem.
2010, 2035–2056.
(27) Lunazzi, L.; Mancinelli, M.; Mazzanti, A. J. Org. Chem. 2009, 74,
1345–1348 report a high barrier (ΔG = 38.0 kcal/mol at 140 °C) for
racemization of chiral anthraquinone 11. The configurational stability of
anthraquinone 11 was ascribed to steric interaction between the methyl-
naphthyl substituent and anthraquinone carbonyl group. However, this is
only remotely related to the systems studied in the present paper.
˚
length (by 0.14 A) leads to a concomitant increase of the
distance between the Me group that is attached to X and the
Ph group that it encounters at the two transition states where
˚
this situation prevails (at a distance of 3.05 A of the methyl C
atom perpendicular to the phenyl ring in both cases). On the
other hand, the calculated barriers for converting the syn-a
or syn-s to the anti conformers, i.e., for rotation of a lateral
(23) To quantitate this effect, we calculated the potential for the interac-
tion of dimethylsulfide or dimethylselenide, respectively, with a benzene
˚
H-atom as depicted. The equilibrium distance was 2.8 A for S and 3.0 A for
˚
Se. At the transition states for rotation around a lateral phenyl group in 1a
˚
the S H distance is ca. 2.1 A and in 2a it is 2.3 A. At these distances the
˚
3 3 3
energy is higher by 13.2 kJ/mol in the model complex for X = S and by
16.3 kJ/mol for X = Se. The difference between these two numbers (3.1 kJ/mol)
accounts for a large part of the differences in the enthalpies of activation for
rotation of a lateral phenyl group in 1a and 2a.
(24) House, H. O.; Hrabie, J. A.; VanDerveer, D. J. Org. Chem. 1986, 51,
921–929.
(28) Still, W. C.; Kahn, M.; Mitra, A. J. Org. Chem. 1978, 43, 2923–2925.
J. Org. Chem. Vol. 75, No. 24, 2010 8369

Zakai et al.
JOCArticle
elsewhere suggest that unsubstituted and substituted
(CH), 130.70 (Cq), 131.34 (CH), 138.12 (Cq), 142.86 (Cq), 147.23
(Cq); IR υmax (film)/cm-1 3054 (C-H), 2922 (C-H), GC-MS
m/z 290 (Mþ); HRMS (MALDI) calcd for C20H18S: 290.11237,
found 290.11238.
m-terphenylthioethers 1 and selenoethers 2 may provide
insight into through-space S and Se π interaction. Conse-
3 3 3
quently, studies are underway to elucidate the structure and
bonding in their one-electron oxidation products.
Synthesis of 1-3 via Grignard Reagent 9. The methodology
used will be exemplified by the synthesis of thioether 1b. A
solution of 2,6-dichlorophenyllithium was prepared follow-
ing the procedure of Saednya and Hart.7 To a solution of
1,3-dichlorobenzene 8 (1.47 g, 10 mmol) in dry THF (25 mL) under
argon and cooled in a Dry ice-acetone bath was added a 1.6 M
solution of n-BuLi in hexanes (10 mmol) dropwise by syringe
over 20 min. After stirring at this temperature for 1.5-2 h, the
resulting white slurry was rapidly added to a warmed solution
of m-MeOC6H4-MgBr, prepared by stirring a solution of
m-MeOC6H4Br (30 mmol) in dry THF (30 mL) with dried Mg
turnings (30-40 mmol) for 1 h at room temperature. A portion
of this solution of Grignard 9, Ar = m-MeOC6H4 (27.5 mL, 5.00
mmol) was placed in a flask under argon and cooled in a Dry ice-
acetone bath. S-Methylthiosulfonate29 (0.94 mL, 1.26 g, 10 mmol)
was added, and the reaction misture was stirred for 3 h with
cooling. The mixture was allowed to warm to room temperature
overnight. Water was added, and the mixture was extracted with
Et2O. After drying and evaporating, the solid was recrystallized
from CH2Cl21 to give thioether 1b (2.48 g, 74% yield): mp
110-112 °C; H NMR (CD2Cl2) δ 1.723 (s, 3H, SMe), 3.847
(s, 6H, OMe), 6.925 (ddd, J = 8.3, 2.7, 0.9 Hz, 2H), 7.053 (ddd,
J = 2.7, 1.5, 0.8 Hz, 2H), 7.07 (ddd, J = 7.2, 1.5, 0.9 Hz, 2H),
7.27-7.40 (m, 5H); 13C NMR (CD2Cl2) δ 19.60 (CH3), 55.81
(CH3), 113.08 (CH), 115.84 (CH), 122.52 (CH), 127.83 (CH),
129.39 (CH), 130.43 (CH), 134.26 (Cq), 144.04 (Cq), 146.91 (Cq),
159.78 (Cq); IR υmax (KBr)/cm-1 3047 (C-H), 2997 (C-H),
2955 (C-H), 2945 (C-H), 2827 (C-H), MS (EI) m/z (rel
intensity %) 336 (Mþ, 100), 321 (93), 306 (88), 247 (53), 234
(67), 202 (63); HRMS (EI) calcd for C21H20O2S: 336.1184,
found 336.1181.
Experimental Section
General. All reactions were performed using standard
Schlenk techniques under an atmosphere of argon. THF was
purified by distillation under N2 from potassium-benzophenone
ketyl. Column chromatography was done using 32-63 μm flash
silica gel following the method of Still et al.28 Reactions were
monitored by GC-MS using a gas chromatograph with a ZB-5 ms
column interfaced with a mass spectrometer. All mp are un-
1
corrected. All H variable-temperature, 13C, and APT NMR
spectra were obtained using an NMR spectrometer operating at
a H frequency of 299.956 MHz, using a 5 mm Four-Nucleus
1
probe. The ambient temperature without heating or cooling was
22-23 °C. Carbon types (CH3 /CH or CH2/Cq) were determined
from APT spectra. NMR chemical shifts and coupling patterns
in the aromatic rings were elucidated by simulation and curve
fitting using Win-DNMR version 7.1.12 (Reich, H. J., Univer-
sity of Wisconsin, Madison, WI). Low-temperature experiments
used dry nitrogen gas-cooled to 77 K in a heat exchanger, and
temperatures were calibrated using the 1H shift separation of a
low-volume methanol sample. All 1H spectra are referenced to
residual solvent at δ 2.49 ppm (DMSO-d6) or δ 5.32 ppm
(CD2Cl2), and all 13C spectra are referenced to deuterated
solvent at δ 39.51 ppm (DMSO-d6) or 54.00 (CD2Cl2). 77Se
NMR were acquired on a spectrometer operating at a 77Se
frequency of 95.2775 MHz using a 5 mm broadband inverse
1
3-axis gradient probe at 25 °C, with inverse-gated H decou-
pling. All 77Se spectra are referenced to external selenomethio-
nine at δ 268.6 ppm. High resolution MS were obtained by direct
insertion.
The other reactions reported in Table 1 were done in a similar
way as that reported for thioether 1b except Ph2S2, Me2Se2,
Ph2Se2, Me2Te2, or Ph2Te2 were used instead of S-methyl
thiosulfonate and the appropriate Grignard reagent used in
place of m-MeOC6H4MgBr. The reactions were monitored by
GC-MS after quenching with water and usually required col-
umn chromatography on silica gel for purification after com-
pletion of the reaction. Yields, physical properties, and spectro-
scopic details for 1-3 are found in the Supporting Information.
Synthesis of Selenoether 2d from m-Terphenyliodide, Ph2Se2,
Synthesis of Thioether 1d from 2,6-Diphenylaniline 5. The
procedure of Bryant et al.2 was adopted for the synthesis of
thioether 1d. Isoamyl nitrite (0.12 mL, 0.91 mmol) was added to
a solution of 2,6-diphenylaniline 5 (83 mg, 0.34 mmol) in
dimethyl disulfide (0.58 mL) heated to 75 °C. After the addition,
the temperature was raised to 90 °C and held at this temperature
for 2-3 h. The solution was concentrated under reduced pres-
sure and the residue chromatographed on preparative TLC
plates eluting with hexanes to give thioether 1d (39 mg, 42%
yield): 1H NMR (CD2Cl2) δ 1.664 (s, 3H), 7.29 (br d, J = 7.4 Hz,
2H), 7.33-7.53 (m, 11H); 13C NMR (CD2Cl2) δ 19.45 (CH3),
127.62 (CH), 127.98 (CH), 128.36 (CH), 130.12 (CH), 130.49
(CH), 134.23 (Cq), 142.72 (Cq), 147.20 (Cq); IR υmax (film)/cm-1
3054 (C-H), 2918 (C-H), 2849 (C-H), MS (EI) m/z (rel
intensity %) 276 (Mþ, 100), 261 (64), 260 (35), 228 (10), 184
(19); HRMS (EI) calcd for C19H16S: 276.0973, found 276.0966.
Synthesis of Thioether 1e by Suzuki Coupling with 2,6-Diiodo-
4-methylthioanisole 7. Following the procedure of Bryant et al.,2
a solution of Pd(PPh3)4 (160 mg) and phenylboronic acid (185
mg, 1.52 mmol) in ethanol (1 mL) was added to a mixture of
2,6-diiodo-4-methylthioanisole2 7 (200 mg, 0.512 mmol) in
benzene (3 mL) and aqueous sodium carbonate (1.28 mL) at
room temperature under a nitrogen atmosphere. The vigorously
stirred mixture was heated at reflux for 7 d with the addition of
more catalyst (10-15%). The mixture was filtered, and the
organic layer was separated, dried, concentrated, and column
chromatographed on silica gel to give thioether 1e as a colorless
solid (27% yield): mp 86-87 °C; 1H NMR (600 MHz, CD2Cl2) δ
1.666 (s, 3H), 2.397 (s, 3H), 7.147 (br s, 2H), 7.379 (C, 2H), 7.434
0
and CsOH H2O. The procedure used was adapted from that
3
reported by Verala et al.16 A mixture of m-terphenyliodide
(89 mg, 0.25 mmol), Ph2Se2 (117 mg, 0.375 mmol), CsOH H2O
3
(84 mg, 0.50 mmol), and DMSO (0.75 mL) was sealed in a Pyrex
glass tube and placed in a sand bath maintained at 110 °C for
24 h. After cooling, the reaction mixture was added to crushed
ice, and then aqueous NH4Cl solution was added. The mixture
was stirred for 5-10 min and then repeatedly extracted with
Et2O. The combined extracts were dried (MgSO4) and filtered.
Silica gel was then added, and the mixture was concentrated
under vacuum. The residue was put on a silica gel column and
chromatographed to obtain 2d (55.6 mg, 58% yield). The
spectra (IR, 1H and 13C NMR) were the same as that prepared
before and an intimate mixture with the compound prepared
before showed an undepressed mp.
The other reactions reported in Table 2 were conducted
similarly but with the appropriate m-terphenyliodide and di-
phenyldichalcogenide.
Variable-Temperature NMR Studies. For temperatures above
room temperature, DMSO-d6 was used as the solvent, and for
temperatures below room temperature, CD2Cl2 was used as the
(B, 4H), 7.495 (A, 4H) (AA0BB0C, Jab = 7.6, Jbc = 7.5, Jaa
=
1.5, Jbb = 1.9, Jac = 1.3 Hz); 13C NMR (150 MHz, CD2Cl2) δ
0
19.61 (CH3), 21.27 (CH3), 127.54 (CH), 128.30 (CH), 130.13
(29) Scholz, D. Synthesis 1983, 944–945.
8370 J. Org. Chem. Vol. 75, No. 24, 2010

Zakai et al.
JOCArticle
solvent. The 1H NMR exchange line fitting results were ob-
tained using WinDNMR version 7.1.12 (Reich, H. J., Univer-
sity of Wisconsin, Madison, WI). The spectra used for fitting
were acquired just below the coalescence temperature for the
resonance being fitted in order to maximize the information
content. Line widths were determined by fitting peaks well
below the coalescence temperature where exchange broadening
is negligible. Chemical shifts and relative peak areas of the two
conformers, as well as the exchange rate constants were ob-
tained by visual fit to the experimental data. The conformer
present in greater concentration just below coalescence is re-
ferred to as the “A” conformer, and the other conformer is
described as “B”. The free energy of activation was calculated
using the equation22
using the GIAO method and the WP04 functional that was
developed with that purpose in mind,30 using the cc-pVDZ basis
set. SCRF calculations were again carried out in DMSO as a
solvent, and the raw isotropic magnetic shieldings (IMS) were
converted to chemical shifts relative to TMS δ by scaling them
with the parameters elaborated recently by Jain et al.21 (δ =
31.844 - (IMS/1.0205)). All calculations were done with the
Gaussian program packages.32
Acknowledgment. The authors gratefully acknowledge
financial support of this work by the National Science Foun-
dation (CHE-0455575, R.S.G., CHE-9610347) and the Swiss
National Science Foundation (Project 2000-121747, T.B.).
The authors also thank Mr. Takuhei Yamamoto for the
preparation and characterization of thioether 1e.
ΔGq ¼ ½23:76 - lnðk=TÞꢀRT
Supporting Information Available: Yields, physical proper-
ties and spectroscopic details for 1-3. Variable-temperature 300
MHz H NMR spectroscopic studies showing the full spectral
Quantum Chemical Calculations. Geometry optimizations
and frequency calculations were first done with the B3LYP/6-31G*
method, where all stationary points (minima, transition states)
were properly characterized. Geometries were then reoptimized
with the recently introduced “double hybrid” B2PLYP func-
tional,31a corrected for contributions of dispersion energy,31b
which accounts much better for steric interactions than pure
DFT functionals. However, analytic second derivatives are not
yet available for the B2PLYP method, so we had to use thermal
corrections and entropies from B3LYP calculations in the
evaluation of the relative enthalpies and free energies listed in
Table 6. Finally, single-point SCRF calculations in DMSO as a
continuum solvent were done with the B2PLYP-D functional at
the corresponding geometries to evaluate relative enthalpies and
free energies in solution. NMR chemical shifts were calculated
1
range for thioether 1a in CD2Cl2 and DMSO-d6, thioether 1b in
CD2Cl2, thioether 1h in CD2Cl2 and DMSO-d6, selenoether 2a
in CD2Cl2 and DMSO-d6, selenoether 2e in DMSO-d6, tell-
uroether 3a in CD2Cl2, and DMSO-d6, telluroether 3d in
DMSO-d6. Fits of calculated and experimental NMR spectra
for selenoether 2a and telluroether 3a. X-ray crystallographic
data for thioether 1a in CIF format. Complete citation for the
Gaussian Program.32 Calculated structures of the three atrop-
isomers of thioether 1a and selenoether 2a. This material is
available free of charge via the Internet at http://pubs.acs.org.
(31) (a) Grimme, S. J. Chem. Phys. 2006, 124, 034108/1–034108/16.
(b) Schwabe, T.; Grimme, S. Phys. Chem. Chem. Phys. 2007, 9, 3397–3406.
For an overview of these methods, see: Schwabe, T.; Grimme., S. Acc. Chem.
Res. 2008, 41, 569–579.
(32) Frisch, M. J. et al. Gaussian 03, Revision E.01 and Gaussian 09,
Revision A.02; Gaussian, Inc.: Wallingford, CT, 2003 (for complete refer-
ence, see Supporting Information).
(30) Wiitala, K. W.; Hoye, T. R.; Cramer, C. J. J. Chem. Theory Comput.
2006, 2, 1085–1092. In Gaussian, the WP04 functional is invoked by
specifying the BLYP keyword and adding iop(3/76=1000001189,3/77=
0961409999,3/78=0000109999) to the keyword line.
J. Org. Chem. Vol. 75, No. 24, 2010 8371