New and Facile Synthesis of Symmetrical and Unsymmetrical Diaryl Chalcogenides Using Trivalent Organobismuth Derivatives

Full text of DOI:10.1021/jo982093x

3722
J . Org. Chem. 1999, 64, 3722-3725
Notes
xylenes, mesitylene, or DMSO yields in a few hours a
mixture of phenyl telluride and diphenyl telluride (eq 1).
New a n d F a cile Syn th esis of Sym m etr ica l
a n d Un sym m etr ica l Dia r yl Ch a lcogen id es
Usin g Tr iva len t Or ga n obism u th
Der iva tives
Ph3Bi + Te0 heating8 Ph2Te + Ph2Te2
(1)
It was found that despite the long reaction time (24 h)
in the case of benzene and toluene, some Ph3Bi still
remained in solution. In xylenes and mesitylene, how
ever, Ph3Bi was consumed in a reasonable amount of time
(4.5 and 4 h, respectively) with a higher ratio in favor of
diphenyl telluride (Ph2Te/Ph2Te2 ) 63/37) with the latter
solvent. The ratio was somewhat reversed when heated
under reflux of DMSO for 1 h (Ph2Te/Ph2Te2 ) 32/68).
Arylbismuthine derivatives are best known for the
arylation of amines, enols, and phenols, which they are
able to perform with high effectiveness.5 However, only
in a very few cases,6 had it been possible to transfer to
the functional organic group more than one aryl group
out of the three that are available on the bismuth
reagent. These arylation reactions could then be consid-
ered wasteful because two aryl groups were not utilized.
However, in the present case, it must be noted that all
the phenyl groups from the bismuth could be accounted
for in the two products, Ph2Te2 and Ph2Te.
Since phosphines react with elemental chalcogen in
boiling toluene or at higher temperatures to yield stable
species such as R3PdX,7 it is reasonable to assume that
the formation of the analogous bismuth(V) intermediate
can constitute the first step in the reaction between Ph3-
Bi and Te0 (Scheme 1). Ph3BidTe is likely to undergo
rapid self-reduction to Ph2BiTePh. The oxidation/self-
reduction process could repeat itself once or twice before
a ligand coupling finally occurs, yielding the mono- or
ditelluride derivative, respectively.
Thomas Arnauld,*,† Derek H. R. Barton,‡ and
J ean-Franc¸ois Normant§
Department of Chemistry, Texas A&M University,
P.O. Box 30012, College Station, Texas 77842-3012, and
Laboratoire de Chimie des Organoe´le´ments,
URA 473 du CNRS, Universite´ Pierre et Marie Curie,
Tour 44-45, 4, Place J ussieu, Boˆıte 183,
75252 Paris Ce´dex 05, France
Received October 19, 1998
In tr od u ction
A number of syntheses of diaryl chalcogenide deriva-
tives are already available,1 but finding an easy and
general method still remains a task of interest. Indeed,
among the methods using lithium derivatives, Grignard
reagents, and aryl diazonium and aryl iodonium salts,
the early syntheses based on Ar2Hg as the aryl donor are
still considered as the most general way to obtain sym-
metrical chalcogenides when reacting on elemental chal-
cogen2 or unsymmetrical chalcogenides when reacting on
a diaryl dichalcogenide.3 However, these methods are
often undesirable because of the high toxicity of the
diarylmercury salts. Whereas numerous examples of
bismuth-chalcogen interactions can be found either in
mineralogy or in inorganic chemistry, very few accounts
have been reported in the field of organic chemistry.4
Wishing to examine the reactivity of triarylbismuthine
derivatives toward chalcogens, we discovered two new
reactions allowing the synthesis of symmetrical and
unsymmetrical diaryl chalcogenides in good to high yield.
Arylbismuthine derivatives are effective aryl donors on
amines, enols, and phenols.5 Now they prove to be
efficient aryl donors in the preparation of diaryl chalco-
genide derivatives.
In both cases, PhBidTe is generated, which could
undergo a redistribution reaction (eq 2), common in
bismuth chemistry.8,9 Ph3Bi will therefore be regener-
3PhBidTe f Ph3Bi + Bi2Te3
(2)
ated,which accounts for the remarkable efficiency of the
reaction. We suspect Bi2Te3 to be the side product of the
reaction. However, since the insoluble product that is
formed during the reaction could not be separated from
the excess of Te0 used in the reaction, we have no
evidence to support this hypothesis.
The procedure was successfully adapted to other tri-
arylbismuthine derivatives (Table 1). The purification
required no column chromatography. Instead, the prod-
ucts could be purified using a reduction-extraction
Resu lts a n d Discu ssion
Heating triphenylbismuthine in the presence of tel-
lurium metal (10 equiv) under reflux of benzene, toluene,
† Texas A&M University until March 1999. Universite´ Pierre et
Marie Curie from this date.
‡ In memory of Derek H. R. Barton (deceased March 16, 1998).
§ Universite´ Pierre et Marie Curie.
(1) The Chemistry of Organic Selenium and Tellurium Compounds;
Patai, S., Rappoport, Z., Eds.; J ohn Wiley & Sons: New York, 1986,
Vol. 1; 1987, Vol. 2. Magnus, P. D. Comprehensive Organic chemistry;
J ones, D. N., Ed.; Pergamon Press: Oxford, 1979; Vol. 3, pp 491-538.
Paulmier, C. Selenium Reagents and Intermediates in Organic Syn-
thesis; Baldwin, J . E., Ed.; Pergamon Press: Oxford, 1986. Petragnani,
N.; Comasseto, J . V. Synthesis 1986, 1-30. Petragnani, N. Tellurium
in Organic Synthesis; Academic Press: London, 1994.
(2) Krafft, F.; Lyons, R. E. Chem. Ber. 1894, 27, 1768-1773.
(3) Okamoto, Y.; Yano, T. J . Organomet. Chem. 1971, 29, 99-103.
(4) (a) Barton, D. H. R.; Dadoun, H.; Gourdon, A. Nouv. J . Chim.
1982, 6, 53-57. (b) Calderazzo, F.; Morvillo, A.; Pelizzi, G.; Poli, R.;
Ungari, F. Inorg. Chem. 1988, 27, 3730-3733. (c) Goel, R. G.; Prasad,
H. S. Spectrochim. Acta 1979, 35A, 339-344.
(5) (a) Barton, D. H. R.; Finet, J . P. Pure Appl. Chem. 1987, 59, 937-
946. (b) Finet, J . P. Chem. Rev. 1989, 89, 1487-1501. (c) Suzuki, H.;
Ikegami, T.; Matano, Y. Synthesis 1997, 249-267.
(6) Arnauld, T.; Barton, D. H. R.; Doris, E. Tetrahedron 1997, 53,
4137-4144.
(7) (a) Screttas, C.; Isbell, A. F. J . Org. Chem. 1962, 27, 2573-2577.
(b) Zingaro, R. A.; Steeves, B. H.; Irgolic, K. J . Organomet. Chem. 1965,
4, 320-323. (c) Du Mont, W. W.; Kroth, H. J . J . Organomet. Chem.
1976, C35-C37.
10.1021/jo982093x CCC: $18.00 © 1999 American Chemical Society
Published on Web 05/14/1999

Notes
J . Org. Chem., Vol. 64, No. 10, 1999 3723
Sch em e 1
run with tris[p-(trifluoromethyl)phenyl]bismuthine. In
most cases, the products were isolated by using only the
reduction-extraction procedure described for the tel-
lurides. But sometimes, column chromatography was
necessary to purify the products further.
Comparing the tellurium and the selenium series, we
observed that the ratio of Ar2Te/Ar2Te2 is constantly
higher or equal to the ratio of Ar2Se/Ar2Se2 for one given
ligand. This could be related to the different dissociation
energies of the Bi-Te (230.8 ( 11.3 kJ ‚mol-1) and Bi-
Se (279.0 ( 5.9 kJ ‚mol-1) bonds.13 Also, in the tellurium
series as well as in the selenium series the ratio of mono/
dichalcogenide is obviously dependent on the electron-
withdrawing force of the ligand transferred. Indeed,
TFMP gives in both cases significantly less of the
monochalcogenide than mesityl or anisyl ligands.
As discussed previously, the ratio of Ph2Te2/Ph2Te was
reversed by changing the solvent to refluxing DMSO.
Consequently, we were interested in exploring the pos-
sibility of controlling the ratio of mono- and dichalco-
genide by modifying the reaction conditions. To see if
Ph2Te2 was produced from Ph2Te or vice versa, a simple
kinetic experiment was carried out. The results show that
Ph2Te2 and Ph2Te appear to be formed simultaneously
and that the ratio of these two products remains constant
after Ph3Bi was consumed.
Nevertheless, it is known that Ph2Te2 can lose a
tellurium atom and yield Ph2Te when heated under high
temperature.14 However, a blank experiment showed
that, under the reaction conditions, only 8% of Ph2Te
could be isolated after 24 h, and 92% of the starting Ph2-
Te2 was recovered. Clearly, the telluride does not arise
from the thermolysis of the ditelluride.
Another pathway could explain the formation of Ph2-
Te as a secondary product from Ph2Te2. We studied the
possibility of the reaction between Ph2Te2 and Ph3Bi (eq
3). We found that the reaction occurs slowly (48 h in
mesitylene under reflux for completion) and, therefore,
can be disregarded as the main pathway for producing
Ph2Te.
Ta ble 1. Ar 3Bi + Te0 (10 Equ iv)a
b
aryl group
time (h)
Ar2Te2 (%)
Ar2Teb (%)
phenyl
12
24
3
5
30
34
<5
<5
<5
61
66
95
89
93
39
mesitylc
anisyl
1-naphthyl
TFMPd
a
Reactions performed in mesitylene under reflux unless other-
b
wise stated. Isolated yields (given in phenyl groups transferred).
c 5 equiv of Te0 and reflux in toluene were sufficient in this
reaction. TFMP, p-(trifluoromethyl)phenyl. 1H NMR yield.
d
Ta ble 2. Ar 3Bi + Se0 (5 Equ iv)a
b
aryl group
time (h)
Ar2Se2 (%)
75
Ar2Seb (%)
phenylc
mesityld
anisylc
1-naphthyl
TFMP
3
24
4
5
24
25
77
73
94
<5
26
<5
78
a
Reactions performed in mesitylene under reflux unless other-
b
Ph3Bi + Ph2Te2 heating8 Ph2Te
(3)
wise stated. Isolated yields (given in aryl groups transferred).
c Performed in xylenes under reflux. Performed in toluene under
d
reflux.
However, it was interesting to note that more than 1
equiv of Ph2Te, compared to Ph2Te2, was obtained in this
reaction. Thus, Ph2Te is not produced by simple detel-
luration of Ph2Te2 as is the case with copper,15 nickel,16
palladium,17 and even cobalt.18 It is more likely that the
two moieties of the ditelluride are used in this reaction.
Therefore, from one diphenyl ditelluride, two arylphenyl
telluride derivatives are expected (eq 4).
procedure.10 However, in the case of TFMP the products
could not be separated, and the ratio was determined by
1H NMR, based upon the assignment of the spectral data
reported in the literature.11,12 In all cases but the latter,
the monotelluride derivative was the major product.
Similar results were obtained with selenium (Table 2).
Fewer equivalents of Se0 were needed to bring the
reaction to completion. Still, good yields of the monose-
lenide were obtained, albeit that the selectivity was
generally lower than in the corresponding monotelluride.
Once again, a strong electron-withdrawing ligand had
a dramatic effect on the product ratio. Indeed, only traces
of the monoselenide were detected when the reaction was
(11) Minoura, M.; Sagami, T.; Miyasato, M.; Akiba, K. Tetrahedron
1997, 53(36), 12195-12202.
(12) Engman, L.; Persson, J . J . Organomet. Chem. 1990, 388, 71-
74.
(13) Uy, O. M.; Drowart, J . Trans. Faraday Soc. 1969, 65, 3221-
3230.
(14) Dunne, S. J .; Summers, L. A.; von Nagy-Felsobuki, E. I. J .
Heterocycl. Chem. 1993, 30, 409-412.
(15) (a) Akiba, M.; Lakshmikantham, M. V.; J en K. Y.; Cava M. P.
J . Org. Chem. 1984, 49, 4819-4821. (b) Sadekov, I. D.; Bushkov, A.
Y.; Minkin, V. I. Zh. Obshch. Khim. 1973, 43, 815-817.
(16) Han, L. B.; Tanaka, M. J . Chem. Soc., Chem. Commun. 1998,
47-48.
(17) Engman, L.; Stern, D.; Pelcman, M.; Andersson, C. M. J . Org.
Chem. 1994, 59, 1973-1979.
(18) Uemura, S.; Takahashi, H.; Ohe, K.; Sugita, N. J . Organomet.
Chem. 1989, 361, 63-72.
(8) Deacon, G. B.; J ackson, W. R.; Pfeiffer, J . M. Aust. J . Chem. 1984,
37, 527-535.
(9) Gilman, H.; Yablunky, H. L. J . Am. Chem. Soc. 1941, 63, 207-
211.
(10) The separation of Ph2Te2 and Ph2Te could not be performed by
column chromatography due to their similar Rf values. However, Ph2-
Te2 can be extracted under its reduced form as phenyl telluride anion,
allowing a quantitative separation and recovery of the two derivatives
Ph2Te2 and Ph2Te.

3724 J . Org. Chem., Vol. 64, No. 10, 1999
Ar3Bi + Ph2Te2 heating8 2ArTePh
Notes
Ta ble 3. Ar 3Bi + P h 2X2 (1 Equ iv)a
(4)
aryl group
X
time (h)
Ar-X-Phb (%)
phenyl
Te
Te
Te
Te
Te
Se
Se
Se
Se
Se
48
48
48
12
12
24
24
48
24
48
88
95
74
71
62
76
62
62
89
65
This would provide a new route for the synthesis of
unsymmetrical tellurides and selenides simply if the aryl
groups borne by the chalcogen and bismuth are different
(Table 3).
mesitylc
anisyl
1-naphthyl
TFMP
phenyl
As predicted, the reaction between diphenyl ditelluride
or diphenyl diselenide with different triarylbismuthine
species gave the unsymmetrical chalcogenides in good to
high yield. No real difference in the yields can be
observed when the reaction was performed either on the
ditelluride or on the diselenide. Furthermore, as men-
tioned previously, from one Ph2X2, two ArXPh are
produced, making this a particularly efficient reaction.
As was mostly the case in Tables 1 and 2, mesitylene
was again the preferred solvent. The choice of a lower
boiling solvent like toluene when trimesitylbismuthine
is used was driven by the fact that a complicated mixture
of products was obtained when the reaction was per-
formed at a higher temperature.
mesitylc
anisyl
1-naphthyl
TFMP
a
Reactions performed in mesitylene under reflux unless other-
b
wise stated. Isolated yields (calculated with respect to the PhX
moiety). c Performed in xylenes under reflux.
(6H, m); 13C NMR δ 118.71, 122.32, 125.93, 127.26, 127.31,
127.36, 127.41, 129.54, 129.90, 130.33, 130.77, 131.20, 137.76;
19F NMR δ -63.38. Anal. Calcd for C21H12BiF9: C, 39.15; H, 1.88.
Found: C, 39.06; H, 1.87.
Rea ction betw een Ar 3Bi a n d Te0 or Se0. A typical proce-
dure is given for Ph3Bi and Te0. Ph3Bi (220 mg, 0.5 mMol) and
Te0 (638 mg, 5 mMol) were stirred together in mesitylene (5 mL)
under reflux for 12 h. After cooling, filtration, and evaporation
of the solvent under oil pump vacuum, the residue was taken
in THF (15 mL) and water (50 mL). A large excess of NaBH4
was added until the red color was completely discharged. Rapid
extraction with ether (50 mL) afforded after drying (MgSO4) the
pure diphenyl telluride (139 mg, 66%) as a colorless oil:22 1H
The reactions carried out in mesitylene had to be
performed under argon atmosphere as we noticed that,
otherwise, a significant amount of the solvent was air-
oxidized to yield 3,5-dimethylbenzaldehyde.
Con clu sion
NMR δ 7.20 ppm (3H, m), 7.70 ppm (2H, dd, J 1 ) 8.1 Hz, J 2
)
1.5 Hz); 13C NMR δ 127.81, 129.48, 137.96; MS m/e 282 (M+).
HCl (3 N) was added to the aqueous layer until no more H2 was
released. Extraction with ether (50 mL), drying (MgSO4), and
evaporation of the solvent afforded pure diphenyl ditelluride (104
mg, 34%) as a red solid:23 mp 62-65 °C; 1H NMR δ 7.20 (3H,
m), 7.80 (2H, dd, J 1 ) 8.1 Hz, J 2 ) 1.5 Hz); 13C NMR δ 128.06,
129.27, 137.60; MS m/e 409 (M+).
Two new and original reactions for the synthesis of
symmetrical and unsymmetrical chalcogenides have been
developed. Once again, triarylbismuthine reagents prove
to be very efficient aryl donors. Their stability to moisture
and air, as well as their low toxicity, give them a real
advantage over most other aryl donors. Even though the
results were not satisfactory in the particular case of the
symmetrical derivatives bearing a strong electron-
withdrawing ligand, this new procedure is easy and
general enough to be compared to the well-known mer-
cury salt method.
Bis[1-n a p h th yl] tellu r id e: mp 108-109 °C (lit.24 mp 126.5
°C, lit.25 123-126 °C); 1H NMR δ 7.22 (1H, m), 7.50 (2H, m),
7.80 (3H, m), 8.15 (1H, m); 13C NMR δ 126.30, 126.60, 126.97,
128.82, 129.22, 131.27, 133.78, 138.01; MS m/e 384 (M+). Anal.
Calcd for C20H14Te: C, 62.90; H, 3.69. Found: C, 62.75; H, 3.64.
Bis[1-n a p h th yl] selen id e: mp 106-107 °C (lit.24 mp 114 °C);
1H NMR δ 7.28 (1H, t, J ) 7.7 Hz), 7.52 (3H, m), 7.79 (1H, d, J
) 8.1 Hz), 7.86 (1H, m), 8.34 (1H, m); 13C NMR δ 126.09, 126.33,
126.81, 127.09, 128.40, 128.61, 132.17, 134.11; MS m/e 334 (M+).
Anal. Calcd for C20H14Se: C, 72.08; H, 4.23. Found: C, 72.05;
H, 4.27.
Rea ction betw een Ar 3Bi a n d P h 2X2. A typical procedure
is described for Mes3Bi and Ph2Te2. Mes3Bi (283 mg, 0.5 mMol)
and Ph2Te2 (205 mg, 0.5 mMol) were stirred in xylenes (5 mL)
under reflux until the red color was discharged (48 h). After
filtration, evaporation of the volatiles, and column chromatog-
raphy on silica gel (eluent: hexanes), mesitylphenyl telluride
was isolated as a solid (307 mg, 95% yield): mp 27-29 °C; 1H
NMR δ 2.31 (3H, s), 2.54 (6H, s), 7.01 (2H, s), 7.12 (3H, m), 7.32
(2H, m); 13C NMR δ 21.00, 29.50, 116.27, 118.28, 126.53, 127.57,
Exp er im en ta l Section
Melting points are uncorrected. 1H, 13C, and 19F NMR were
recorded (300, 75, and 282 MHz, respectively) with Me4Si,
residual solvent peak, and CFCl3 as respective internal refer-
ences. Elemental analyses were performed by Atlantic Microlab,
Inc., Norcross, GA. Ph3Bi, Ph2Te2, Ph2Se2, tellurium powder (200
mesh), and selenium powder (100 mesh) were purchased from
Aldrich Chemical Co. Trimesitylbismuthine,9 tris[p-anisyl]bis-
muthine19 and tris[1-naphthyl]bismuthine20 were synthesized
following the literature procedures. The typical procedure is
given for tris[p-(trifluoromethyl)phenyl]bismuthine21 mentioned
twice in the literature but for which no physical data have been
given.
Tr is[p-(tr iflu or om eth yl)p h en yl]bism u th in e. A solution of
1-bromo-4-(trifluoromethyl)benzene (7.30 mL, 52 mMol) in dry
THF (100 mL) was added to magnesium turnings (2.0 g, 83
mMol) under argon. The solution was transferred via cannula
to a solution of BiCl3 (5 g, 15.86 mMol) in dry THF (50 mL).
After the addition, the solution was quenched with methanol,
and the volatiles were evaporated. The residue was taken in
hexanes, and the solution was filtered and concentrated to yield
the crude product. Crystallization from hexanes afforded the tris-
[p-(trifluoromethyl)phenyl]bismuthine as colorless plates (7.71
g, 75% yield): mp 145-146 °C; 1H NMR δ 7.65 (6H, m), 7.85
129.34, 134.59, 139.41, 145.45; HRMS calcd for
326.0314, found: 326.0324. Anal. Calcd for C15H16Te: C, 55.63;
H, 4.98. Found: C, 55.85; H, 5.00.
C15H16Te
p-(Tr iflu or om eth yl)p h en yl p h en yl tellu r id e: oil; 1H NMR
δ 7.28 (2H, m), 7.40 (3H, m), 7.66 (2H, m), 7.80 (2H, m); 13C
NMR δ 125.84, 125.89, 128.70, 129.84, 136.42, 138.02, 139.41;
19F NMR δ -63.13; HRMS calcd for C13H9F3Te: 351.9719,
found: 351.9746. Anal. Calcd for C13H9F3Te: C, 44.64; H, 2.59.
Found: C, 44.84; H, 2.83.
p-(Tr iflu or om eth yl)p h en yl p h en yl selen id e: oil; 1H NMR
δ 7.35 (3H, m), 7.44 (4H, m), 7.57 (2H, m); 13C NMR δ 122.30,
(22) Detty, M. R.; Seidler, M. D. J . Org. Chem. 1982, 47, 1354-
1356.
(19) Gillmeister, A. Chem. Ber. 1897, 30, 2843-2850.
(20) Challenger, F. J . Chem. Soc. 1914, 105, 2210-2218.
(21) (a) Chen, X.; Ohdoi, K.; Yamamoto, Y.; Akiba, K. Y. Organo-
metallics 1993, 12, 1857-1864. (b) Frohn, H. J .; Mauer, H. J . Fluor.
Chem. 1986, 34, 129-145.
(23) Dictionary of Organic Compounds; Chapman & Hall, Eds.;
London, 1996; Vol. 3, p 2840.
(24) Lyons, R. E.; Bush, G. C. J . Am. Chem. Soc. 1908, 30, 831-
836.
(25) Suzuki, H.; Nakamura, T. Synthesis 1992, 549-551.

Notes
J . Org. Chem., Vol. 64, No. 10, 1999 3725
125.35, 125.40, 125.82, 125.87, 125.91, 125.97, 128.24, 128.52,
128.75, 129.71, 129.78, 130.95, 131.03, 134.83, 135.59, 137.21,
137.79; 19F NMR δ -63.13; HRMS calcd for C13H9F3Se 301.9822,
found: 301.9831.
Su p p or tin g In for m a tion Ava ila ble: Melting point and
1H, 13C, and 19F (when applicable) NMR data for all compounds
except bis[p-(trifluoromethyl)phenyl] telluride and bis[p-(tri-
fluoromethyl)phenyl] ditelluride, for which only 1H NMR data
are available. Tables containing the optimization of the
reaction conditions and the kinetic experiment. This material
is available free of charge via the Internet at http://pubs.acs.org.
Ack n ow led gm en t. We thank Schering Plough, Uni-
lever, and Merck Research Laboratories for their timely
support of this work. We also thank F. Taran, E. Doris,
and J . A. Smith for discussion and for reviewing the
manuscript.
J O982093X