Fluoride-triggered indium-mediated synthesis of (hetero)biaryls

Article abstract of DOI:10.1039/b901722d

A simple method for the preparation of triorganoindium reagents under mild conditions based in indium-silicon exchange is described.

Full text of DOI:10.1039/b901722d

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/dalton | Dalton Transactions
Fluoride-triggered indium-mediated synthesis of (hetero)biaryls†‡
´
Enrique Font-Sanchis, Angela Sastre-Santos and Fernando Ferna´ndez-La´zaro*
Received 27th January 2009, Accepted 13th February 2009
First published as an Advance Article on the web 20th February 2009
DOI: 10.1039/b901722d
Table 1 (Hetero)aromatics via triorganoindium reagents formed in situa
A simple method for the preparation of triorganoindium
reagents under mild conditions based in indium–silicon ex-
change is described.
TBAF/ InCl3/ Temp.b/
Entry Ar mmol
mmol timeb
Ar¢ Ar–Ar¢ Yield (%)c
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
1
1
1
1
1
1
2
2
2
2
3
3
3
1
2
4
4
4
0
0.7
0.7
0.7
0.0
0.7
0.7
0.7
0.7
0.7
0.0
0.7
0.7
0.0
0.7
0.7
0.7
0.7
0.0
rt/1 h
rt/1 h
rt/1 h
rt/1 h
rt/0.5 h
reflux/2 h
rt/0.5 h
rt/1 h
reflux/2 h
rt/1 h
rt/0.5 h
rt/1 h
rt/0.5 h
rt/1 h
a
a
a
a
a
a
a
a
a
a
a
a
a
bd
c
1a
1a
1a
1a
1a
1a
2a
2a
2a
2a
3a
3a
3a
1b
2c
2c
2c
2c
0
68
31
0
46
68
49
89
87
0
90
66
25
54
82
71
45
0
Coupling of organometallic compounds with organic electrophiles
in the presence of a palladium or nickel catalyst is a widespread
and useful way for carbon–carbon bond formation.1 Thus,
organoboron, organotin, organosilicon and organozinc com-
pounds, among others, have been extensively studied. A less
common strategy is the use of triorganoindium reagents, although
they benefit from high reactivity, efficiency, versatility and chemos-
electivity, together with a low toxicity of indium derivatives.2,3
The preparation of these reagents is mainly achieved by lithium–
halogen exchange, using butyllithium, followed by indium–lithium
exchange on the organolithium species,4 which limits the func-
tional groups that can be present in the organometallic compound.
The direct indium–halogen exchange can also be problematic
due to the reducing character of indium metal.5 As part of our
interest in this field, we were involved in developing an alternative
method to prepare triorganoindium reagents mild enough to allow
the presence of “sensitive” groups. Here, we wish to report the
fluoride promoted indium–silicon exchange as a versatile and
simple method to prepare triorganoindium reagents, even in the
presence of organolithium-sensitive functional groups (Scheme 1).
2.1
8.4
2.1
2.1
2.1
2.1
2.1
2.1
2.1
2.1
2.1
2.1
2.1
2.1
2.1
2.1
2.1
rt/1 h
rt/0.5 h
rt/1 h
a
a
a
rt/1 h
a Reagents and conditions: ArSi(OEt)3 2.1 mmol, TBAF, InCl3, THF;
Ar¢–I (2.1 mmol), Pd(PPh3)4 (0.06 mmol), reflux, 14 h. b Conditions
for the transmetallation step. c Yield of isolated, purified products.
d p-Bromobenzaldehyde was used instead of the p-iodo analogue.
tetrakis(triphenylphosphine)palladium(0) at reflux (Scheme 2,
Table 1). The palladium catalyst was selected as result of previous
work on heterobiaryl synthesis using triorganoindium reagents.3
Scheme 1 Proposed synthesis of organoindium reagents.
We decided to study the preparation of trisphenylindium
starting from triethoxyphenylsilane as a model reaction. Thus,
we reacted the trialkoxysilane with indium trichloride and
tetrabutylammonium fluoride (TBAF) in THF to prepare the
triorganoindium species, which was subsequently reacted in
a one pot process with 2-iodothiophene in the presence of
Divisio´n de Qu´ımica Orga´nica, Instituto de Bioingenier´ıa, Universidad
Miguel Herna´ndez, 03202, Elche, Spain. E-mail: fdofdez@umh.es; Fax: +34
96 665 8351; Tel: +34 96 665 8405
† Electronic supplementary information (ESI) available: Experimental
1
procedures and H- and 13C-NMR data and spectra for all compounds.
See DOI: 10.1039/b901722d
‡ General procedure: A solution of TBAF (1 M in THF, 2.1 mmol) was
added to a stirring solution of InCl3 (0.7 mmol) and the corresponding
trialkoxysilane (2.1 mmol) in THF (20 mL). After 1 h at room temperature,
the electrophile (2.1 mmol) and Pd(PPh3)4 (0.06 mmol) were added. The
resulting mixture was refluxed under argon atmosphere overnight (14 h)
and then, the reaction was quenched by addition of water (25 mL).
The crude reaction was extracted with diethyl ether (3 ¥ 25 mL). The
combined organic extracts were dried over anhydrous Na2SO4, filtered, and
concentrated. The residue was purified by flash column chromatography.
Scheme 2 General scheme of the fluoride-triggered organoindium prepa-
ration and its palladium-catalyzed coupling.
Entries 1–3 show the effect of the amount of TBAF. Thus,
in the absence of fluoride anions the reaction does not proceed,
the best ratio being a 1 : 1 trialkoxysilane : fluoride. The presence
2470 | Dalton Trans., 2009, 2470–2473
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
of indium(III) chloride is necessary for the reaction to proceed
as pointed out in entry 4, thus indicating that this is not just
an example of Hiyama coupling.6 Finally, the effect of time
and temperature was studied (entries 2, 5 and 6), where room
temperature and 60 min were the best choices. It is noteworthy
that the use of indium trifluoride instead of the mixture of TBAF
and indium(III) chloride did not afford the coupling product.
After examination of these results, one might be tempted to
consider that the Hiyama coupling is a more straightforward
synthesis when starting from a trialkoxysilane. However, both
trialkoxysilanes and organoindium compounds have their own
reactivities and limitations, and thus it could be very useful to
have a second choice to select from. In this regard, it is interesting
to note that yields for Hiyama couplings can vary from modest to
excellent, the latter being frequently obtained with sophisticated
palladium complexes.7 Moreover, a typical Hiyama reaction is
carried out using 2 equivalents of trialkoxysilane, 2 equivalents
of fluoride-releasing agent and up to 10% of palladium catalyst
with respect to the aromatic halide,8 while our procedure just
employs 1 equivalent of trialkoxysilane, 1 equivalent of TBAF
and 3% of Pd(PPh3)4 catalyst, thus being a more efficient process.
Be this as it may, it is clear that the process here described is
not just an example of Hiyama coupling, and can be syntheti-
cally useful depending on the complexity of the trialkoxysilane
derivative.
For the study of this methodology on an electron-rich het-
eroaromatic, we selected 2-(triethoxysilyl)thiophene as starting
material (Table 1, entries 7–10). Based on the previous set of
experiments, we fixed the trialkoxysilane : fluoride : indium ratio to
1 : 1 : 0.33, although control experiments without indium salt were
performed. Results follow the same trends as in the phenylsilane
case. Thus, the better yield is obtained after one hour at room
temperature in the presence of indium(III) chloride (entry 8). Once
again, in the absence of indium no reaction takes place (entry 10),
thus discarding any contribution from an Hiyama coupling. It is
noteworthy that yields are better for the thienylsilane derivative,
when compared with the phenylsilane one.
Then, we moved to electron-poor heteroaromatics, select-
ing 3-(triethoxysilyl)pyridine as model compound (Table 1,
entries 11–13). Also in this case, we fixed the trialkoxysi-
lane : fluoride : indium ratio to 1 : 1 : 0.33, although control experi-
ments without indium salt were performed. To our surprise, in the
pyridine series the best result was obtained with the shorter reac-
tion time, just 30 min (entry 11), and for this condition a certain
contribution of Hiyama coupling was observed (entry 13). Longer
reaction times gave a lower yield. Perhaps this can be attributed to
a lower stability of the tris(pyridyl)indium when compared with
the analogous indium compounds bearing phenyl or thienyl rings.
With the previous results in mind, we tried to perform
the reaction using an organic electrophile bearing a sensitive
functional group, such as the carbonyl of p-bromobenzaldehyde
(Table 1, entry 14). Thus, reaction of the phenylsilane (2.1 mmol)
with TBAF (2.1 mmol) and indium trichloride (0.7 mmol) in
THF for 1 h at room temperature, followed by addition of
p-bromobenzaldehyde and the palladium catalyst (0.06 mmol) and
heating at reflux for 14 h yielded the desired compound in 54%.9
This relatively low yield is not unexpected, as it has been previously
reported that bromoaromatics do not couple with organoindium
reagents as well as iodoaromatics do.3 Moreover, as mentioned
above, the phenylsilane series leads to poorer results than the
thienylsilane one.
We attempted the synthesis of 4-(2-thienyl)acetophenone10
starting from 2-(triethoxysilyl)thiophene using the standard con-
ditions for the thienyl series (Table 1 entry 15). The desired
compound was obtained in 82% yield, thus showing the usefulness
of this methodology.
Finally, we also carried out the synthesis of 4-(2-
thienyl)acetophenone by preparing the triorganoindium derivative
of the acetophenone (Table 1, entries 16–18). The results resembled
those of the pyridine series. Thus, the best conditions are achieved
by using shorter periods of time (entry 16). No product was
detected in the absence of indium trichloride (entry 18). These
results indicated that this methodology can be applied to the
preparation of organoindium reagents bearing groups which are
sensitive to organolithium species.
To develop a better understanding of the role of the indium
salt, i.e. whether it forms the triorganoindium reagent or just
acts as a Lewis acid inducing the cross-coupling,11 we performed
some experiments to record the NMR of the intermediate species.
Thus, 2-(triethoxysilyl)thiophene was treated with TBAF and
indium(III) chloride in THF at room temperature for one hour,
the solvent was distilled off and the residue was dissolved in
CDCl3. To our satisfaction, comparison of the 1H-NMR spectrum
of the starting trialkoxysilane (A) with that of the residue
(B) showed the disappearance of the former (Fig. 1 and ESI†).
Thus, in (B) there are not even traces of signals from the starting
compound at 7.21 ppm, and the signals at 3.89 and 1.25 ppm
corresponding to the ethoxy groups of the starting trialkoxysilane
have completely disappeared. The only clear signals in (B) are those
of the thienyl group, tetrabutylammonium chloride (formed as
byproduct) and some traces of THF (ESI). Moreover, we prepared
tris(thienyl)indium by an accepted standard procedure, namely
treatment of thienyllithium with indium(III) chloride in THF; after
1
distillation of the solvent, the H-NMR in CDCl3 was recorded
(Fig. 1C and ESI), showing a reasonable match, within 0.01 ppm,
with the aromatic signals of Fig. 1B (spectrum 1B contains the
intermediate compound and tetrabutylammonium chloride, while
spectrum 1C contains only the intermediate species).
We were not able to record a 13C-NMR spectrum in CDCl3,
because the compound decomposed generating thiophene. Then,
we performed the same experiments but dissolving the residue
in THF-d8. In this way we could obtain a 13C-NMR spectrum
of the intermediate obtained from the In–Si exchange (ESI),
which showed the complete disappearance of the starting material.
Thus, the signals at 137.5, 132.6, 130.6 and 128.7 corresponding
to 2-(triethoxysilyl)thiophene have vanished, new ones emerging
at 135.9, 130.0 and 127.6. We could not find the quaternary
carbon atom, because prolonged acquisition times resulted in
decomposition of the intermediate in this complex mixture, with
many new signals appearing, while the intensity of the three
signals of the tris(thienyl)indium decreased. At higher field we
could only observe signals corresponding to tetrabutylammonium
chloride and some traces of THF and triethoxyfluorosilane (ESI).
1
As in the H-NMR experiments, we recorded the 13C-NMR of
tris(thienyl)indium prepared via In–Li exchange (ESI), obtaining
once again a reasonable match of the aromatic signals. Taking
into account that (a) the aromatic signals in both 1H- and
13C-NMR are the same, no matter whether the intermediate is
This journal is
The Royal Society of Chemistry 2009
Dalton Trans., 2009, 2470–2473 | 2471
©

View Article Online
Fig. 1 Expanded aromatic region of the 1H-NMR of commercially available 2-(triethoxysilyl)thiophene (A), tris(thienyl)indium prepared by
indium–silicon exchange (B) and tris(thienyl)indium prepared by indium–lithium exchange (C).
prepared by indium–silicon exchange or by the well-precedented
indium–lithium exchange, (b) the NMR signals seem to corre-
spond to just one compound, not a mixture, and (c) our previous
work3 and the amount of InCl3 used, we propose that the active
species is the tris(thienyl)indium.
In conclusion, we have devised a new strategy for the prepa-
ration of triorganoindium reagents, which is mild enough to
allow the presence of organolithium-sensitive functional groups.
Moreover, yields ranging from good to excellent have been
obtained although equimolar amounts of trialkoxysilane, TBAF
and aromatic halide were used, as opposed to Hiyama couplings
were 100% excess of trialkoxysilane and TBAF is usually em-
ployed. Further work to extend the variety of organosilanes
used as precursors, the functional group tolerance and the type
of organometallics available with this methodology is currently
underway.
Reactions, 2nd edn, ed. A. de Meijere and F. Diederich, Wiley-VCH,
Weinheim, 2004.
2 I. Pe´rez, J. Pe´rez Sestelo and L. A. Sarandeses, Org. Lett., 1999, 1, 1267–
1269; K. Takami, H. Yorimitsu, H. Shinokubo, S. Matsubara and K.
Oshima, Org. Lett., 2001, 3, 1997–1999; T. Hirashita, Y. Hayashi, K.
Mitsui and S. Araki, J. Org. Chem., 2003, 68, 1309–1313; L. Baker and
T. Minehan, J. Org. Chem., 2004, 69, 3957–3960; P. H. Lee, K. Lee
and Y. Kang, J. Am. Chem. Soc., 2006, 128, 1139–1146; M. Barbero,
S. Cadamuro, S. Dughera and C. Giaveno, Eur. J. Org. Chem., 2006,
4884–4890; M. A. Pena, J. Pe´rez Sestelo and L. A. Sarandeses, J. Org.
Chem., 2007, 72, 1271–1275.
3 E. Font-Sanchis, F. J. Ce´spedes-Guirao, A. Sastre-Santos and F.
Ferna´ndez-La´zaro, J. Org. Chem., 2007, 72, 3589–3591.
4 Other procedures for preparation of organoindium reagents have been
described. See, for example: T. Hirashita, H. Yamamura, M. Kawai
and S. Araki, Chem. Commun., 2001, 387–388; P. H. Lee, S.-Y. Sung
and K. Lee, Org. Lett., 2001, 3, 3201–3204; K. Takami, H. Yorimitsu
and K. Oshima, Org. Lett., 2002, 4, 2993–2995; K. Takami, S. Mikami,
H. Yorimitsu, H. Shinokubo and K. Oshima, J. Org. Chem., 2003,
68, 6627–6631; T. Hirashita, Y. Hayashi, K. Mitsui and S. Araki,
Tetrahedron Lett., 2004, 45, 3225–3228; P. H. Lee, E. Shim, K. Lee,
D. Seomoon and S. Kim, Bull. Korean Chem. Soc., 2005, 26, 157–160;
P. H. Lee and K. Lee, Angew. Chem., Int. Ed., 2005, 44, 3253–3256.
5 M. R. Pitts, J. R. Harrison and C. J. Moody, J. Chem. Soc., Perkin
Trans. 1, 2001, 955–977; J. Podlech and T. C. Maier, Synthesis, 2003,
633–655; Y.-H. Chen and P. Knochel, Angew. Chem., Int. Ed., 2008, 47,
7648–7651.
Acknowledgements
This work was partially supported by the Spanish Government-
MEC-CICYT (Consolider-Ingenio 2010 project HOPE
CSD2007–00007, and Grants No. SAF2003–08140-C02–01
and CTQ2007–67888/BQU), as well as Generalitat Valenciana
(ACOMP06/057).
6 See, for example: T. Hiyama, Metal-Catalyzed Cross-Coupling Re-
actions, ed. F. Diederich and P. J. Stang, Wiley-VCH, Weinheim,
1998, chapter 10; T. Hiyama, J. Organomet. Chem., 2002, 653, 58–
61; T. Hiyama and Y. Hatanaka, Pure Appl. Chem., 1994, 66, 1471–
1478.
7 See, for example: D. Domin, D. Benito-Garagorri, K. Mereiter, J.
Fro¨hlich and K. Kirchner, Organometallics, 2005, 24, 3957–3965; L.
Ackermann, C. J. Gschrei, A. Althammer and M. Riederer, Chem.
Commun., 2006, 1419–1421; D. Srimani, S. Sawoo and A. Sarkar, Org.
Lett., 2007, 9, 3639–3642.
Notes and references
1 Handbook of Organopalladium Chemistry for Organic Synthesis, ed.
E. Negishi, Wiley, New York, 2002; Metal-Catalyzed Cross-Coupling
2472 | Dalton Trans., 2009, 2470–2473
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
8 See, for example: M. E. Mowery and P. DeShong, Org. Lett., 1999, 1,
2137–2140; P. Pierrat, P. Gros and Y. Fort, Org. Lett., 2005, 7, 697–
700.
9 This compound, p-phenylbenzaldehyde, has been previously described.
See, for example: Y.-H. Cheng, C.-M. Weng and F.-E. Hong, Tetra-
hedron, 2007, 63, 12277–12285; C.-P. Chang, Y.-L. Huang and F.-E.
Hong, Tetrahedron, 2005, 61, 3835–3839.
10 This compound has been previously described. See, for example: W.
Solodenko, H. Wen, S. Leue, F. Stuhlmann, G. Sourkouni-Argirusi, G.
Jas, H. Scho¨nfeld, U. Kunz and A. Kirschning, Eur. J. Org. Chem.,
2004, 3601–3610; S. Kotha, D. Kashinath, K. Lahiri and R. B. Sunoj,
Eur. J. Org. Chem., 2004, 4003–4013.
11 See, for example: T. Saeki, E.-C. Son and K. Tamao, Org. Lett., 2004,
6, 617–619.
This journal is
The Royal Society of Chemistry 2009
Dalton Trans., 2009, 2470–2473 | 2473
©