Palladium-catalyzed carbon-carbon coupling reactions using aryl Grignards

Article abstract of DOI:10.1016/S0040-4039(02)01538-1

Coupling reactions using Pd(PPh₃)₄ were investigated with a number of electron donating and electron withdrawing substituents. High yields were obtained with both types of substituents. In competitive reactions the electron-withdrawi

Full text of DOI:10.1016/S0040-4039(02)01538-1

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 43 (2002) 7091–7094
Palladium-catalyzed carbonꢀcarbon coupling reactions using
aryl Grignards
Christine Gottardo* and Andrea Aguirre
Department of Chemistry, Lakehead University, Thunder Bay, Ont., Canada P7B 5E1
Received 28 June 2002; accepted 22 July 2002
Abstract—Coupling reactions using Pd(PPh3)4 were investigated with a number of electron donating and electron withdrawing
substituents. High yields were obtained with both types of substituents. In competitive reactions the electron-withdrawing group
-NO2 reacted preferentially over the electron donating groups. When the starting halides were converted to Grignard reagents,
high yields were obtained for some hindered electron-withdrawing groups. © 2002 Elsevier Science Ltd. All rights reserved.
Preparations of aryl alkynes have been described in the
literature using the Sonogashira coupling reaction.1
These coupling reactions have been used in natural
product synthesis2 and materials chemistry.3 As these
types of reactions have become an integral part of the
synthetic toolbox, and the number of reports of their
use have increased,2,3 variations in reaction conditions
have also been introduced in the literature. The initial
report of Sonogashira et al. with the palladium catalyst1
described the requirement for a base and cuprous
iodide. The nature of the base,4 the solvent5 and the
palladium catalyst have been investigated, with changes
in these resulting in improved yields and decreased
reaction times.
completed with Pd(PPh3)4 as the catalyst using identical
reaction conditions, i.e. the reactions have not been
optimized. The compounds investigated were chosen
due to the availability of all three isomers, with the
exception of the ethyliodobenzenes.
For most substituents, substitution in the para position
provided the highest yield. The reactivity pattern that
Stephens and Castro6 first noted for electron withdraw-
ing versus electron rich groups is not evident from the
yields for the trifluoromethyl and ethyl groups, with
exceptional yields in both cases. To determine if the
electron withdrawing substituents were indeed the most
reactive, competitive reactions were run between 4-
iodonitrobenzene and both 4-iodotoluene and 4-
iodoanisole.8 The competitive reaction of toluene and
nitrobenzene was monitored by GC; it was found that
after 1 h there was no remaining 4-iodonitrobenzene,
and no disappearance of 4-iodotoluene. Over a pro-
longed reaction time (2 h) the corresponding coupled
nitro-product was obtained with no evidence of the
coupled toluene product. The competitive reaction
between 4-iodonitrobenzene and 4-iodoanisole resulted
in a similar pattern of reactivity with the disappearance
of the 4-iodonitrobenzene and no reaction of the substi-
tuted anisole. The reactivity pattern originally described
by Castro and Stephens6 is also seen in these palladium-
catalyzed, cross-coupling reactions.
Previous cross-coupling reactions reported by Stephens
and Castro,6 using cuprous acetylides, indicate that
electron withdrawing groups in the para position
increase the ease of substitution relative to electron-
donating groups in the following manner: p-nitro>H>
p-methoxy. Subsequent reports that have described
improved yields, focused on electron withdrawing sub-
stituents such as carboxyaldehydes, nitro groups and
esters.5,7
We prepared a number of substituted trimethyl silyl
ethynyl benzenes using Sonogashira coupling (Table 1,
Method A), and then compared these results to a novel
method of coupling aryl iodides to terminal alkynes
developed in our laboratory. All of the reactions were
Many of the carbonꢀcarbon bond forming reactions
that use a palladium catalyst, such as Stille coupling9 or
the Heck reaction of aryl mercury compounds,10 result-
ing in the formation of an alkenyl aryl compound,
involve a transmetallation reaction. This type of
transmetallation has not been extended to the forma-
Keywords: Sonogashira; coupling; Grignard reagent.
* Corresponding author. Tel.: (807) 343-8466; fax: (807) 346-7775;
e-mail: cgottard@gale.lakeheadu.ca
0040-4039/02/$ - see front matter © 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(02)01538-1

7092
C. Gottardo, A. Aguirre / Tetrahedron Letters 43 (2002) 7091–7094
Table 1. Palladium-catalyzed cross-coupling reactions
25) interferes with Grignard reactions, it was antici-
pated that only 1 would form (Scheme 1). When the
reaction was attempted using methyl 4-iodobenzoate, a
trace amount of 1 was obtained and a 43% yield of 2
was also obtained.
Entry
R
Yield (%)a
3-
2-
4-
1
2
3
4
5
6
7
8
CF3
Ab
Bb
Ac
Bc
Ac
Bc
Ac
Bc
Ab
Bb
Ac
Bc
Ac
Bc
Ac
Bc
\99
18
73
9
90
7
37
94
\99
75
91
59
74
59
61
56
\99
88
87
65
62
97
30
92
–
\99
97
89
43
94
41
54
69
\99
75
78
20
99
20
42
38
This surprising result was further investigated to deter-
mine the effects of different substituents (Table 1,
Method B) and the reaction mechanism. With the
electron-donating methyl group, significantly higher
yields were obtained than those from the corresponding
reaction using standard Sonogashira conditions (Table
1, entry 4). In particular, neither the steric nor elec-
tronic effects that result from the presence of the ortho-
methyl group seem to hinder reactivity in this Grignard
modified reaction, with a high yield being obtained in
comparison to the standard Sonogashira conditions.
This high yield is in direct contrast with many reported
yields, using different starting reagents and alkynes.
For example, the preparation of trimethyl [(2-
methylphenyl)ethynyl]silane from the aryl cuprate and
haloacetylene proceeded in a 48% yield.11 A recent
report using 2-bromotoluene does show the preparation
of 2-tolylalkynes in high yields (94% and 84%) using a
catalyst optimized for aryl bromides.12 Our methodol-
ogy shares the same mild (room temperature) condi-
tions reported therein.12 Another surprising aspect of
our investigations was that functional groups which
usually interfere with Grignard reagents (NO2, OH,
CO2CH3) still resulted in the formation of product,
albeit at lower yields than those obtained for the stan-
dard conditions.
CO2Me
NO2
CH3
CH2CH3
OH
–
87
49
48
49
65
73
OCH3
NH2
a Reaction conditions: Method A: Pd(PPh3)4 (46 mg, 0.04 mmol), CuI
(15 mg, 0.08 mmol) and Et3N (303 mg, 3.0 mmol) were combined in
THF (5.0 mL) under a N2 atmosphere. Substituted iodobenzene (2.0
mmol) was added and the reaction mixture was cooled to 0°C. After
stirring for 10 min, trimethylsilylacetylene (206 mg, 2.1 mmol) was
added dropwise over 30 min. The reaction mixture stirred at room
temperature overnight and was filtered through Celite to remove Pd
and Cu catalysts; Method B: Magnesium (2.0 mmol), the substi-
tuted iodide (2.0 mmol) and I2 (catalytic) were combined in dry
THF (4.0 mL) and heated to 40°C overnight to generate the
Grignard. Pd(PPh3)4 (46 mg, 0.04 mmol), CuI (15 mg, 0.08 mmol)
and Et3N (0.42 mL, 3.0 mmol) were added. After stirring for 10
min, TMS–acetylene (0.30 mL, 2.0 mmol) was added dropwise over
10 min. The reaction mixture was reacted for 4 h, under N2,
quenched with water, saturated NaCl and dried over MgSO4.
b Isolated yields from an average of two runs.
A series of reactions using 2-iodotoluene was run to
determine which reagents were necessary for the cou-
pling reaction to proceed. In the absence of Pd(PPh3)4
the coupling does not proceed.13 In the absence of the
CuI co-catalyst the reaction also does not proceed.14
With neither catalyst present the reaction does not
proceed. When Mg, Pd(PPh3)4 and CuI are added
simultaneously to the reaction mixture, the yield of
trimethyl[(2-methylphenyl)ethynyl]silane decreased with
respect to that obtained under standard Sonogashira
conditions. It has been postulated that Grignard
c Yield obtained by GC from an average of two or more runs.
tion of alkynyl aryl compounds and we were interested
in investigating this type of chemistry. We chose to
convert the starting iodides to Grignard reagents,
before addition of the palladium and copper catalysts
and alkyne, and compared these coupling reactions to
our standard conditions. Since it is known that the
acidic nature of acetylenic hydrogens (pKa’s from 20–
Pd(PPh3)4
CuI / Et3N / THF
MgI
I
Mg
TMS
CO2Me
CO2Me
TMS
+
CO2Me
CO2Me
1
2
Scheme 1.

C. Gottardo, A. Aguirre / Tetrahedron Letters 43 (2002) 7091–7094
7093
R
TMS
2PPh3
Pd(PPh3)4
R
TMS
IMg
R
(Ph3P)2Pd
Mg + 2 PPh3
CuX
R
Cu
(Ph3P)2Pd
X
TMS
[Et3NH]+-I
3
Et3N / CuI
TMS
Figure 1.
and that functional groups such as the nitro-substituent
still give high yields of the coupled compound. This
modified Grignard–Sonogashira reaction provides a
facile and moderate method for the formation of car-
bonꢀcarbon bonds.
reagents have radical behavior but the presence of a
radical trap (TEMPO) did not interfere in the progress
of the reaction.
We propose that our coupling reactions are indeed
undergoing a transmetallation reaction (Fig. 1). The
more reactive carbonꢀmagnesium bond, relative to the
carbonꢀhalogen bond, can exchange with palladium.
This carbonꢀmagnesium bond shows a higher degree of
polarization that assists with the Pd insertion. The
insertion of Pd0 occurs with an oxidation state change
to Pd2+ and we propose that this oxidation is coupled
to a reduction of the Mg2+ to Mg0. In 31P NMR
studies15 we observed identical intermediates in both
the standard Sonogashira reaction and our Grignard
modified reaction. An observed 31P NMR signal (rela-
tive to H3PO4 as the internal standard) at 23.8 ppm is
indicative of a trans-ArPdX(PPh3)2 species (3).16 Due to
the aqueous work-up in the reaction we have been
unable to determine the nature of the magnesium by-
product in the reaction, to date. We are continuing to
investigate the mechanism of this Mg–Pd exchange.
Acknowledgements
Financial support by the Natural Sciences and Engi-
neering Research Council of Canada is gratefully
acknowledged.
References
1. Sonogashira, K.; Tohda, Y.; Hagihara, K. Tetrahedron
Lett. 1975, 16, 4467–4470.
2. (a) Nicolaou, K. C.; Dai, W.-M. Angew. Chem., Int. Ed.
Engl. 1991, 30, 1387–1416; (b) Maier, M. E.; Boße, F.;
Niestroj, A. J. Eur. J. Org. Chem. 1999, 1–13.
3. Ravikanth, M.; Strachan, J.-P.; Feirong, L.; Lindsey, J.
S. Tetrahedron 1998, 54, 7721–7734 and references cited
therein.
4. McGaffin, G.; de Meijere, A. Synthesis 1994, 583–591
and references cited therein.
5. Thorand, S.; Krause, N. J. Org. Chem. 1998, 63, 8551–
8553.
In summary, we have demonstrated a modified set of
Pd-coupling reaction conditions that use Grignard
reagents as the source of the aryl compounds that react
with an alkyne to give coupled products in good to
exceptional yields. It is particularly significant to note
that the mild reaction conditions do not result in the
alkyne reacting with the Grignard to give the acetylide

7094
C. Gottardo, A. Aguirre / Tetrahedron Letters 43 (2002) 7091–7094
6. Stephens, R. D.; Castro, C. E. J. Org. Chem. 1963, 28,
Chem. Soc. 1996, 118, 1215–1216.
3313–3315.
14. Stille coupling proceeds in the absence of a Cu+ catalyst.
See Ref. 11 and Scott, W. J.; Stille, J. K. J. Am. Chem. Soc.
1986, 108, 3033–3040.
7. Takashi, S.; Kuroyama, Y.; Sonogashira, K.; Hagihara, N.
Synthesis 1980, 627–630.
15. Two reactions were compared: (a) Pd(PPh3)4 (46 mg, 0.04
mmol) and 2-iodotoluene (5 mL, 0.04 mmol) were combined
in dry THF (2.5 mL) and stirred overnight; (b) 2-Iodo-
toluene (10 mL, 0.08 mmol) and Mg (40 mg, 1.6 mmol) were
combined in dry THF (2.5 mL) and stirred until the
Grignard formed. The Grignard reagent was transferred via
cannula to a flask containing Pd(PPh3)4 (46 mg, 0.04 mmol)
and the reaction stirred overnight. Work-up for both
reactions: The reaction mixture was filtered through Celite,
the solvent was removed in vacuo and NMR spectra were
obtained. 31P NMR (C6D6): 23.8, −4.4 ppm.
8. Reaction conditions similar to Table 1 but both 4-
iodonitrobenzene (2.0 mmol) and 4-iodoanisole or 4-iodo-
toluene (2.0 mmol) were added. The progress of the reaction
was monitored by GC, with sample injections occurring
every 30 min.
9. Scott, W. J.; Crisp, G. T.; Stille, J. K. J. Am. Chem. Soc.
1984, 106, 4630–4632.
10. Heck, R. F. J. Am. Chem. Soc. 1968, 90, 5518–5531.
11. Oliver, R.; Walton, D. R. M. Tetrahedron Lett. 1972, 13,
5209–5212.
12. Hundertmark, T.; Littke, A. F.; Buchwald, S. L.; Fu, G.
C. Org. Lett. 2000, 2, 1729–1732.
13. The coupling of alkenyltrialkylstannanes in the absence of
16. Verkade, J. G.; Quin, L. D. Phosphorus-31 NMR Spec-
troscopy in Stereochemical Analysis, Methods in Stereo-
chemical Analysis; VCH: Weinheim, 1987; p. 509.
Pd has been reported. See: Piers, E.; Romero, M. A. J. Am.