Air-stable, phosphine-free anionic palladacyclopentadienyl catalysts: Remarkable halide and pseudohalide effects in Stille coupling

Article abstract of DOI:10.1002/adsc.200505325

The Stille cross-coupling of allylic and benzyl bromides is shown to proceed efficiently using phosphine-free dinuclear anionic palladacyclopentadienyl catalysts possessing bridging (N,O)-imidate ligands. The type of bridging anion influences the catalytic activity considerably. Halide anions such as chloride, bromide or iodide also influence the catalytic activity but to a far lesser extent than the pseudohalide imidate anions (from succinimide or phthalimide). A Baldwin-type cooperative effect is seen with 7a using CuI as a co-catalyst, in the presence of two equivalents of CsF in DMF at 40°C. In toluene, these additives slow down substrate turnover. Wiley-VCH Verlag GmbH & Co. KGaA.

Full text of DOI:10.1002/adsc.200505325

COMMUNICATIONS
DOI: 10.1002/adsc.200505325
Air-Stable, Phosphine-Free Anionic Palladacyclopentadienyl
Catalysts: Remarkable Halide and Pseudohalide Effects in Stille
Coupling
Catherine M. Crawforth,a Ian J. S. Fairlamb,a,* Anant R. Kapdi,a
b
a
c
´
Jose Luis Serrano, Richard J. K. Taylor, Gregorio Sanchez
a
Department of Chemistry, University of York, Heslington, York, YO10 5DD, U.K.
Fax: (þ44)-1904-432-515, e-mail: ijsf1@york.ac.uk
b
´
´
´
´
´
Departamento de Ingeniería Minera, Geologica y Cartografica, Area de Química Inorganica, Universidad Politecnica de
Cartagena, 30203 Cartagena, Spain
E-mail: jose.serrano@upct.es
c
´
Departamento de Química Inorganica, Universidad de Murcia, 30071 Murcia, Spain
E-mail: gsg@um.es
Received: August 22, 2005; Accepted: December 30, 2005
Abstract: The Stille cross-coupling of allylic and ben-
zyl bromides is shown to proceed efficiently using
phosphine-free dinuclear anionic palladacyclopenta-
dienyl catalysts possessing bridging (N,O)-imidate li-
gands. The type of bridging anion influences the cat-
alytic activity considerably. Halide anions such as
chloride, bromide or iodide also influence the cata-
lyticactivity but to a far lesser extent than the pseu-
dohalide imidate anions (from succinimide or phthal-
imide). A Baldwin-type cooperative effect is seen
with 7a using CuI as a co-catalyst, in the presence
of two equivalents of CsF in DMF at 408C. In tol-
uene, these additives slow down substrate turnover.
the catalytically active species – the latter species con-
tains a halide ligand.[7]
Generally, the manipulation of the metal halide func-
tionality is a highly valuable method for tuning the reac-
tivity of transition metal complexes (in catalysts or in
stoichiometric processes).[8] Importantly, alteration of
the halide does not usually perturb the reaction system
such that catalysis inhibition is seen, but more often pro-
motes changes in reactivity and/or selectivity. The steric
and electronic effects of the specific type of halide will
delicately alter the electronic properties of other ligands
bound to the metal, thereby producing changes in metal
reactivity. Halides can also participate in bonding with a
metal centre, e.g., Pd(0) or neutrÀal Pd(II), generating
À
[9]
À
À
À
Keywords: C C bond formation; copper(I) salts;
anionic species such as “X Pd(0) ” or “X Pd(II) ”,
cross-coupling; halide effects; palladium
which ultimately increases the nucleophilicity of the
metal centre, but potentially offers the opportunity to
influence the metal reactivity through changes in halide
(X).
As the nucleofuge concentration increases during the
course of a cross-coupling reaction, one predicts that
Palladium-catalysed cross-coupling processes have anionic Pd(0) adducts will contain halides derived
solved numerous syntheticproblems, notably in natural
from the organohalide substrate. However, sub-stoi-
product synthesis,[1] pharmaceutical targets,[2] fluores- chiometric quantities of halide and pseudohalide addi-
cent DNA base analogues and biological probes.[3] The tives (as low as 10 mol %) often affect catalyst activity
À
À
C C and C X (X¼heteroatom) bond-forming technol- and lifetime in these reactions – the degree of halide
ogies, such as Buchwald–Hartwig, Heck, Kumada, Ne- binding to Pd therefore depends on the type of organo-
gishi, Sonogashira, Suzuki and Stille, represent just a halide substrate employed and any additives used.[10]
few of the key transformations widely used.[4] In recent
The influence of imidate-type ligands, derived from
years, broader substrate scope has become accessible succinimide (succ), phthalimide (phal) and maleimide
through the development of highly active Pd catalysts,[5] (mal), and their influence in cross-coupling processes
particularly for aryl chlorides, where palladacycles have has been of recent interest to us, particularly in Stille
been studied extensively (as precatalysts).[6] The palla- coupling.[11] The discovery of the useful precatalyst cis-
dacyclic unit is believed to degrade under the reaction bis(triphenylphosphine)palladium(II)(bromo)succini-
conditions to slowly release neutral or anionic Pd(0) as mide (cis-1, Figure 1), for Stille coupling of allylic and
Adv. Synth. Catal. 2006, 348, 405 – 412
ꢀ
Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
405

Catherine M. Crawforth et al.
COMMUNICATIONS
Figure 1. Neutral and anionic palladium(II) complexes containing imidate ligands.
benzylicbromides with organostannanes has led to the
action requires preservation of the geometry and posi-
finding that the succinimide ligand, a rather unusual tion of the double bond. Organostannane Z-9 is also rel-
pseudohalide with a subtle blend of s-donating and p- atively deactivated.[12d]
accepting ability, imparts interesting selectivity for ben-
Complexes 4–7 were tested using standard conditions
zylic substrates over activated aryl bromides, when com- developed previously[12e] by our group (toluene,
pared to common Pd(0)/(II) catalysts/precatalysts.[12]
Second generation precatalysts have been synthesised
and evaluated in Stille (2),[13] Suzuki and Sonogashira re-
actions (3–6).[14] Differences in reactivity were observed
against small changes in structure of the Pd/L assembly.
Overall, it appears that the ligand backbone plays a ka-
mikaze role in initiating the active palladium catalyst.
Our first studies in this specific report are focused to-
ward the application of phosphine-free neutral and
anionic dinuclear Pd(II) imidate complexes (4–7) as
catalysts/precatalysts for Stille coupling to study further
halide and pseudohalides effects.
1 mol % Pd, 608C – referred to as method A) (Table 1).
The reaction of benzyl bromide 8 with cis-vinylstannyl
ester Z-9a to give Z-10a represents the benchmark for
initial catalyst screening (Scheme 1).[12] Crucially, the re- Scheme 1. Stille coupling mediated by cis-1.
406
asc.wiley-vch.de
ꢀ
Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2006, 348, 405 – 412

Halide and Pseudohalide Effects in Stille Coupling
COMMUNICATIONS
Table 1. Catalyst precursor screening for Stille coupling of
8þZ-9a!Z-10a.[a]
All reactions were analysed after 0.75 hours and then
followed to completion. With catalyst cis-1, 48% conver-
sion in Z-10a was observed (entry 1). The reaction is es-
sentially quantitative after 3 hours. For the palladacyclic
dimers containing an azobenzene backbone (4a–c), low
conversions were observed after 0.75 hours, although
reasonable yields resulted after 18 hours (entries 2–4).
The yields improved for the 2-phenylpyridine-derived
palladacyclic dimers (5a–c) (entries 5–7). Notable dif-
ferences were seen between the imide ligands after 18
hours (for 5a, 86%; 5b, 73%; 5c, 56%). Alteration of
the palladacyclic backbone to a phenanthroline im-
proved the yields after 0.75 hours (entries 8–10), show-
ing similar catalytic activity to 1. Turning our attention to
Entry
[Pd] precursor
catalyst
Reaction time
[hours]
Yield
[%][b]
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
1
4a
4b
4c
5a
5b
5c
6a
6b
6c
7a
7b
0.75 (3)
0.75 (18)
0.75 (18)
0.75 (18)
0.75 (18)
0.75 (18)
0.75 (18)
0.75 (18)
0.75 (18)
0.75 (18)
0.75 (3)
0.75 (3)
0.75 (3)
0.75 (3)
0.75 (3)
0.75 (3)
0.75 (3)
48 (98)
13 (57)
18 (64)
6 (31)
23 (86)
18 (73)
21 (56)
45 (85)
53 (81)
38 (67)
65 (99)
77 (99)
48 (94)
9 (49)
the anionicpalladayclcopentadienyl ocmplexes (
7a–c),
significantly higher catalytic activity was observed
(entries 11–13). For example, for the succinimide com-
plex 7a, 65% conversion was observed after 0.75 hours,
and 99% after 3 hours (entry 11). The phthalimide com-
plex 7b, showed higher activity (77% after 0.75 hours
and 99% after 3 hours) (entry 12). The yield dropped
for the maleimide complex 7c (entry 13). At this stage
we were intrigued by the differences observed on chang-
ing the pseudohalide from succinimide, through to
phthalimide and maleimide. Co-additives, in the form
of soluble organichalides, were added to 7a to assess
their affect on catalytic activity (entries 14–17). We
were surprised to find that TBAF (0.05 equivs.) had a
dramatic affect causing the conversion to drop from
65% to 9% after 0.75 hours (entry 14).[15] Conversion
was better for TBAC, TBAB and TBAI, but essentially
catalysis was slowed by addition of the halide co-addi-
tives –fluoride retarded catalysis the most, with bromide
having the minimum effect of the halides studied.
Interested by the effect of the halide co-additives, the
m2-hydroxy- (7d) and m2-halo-anionic palladacyclopen-
tadienyl complexes (7e–g, Figure 2) were evaluated in
the benchmark reaction (Table 2).
7c
7aþTBAF[c]
7aþTBAC[c]
7aþTBAB[c]
7aþTBAI[c]
22 (74)
38 (87)
29 (80)
[a]
Reaction conditions as for Scheme 1; The Pd catalyst
loading¼1 mol % Pd (0.01 equiv.).
[b]
[c]
Numbers in brackets are yields after KF work-up and col-
umn chromatography.
Tetrabutylammonium salt added (0.05 equivs.); TBAF¼
tetrabutylammonium fluoride; TBAC¼tetrabutylammo-
nium chloride; TBAB¼tetrabutylammonium bromide;
TBAI¼tetrabutylammonium iodide.
Figure 2. Anionic palladium(II) complexes containing m2-hy-
droxy- and m2-halide bridging ligands.
The hydroxy complex 7d showed poor catalytic activ-
ity (entry 1). Product conversion was higher for the
chloro, bromo and iodo complexes (7e–g) relative to
7d, with the bromide showing the best catalytic activity
(entry 3). However, these yields are notably lower
than the succinimide and phthalimide relatives (7a and
7b). This series of results indicates the importance asso-
ciated with the specific type of halide/pseudohalide for
catalytic activity.
In screening various organostannanes for the reaction
we determined that the reactivity of Z-9 was greatly
affected by the size of the ester group. For example, in
comparison with the ethyl ester in Z-9a, the methyl ester
(Z-9b) reacted more quickly (Table 3).
Table 2. Effect of m2-hydroxy- and m2-halide-bridging ligands
in Stille coupling of 8þZ-9a!Z-10a.[a]
Entry
[Pd] precursor
catalyst
Reaction time
[hours]
Conversion
[%][b]
1
2
3
4
7d
7e
7f
0.75 (3)
0.75 (3)
0.75 (3)
0.75 (3)
15 (33)
34 (78)
42 (85)
31 (69)
7g
[a]
Reaction conditions as for Scheme 1 and Table 1.
Conversion calculated by GC analysis against an internal
standard (dodecane).
[b]
In this series, the succinimide (7a), phthalimide (7b),
chloro (7e), bromo (7f) and iodo (7g) functions again
showed significant differences in reactivity (entry 1).
The reaction of 8 with phenylstannane 11 to give di-
In stark contrast, the i-propyl ester (Z-9c) was even phenylmethane 12 was evaluated next (Scheme 2 and
less reactive than Z-9a and Z-9b
Table 4).
Adv. Synth. Catal. 2006, 348, 405 – 412
ꢀ
Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
407

Catherine M. Crawforth et al.
COMMUNICATIONS
Table 3. Substituent effect in the vinylstannyl ester (Z-9).[a]
Table 5. Cooperative effect with CsF and CuI in the reaction
of 8þZ-9a!Z-10a.[a]
¼
Entry Bu3SnCH CHCO2R Reaction time Conversion
[hours]
[%][b]
Entry [Pd] precursor
catalyst
Reaction time Conversion
[hours]
[%][b]
1
R¼Me (Z-9b)
7a
0.75 (3)
0.75 (3)
0.75 (3)
0.75 (3)
0.75 (3)
85 (99)
93 (99)
59 (88)
71 (99)
62 (94)
1
2
3
4
5
6
7
8
9
7a
0.75 (3)
0.75 (3)
0.75 (3)
0.75 (3)
0 (10)
10 (48)
2 (12)
89 (99)
62 (99)
5 (27)
0 (8)
7b
7e
7f
7 g
7aþCuI[c]
7aþCsF[d]
7aþCuIþCsF[e]
Pd(PPh3)4 þCuIþCsF[f] 0.75 (3)
2
R¼i-Pr (Z-9c)
7aþCuI[g]
0.75 (3)
0.75 (3)
0.75 (3)
0.75 (3)
7a
7b
0.75 (3)
0.75 (3)
40 (84)
34 (91)
7aþCsF[h]
7aþCuIþCsF[i]
8 (33)
17 (57)
7aþCuIþCsF[j]
[a]
[b]
Reaction conditions as for Scheme 1 and Table 1.
Conversion calculated by GC analysis against an internal
standard (dodecane).
[a]
As for Scheme 1, using DMF as the solvent, 5 mol % Pd,
408C.
Conversion calculated by GC analysis against an internal
standard (dodecane).
10 mol % CuI added.
2 equivalents of CsF added.
Both CuI (10 mol %) and CsF (2 equivs.) added.
Identical catalyst system and conditions employed by
Baldwin and co-workers.
[b]
[c]
[d]
[e]
[f]
[g]
[h]
Reaction conditions as for Scheme 1 (Method A); The Pd
catalyst loading¼1 mol % Pd (0.01 equiv.). CuI was ad-
ded (0.05 equivs.).
Scheme 2. Stille coupling of benzyl bromide 8 with phenyl-
stannane 11 to give 12.
As for [g], using catalytic CsF (0.05 equivs.) (without added
CuI).
[i]
[j]
As for [g], using CuI (0.05 equivs.) and CsF (0.05 equivs.).
As for [g], CuI (0.05 equivs.) and CsF (2 equivs.).
Table 4. Reaction of 8 with PhSnBu3 11 to give 12.[a]
Entry
[Pd] precursor
catalyst
Reaction time
[hours]
Conversion
[%][b]
1
2
3
4
5
6
7a
7b
7c
7e
7f
6 (12)
6 (12)
6 (12)
6 (12)
6 (12)
6 (12)
37 (81)
44 (93)
32 (78)
14 (41)
31 (67)
24 (62)
7g
Scheme 3. Stille coupling of 3-nitrobenzyl bromide 13 with
phenylstannane 11 to give 14.
[a]
Reaction conditions as for Scheme 1 and Table 1.
Conversion calculated by GC analysis against an internal
standard (dodecane).
[b]
In the absence of CuI and CsF, complex 7a (5 mol %
Pd) was a poor catalyst in DMF (entry 1). Lower reac-
The reaction was monitored after 6 and 12 hours. The tion concentrations gave similar results – ruling out the
best yields were observed for 7a and 7b (entries 1 and 2). formation of Pd clusters. Sluggish substrate turnover
Of the halide complexes, the poorest activity was seen was seen with either CuI or CsF (entries 2 and 3). How-
for 7e (entry 4) and the best for 7f (entry 5).
ever, the cooperative effect clearly exists when 7a, CuI
We were interested in evaluating whether cooperative and CsF were used in tandem (entry 4). For comparison
effects with CuI and CsF would be observed in DMF at purposes, a yield of 89% conversion was attained for 7a
408C (Method B). In a recent study, Baldwin and co- after 0.75 hours, whereas using Pd(PPh3)4, 62% conver-
workers have revealed a remarkable cooperative effect sion was seen after the same time (entries 4 and 5). The
between CuI and CsF, when using Pd(PPh3)4 (5 mol %) effect of a Cu(I) co-catalyst was also assessed using
as the catalyst, which was most striking when employing Method A, which resulted in negligible conversion (en-
DMFas the solvent in Stille coupling.[16] The benchmark try 6). The addition of CsF (0.05 equivs.) also resulted in
reaction (8þZ-9a!Z-10a) was used to probe whether very low conversion (entry 7). Perhaps unsurprisingly,
the cooperative effect was operative for the phosphine- when CuI and CsF were both added low conversions
free complex 7a using method B (Table 5).
were also seen (entries 8 and 9).
408
asc.wiley-vch.de
ꢀ
Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2006, 348, 405 – 412

Halide and Pseudohalide Effects in Stille Coupling
COMMUNICATIONS
Table 6. Other examples.[a]
Table 7. Products from the Stille coupling of allylic or benzyl-
icsubstrates with organostannanes, mediated by 7a (using
Method A).[a]
Entry
[Pd] precursor
catalyst
Reaction time
[hours]
Conversion
[%][b]
1
2
3
4
5
6
7a[c]
7b[c]
7a[d]
7b[d]
7a[d]
7b[e]
3 (6)
3 (6)
2 (4)
2 (4)
3 (6)
3 (6)
50 (97)
66 (99)
85 (99)
89 (99)
32 (81)
37 (84)
Entry
1
Coupled Product
Yield [%][b]
95
[a]
As for Scheme 3.
Conversion calculated by GC analysis against an internal
standard (dodecane).
Using Method A with no additional additives (1 mol%
Pd).
Using Method B.
[b]
[c]
[d]
[e]
2
3
82
75
Using Method B, but using 1 mol% Pd.
We also tested both Methods A and B for the reaction
of 3-nitrobenzylbromide 13 with 11 to give 14 (Scheme 3
and Table 6).
The conversions for 7a and 7b, using Method A with
no additional additives, were 50% and 66% after 3
hours, and essentially quantitative conversion after 6
hours (entries 1 and 2). Conversions were higher for
both 7a and 7b using Method B, where again quantita-
tive conversions were seen (entries 3 and 4). For lower
Pd catalyst loadings, for direct comparison with Method
A, inferior catalyst activity was noted (entries 5 and 6).
It is clear from the above results that 7a and 7b are the
best catalysts (using Method A). A library of substrates
were screened using 7a as the catalyst (similar yields are
seen for 7b for selected examples given in Table 7).
In this series of compounds, good to excellent yields
were observed. In all cases, the alkene regio- and stereo-
selectivity were preserved from the vinylstannyl starting
material. The yields for these products are higher, and
the Pd catalyst concentrations lower, than the best pre-
viously reported results for these reactions.[12c,12d]
4
5
6
90
81
92 (85)[c]
7
86 (81)[c]
[a]
Reaction
conditions:
allylic/benzylic bromide
(0.25 mmol), organostannane (0.3 mmol), 7a (0.5 mol %;
1 mol % Pd), C6H5CH3 (2.5 mL) at 608C, under an inert
atmosphere of N2, 6 hours.
Yields after KF work-up and column chromatography.
Yields in brackets are using 7b instead of 7a.
[b]
[c]
A series of other substrates was screened with 7a using tions mediated by 7a or related complexes, remain un-
Method B (Table 8). Again, high yields were observed known. The dominant pathway cannot be confirmed at
throughout the series of examples (entries 1–7). In en- this stage. However, the pronounced halide/pseudoha-
tries 3 and 7, 7a compares favourably against Pd(PPh3)4. lides effect detailed vide supra undoubtedly indicates a
A mechanism accounting for the CuI/CsF cooperative role for anionic ligands in the catalytic cycle. Alteration
effect in reactions mediated by 7a (X’¼succinimide) in of the halide/pseudohalide ligand (as in 7b–7g), and
DMF is tentatively suggested in Scheme 4 (as a modifi- their affect on catalytic activity with respect to the coop-
cation to that previously postulated for Pd(PPh3)4).[16] erativity of catalytic Cu(I) and stoichiometric CsF, will
We can assume the in situ generation of an organocup- form part of a more detailed study that will be reported
rate in DMF, which is expected to be more reactive in due course.
than the organostannane in the transmetallation step
Overall, we have identified that dinuclear anionic pal-
of the catalytic cycle. Two related cycles are presented: ladacyclopentadiene complexes 7a–c are active cata-
(1) a neutral catalytic cycle (Pathway A) and (2) an lysts for Stille coupling. They represent the first anionic
anionic catalytic cycle (Pathway B). Given that we Pd(II) catalysts that do not possess a donor ligand, e.g.,
have not detected any side-products that would result PR3 or NR3. Of particular note was the finding that re-
from degradation of the palladacyclopentadienyl unit markable halide and pseudohalide effects were ob-
via reaction with the organostannane, the oxidation served with these complexes, indicating that the anionic
states (Y) of the various Pd intermediates, in the reac- ligand either stabilises or enhances the reactivity of the
Adv. Synth. Catal. 2006, 348, 405 – 412
ꢀ
Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
409

Catherine M. Crawforth et al.
COMMUNICATIONS
nuclear species.[13] Baldwin-type cooperative effects
(CuI and CsF)[16] were observed in reactions conducted
in DMF, but not in toluene. In the latter solvent, the ad-
ditives reduce the rate of catalysis. It should be noted
that an increase in catalyst concentration is required
for high catalytic activity using method B [for both 7a
and Pd(PPh3)4] and that it is more cumbersome to re-
move DMF than toluene. Interestingly, the results using
7a and Method B, indicate that the role of Cu(I) is to ac-
tivate the organostannane and not to act as a donor li-
gand scavenger (given the absence of donor ligand in
7a), although one should not rule out the formation of
bimetalliccatalyst intermediates. Future studies will fo-
cus on identifying the active catalyst species in these re-
actions.
Table 8. Selected other Stille couplings of organobromides
with organostannanes mediated by 7a in combination with
CuI (10 mol%) and CsF (2 equivs.) (using Method B).[a]
Entry
1
Coupled Product
Conversion [%][b]
99 (2 h)
2
3
93 (3 h)
Experimental Section
94 (4 h);[c] 97 (2 h)[d]
General Remarks
See reference for general experimental information.[12d] Com-
plexes 4–6[14] and 7a–c[13] have been previously prepared and
characterized. Pd(PPh3)4 was freshly prepared from the reac-
tion of Pd(PPh3)2Cl2 in DMSO with hydrazine at 1208C (the
catalytic activity of commercial material was generally poor
when compared to freshly prepared material, which was stored
under argon).
4
5
6
74 (2 h)
97 (8 h)
98 (6 h)
Stille Coupling Procedure (Method A)[12]
Following a similar procedure to that previously reported by our
group:[12d] To a solution of the organohalide (0.25 mmol, 1
equiv.) and the organostannane (0.3 mmol, 1.2 equivs.) in dry
degassed (freeze-pump-thaw cycles) toluene (2.5 mL) and do-
decane (0.1 equiv.; internal standard) was added the palladium
catalyst (0.0025 mmol, 0.01 equiv.). The mixture was placed un-
der a dry N2 atmosphere and heated to 608C in the dark (the
flask was covered with domestic foil) for the time specified in
the text. All reactions were monitored by TLC, GC or GC/
MS analysis [GC samples were quenched by addition of a
1 M solution of HCl (0.25 mL) to the analysis sample
(0.1 mL), then extracted into EtOAc (2ꢀ0.5 mL) and passed
through a plug of silica ca. 1 g; samples were then analyzed di-
rectly]. On completion, the reaction was cooled to ambient
temperature, then saturated aqueous KF (2.5 mL) added and
the mixture stirred vigorously for 1 hour. The mixture was fil-
tered through Celiteꢁ, and the residue rinsed with Et2O (2ꢀ
5 mL), washed with saturated aqueous NaCl (2ꢀ2.5 mL) and
dried (MgSO4). Concentration under vacuum and subsequent
purification by column chromatography, using EtOAc/hexane
or PE mixtures, gave the products as oils.
7
96 (4 h)[e]
[a]
Reaction
conditions:
allylic/benzylic
bromide
(0.25 mmol), organostannane (0.3 mmol), 7a (2.5 mol %;
5 mol % Pd), CuI (5 mol %), CsF (2 equivs.), DMF
(2.5 mL) at 608C, under an inert atmosphere of N2, 6
hours.
Yields are conversion by GC analysis against an internal
standard (dodecane).
[b]
[c]
Using 4-methoxybromobenzene.
[d]
Using 4-methoxyiodobenzene. After one hour, 60% con-
version was seen. Pd(PPh3)4 showed 34% conversion after
1 hour, and 87% after 2 hours.
After 2 hours, 58% conversion observed. Pd(PPh3)4
showed 41% conversion after 2 hours, and 83% after 4
hours.
[e]
active Pd catalyst species. For the reaction of 8þZ-9a!
Z-10, the order of reactivity was 7b (phthal)¼7a
(succ)>7c (mal)>7f (Br)>7e (Cl)>7g (I)>7d (OH).
Previous studies with mono-nuclear relatives, contain-
ing various activating donor ligands (PR3, NR3 etc.), in-
Stille Coupling Procedure (Method B)
Following a similar procedure to that reported by Baldwin and
co-workers:[16] To a solution of the organohalide (0.25 mmol, 1
dicate that they are considerably less active than the di- equiv.) and the organostannane (0.3 mmol, 1.2 equivs.) in de-
410
asc.wiley-vch.de
ꢀ
Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2006, 348, 405 – 412

Halide and Pseudohalide Effects in Stille Coupling
COMMUNICATIONS
Scheme 4. Proposed mechanism for Stille coupling in the presence of catalytic CuI and stoichiometric CsF in reactions medi-
ated by 7a (X’¼succinimide) in DMF (Pathway A is a neutral catalytic cycle and Pathway B is an anionic cycle; the oxidation
state “Y” is not specified).
gassed (freeze-pump-thaw cycles) DMF (2.5 mL), was added
the palladium catalyst (0.0125 mmol, 0.05 equivs.), CuI
(0.0125 mmol, 0.05 equivs.) and CsF (0.5 mmol, 2 equivs.).
The mixture was heated to 408C for the times indicated in
the text. GC analysis and work-up were similar to Method A.
The following compounds have been previously character-
ised: ethyl 3-(tributylstannyl)-2Z-propenoate (Z-9a),[12d] meth-
yl 3-(tributylstannyl)-2Z-propenoate (Z-9b),[17] ethyl (2Z)-4-
phenyl-2-butenoate (Z-10a),[12d] methyl (2Z)-4-phenyl-2-but-
enoate (Z-10b),[18] isopropyl (2Z)-4-phenyl-2-butenoate (Z-
10c),[19] diphenylmethane (12),[12d] 3-nitrophenyl(phenyl)me-
thane (14), ethyl (2Z)-4-(4’-nitrophenyl)-2-butenoate (entry
2, Table 7),[12d] ethyl (2Z)-4-(4’-methoxyphenyl)-2-butenoate
(entry 3, Table 7),[12d] ethyl (1E)-2-phenylpropenoate (entry
4, Table 7),[12d] ethyl (2Z,5E)-6,10-dimethyl-2,5,9-undecatri-
enoate (entry 6, Table 7),[12d] ethyl (2Z,5E)-6-phenyl-2,5-hexa-
dienoate (entry 7, Table 7),[12d] 3-nitrobiphenyl (entry 1, Ta-
ble 8),[20] 4-nitrobiphenyl (entry 2, Table 8),[21] 4-methoxybi-
phenyl (entry 3, Table 8),[14] methyl 4-vinylbenzoate (entry 5,
Table 8),[22] 3-methoxyphenyl(phenyl)methane (entry 7, Ta-
ble 8).[12d]
served: J
¼58.0, J
¼59.0 Hz); 13C NMR (100 MHz,
117Sn-H
119Sn-H
CDCl3): d¼11.0, 13.7, 20.4, 27.3, 29.0, 67.7, 135.6, 156.6,
167.3; MS (EI): m/z¼405 (25, Mþ), 347 (100), 305 (30), 291
(10); HR-MS (EI): m/z¼401.18109, calculated for (Mþ):
401.18045.
1
Data for E-9c: Rf ¼0.60 (PE:EtOAc, 9:1, v:v); H NMR
(400 MHz, CDCl3): d¼0.78–1.09 (m, 21H), 1.24–1.34 (m,
6H), 1.43–1.56 (m, 6H), 4.99 (septet, 1H), 6.22 (d, 1H, J¼
12.9 Hz, satellite peaks observed: J
¼58.0 Hz, J
¼
117Sn-H
119Sn-H
61.0 Hz), 7.67 (d, 1H, J¼12.9 Hz, satellite peaks observed:
J
¼53.0, J
¼56.0 Hz); 13C NMR (100 MHz, CDCl3):
117Sn-H
119Sn-H
d¼11.0, 13.7, 20.4, 27.3, 29.0, 67.7, 137.6, 153.2, 167.3; MS
(EI): m/z¼405 (25, Mþ), 347 (100), 305 (30), 291 (10); HR-
MS (EI): m/z¼401.18109, calculated for(Mþ): 401.18045.
Acknowledgements
I. J. S. F. thanks the Royal Society for a University Research Fel-
lowship and generous equipment grant. We are grateful to Drs.
Jason M. Lynam and John E. McGrady (York) for informative
´
discussions. Financial support of this work by Direccion Gener-
´
al de Investigacion (Spain) (project-BQU2001-0979-C02-01/
Isopropyl 3-(Tributylstannyl)-2Z-propenoate (Z-9c)
02) is gratefully acknowledged.
AIBN (0.073 g, 1.4 mmol) was added to a neat mixture of iso-
propyl propynoate (5 g, 44 mmol) and tributyltin hydride
(13.2 mL, 49 mmol) and heated to 608C for 24 hours. The mix-
ture was then cooled to room temperature and purified by col-
umn chromatography using petroleum ether (40–
608C):EtOAc(95:5, v:v) as the eluent. This gave the Z-iso-
mer (6.1 g, 34%), followed by the E-isomer (9.7 g, 54%) as col-
ourless oils.
References and Notes
[1] K. C. Nicolaou, E. J. Sorensen, Classics in Total Syn-
thesis, VCH, New York, 1996, Chapter 31.
[2] a) N. Yasuda, J. Organomet. Chem. 2002, 653, 279; b) C.
Mauger, G. Mignani, Adv. Synth. Cat. 2005, 347, 773;
c) E. Burello, G. Rothenberg, Adv. Synth. Cat. 2005,
345, 1334.
[3] a) J. Q. Liu, M. J. Robins, Org. Lett. 2005, 7, 1149; b) F.
Amblard, S. P. Nolan, R. F. Schinazi, L. A. Agrofoglio,
Tetrahedron 2005, 61, 537; c) E. Mayer, L. Valis, R. Hub-
1
Data for Z-9c: Rf ¼0.70 (PE:EtOAc, 9:1, v:v); H NMR
(400 MHz, CDCl3): d¼0.78–1.09 (m, 21H), 1.24–1.34 (m,
6H), 1.43–1.56 (m, 6H), 4.99 (septet, 1H, J¼7.1 Hz), 6.64 (d,
1H, J¼12.7 Hz, satellite peaks observed: J
¼112.0 Hz,
117Sn-H
J
¼114.0 Hz), 7.07 (d, 1H, J¼12.7 Hz, satellite peaks ob-
119Sn-H
Adv. Synth. Catal. 2006, 348, 405 – 412
ꢀ
Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
411

Catherine M. Crawforth et al.
COMMUNICATIONS
er, N. Amann, H. A. Wagenknecht, Synthesis 2003, 2335;
d) N. Amann, H. A. Wagenknecht, Synlett 2002, 687;
e) I. J. S. Fairlamb, Tetrahedron 2005, 61, 9661 (special is-
sue on palladium catalysis in synthesis).
[11] a) J. K. Stille, Angew. Chem. Int. Ed. 1986, 25, 508; b) V.
Farina, V. Krishnamurthy, W. J. Scott, Org. React. 1997,
50, 1; c) T. N. Mitchell, in: Metal-catalyzed Cross-cou-
pling Reactions, (Eds.: A. de Meijere, F. Diederich), Wi-
ley-VCH, New York, 2004, Chapter 3.
[12] a) C. M. Crawforth, S. Burling, I. J. S. Fairlamb, R. J. K.
Taylor, A. C. Whitwood, Chem. Commun. 2003, 2194;
b) C. M. Crawforth, I. J. S. Fairlamb, R. J. K. Taylor, Tet-
rahedron Lett. 2004, 45, 461; c) I. J. S. Fairlamb, A. R.
Kapdi, J. M. Lynam, R. J. K. Taylor, A. C. Whitwood,
Tetrahedron 2004, 60, 5711; d) C. M. Crawforth, S. Burl-
ing, I. J. S. Fairlamb, A. R. Kapdi, R. J. K. Taylor, A. C.
Whitwood, Tetrahedron 2005, 61, 9736.
[4] a) Metal-catalyzed Cross-coupling Reactions, (Eds.: A.
de Meijere, F. Diederich), Wiley-VCH: New York,
2004; b) J. Tsuji, Palladium Reagents and Catalysts, Inno-
vations in Organic Synthesis, Wiley, New York, 1995.
[5] a) A. F. Littke, G. C. Fu, Angew. Chem. Int. Ed. 2002, 41,
4176; b) R. B. Bedford, Chem. Commun. 2003, 1787.
[6] a) I. P. Beletskaya, A. V. Cheprakov, J. Organomet.
Chem. 2004, 689, 4055; b) J. Dupont, C. S. Consorti, J.
Spencer, Chem. Rev. 2005, 105, 2527.
´
[7] a) W. A. Herrmann, V. P. W. Bçhm, C.-P. Reisinger, J.
Organomet. Chem. 1999, 576, 23; b) I. P. Beletskaya,
V. A. Cheprakov, Chem. Rev. 1999, #99#100, 3009; c) J.
Dupont, M. Pfeffer, J. Spencer, Eur. J. Inorg. Chem.
2001, 1917; d) W. A. Herrmann, K. Ölefe, D. V. Preysing,
S. K. Schneider, J. Organomet. Chem. 2003, 687, 229. The
important role played by both phosphine and the anionic
ligand “X” in the overall performance of palladacyclic
catalysts has been noted, see: e) R. B. Bedford, C. S. J.
Cazin, S. J. Coles, T. Gelbrich, P. N. Horton, M. B. Hurst-
house, M. E. Light, Organometallics 2003, 22, 987;
f) R. B. Bedford, C. S. J. Cazin, S. J. Coles, T. Gelbrich,
M. B. Hursthouse, V. J. M. Scordia, Dalton Trans. 2003,
3350.
[13] J. L. Serrano, I. J. S. Fairlamb, G. Sanchez, L. García, J.
´
´
Perez, J. Vives, G. Lopez, C. M. Crawforth, R. J. K. Tay-
lor, Eur. J. Inorg. Chem. 2004, 2706.
[14] I. J. S. Fairlamb, A. R. Kapdi, A. F. Lee, G. Sanchez, G.
´
´
´
´
Lopez, J. L. Serrano, L. García, J. Perez, E. Perez, Dalton
Trans. 2004, 3970.
[15] Halide effects (and the use of tetralkylammonium salts)
have been studied in detail for Heck alkenylation, see:
C. S. Consorti, F. R. Flores, J. Dupont, J. Am. Chem.
Soc. 2005, 127, 12054.
[16] S. P. H. Mee, V. Lee, J. E. Baldwin, Angew. Chem. Int.
Ed. 2004, 43, 1132.
[17] G. A. Russell, P. Ngoviwatchai, H. Tashtoush, J. Hersh-
berger, Organometallics 1987, 6, 1414.
[8] K. Fagnou, M. Lautens, Angew. Chem. Int. Ed. 2002, 41,
27, and references cited therein.
[9] a) C. Amatore, A. Jutand, Acc. Chem. Res. 2000, 33, 314;
b) C. Amatore, A. Jutand, J. Organomet. Chem. 1999,
576, 254, and references cited therein.
[18] P. Mueller, P. Polleux, Helv. Chim. Acta 1998, 81, 317.
[19] K. Itoh, N. Hamaguchi, M. Miura, M. Nomura, J. Mol.
Catal. 1992, 75, 117.
[20] M. Dai, B. Liang, C. Wang, J. Chen, Z. Yang, Org. Lett.
2004, 6, 221.
[10] V. Farina, G. P. Roth, in: Advances in Metal-Organic
Chemistry; (Ed.: L. S. Liebeskind), JAI, London, 1996,
Vol. 5, p. 45.
[21] N. E. Leadbeater, M. Marco, Org. Lett. 2002, 4, 2973.
[22] W. Su, S. Urgaonkar, J. G. Verkade, Org. Lett. 2004, 6,
1421.
412
asc.wiley-vch.de
ꢀ
Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2006, 348, 405 – 412