Ligand effects on the Pd-Catalyzed cross-coupling reaction of 3-Iodoalk-2-enoates with Propargyl/1,2-Allenylie metallic species: An efficient regiodivergent synthesis of 2,4,5-Trienoates

Article abstract of DOI:10.1002/chem.200901287

"Chemical Equation Presented" A ligand makes a big difference : The 1,3-lithium shift of propargyl/1,2-allenyl lithium has been tuned and applied in a subsequent transmetalation and Pd-catalyzed Negishi coupling reaction with 3-iodoalk-2-enoates to give a highly regioselective and regiodivergent synthesis of 2,4,5-trienoates. It was discovered that ligands play a very important role in the cross-coupling reactions with respect to both regioselectivity and reactivity.

Full text of DOI:10.1002/chem.200901287

DOI: 10.1002/chem.200901287
Ligand Effects on the Pd-Catalyzed Cross-Coupling Reaction of 3-Iodoalk-2-
enoates with Propargyl/1,2-Allenylic Metallic Species: An Efficient
Regiodivergent Synthesis of 2,4,5-Trienoates
Jinbo Zhao,[a] Yihua Yu,[b] and Shengming Ma*[a]
Transition-metal-catalyzed cross-coupling reactions be-
tween organic halides or pseudohalides with main-group
propargyl/1,2-allenyl zinc species and aryl iodides for the
formation of allenes or alkynes and the unique selectivity in
the related reaction of propargyl/1.2-allenyl zinc reagents
with 3-iodoalk-2-enoates to form substituted 2,4,5-alkatri-
metal species has become a powerful tool for the efficient
[1]
À
construction of C C bonds both in academia and industry.
In most cases, these cross-coupling reactions do not have re-
gioselectivity problems. For cross-coupling reactions involv-
ing propargyl/1,2-allenyl metal species, however, the issue of
regioselectivity arises.[2,3] Over the past few years, we have
studied the Negishi coupling reaction[4] involving propargyl/
1,2-allenyl metal species to afford allenes or alkynes in a
highly selective manner (Scheme 1). We found a dramatic
dependence of regioselectivity on the steric factors of both
enoates.[5] We have also regioselectively tuned the Pd-cata-
ACHTUNGTRENNUNG
lyzed cross-coupling of propargyl mesylates with 2-(alkoxy-
carbonyl)ethenyl zinc iodides by using the ligand effect
(Mode II, Scheme 1).[6]
On the other hand, we observed an interesting “1,3-H/Li
shift” phenomenon in the lithiation of 1-arylalk-1-ynes[7]
(Scheme 2): the initially formed 1-arylalka-1,2-dien-1-lithi-
Scheme 1. The two complementary modes of Negishi cross-coupling in-
volving propargyl/1,2-allenyl metallic species.
Scheme 2. The 1,3-H/Li shift in the Pd-catalyzed cross-coupling reaction
of propargyl/1,2-allenyl zinc species with different types of organic hal-
ides. 1) Lithium reagent, 2) ZnBr2, 3) [Pd], R’I.
[a] J. Zhao, Prof. Dr. S. Ma
State Key Laboratory of Organometallic Chemistry
Shanghai Institute of Organic Chemistry
Chinese Academy of Sciences
354 Fenglin Lu, Shanghai 200032 (P. R. China)
Fax : (+86)21-6260-9305
E-mail: masm@mail.sioc.ac.cn
um intermediate A underwent a formal H/Li shift to afford
the thermodynamically more stable 1-arylalka-1,2-dien-3-
lithium intermediate B. By using nBuLi (1.2 equiv) at
À208C for 1 h (lithiation conditions A) or lithium diisopro-
pylamide (LDA; 2 equiv) at À788C to room temperature
for 1 h (lithiation conditions B) as the lithiation conditions,
the two types of propargyl/1,2-allenyl lithium reagents A/A’
or B/B’ could be obtained selectively. Subsequent transmeta-
[b] Prof. Dr. Y. Yu
Shanghai Key Laboratory of Functional Magnetic
Resonance Imaging
Department of Physics, East China Normal University
Shanghai 200062 (P. R. China)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.200901287.
74
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 74 – 80

COMMUNICATION
Table 1. Lithiation of 1-phenylhex-1-yne (1a) with LDA, transmetalation
with ZnBr2, and the Pd-catalyzed cross-coupling reaction of the in situ-
formed propargyl/1,2-allenyl zinc species with methyl 3-iodoprop-2(E)-
enoate [(E)-2a] under different conditions.[a]
lation and [PdACHTUNGTRENNUNG(PPh3)4]-catalyzed cross-coupling with aryl io-
dides afforded the differently trisubstituted allenes starting
from simple, readily available 1-arylalk-1-ynes and aryl hal-
ides.[8] As an extension of this strategy, we wished to realize
the 1,3-H/Li shift/transmetalation/cross-coupling reaction
with 3-iodoalk-2-enoates. However, primary results showed
that the coupling reaction with 3-iodoalk-2-enoates is very
different from that with aryl iodides. Under the conditions
developed for the cross-coupling with aryl iodides, either a
mixture of allene (E)-3 and conjugated enyne (Z)-4 was ob-
tained with poor selectivity (under the “1,3-H/Li shift” con-
ditions) or the direct propargyl/alkenyl cross-coupling prod-
uct enyne (E)-5 was isolated with a low yield (under the
“non-1,3-H/Li shift” conditions). Thus, new protocols had to
be established. In this communication, we wish to report the
pursuit of this strategy, which has led to the discovery of
dramatic ligand effects on the regioselectivity[9] and reactivi-
ty in the cross-coupling of propargyl/1,2-allenyl zinc species
with 3-iodoalk-2-enoates, and the one-pot regiodivergent
synthesis of two types of differently substituted 2,4,5-tri-
Entry
T [8C]
Catalyst
t [h]
Combined
yield [%]
(E)-3aa/
(Z)-4aa[b]
1
2[c]
3[e]
4[f]
5
6
7
8
28
28
28
27
0
25
25
0
[Pd
[Pd
[Pd
[Pd
[Pd
[Pd
[Pd
[Pd
N
2
2
2
2
17
3
58
81:19
81:19
80:20
79:21
89:11
78:22
94:6
[d]
–
–
[d]
44
22
45
64
77
[g]
[h]
[h]
0.67
1
98:2
1) LDA, À788C–RT, 1 h; 2) ZnBr2, RT, 0–10 min, then RT to T, 10–
40 min; 3) catalyst, THF. [a] Reaction conditions unless otherwise speci-
fied: 1a (0.50 mmol, 1.7 equiv), LDA (1.0 mmol, 3.3 equiv), ZnBr2
(2.0 mmol, 6.7 equiv), [PdACHTUNGTRENNG(U PPh3)4] (5 mol%) or [PdAHCTUGNTREN(NGUN p-allyl)Cl]2
(2.5 mol%)/ligand (10 mol%), (E)-2a (0.3 mmol). [b] Due to the compli-
enoates.
1
cated signals in the H NMR spectrum of the crude reaction mixture and
the inseparability of the two products [(E)-3aa and (Z)-4aa], the selec-
tivity was determined based on isolation. [c] 2.5 equiv of LDA was used.
[d] The isolated products were contaminated with an unidentified impuri-
ty. [e] 2.0 equiv of LDA was used. [f] The lithiation was carried out at
08C for 1 h. [g] TriACTHNUTRGNENGU(o-tol)phosphine was used as the ligand. [h] TriACHTUNGTRENNUNG(2-fur-
yl)phosphine was used as the ligand.
Synthesis of 4-alkyl-6-arylhexa-2(E),4,5-trienoates: In previ-
ous studies in our group, we developed a protocol for the ef-
ficient synthesis of alka-2,4,5-trienoates from the cross-cou-
pling reaction of 3-iodoalk-2-enoates with the propargyl/1,2-
allenyl zinc species generated from the lithiation of 1-ary-
lalk-1-yne with nBuLi followed by transmetalation with
ZnBr2.[5a] The fact that addition of HgCl2 is necessary to
achieve good yields is expected to cause environmental
problems. Furthermore, it was found that this procedure is
not easily reproducible in terms of using commercially avail-
able nBuLi from different sources, and the structures of the
products from the reaction of 1a with (E)-2b, (Z)-2a, (Z)-
2b, (Z)-2c and (Z)-2d according to the reported procedure
are not the proposed ones, but their regioisomers based on
our most recent studies [Eq. (1)].[5a,8]
Evidently, new reaction conditions needed to be devel-
oped to solve the issue of regioselectivity. When we applied
Figure 1. NOE effects of compounds (Z)-4aa and (E)-3ac.
formation of an inseparable im-
purity together with the prod-
ucts (E)-3aa and (Z)-4aa
(Table 1, entries 2 and 3); this
indicated that using an exces-
sive amount of LDA is essential
for a complete reaction, as was
also observed in our previous
studies. The reason why 2 equiv
the conditions developed for the 1,3-H/Li shift/transmetala-
tion/cross-coupling with aryl halides,[8] a mixture of the ex-
pected “1,3-H/Li shifted” allene (E)-3aa and the conjugate
enyne (Z)-4aa[5a] was formed disappointingly in a combined
yield of 58% (Table 1, entry 1). The structure and the con-
figuration of the double bond of (Z)-4aa (Figure 1) were un-
of LDA (relative to 1) is needed is still not clear, but is sup-
posed to be for mechanistic reasons. Lowering the cross-
coupling temperature led to a decrease in the yield (Table 1,
entry 5). To our disappointment, the use of [Pd
ACHTUNGTRENNUNG(p-allyl)-
Cl]2/A(o-tol)3P as the catalyst reduced the ratio of (E)-3aa/
CTHUNGTRENNUNG
(Z)-4aa to 78:22, although the impurity was not formed
(Table 1, entry 6). However, we were fortunate to observe
1
ambiguously determined by analysis of its H/13C NMR and
NOESY spectra. Reducing the amount of LDA led to the
that the use of TFP [triACTHUGNTERN(UNG 2-furyl)phosphine] as the ligand in
Chem. Eur. J. 2010, 16, 74 – 80
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
75

S. Ma et al.
combination with [Pd
(p-allyl)Cl]2 improved the ratio of (E)-
no isomerization was observed at 08C (Scheme 3, entries 1
and 3); when the isomerization reactions were carried out at
room temperature, a dramatically higher extent of isomeri-
3aa/(Z)-4aa to 94:6 (Table 1, entry 7). Lowering the cross-
coupling temperature to 08C finally enabled us to improve
this selectivity to 98:2 (Table 1, entry 8).
With these optimized conditions (conditions A: Table 1,
entry 8) in hand, the scope of the reaction was then exam-
ined. As shown in Table 2, the reactions with both (Z)- and
Table 2. The cross-coupling reaction of 3-iodopropenoates 2a and 2b
under Conditions A[a]
Entry R (1)
2
T [8C] t [h] Combined
(E)-3/
yield [%]
(Z)-4[b]
1
2
3
4
5
nC3H7 (1a) (E)-2a
0
1
1
77
98:2
[(E)-3aa+(Z)-4aa]
80
[(E)-3ab+(Z)-4ab]
nC3H7 (1a) (E)-2b 21
nC3H7 (1a) (Z)-2a 15
nC3H7 (1a) (Z)-2b 20
96:4
0.5 63
[(E)-3aa+(Z)-4aa]
0.5 60
[(E)-3ab+(Z)-4ab]
64
[(E)-3ba+(Z)-4ba]
98:2
>98:2
98:2
nC5H11 (1b) (E)-2a
0
1
Scheme 3. Isomerization between (E)-3aa and (Z)-4aa. 1) LDA,
2) ZnBr2.
1) LDA, À788C–RT, 1–1.5 h; 2) ZnBr2, RT, 20–60 min; 3) cat. [Pd
ACHTUNGERTN(NUNG p-
allyl)Cl]2/TFP, 2, THF. [a] Unless otherwise specified, the reaction was
conducted by using conditions A: alkyne 1 (0.50 mmol, 1.7 equiv), LDA
(1.0 mmol, 3.3 equiv), ZnBr2 (2.0 mmol, 6.7 equiv), iodide 2 (0.30 mmol),
zation was observed (Scheme 3, entries 2 and 4). Further-
more, when the reactants were treated with 9.3 equiv of an
organozinc species at room temperature for 24 h, both reac-
tions afforded a mixture of (E)-3aa and (Z)-4aa in almost
the same ratio (Scheme 3, entries 5 and 6).
[PdACHTUNGTRENNUNG(p-allyl)Cl]2 (2.5 mol%), and TFP (10 mol%). [b] Ratio determined
by NMR analysis after chromatography due to complicated signals on
the 1H NMR spectrum of the crude reaction mixture and the inseparabili-
ty of the two products.
This observation indicates that an equilibrium exists be-
tween the deprotonated isomers C and D in the presence of
organozinc species that caused the isomerization between
(E)-3aa and (Z)-4aa at the higher temperature (Scheme 4).
However, it is interesting to observe that, in the case of
the 3-substituted 3-iodoalk-2-enoates, the formation of
enyne (Z)-4 was not observed (Table 3); that is, for all the
cases studied, the corresponding allenes (E)-3 were formed
exclusively in moderate to good yields. Notably, naphthyl
substitution on the alkyne also led to exclusive formation of
the corresponding allene (E)-3gd (Table 3, entry 8). With
the increasing steric hindrance of the substitution on the io-
dides, the reaction rate decreased dramatically, that is,
methyl 3-iodo-3-phenyl-2(Z)-propenoate [(Z)-2c] was not
fully consumed (Table 3, entry 1). Cross-coupling for all the
Z-configured iodides (Z)-2a–e examined afforded the E
product (Table 2 entries 3 and 4 and Table 3), as determined
by NOE study of the product (E)-3ac (Figure 1).
(E)-3-iodopropenoates proceeded smoothly, with the minor
isomers (Z)-4aa, (Z)-4ab and (Z)-4ba being obtained in
ꢀ4% yield. It is worth noting that the cross-coupling tem-
perature is crucial for the selectivity and reaction rate in
each case: lowering the temperature led to a much slower
reaction, while raising it resulted in a dramatic decrease of
chemical yield. No matter what the configuration of the
double bond in the starting iodides 2 was, the configuration
of the double bond in the products (E)-3 is exclusively trans,
as unambiguously demonstrated by the coupling constant of
the vinylic hydrogen signals.[6,10,11] The improved selectivity
of (E)-3/(Z)-4 may be explained by the electron-rich nature
of TFP, which increases the reactivity and allows the cou-
pling to occur under milder conditions, thus suppressing the
isomerization from (E)-3 to (Z)-4.[12] To confirm this possi-
bility, both (E)-3aa [(E)-3aa/(Z)-4aa=96:4] and (Z)-4aa
[(Z)-4aa/(E)-3aa ꢁ97:3 or 99:1] were treated separately
with different amounts of propargyl/1,2-allenyl zinc species
at different temperature. After the usual workup, essentially
Synthesis of 4-arylalka-2(E),4,5-trienoates: At the onset of
the research into this subject, we examined the conditions
76
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 74 – 80

Regiodivergent Synthesis of 2,4,5-Trienoates
COMMUNICATION
We have observed a dramatic
ligand effect on regioselectivity
in the Pd-catalyzed Negishi
cross-coupling of propargyl me-
sylates with the (alkoxylcarbo-
nyl)ethenylzinc iodides.[6] We
reasoned that, considering the
similarity of the two reaction
types, a similar ligand effect
might be operative in the pres-
ent transformation. To this end,
Scheme 4. Possible mechanism for the interconversion between (E)-3aa and (Z)-4aa.
Table 3. The cross-coupling reaction with 3-substituted 3-iodoalk-2-
Table 4. The effect of different reaction parameters on the Pd-catalyzed
coupling reaction of methyl 3-iodoprop-2(E)-enoate (E)-2a with the
propargyl/allenyl zinc species generated from lithiation of 1-phenylhex-1-
enoates (Z)-2c–e under conditions A[a]
[a]
yne (1a) with nBuLi and subsequent transmetalation with ZnBr2
Entry Ar/R1 (1)
R2 (2)
T [8C] t [h] Yield [%][b]
1
2
3
4
5
6
7
8
Ph/nC3H7 (1a)
Ph
[(Z)-2c]
Me
[(Z)-2d]
Bn
[(Z)-2e]
Me
[(Z)-2d]
Me
[(Z)-2d]
20
15
19
20
24
15
18
15
8
48[c] [(E)-3ac]
71 [(E)-3ad]
34 [(E)-3ae]
54 [(E)-3cd]
74 [(E)-3dd]
72 [(E)-3ed]
53 [(E)-3 fd]
70 [(E)-3gd]
Ph/nC3H7 (1a)
Ph/nC3H7 (1a)
Ph/Me (1c)
1.5
4.7
1
Entry
T1 [8C]
t1 [h]
T2 [8C]
t2 [h]
Yield (E)-5aa [%][b]
1
2
3
4
5
6
7
0
À3
0
À3
À3
0
1
1
RT
0
2
1
23
22
32
26
1.5
2
1.5
1.5
1.5
À20
À20
À20
À20
À20
11
10.5
16
16.5
9
p-PhC6H4/nC3H
7A
2
18[c,d]
29[e]
20
p-MeOC6H4/nC3H7 (1e) Me
[(Z)-2d]
Me
1.5
2
À20
p-PhC6H4/nC5H11 (1 f)
a-naphthyl/nC3H7 (1g)
[(Z)-2d]
Me
[(Z)-2d]
1) nBuLi, T1, t1; 2) ZnBr2, T1 to T2, 20–42 min; 3) cat. [PdACHTNUGRTNEUNG(PPh3)4], THF,
T2, t2. [a] Unless otherwise specified, the reaction was carried out with 1a
(0.50 mmol, 1.7 equiv), nBuLi (0.60–0.64 mmol, 2.0–2.1 equiv), ZnBr2
1
(1.0 mmol, 6.7 equiv), [Pd
[b] Isolated yields. [c] Pd(PPh3)2Cl2 (5 mol%) was used as the catalyst.
[d] 0.40 mmol of (E)-2a and 0.020 mmol (5 mol%) of [Pd(PPh3)4] were
used. [e] 0.20 mmol of (E)-2a and 0.010 mmol (5 mol%) of [PdACHTUNGTRENNUNG(PPh3)4]
were used.
ACHTUNGTERN(NUNG PPh3)4] (5 mol%), and (E)-2a (0.30 mmol).
1) LDA, À788C–RT, 1 h; 2) ZnBr2, RT, 20–50 min; 3) cat. [Pd
ACHTUNGTRNE(NUNG p-al-
AHCTUNGTRENNUNG
lyl)Cl]2/TFP, THF. [a] Unless otherwise specified, the reaction was con-
ducted under conditions A. [b] Isolated yields. [c] The reaction was car-
ried out with 1a (0.44 mmol), LDA (1.0 mmol), ZnBr2 (2.0 mmol), iodide
AHCTUNGTRENNUNG
(Z)-2 (0.20 mmol), [PdACHTUNGTRENNUNG(p-allyl)Cl]2 (5 mol%), and TFP (20 mol%), and
the yield was calculated based on 64% conversion of the iodide.
several ligands with distinct electronic properties were ex-
amined. As depicted in Table 5, when triphenyl phosphite
(L1) was used as the ligand, the propargyl-alkenyl coupling
product (E)-5aa was obtained exclusively in a low yield.
developed for the cross-coupling reaction of aryl iodides
with propargyl/1,2-allenyl zinc species derived from the lith-
iation of 1-arylalk-1-ynes with nBuLi to form trisubstituted
aryl allenes[7] (Table 4, entry 1). Unfortunately, it was ob-
served that methyl 6-phenyl-4-propyl-5-yn-2(E)-enoate [(E)-
5aa] was obtained as the only regioisomer in 23% yield.[13]
We at first attributed this observation to insufficient lithia-
tion of substrate 1a and/or unsuitable cross-coupling tem-
perature, so we screened experimental parameters such as
lithiation temperature, lithiation time, cross-coupling tem-
perature, and additives, aiming to find the right conditions
for highly selective synthesis of (E)-5aa or methyl 4-phenyl-
nona-2(E),4,5-trienoate [(E)-6aa]. As briefly summarized in
Table 4, many different reaction conditions were screened
and found to be fruitless. In all the cases listed, only the
propargyl-alkenyl coupling product (E)-5aa was isolated in
low yields (Table 4).
Using triACTHNUTRGNE(UNG 2-furyl)phosphine (L2), the ratio of (E)-5aa/(E)-
6aa was improved to <7:1. Surprisingly, when the ligand
was changed to (R)-2-diphenylphosphino-2’-hydroxy-1,1’- bi-
naphthalene [(R)-L3], which was used in our previous
study,[6] we were very happy to find that the reaction
reached completion within 5 min to afford (E)-6aa exclu-
sively (Table 5, entry 3). However, essentially no ee was ob-
served. The regioselectivity reversal is speculated to be in-
fluenced by the electronic properties of the ligand. As seen
in Scheme 5, since the sp2-hybrized allenic carbon is more
electronegative than the sp3-hybrized propargyl carbon, the
electron-donating ability of the ligand increases the electron
density of palladium through coordination of the phenolate
moiety, driving the equilibrium to the formation of allenylic
Chem. Eur. J. 2010, 16, 74 – 80
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
77

S. Ma et al.
Table 5. Ligand effect on the cross-coupling reaction of methyl (E)-3-io-
dopropenoate [(E)-2a] with the propargyl/1,2-allenyl zinc species gener-
ated from lithiation of 1-phenylhex-1-yne (1a) with nBuLi followed by
transmetalation with ZnBr2.[a]
Table 6. The effect of ligand on reactivity and an examination of the cat-
alytic efficiency of the [PdACTHNUTRGNE(UNG p-allyl)Cl]2/ligand-catalyzed Negishi cross-
coupling of methyl 3-iodobut-2(Z)-enoate (Z)-2d with the propargyl/1,2-
allenyl zinc species generated from lithiation of 1-phenylhex-1-yne with
nBuLi followed by transmetalation with ZnBr2.[a]
Entry Catalyst [loading]
t
Yield
[%][b]
1
2
5 mol% [Pd
2.5 mol% [Pd
0.5 mol% [Pd
G
4 h
16[c]
3[d]
4[d]
0.17 mol% [PdCAHTUNGTRENNNUG
L3
Entry
Ligand
t
Combined yield [%]
(E)-5aa/(E)-6aa
1) nBuLi, À208C, 1 h; 2) ZnBr2, À208C, 10 min then À208C to RT,
20 min; 3) catalyst, THF, RT. [a] Unless otherwise specified, the reaction
was carried out with 1a (0.50 mmol, 1.7 equiv), nBuLi (0.60 mmol,
2.0 equiv), ZnBr2 (1.0 mmol, 3.3 equiv), (Z)-2d (0.30 mmol), and catalyst
(loading indicated for each case). [b] Isolated yields. [c] The yield is
based on the isolated product after flash chromatography and preparative
TLC. [d] 1a (2.5 mmol, 1.7 equiv), nBuLi (3.3 equiv), ZnBr2 (3.3 equiv),
and (Z)-2d (1.5 mmol) were used.
1
2
3
L1
L2
(R)-L3
overnight
3 h
5 min
32
26
100:0[b]
<7:1[c]
0:100[e]
[d]
–
1) nBuLi, À208C, 1 h; 2) ZnBr2, À208C to RT, 10–34 min; 3) cat. [Pd
ACHTUNGERTN(NUNG p-
allyl)Cl]2/ligand. [a] Unless otherwise specified, the reaction conditions:
1a (0.50 mmol, 1.7 equiv), nBuLi (0.60 mmol, 2.0 equiv), ZnBr2
(1.0 equiv, 3.3 equiv), [Pd
ACHTUNGTRENNUNG
(E)-2a (0.30 mmol). [b] [PdAHCTNUGTRENNNUG
(13 mg, 0.040 mmol, 10 mol%), and (E)-2a (0.20 mmol) were used. The
ratio was determined by NMR analysis of the isolated product, and the
yield was based on 90% conversion of iodide (E)-2a. [c] (E)-6aa is not
stable during separation, thus the ratio is inaccurate. [d] Due to the insta-
bility of (E)-6aa, no information on the yield is available. [e] Determined
by NMR analysis of the reaction mixture.
of this reaction was reduced to 0.17 mol% [PdACTHUNTRGNE(NUG p-allyl)Cl]2/
0.67 mol% (Æ)-L3, the reaction could also finish within
30 min to afford (E)-6ad in 59% yield, indicating a turnover
number (TON) of over 300 and a turnover frequency (TOF)
of over 600 hÀ1 (Table 6, entry 4).
The scope of the reaction conditions was then examined.
As summarized in Table 7, under conditions B (Table 6,
entry 2), the reaction of differently substituted 1-arylalk-1-
ynes 1 with 3-iodoalk-2(Z) or (E)-enoates 2 proceeded read-
ily to afford the corresponding 4-arylalka-2(E),4,5-trienoates
(E)-6 efficiently. Generally, the reaction rate depends large-
ly on the steric hindrance of R2. For example, when R2 is
alkyl (Table 7, entries 1, 2, 4, 6–9) or benzyl (entry 5), the re-
action usually reached completion within 2 h to afford the
products (E)-6 in good to excellent yields; for bulkier sub-
stituent, such as Ph (Table 7, entry 3), the reaction was com-
plete after a prolonged reaction time to afford the corre-
sponding product (E)-6ac in a low yield. Notably, for all the
3-iodoalk-2(Z)-enoates (Z)-2, the C=C double bonds in the
corresponding products was also in the more stable E con-
figuration, as illustrated by an NOE study of product (E)-
6ae and the alcohol (E)-7 formed by the reduction of (E)-
6ae with diisobutylaluminum hydride (DIBALH) [Eq. (2)].
The control experiment in Scheme 3 excluded the exclu-
sive formation of (E)-3aa via (Z)-4aa, thus, we reasoned
that the exclusive formation of the more stable (E)-3 and
(E)-6 from (Z)-2 might occur either by path a: isomerization
of double bond in the organopalladium intermediate H to
form I, or path b: the initially formed (Z)-3 or (Z)-6 would
coordinate with palladium and undergo a rapid isomeriza-
tion[10,14–17] (J to K) to form the final products (E)-3 or (E)-
6, during which racemization of the axial chirality of the
allene would also be expected (Scheme 6).[11]
Scheme 5. Proposed ligand effect on regioselectivity.
intermediate G, thereby favoring the formation of allenic
product 6.
It was soon observed that substitution of the iodide has a
dramatic influence on the regioselectivity. For example, the
cross-coupling reaction of methyl 3-iodobut-2(Z)-enoate
(Z)-2d catalyzed by [PdACHTNUTRGNEN(UG p-allyl)Cl]2/PACHTUNTGREN(NGUN OPh)3 also yielded
allene (E)-6ad as the only isomer, albeit in a very low yield
of 16%, which is stable enough under preparative TLC or
normal flash chromatography conditions (cf. Tables 5 and 6,
entries 1). Finally, it was observed that, under the catalysis
of [Pd
G
6ad in a much higher (91%) yield within 5 min (Table 6,
entry 2). It is also noteworthy that when the catalyst loading
78
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 74 – 80

Regiodivergent Synthesis of 2,4,5-Trienoates
COMMUNICATION
differently substituted regioisomeric alka-2,4,5-trienoates.
During this process, we have observed a dramatic ligand
effect on the regioselectivity and reactivity. This observation
will no doubt spur further interest in the ligand effect on re-
gioselectivity in other reactions involving propargyl/1,2-al-
lenyl species. Although at present essentially no chiral in-
duction is observed when using (R)-L3 as the ligand, the ob-
servations presented in this study should stimulate further
interest in the development of this type of asymmetric trans-
formation by using modified 2-diphenylphosphino-2’-hy-
droxy-1,1’-binaphthalenes. Further studies in this area are
currently underway in our laboratory.
Table 7. Synthesis of 4-arylalka-2(E),4,5-trienoates [(E)-6] from the Ne-
gishi cross-coupling of 3-iodoalk-2-enoates 2 with propargyl/1,2-allenyl
zinc species generated from the lithiation of 1-arylalka-1-yne 1 with
nBuLi followed by transmetalation with ZnBr2 (conditions B).[a]
Experimental Section
Entry Ar/R1 (1)
R2/R3 (2)
t
Yield [%][b]
5 min 91 [(E)-6ad]
Typical procedure I (conditions A) for the synthesis of 4-alkyl-6-aryl-
hexa-2(E),4,5-trienoates (E)-3
1
Ph/nC3H7 (1a)
Me/Me [(Z)-
2d]
Me/Et [(E)-2h] 5 min 78 [(E)-6ah]
Ph/Me [(Z)-2c] 16 h 32 [(E)-6ac]
nC4H9/Me [(Z)- 1 h
2 f]
Bn//Me [(Z)-
2e]
Me//Me [(Z)-
2d]
Me/Me [(Z)-
2d]
nC6H13/Me
[(Z)-2g]
Me/Me [(Z)-
2d]
Methyl 4-propyl-6-phenylhexa-2(E),4,5-trienoate [(E)-3aa]: Under an Ar
atmosphere, a flame-dried Schlenk tube was charged with 1-phenylhex-1-
yne (1a; 80 mg, 0.5 mmol) and THF (2 mL). The vessel was cooled to
À788C, and a solution of LDA (0.50 mL, 2.0m) in THF/ethylbenzene/
heptane (1.0 mmol) was added dropwise at À788C. After the addition,
the cooling bath was removed, and the mixture was allowed to warm up
to room temperature within 1.5 h, then a solution of ZnBr2 (2.0 mmol,
450 mg) in THF (2 mL) was added to prepare the zinc reagent. In anoth-
er flame-dried Schlenk tube, a solution of catalyst and substrate in THF
2
3
4
Ph/nC3H7 (1a)
Ph/nC3H7 (1a)
Ph/nC3H7 (1a)
80 [(E)-6af]
5[d]
6
Ph/nC3H7 (1a)
2 h
1 h
1 h
2 h
1 h
74[c] [(E)-6ae]
72 [(E)-6cd]
86 [(E)-6dd]
75[c,e] [(E)-6dg]
77[c] [(E)-6hd]
Ph/Me (1c)
was prepared by adding sequentially [Pd
ACHTUNGTRENN(NUG p-allyl)Cl]2 (3 mg,
7
p-PhC6H4/nC3H7 (1d)
p-PhC6H4/nC3H7 (1d)
0.0075 mmol), THF (1 mL), tri(2-furyl)phosphine (7 mg, 0.030 mmol),
AHCTUNGTRENNUNG
8[d]
9
THF (1 mL), methyl (E)-3-iodopropenoate [(E)-2a; 64 mg, 0.30 mmol],
and THF (1 mL) with stirring. The solution was cooled to 08C with stir-
ring while the above zinc reagent was added with a syringe. After the re-
action was complete, as monitored by TLC, it was quenched with saturat-
ed NH4Cl solution and extracted with diethyl ether, and the combined or-
ganic phase was dried over MgSO4. After filtration the solvent was re-
moved in vacuo, and the residue was subjected to flash chromatography
to afford a mixture of (E)-3aa and (Z)-4aa (56 mg, 77%, (E)-3aa/(Z)-
4aa=98:2). (E)-3aa: oil, 1H NMR (300 MHz, CDCl3): d=7.31–7.10 (m,
6H), 6.35 (brs, 1H), 5.89 (dd, J=15.9, 1.2 Hz, 1H), 3.67 (s, 3H), 2.29–
2.07 (m, 2H), 1.59–1.38 (m, 2H), 0.88 (t, J=7.5 Hz, 3H); 13C NMR
(75.4 MHz, CDCl3): d=212.3, 167.2, 143.6, 133.1, 128.7, 127.4, 127.0,
117.5, 108.2, 96.1, 51.5, 30.6, 20.8, 13.9; IR (neat): n˜ =1926, 1721, 1621,
1598, 1495, 1274, 1170 cmÀ1; MS (70 eV, EI) m/z (%): 242 (17.14) [M]+,
153 (100); HRMS (EI) calcd for C16H18O2: 242.1307; found: 242.1313.
p-C4H9CꢂCC6H4/
nC3H7 (1h)
1) nBuLi, À208C, 1 h; 2) ZnBr2, À208C, 10 min then À208C to RT,
20 min; 3) cat. [Pd
(p-ally)Cl]2/(Æ)-L3, THF, RT. [a] Unless otherwise
specified, the reaction was conducted under conditions B. [b] Isolated
yields. [c] 0.20 mmol of iodide was used. [d] [Pd(p-allyl)Cl]2 (3.75 mol%)
and (Æ)-L3 (15 mol%) were used. [e] The isolated yield of the pure
product obtained after flash chromatography on silica gel and preparative
TLC.
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
In conclusion, we have realized the Pd-catalyzed Negishi
cross-coupling reaction of 3-iodoalk-2-enoates with the
propargyl/1,2-allenyl zinc species derived from the lithiation
of 1-arylalk-1-ynes and transmetalation with ZnBr2 to afford
Typical procedure II (conditions B) for the synthesis of 4-arylalka-
2(E),4,5-trienoates (E)-6
Methyl 3-methyl-4-phenylnona-2(E),4,5-trienoate [(E)-6ad]: Under an Ar
atmosphere, a flame-dried Schlenk tube was charged with 1-phenylhex-1-
yne (1a; 79 mg, 0.50 mmol) and THF
(3 mL). The reaction vessel was cooled
to À208C, and a solution of nBuLi
(0.38 mL,
1.6m)
in
hexanes
(0.60 mmol) was added at this temper-
ature. After the mixture had been
stirred at this temperature for 1 h, a
solution of ZnBr2 (1.0 mmol, 225 mg)
in THF (2 mL) was added dropwise.
After 10 min, the cooling bath was re-
moved, and the mixture was allowed
to warm up to room temperature over
20 min to afford the zinc reagent. In
another flame-dried Schlenk tube, a
solution of catalyst and substrate in
THF was prepared by adding se-
Scheme 6. Possible pathways for the observed Z/E isomerization.
quentially
[Pd
G
(3 mg,
Chem. Eur. J. 2010, 16, 74 – 80
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
79

S. Ma et al.
0.0075 mmol), THF (1 mL), rac-L3 (14 mg, 0.030 mmol), THF (1 mL),
methyl 3-iodobut-2(Z)-enoate [(Z)-2d; 68 mg, 0.30 mmol], and THF
(1 mL) with stirring. The prepared zinc reagent was then transferred to
this mixture with a syringe at room temperature. After the reaction was
complete (TLC), it was quenched with saturated aqueous NH4Cl and ex-
tracted with diethyl ether, and the combined organic phase was dried
over MgSO4 and filtrated. The solvent was evaporated under vacuo, and
the residue was subjected to flash chromatography to afford (E)-6ad
(70 mg, 91%) as an oil. 1H NMR (400 MHz, CDCl3): d=7.28–7.21 (m,
2H), 7.21–7.13 (m, 3H), 5.59 (s, 1H), 5.54 (t, J=6.6 Hz, 1H), 3.57 (s,
3H), 2.31 (s, 3H), 2.07–2.00 (m, 2H), 1.47–1.36 (m, 2H), 0.85 (t, J=
7.2 Hz, 3H); 13C NMR (75.4 MHz, CDCl3): d=207.4, 167.4, 153.1, 136.1,
129.1, 128.4, 127.3, 116.9, 112.8, 94.5, 50.9, 30.7, 22.2, 17.5, 13.8; IR
(neat): n˜ =1938, 1717, 1615, 1491, 1167, 699 cmÀ1; MS (70 eV) m/z (%):
256 (13.76) [M]+, 227 (100); HRMS (EI) calcd for C17H20O2: 256.1463;
found: 256.1462.
The Chemistry of Organozinc Compounds (Eds.: Z. Rappoport, I.
Marek), Wiley, New York, 2006, Chapter 10.
[3] a) S. Ma, Eur. J. Org. Chem. 2004, 1175; b) N. Yasuda, J. Organomet.
Chem. 2002, 653, 279.
[4] For a review of Negishi cross-coupling reactions, see: E. Negishi, Q.
Hu, Z. Huang, G. Wang, N. Yin in The Chemistry of Organozinc
Compounds (Eds.: Z. Rappoport, I. Marek), Wiley, New York, 2006,
Chapter 11.
[5] a) S. Ma, A. Zhang, Y. Yu, W. Xia, J. Org. Chem. 2000, 65, 2287;
b) S. Ma, A. Zhang, J. Org. Chem. 2002, 67, 2287.
[6] S. Ma, G. Wang, Angew. Chem. 2003, 115, 4347; Angew. Chem. Int.
Ed. 2003, 42, 4215.
[7] For definitive studies on the structure of propargyl/allenyl lithium
species, see: a) H. J. Reich, J. E. Holladay, T. G. Walker, J. L.
Thompson, J. Am. Chem. Soc. 1999, 121, 9769; b) H. J. Reich, J. L.
Thompson, Org. Lett. 2000, 2, 783.
[8] a) S. Ma, Q. He, Angew. Chem. 2004, 116, 1006; Angew. Chem. Int.
Ed. 2004, 43, 988; b) S. Ma, Q. He, X. Zhang, J. Org. Chem. 2005,
70, 3336; c) S. Ma, Q. He, X. Jin, Synlett 2005, 514.
[9] For examples of the ligand effect on regioselectivity, see: a) S. Ku-
mar Nune, M. Tanaka, Chem. Commun. 2007, 2858; b) E. Zuidema,
E. Daura-Oller, J. J. Carbꢁ, C. Bo, P. W. N. M. van Leeuwen, Orga-
nometallics 2007, 26, 2234; c) C. Lꢁpez, S. Pꢂrez, X. Solans, M. Font-
Bardꢃa, A. Roig, E. Molins, P. W. N. M. van Leeuwen, G. P. F. van
Strijdonck, Organometallics 2007, 26, 571; d) T.-S. Lai, S. K. S. Lee,
L.-L. Yeung, H.-Y. Liu, I. D. Williams, C. K. Chang, Chem.
Commun. 2003, 620; e) R. J. van Haaren, P. H. Keeven, L. A. van
der Veen, K. Goubitz, G. P. F. van Strijdonck, H. Oevering, J. N. H.
Reek, P. C. J. Kamer, P. W. N. M. van Leeuwen, Inorg. Chim. Acta
2002, 327, 108; f) H. K. Reinius, A. O. I. Krause, Catal. Lett. 2000,
70, 149; g) Y. Xia, A. S. Dudnik, V. Gevorgyan, Y. Li, J. Am. Chem.
Soc. 2008, 130, 6940; h) W. Cabri, I. Candiani, A. Bedeschi, R. Santi,
J. Org. Chem. 1992, 57, 3558; i) W. Cabri, I. Candiani, A. Bedeschi,
R. Santi, J. Org. Chem. 1990, 55, 3654; j) S. G. Pyne, Z. Dong, Tetra-
hedron Lett. 2000, 41, 5387; k) J. W. Faller, J. C. Wilt, J. Parr, Org.
Lett. 2004, 6, 1301.
[10] P. Knochel, C. J. Rao, Tetrahedron 1993, 49, 29.
[11] G. A. Molander, E. M. Sommers, S. R. Baker, J. Org. Chem. 2006,
71, 1563.
[12] a) A. F. Littke, C. Dai, G. C. Fu, J. Am. Chem. Soc. 2000, 122, 4020;
b) J. P. Wolfe, R. A. Singer, B. H. Yang, S. L. Buchwald, J. Am.
Chem. Soc. 1999, 121, 9550.
[13] A similar coupling product was observed for the reaction of 1-phe-
nylprop-1-yne. See ref. [5a].
Acknowledgements
Financial support from the National Natural Science Foundation of
China (20732005 and 20423001), the State Basic and Research Develop-
ment Program (Grant No. 2006CB806105), and the Shanghai Municipal
Committee of Science and Technology are gratefully appreciated. We
thank Jinqiang Kuang in our group for reproducing the results listed in
Tables 3 (entries 1 and 5) and 7 (entries 3 and 7).
Keywords: allenes
· cross-coupling · ligand effects ·
palladium · regioselectivity
[1] For reviews and some selected examples, see: a) Metal-Catalyzed
Cross-Coupling Reactions (Eds.: F. Diederich, P. J. Stang), Wiley-
VCH, Weinheim, 1998; b) Transition Metal Reagents and Catalysts:
Innovations in Organic Synthesis (Ed.: J. Tsuji), Wiley, New York,
2003; c) N. Miyaura, A. Suzuki, Chem. Rev. 1995, 95, 2457; d) J.-P.
Corbet, G. Mignani, Chem. Rev. 2006, 106, 2651; e) E. Negishi, Acc.
Chem. Res. 1982, 15, 340; f) C. Amatore, A. Jutand, Acc. Chem. Res.
2000, 33, 314; g) C. E. Tucker, P. Knochel, J. Org. Chem. 1993, 58,
4781; h) P. Knochel, R. D. Singer, Chem. Rev. 1993, 93, 2117; i) R.
Giovannini, P. Knochel, J. Am. Chem. Soc. 1998, 120, 11186; j) T. J.
Korn, M. A. Schade, S. Wirth, P. Knochel, Org. Lett. 2006, 8, 725;
k) S. Sase, M. Jaric, A. Metzger, V. Malakhov, P. Knochel, J. Org.
Chem. 2008, 73, 7380; l) G. C. Fu, Acc. Chem. Res. 2008, 41, 1555;
m) C. Fischer, G. C. Fu, J. Am. Chem. Soc. 2005, 127, 4594; n) S.
Son, G. C. Fu, J. Am. Chem. Soc. 2008, 130, 2756; o) Y.-Q. Fang, M.
Lautens, J. Org. Chem. 2008, 73, 538; p) G. Dai, R. C. Larock, J.
Org. Chem. 2003, 68, 920.
[14] M. Zembayashi, K. Tamao, M. Kumada, Tetrahedron Lett. 1975, 16,
1719.
[15] The Sonagashira coupling reaction of iodopropenoates with alkynes
is stereoselective; see: R. Takeuchi, K. Tanabe, S. Tanaka, J. Org.
Chem. 2000, 65, 1558.
[16] Retention of configuration was observed in the propargyl/alkenyl
coupling product (Z)-5 in an analogous reaction, see ref. [6].
[17] A. S. Batsanov, J. P. Knowles, B. Samsam, A. Whiting, Adv. Synth.
Catal. 2008, 350, 227.
[2] a) K. Tsutsumi, S. Ogoshi, K. Kakiuchi, S. Nishiguchi, H. Kurosawa,
Inorg. Chim. Acta 1999, 296, 37; b) J. Tsuji, T. Mandai, Angew.
Chem. 1995, 107, 2830; Angew. Chem. Int. Ed. Engl. 1995, 34, 2589,
and refrences therein; c) For a recent monograph on the preparation
and reactivity of propargyl/allenyl zinc species, see: J. A. Marshall in
Received: May 14, 2009
Revised: August 18, 2009
Published online: November 17, 2009
80
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 74 – 80