Electron-deficient phosphines accelerate the heck reaction of electronrich olefins in ionic liquid

Article abstract of DOI:10.2174/157017809787003052

Using various substrates and ligands, we show that electron-deficient, bidentate phosphines are the ligands of choice for palladium-catalyzed arylation of electron-rich olefins. This is in contrast to the reaction of electron-deficient olefins, which bene

Full text of DOI:10.2174/157017809787003052

60
Letters in Organic Chemistry, 2009, 6, 60-64
Electron-Deficient Phosphines Accelerate the Heck Reaction of Electron-
rich Olefins in Ionic Liquid
Shifang Liua, Ourida Saidia, Neil Berrya, Jiwu Ruana, Alan Pettmanb, Nick Thomsonb and Jianliang
Xiao*,a
aLiverpool Centre for Materials and Catalysis, Department of Chemistry, University of Liverpool, Liverpool L69 7ZD,
UK
bPfizer Global Research and Development, Chemical Research and Development Department, Ramsgate Road, Sand-
wich, Kent, CT13 9NJ, UK
Received July 16, 2008: Revised October 27, 2008: Accepted October 31, 2008
Abstract: Using various substrates and ligands, we show that electron-deficient, bidentate phosphines are the ligands of
choice for palladium-catalyzed arylation of electron-rich olefins. This is in contrast to the reaction of electron-deficient
olefins, which benefit from electron-rich monodentate phosphines. A tentative explanation is offered based on DFT calcu-
lations.
Keywords: Heck reaction, palladium, phosphine ligand, olefin, catalysis.
The Heck reaction has been well known as one of the
most versatile methods for the formation of C–C bonds in
synthetic chemistry [1, 2]. This chemistry can occur under
diverse conditions and catalysts. Depending on the condi-
tions, the Heck reaction is believed to follow one of the two
different mechanisms, neutral pathway A or ionic pathway
B, as shown in Scheme 1.
generally favored, particularly when deactivated aryl bro-
mides and chlorides are concerned [5-9]. We now report that
contrary to this belief, the palladium-catalyzed Heck reaction
of electron-rich olefins is accelerated with electron-deficient,
bidentate phosphine ligands.
Ever since the pioneering work of Cabri [10], the easily-
available 1,3-bis(diphenylphosphino)propane (DPPP) has
established itself as the prototype ligand in the Heck reaction
of electron-rich olefins [3, 10, 11]. However, there appears to
In general, the Heck arylation of electron-deficient ole-
fins provides a terminally arylated linear product via the neu-
+
L
Ar
L
L
L
L
Ar
L
L
X
X-
B
Pd
R
Pd
Ar
R
Pd
Pd
Ar
R
L
X
X
Ar
R
R
X-
Ar
R
L
A
R
R
PdArXL2
ArX
PdArXL2
ArX
PdHXL2
base
PdHXL2
base
HX.base
PdL2
PdL2
HX.base
Scheme 1. The Heck reaction and its neutral (A) and ionic (B) pathways.
tral pathway A, while that of electron-rich olefins provides
an internally arylated branched product via the ionic path-
way B [1a, 3]. A variety of factors are known to affect the
outcome of the Heck reaction including its regioselectivity
[4-9], and the property of ligand is no exception. In the case
of electron-deficient olefins reacting via pathway A, studies
from a number of groups have shown that electron-rich, ba-
sic, sterically demanding monodentate phosphine ligands are
be no study into how the electronic properties of this type
ligand may impact on the Heck reaction [12]. In continuing
our research in using ionic liquids to promote the ionic Heck
pathway [11a-f, 13], we synthesized a series of substituted
DPPP ligands, in which the phenyl rings are para-substituted
with -Me, -OMe, -CF3, and -CN [14], as shown in Fig. (1).
This substitution would ensure that the observed effects on
the Heck reaction stem from ligand electronic rather than
steric properties. The Hammett constants are also provided,
reflecting the electron-withdrawing/donating capabilities of
the substituents used. With these ligands in hand, we subse-
quently studied the Heck reaction of a series of electron-rich
olefins in an ionic liquid, [bmim][PF6] (bmim: 1-butyl-3-
*Address correspondence to this author at the Liverpool Centre for Materi-
als and Catalysis, Department of Chemistry, University of Liverpool, Liver-
pool L69 7ZD, UK; Tel: +44 151 794 2937; Fax: +44 151 794 3589;
E-mail: j.xiao@liv.ac.uk
1570-1786/09 $55.00+.00
© 2009 Bentham Science Publishers Ltd.

Electron-Deficient Phosphines Accelerate the Heck Reaction
Letters in Organic Chemistry, 2009, Vol. 6, No. 1 61
methylimidozolium). Previous studies by our group and that
of Hallberg and Larhed have shown that electron-rich olefins
can be readily regioselectively arylated with aryl halides in
imidazolium ionic liquids without recourse to any halide
scavengers [11a-f, 15].
rich olefin 2, faster and regioselective arylation is made pos-
sible with electron-deficient bidentate phosphines.
The observation above is not an isolated case. In an effort
to determine if the chemistry could be extended to other sub-
strates, we investigated the reaction of 1 with four other elec-
tron-rich olefins. The results are summarized in Table 1. As
is clear from the table, for all the electron-rich olefins pre-
sented here, there is again a significant ligand electronic ef-
fect – the more electron-deficient the phosphines, the higher
the conversions.
ꢀp
R
Ligand
p-CNDPPP
p-CF3DPPP
p-CN
p-CF3
0.70
0.54
R
R
DPPP
H
p-Me
0
P
P
p-MeDPPP
-0.17
We recently reported the arylation of ꢀ-substituted allylic
alcohols in ionic liquids [17]. Using the electron-deficient
DPPP type ligands, the reaction was again made easier. An
example is the arylation of but-1-en-3-ol 9 with 2-
bromonaphthalene 8, shown in Table 2. As can be seen, a
significantly faster reaction along with slightly enhanced
regioselectivity was achieved with p-CNDPPP and p-
CF3DPPP. Whilst still unsatisfactory, the regioselectivity
observed with p-CNDPPP represents the best value for this
type of reaction [17].
p-MeODPPP
p-HODPPP
p-OMe
p-OH
-0.28
-0.38
R
R
Fig. (1). DPPP and derivatives.
In our initial investigation, the arylation of the bench-
mark electron-rich olefin n-butyl vinyl ether was examined
using the ligands shown in Fig. (1) [16]. In a typical reaction,
a mixture of 4-bromoacetophone 1 (1.0 mmol), n-butyl vinyl
ether 2 (2.0 mmol), Pd(OAc)2 (2.0 mol%), ligand (4.0 mol%)
and Et3N (1.5 mmol) was heated in [bmim][PF6] (1.0 mL) at
115 °C under N2 for 6 h. For comparison, the reaction was
also performed with one of the most common monodentate
ligands, PPh3 (8.0 mol%). The results are presented in eq 1.
To shed more light on ligand effects in the Heck reaction,
we also examined the arylation of a benchmark electron-
deficient olefin, methyl acrylate, as shown in eq 2. In a man-
ner similar to that for eq 1, a reaction mixture of 12 (2.0
mmol), 1 (1.0 mmol), Pd(OAc)2 (2.0 mmol%), ligand (4.0
mol%) and Et3N (1.5 mmol) in [bmim][PF6] (1.0 mL) was
heated at 115 °C under N2 for 12 h. The monodentate PPh3
(8 mol%) was again used for the purpose of comparison. The
results show that the bidentate ligands are much less effec-
tive than PPh3 for the palladium catalysis and as might be
expected, electron-rich diphosphines significantly outper-
form their electron-deficient analogues. As with other Heck
reaction of electron-deficient olefins, the regioselectivity is
in favor of the linear product 14 regardless of the ligands
used. These results highlight the difference in ligand re-
quirement for the Heck reaction of electron-rich vs -deficient
olefins, and the former is facilitated with electron-deficient,
bidentate phosphines.
As can be seen, all of the bidentate ligands are capable of
promoting the regioselective arylation of 2 with 1, affording
exclusively the ꢀ arylated product 3. This is in line with our
previous findings, that is ionic liquids promote the formation
of branched olefins by facilitating the ionic pathway B [3,
11a-f]. Contrary to one’s intuition is, however, the observa-
tion that the conversion increases with the increase in the
electron deficiency of phosphine; the conversion of the reac-
tion rose from 29% with p-MeODPPP to 74% with p-
CNDPPP. On the basis of previous studies concerning the
Heck reaction proceeding via the neutral pathway A, one
might expect a faster rate with Pd-p-MeODPPP and Pd-p-
MeDPPP. Clearly, this is not the case. Still further, using a
p-OH (ꢁp = -0.38) substituted DPPP led to a conversion of
40%, and surprisingly a mixture of regioisomer was ob-
tained, with the branched product accounting for only 37%.
Similarly, when using PPh3 as ligand, the reaction becomes
less regioselective and slower in comparison with using the
p-CF3DPPP and p-CNDPPP (eq 1). Thus, for the electron-
So why are electron-deficient, bidentate ligands better for
the Heck reaction of electron-rich olefins? The use of a bi-
dentate ligand to promote the ionic pathway B has long been
established [3, 10-15, 17]. However, the electronic effects of
such ligands have rarely been investigated and those
Pd(OAc)2
Ligand
OnBu
Br
OnBu
OnBu
+
+
(1)
Et3N
[Bmim][PF6]
115 oC, 6 h
O
O
O
3
1
2
4
Conv. (%)
3/4
p-MeODPPP
29 >99/1
41 >99/1
p-MeDPPP
DPPP
>99/1
>99/1
>99/1
86/14
47
60
p-CF3DPPP
p-CNDPPP
PPh3
74
49

62
Letters in Organic Chemistry, 2009, Vol. 6, No. 1
Liu et al.
Table 1. Arylation of Electron-Rich Olefins with 4-Bromoacetophenone Catalyzed by Pd-diphosphinea
R
Pd(OAc)2
Ligand
Br
R
R
+
+
Et3N
Solvent
O
O
O
1
5
7
6
R
Ligands
Conversion (%)f
6/7f
p-CNDPPP
p-CF3DPPP
DPPP
40
34
28
20
15
90
72
44
21
15
88
69
56
39
31
93
89
84
82
77
>99
>99
N(Me)C(O)Meb
>99
p-MeDPPP
p-MeODPPP
p-CNDPPP
p-CF3DPPP
DPPP
>99
>99
93/7
94/6
94/6
94/6
94/6
>99
c
CH2Si(Me)3
p-MeDPPP
p-MeODPPP
p-CNDPPP
p-CF3DPPP
DPPP
>99
CH2OHd
>99
p-MeDPPP
p-MeODPPP
p-CNDPPP
p-CF3DPPP
DPPP
>99
>99
74/26
74/26
74/26
75/25
75/25
(CH2)2OHe
p-MeDPPP
p-MeODPPP
aReaction conditions: 1 (1.0 mmol), Et3N (1.5 mmol), Pd(OAc)2 (2.0 mol%), ligand (4.0 mol%), at 115 oC, N2.
b5 (1.1 mmol), [bmim][PF6] (0.5 mL)-DMSO (0.5 mL), 24 h.
c5 (2.0 mmol), [bmim][PF6] (1.0 mL), 24 h.
d5 (1.5 mmol), [bmim][PF6] (0.5 mL)-DMSO (0.5 mL), 10 h.
ePd(OAc)2 (1.0 mol%), 5 (2.0 mmol), [bmim][PF6] (0.5 mL)-DMSO (0.5 mL), 6 h.
fDetermined by 1H NMR.
Table 2. Arylation of But-1-en-3-ol with 2-Bromonaphthalenea
O
Pd(OAc)2
Br
OH
Ligand
+
+
OH
Et3N
[Bmim][PF6]
8
9
10
11
Ligand
Conversion (%)b
10/11b,d
p-CNDPPP
p-CF3DPPP
DPPP
78c
94 (67c)
80
75/25
70/30
70/30
69/31
68/32
p-MeDPPP
69
p-MeODPPP
43
aReaction conditions: 8 (1.0 mmol), 9 (1.2 mmol), Et3N (1.5 mmol), Pd(OAc)2 (1.0 mol%), ligand (2.0 mol%), in [bmim][PF6] (1.0 mL), at 115 oC for 8 h under N2.
bDetermined by 1H NMR.
c2 h reaction time.
dIn some cases, product 11 includes <3% of the original alcohol.

Electron-Deficient Phosphines Accelerate the Heck Reaction
Letters in Organic Chemistry, 2009, Vol. 6, No. 1
63
Br
Pd(OAc)2
CO2Me
CO2Me
Ligand
+
CO2Me
(2)
+
Et3N
[Bmim][PF6]
O
O
O
1
12
13
14
Conv. (%)
26
13/14
p-MeODPPP
<1/99
p-MeDPPP
35
13
<1/99
<1/99
DPPP
p-CF3DPPP
p-CNDPPP
PPh3
5.2
4.0
<1/99
<1/99
<1/99
100
observed here are not immediately clear. Previous studies
including our own suggest that the ionic pathway is rate-
limited by the insertion step when using aryl iodides and
bromides and there may be a pre-equilibrium between the
neutral [Pd(Ar)XL2] and the ionic [Pd(Ar)(olefin)L2]+ spe-
cies (Pathway B, Scheme 1) [3, 11c, 18]. Assuming this to
be the case, one may partly attribute the higher rates associ-
ated with the electron-deficient DPPP analogues to a reduced
olefin insertion barrier. In a previous DFT calculation using
a model Pd-diimine catalyst depicted below, it was shown
that electron-withdrawing groups on the diimine ligand
lower the insertion barrier and render the Pd(II)-olefin inter-
mediate more stable [12d].
there is indeed a good correlation between the conversion
and activation energy (Fig. (3)).
+
R2
R2
R1
N
N
R1
Fig. (3). Correlation of conversion with olefin insertion barrier
[ln(conv) = ln (conversion with respect to that obtained with p-
OMeDPPP)].
Pd(II)
In comparison with DPPP, the electron-deficient p-
CNDPPP lowers the insertion barrier whilst the electron-rich
p–MeODPPP does the opposite. This lowering of the inser-
tion barrier could stem from an easier rotation of the olefin
from the out-of-plane to in-plane coordination due to re-
duced ꢀ back donation from Pd(II) to the olefin, and from an
enhanced positive charge on the coordinated olefin, which is
expected to facilitate the intramolecular nucleophilic attack
by the aryl group [12, 21].
Focusing on the insertion step, we carried out similar cal-
culations on three catalytic species bearing the real ligands,
[Pd(MeOCH=CH2)(Ph)L2]+ (L2 = DPPP, p-CNDPPP and p-
MeODPPP) (Fig. (2)). The calculations were carried out us-
ing density functional theory at the B3LYP/6-31* level as
implemented in PCGAMESS [19, 20]. These revealed that
In summary, the results presented in this study establish
that, contrary to the Heck reaction proceeding via the neutral
pathway A that benefits from electron-rich monodentate
ligands, the Heck reaction of electron-rich olefins character-
ized by the ionic pathway B necessitates electron-deficient,
bidentate phosphines.
ACKNOWLEDGEMENTS
Fig. (2). Structures of [Pd(MeOCH=CH2)(Ph)(DPPP)]+ minimum
(left) and transition state (right) obtained from DFT (B3LYP/6-
31*) calculations with hydrogen atoms omitted for clarity. Selected
bond lengths (Å): Minimum: Pd-P1, 2.460; Pd-P2, 2.358; Pd-C1,
2.323; Pd-C2, 2.653; Pd-C3, 2.077; C1-C2, 1.374; C2-C3, 3.091.
Transition state: Pd-P1, 2.308; Pd-P2, 2.448; Pd-C1, 2.107; Pd-C3,
2.209; C1-C2, 1.437; C2-C3, 2.109.
We thank Pfizer for financial support and Jonhson Mat-
they for the loan of palladium.
REFERENCES
[1]
(a) Larhed, M.; Hallberg, A. In Handbook of Organopalladium
Chemistry for Organic Synthesis; Negishi, E.-I., Ed.; Wiley-

64
Letters in Organic Chemistry, 2009, Vol. 6, No. 1
Liu et al.
Interscience: New York, 2002; Vol. 1: pp. 1133; (b) Knowles, J. P.;
Whiting, A. Org. Biomol. Chem., 2007, 5, 31; (c) Whitcombe, N.
J.; Hii, K. K.; Gibson, S.E. Tetrahedron, 2001, 57, 7449; (d)
Alonso, F.; Beletskaya, I.P.; Yus, M. Tetrahedron, 2005, 61, 11771;
e) Oestreich, M. Eur. J. Org. Chem., 2005, 783; (f) Crisp, G.T.
Chem. Soc. Rev., 1998, 27, 427.
J. Am. Chem. Soc., 2003, 125, 3503; (d) Von Schenck, H.; Aker-
mark, B.; Svensson, M. Organometallics, 2002, 21, 2248; (e) Von
Schenck, H.; Stromberg, S.; Zetterberg, K.; Ludwig, M.; Aker-
mark, B.; Svensson, M. Organometallics, 2001, 20, 2813.
General reviews on ionic liquid: (a) Welton, T. Coord. Chem. Rev.,
2004, 248, 2459; (b) Welton, T.; Smith P.J. Adv. Organomet.
Chem., 2004, 51, 251.
[13]
[2]
[3]
Heck, R.F. J. Am. Chem. Soc., 1971, 93, 6896.
Amatore, C.; Godin, B.; Jutand, A.; Lemaître, F. Organometallics,
2007, 26, 1757.
[14]
[15]
[16]
Vallin, K.S.A.; Zhang, Q.S.; Larhed, M.; Curran, D.P.; Hallberg,
A. J. Org. Chem., 2003, 68, 6639.
Vallin, K.S.A.; Emilsson, P.; Larhed, M.; Hallberg, A. J. Org.
Chem., 2002, 67, 6243.
[4]
(a) Amatore, C.; Godin, B.; Jutand, A.; Lemaître, F. Chem. Eur. J.,
2007, 13, 2002; (b) Datta, G.K.; Von Schenk, H.; Hallberg, A.;
Larhed, M. J. Org. Chem., 2006, 71, 3896; (c) Deeth, R.J.; Smith,
A.; Brown, J.M. J. Am. Chem. Soc., 2004, 126, 7144; (d) Popp,
B.V.; Thorman, J.L.; Morales, C.M.; Landis, C.R.; Stahl, S.S. J.
Am. Chem. Soc., 2004, 126, 14832; (e) Fristrup, P.; Le Quement,
S.; Tanner, D.; Norrby, P.-Q. Organometallics, 2004, 23, 6160; (f)
Alcazar-Roman, L.M.; Hartwig, J.F. Organometallics, 2002, 21,
491.
(a) Littke, A.F.; Fu, G.C. J. Org. Chem., 1999, 64, 10; (b) Littke,
A.F.; Fu, G.C. Angew. Chem. Int. Ed., 2002, 41, 4176.
Strieter, E.R.; Blackmond, D.G.; Buchwald, S.L. J. Am. Chem.
Soc., 2003, 125, 13978; (b) Yin, J.; Rainka, M.P.; Zhang, X.;
Buchwald, S.L. J. Am. Chem. Soc., 2002, 124, 6334; (c) Wolfe, J.
P.; Tomori, H.; Sadighi, J.P.; Yin, J.; Buchwald, S.L. J. Org.
Chem., 2000, 65, 1158.
General procedure for the Heck arylation in ionic liquid: An oven-
dried carousel reaction tube containing a stirrer bar was charged
with 4-bromoacetophone 1 (1.0 mmol), Pd(OAc)2 (0.02 mmol),
ligand (0.04 mmol), and [bmim][PF6] (1 mL) under nitrogen at
room temperature. After the mixture was degassed three times, n-
butyl vinyl ether 2 (2.0 mmol) and Et3N (1.5 mmol) were injected
sequentially. The mixture was stirred and heated at 115 °C under
N2 for 6 h. After cooled to room temperature, a small sample was
taken for NMR analysis. All of the products have previously been
fully characterised, and for the purpose of comparison of conver-
sions, the aim of this study, 1H NMR was found to be sufficient, as
the products generally contained only trace undesired products.11c
Liu, S.; Thomson, N.; Pettman, A.; Hyder, Z.; Mo, J.; Xiao, J. J.
Mol. Catal. A Chem., 2008, 279, 210.
[5]
[6]
[17]
[18]
[19]
[20]
[7]
Shaughenessy, K.H.; Kim, P.; Hartwig, J.F. J. Am. Chem. Soc.,
1999, 121, 2123; (b) Culkin, D.A.; Hartwig, J.F. Acc. Chem. Res.,
2003, 36, 234.
Portnoy, M.; Ben-David, Y.; Rousso, I.; Milstein, D. Organometal-
lics, 1994, 13, 3465.
Granovsky A.A. PC GAMESS version 7.0; http://classic.chem.
msu.su/gran/gamess/index.html
[8]
[9]
Ehrentrant, A.; Zapf, A.; Beller, M. Synlett, 2000, 1589.
Van Strjdonck, G.P.F.; Boele, M.D.K.; Kamer, P.C.J.; Vries, J.G.;
Van Leeuwen, P.W.N.M. Eur. J. Inorg. Chem., 1999, 1073.
(a) Cabri, W.; Candiani, I.; Bedeschi, A. J. Org. Chem., 1992, 57,
3558; (b) Cabri, W.; Candiani, I. Acc. Chem. Res., 1995, 28, 2.
(a) Mo, J.; Xiao, J. Angew. Chem. Int. Ed., 2006, 45, 4152; (b) Mo,
J.; Xu, L.; Ruan, J.; Liu, S.; Xiao, J. Chem. Commun., 2006, 3591;
(c) Mo, J.; Xu, L.; Xiao, J. J. Am. Chem. Soc., 2005, 127, 751; (d)
Mo, J.; Liu, S.; Xiao, J. Tetrahedron, 2005, 61, 9902; (e) Pei, W.;
Mo, J.; Xiao, J. J. Organomet. Chem., 2005, 690, 3546; (f) Xu, L.;
Chen, W.; Ross, J.; Xiao, J. Org. Lett., 2001, 3, 295; (g) Sun, L.;
Pei, W.; Shen, C. J. Chem. Res., 2006, 388; (h) Olofsson, K.;
Sahlin, H.; Larhed, M.; Hallberg, A. J. Org. Chem., 2001, 66, 544;
(i) Larhed, M.; Andersson, C.M.; Hallberg, A. Tetrahedron, 1994,
50, 285; (j) Andersson, C.M.; Larsson, J.; Hallberg, A. J. Org.
Chem., 1990, 55, 5757; (k) Davis, G.D.; Hallberg, A. Chem. Rev.,
1989, 89, 1433; (l) Cabri, W.; Candiani, I.; Bedeschi, A.; Penco, S.;
Santi, R. J. Org. Chem., 1992, 57, 1481.
The 6-31G* basis set was used for all atoms except Pd. Relativistic
effects for Pd were addressed by using SBKJC effective core po-
tential (ECP) together with the SBKJC basis set. Geometry optimi-
zations were performed with no symmetry restrictions. Vibrational
frequency calculations at B3LYP/6-31G* level of theory were used
to derive zero-point-energy and entropy contributions at 413.15K
using unscaled frequencies. Energy minima were confirmed by no
imaginary frequencies. Transition states were confirmed by only
one imaginary frequency with associated atomic motion consistent
with the mechanism. The free energy G(gas) has been calculated as
follows: G(gas) = H(gas) – TS(gas), H(gas) = H(SCF) + ZPE.
G(gas) = free energy in gas phase, H(gas) = enthalpy in gas phase,
T = Temperature, 413.15K, S(gas) = entropy in gas phase, H(SCF)
= self consistent field energy, ZPE = zero point energy assuming
harmonic oscillator approximation. Structures were visualized us-
ing Molekel.
[10]
[11]
[21]
Hahn, C. Chem. Eur. J., 2004, 10, 5888.
[12]
(a) Ambrogio, I.; Fabrizi, G.; Cacchi, S.; Henriksen, S.T.; Fristrup,
P.; Tanner, D.; Norrby, P.-O. Organometallics, 2008, 27, 3187; (b)
Surawatanawong, P.; Fan, Y.; Hall, M.B. J. Organomet. Chem.,
2008, 693, 1552; (c) Von Schenck, H.; Akermark, B.; Svensson, M.