Reduction of allylpalladium(II)chloride dimer by formation of allyloxysilanes

Article abstract of DOI:10.1055/s-2006-951516

The reduction of allylpalladium(II)chloride dimer (APC) to a Pd(0) species can be effected by reaction with alkali metal silanolates. The reduction is extremely rapid in the presence of chelating bisphosphine ligands and for a variety of silanolates. Geor

Full text of DOI:10.1055/s-2006-951516

LETTER
2921
Reduction of Allylpalladium(II)chloride Dimer by Formation of Allyloxy-
silanes
Silanolate Red
c
ution of
A
P
o
C
tt E. Denmark,* Russell C. Smith
245 Roger Adams Laboratory, Box 18, Department of Chemistry, University of Illinois, 600 S. Mathews Avenue, Urbana, IL 61801, USA
Fax +1(217)3333984; E-mail: denmark@scs.uiuc.edu
Received 16 May 2006
all of these processes, the Pd(II) complex that is charged
Abstract: The reduction of allylpalladium(II)chloride dimer (APC)
into the reaction must be reduced to a catalytically active
Pd(0) species to initiate the Pd(0)/Pd(II) cycle. Each of
to a Pd(0) species can be effected by reaction with alkali metal sil-
anolates. The reduction is extremely rapid in the presence of chelat-
these catalyst systems use different components including
ligands, solvents, bases and other additives which can in-
fluence the reductive pathway of APC to a Pd(0) catalyst.
ing bisphosphine ligands and for a variety of silanolates.
Key words: palladium, silanolate, cross-coupling, reduction,
mechanism
The reduction of Pd(II) complexes and salts to active
Pd(0) catalysts has been studied in depth.7 For example,
PdX2 salts (where X = Br, Cl and OAc) are readily re-
duced to Pd(0) in the presence of common phosphine
ligands with concomitant formation of phosphine oxides
(Scheme 1, a and b).8 In addition, secondary or tertiary
amines are capable of reducing Pd(II) sources to Pd(0)
(Scheme 1, c).9 Other heteroatom nucleophiles such as
hydroxide, alkoxide, fluoride and even water have been
shown to assist in this reduction as well.10 Potassium tert-
butoxide has been shown to act as a nucleophile in form-
ing N-heterocyclic carbene Pd(0) species through forma-
tion of allyl ethers (Scheme 1, d).11 Finally, even
hydrosilanes have been employed for the reduction of
APC in cross-coupling reactions reported from these
Since its initial discovery in 1959, allylpalladium(II)chlo-
ride dimer (APC, [C3H5PdCl]2) has been widely em-
ployed in many catalytic reactions as a palladium(0)
precursor.1 APC was initially used stoichiometrically as
an electrophile for transfer of an allyl group to carbon nu-
cleophiles such as malonate anions and enamines to pro-
vide allylated products.2 This fundamental process has
since been extended to the use of APC as a catalyst in the
presence of chiral ligands to promote asymmetric allylic
alkylations.3 Other synthetic transformations that have
used APC as a precatalyst include: (1) enantioselective
hydrosilylation of olefins,4 (2) carbostannylation of
alkynes,5 and (3) a variety of cross-coupling reactions.6 In
Pd(0)(PPh3)2(OAc)–
+
Ph3P=O
+
AcOH
+
H+
(a)
(b)
Pd(OAc)2
Pd(OAc)2
+
+
3 PPh3
5 PPh3
+
H2O
Pd(PPh3)4
+
Ph3P=O
[Pd(0)]
+
Ac2O
+
2 HCl
+
MeCHO
+
HNEt2
PdCl2[P(OPh)3]2
+
Et3N
+
H2O
(c)
(d)
Aryl
Cl
Aryl
N
N
Pd
PCy3
Ot-Bu
+
N
Cy3P Pd
t-BuOK
KCl
+
+
N
Aryl
Aryl
Me
Me
Aryl =
Me
Me
Scheme 1 Reduction pathways of Pd(II) sources: reduction by formation of (a) and (b) phosphine oxides, (c) aldehyde from imine, (d) tert-
butyl allyl ether
SYNLETT 2006, No. 18, pp 2921–2928
0
3
.1
1
.2
0
0
6
Advanced online publication: 25.10.2006
DOI: 10.1055/s-2006-951516; Art ID: S10906ST
© Georg Thieme Verlag Stuttgart · New York

2922
S. E. Denmark, R. C. Smith
LETTER
Me
Me
Me
Me
O
Si
toluene, r.t.
Si
O–K+
+
[C3H5PdCl]2
+
dppb
K
+ 1a–
2a
Scheme 2 Formation of (allyloxy)dimethylphenylsilane 2a
laboratories.12 Carbon nucleophiles, such as malonate an- purification, the corresponding allyloxysilane was identi-
ions, can also reduce APC to a Pd(0) species by nucleo- fied (Scheme 2). This led to the conclusion that the silano-
philic ‘deallylation’. The process is greatly facilitated by late was acting as a nucleophile, and was attacking the
splitting APC into mononuclear complexes in the pres- allyl group of APC to give Pd(0) species.18
ence of phosphine ligands.13
To establish the generality of this reduction pathway a se-
The development of silicon-based, cross-coupling reac- ries of potassium silanolates was surveyed and the results
tions as practiced in these laboratories has made extensive are summarized in Table 1. The reaction was conducted
use of APC as a Pd(0) precursor.14 The cross-coupling of by preforming the Pd complex by stirring APC and dppb
alkenylsilanols,15a a-alkoxyalkenylsilanols,15b arylsil- in toluene at room temperature, which produced a white
anols,15c,d and cyclic silyl ethers12,15e,f all use APC as the precipitate. The white solid was later determined to be the
Pd source under different reaction conditions. The cross- neutral bidentate h1-allyl complex as described by Jutand
coupling of alkenylsilanols employs APC without the et al.19 Following addition of the silanolate, the reaction
need for external ligands in the presence of n- mixture immediately darkened to a yellow color. The con-
Bu4NF·3H2O.15a–c Under these reaction conditions, the version to the allyloxysilane was monitored by GC analy-
presence of fluoride or water from the TBAF may serve sis using an internal standard. The formation of the
to reduce the Pd(II) catalyst to Pd(0).8a However, in the corresponding phenyl 2a and 4-methoxyphenyl(allyloxy-
coupling of arylsilanols, cesium carbonate is used in con- silane) (2c) (entries 1 and 3) was extremely rapid re-
junction with a phosphine or arsine additive.14 Further- quiring only 5 minutes to reach 100% conversion at
more, recent advances from these laboratories have room temperature. The more bulky 1-naphthylsilanolate
introduced the use of preformed silanolates to promote (K+1b–) reacted more slowly reaching only 61% conver-
cross-coupling reactions, also in the absence of ligands.16 sion in 5 minutes and requiring 45 minutes to reach 88%.
Therefore, the nature of the reduction pathway for APC A similar result was observed with the (E)-1-heptenyl-
using alkali silanolates as the coupling partners remains silanolate (K+1e–), which reacted more slowly than the
unanswered. Herein, we wish to report a study of the re- phenyl silanolate. Though no quantitative kinetic studies
duction of APC using preformed silanolates. This study were performed, the bulkier aromatic and long chain ali-
probes the generality of the reduction for a range of elec- phatic silanolates were less reactive. Finally, the electron
tronically and sterically differentiated silanolates, the ef- poor 4-trifluormethylphenylsilanolate (K+1d–) gave an in-
fect of ligands, and the role that metal counter ions play on conclusive result. The consumption of the silanolate was
the ease of reduction.
rapid, however only 33% of the expected allyloxysilane
was found. The remainder of the mass was identified as
the corresponding disiloxane.20 This discrepancy is dis-
cussed in the mechanistic analysis below.
In the context of an investigation on the mechanism of the
cross-coupling of arylsilanols with aryl halides we were
faced with answering the question of APC reduction to
understand the components in the active catalytic system. With the generality of this reduction pathway for various
Our initial efforts focused on investigating the fate of the electronically and sterically disparate silanolates now es-
ligand under the established reaction conditions {CsCO3 tablished, the focus shifted to the effect that ligands may
+ 3 H2O (3 equiv), APC (5 mol%), dppb [1,4-bis(diphen- play on the reduction. In recent years a number of investi-
ylphosphino)butane, 10 mol%], toluene, 90 °C}. At- gations have addressed the effect of ligand bite angle on
tempts to isolate the ligand following the cross-coupling the rate and selectivity of allylic alkylations.21 This behav-
reaction provided a mixture of products including recov- ior may correlate with the reduction of APC in the pres-
ered dppb as well as the oxidized forms (dppbO and ence of silanolates. It has been shown that as the bite angle
dppbO2) in varying ratios.17 The ambiguity of these results increases from dppe [1,2-bis(diphenylphosphino)ethane]
forced us to consider other possibilities for the reduction. to dppb, the turnover frequency increases dramatically in
Therefore, our attention turned to using preformed potas- the allylic alkylation.21c However, as the bite angle in-
sium dimethylphenylsilanolate (K+1a–) under anhydrous creases further through the use of the dppf [1,2-bis(di-
conditions to eliminate the action of water as a possible phenylphosphino)ferrocene], the turnover frequency
reducing agent. Following stoichiometric complexation drops because of unfavorable steric interactions between
of [C3H5PdCl]2 with dppb, the addition of preformed the phenyl groups of the ligand with the allyl group. To
(K+1a–) afforded a new organic product in addition to examine if this bite angle effect is also relevant in the
copious amounts of palladium black. Careful inspection reduction of APC, we needed to identify conditions in
1
of the H NMR spectrum showed peaks consistent with which the reaction was suitable for these studies (i.e. the
the formation of an allyl ether. Following isolation and Pd complexes must be completely soluble).
Synlett 2006, No. 18, 2921–2928 © Thieme Stuttgart · New York

LETTER
Silanolate Redution of APC
2923
Table 1 Substrate Scope in the Reduction of APC to Form Allyloxysilanes
Me
Me
Me
Me
O
Si
O–K+
Si
toluene, r.t.
+
APC (0.5 equiv)
+
dppb (1.0 equiv)
K+ 1–
2
Entry
Product
Conversion (%), 5 min Completion time (min) Conversion (%)a
1
2a
2b
100
5b
100
Me
Me
(79)c
Si
O
2
23
45d
88
Me
Si
Me
O
3
2c
100
5e
100
Me
Me
O
Si
MeO
4
5
2d
2e
28
69
30d
30e
33f
83
Me Me
Si
O
F3C
Me
Me
Me
Si
O
n
n = 5
a Yields based on GC conversion using an internal standard.
b Reaction was performed at 0.5 M.
c Yield of isolated material.
d Reaction was performed at 0.4 M.
e Reaction was performed at 0.3 M.
f Remaining mass balance was identified as disiloxane.
After a broad survey of solvents, dichloromethane was the dppb complex 5 almost instantaneously. However, the
chosen, as all reagents were soluble and stable in this sol- sodium silanolate Na+1a– was slower compared to the oth-
vent.22 Initial studies were performed at room temperature er metal silanolates (Table 3). This result is consistent
using the preformed dppb complex 5. The reduction of with the expected trend in the nucleophilicity of the differ-
APC was instantaneous upon addition of a solution of ent metal silanolates.23
K+1a– (Table 2). To facilitate useful comparisons, both
the temperature and the concentration of the reaction were
The complexes derived from the reaction of APC with a
variety of ligands have provided useful insights into the
decreased. Surprisingly, even at 0.01 M and 0 °C the re-
reactivity of these compounds.19,24 As early as 1964, it
duction was complete within 1 minute! The results of re-
was observed that these complexes were h1-allyl-
duction of complexes of other bidentate phosphines are
Pd(II)ClL2 and that the neutral complexes exist in rapid
shown in Table 2. It was found that, independent of the
metallotropic equilibrium in solution.24a This behavior
ligand used in the preformed complex, the reduction was
was contrary to earlier suppositions that the complexes
rapid and quantitative. Further, we found that APC itself
were cationic Pd(II) species. The relative reactivity of the
did not reduce in the absence of ligands at room tempera-
cationic complexes and the neutral Pd complexes were
ture (Table 2, entry 1), which is consistent with literature
later evaluated and it was determined that the cationic
reports.13
complexes react more rapidly.24b The decreased reactivity
To establish if the metal counter ion would have an effect of the neutral complexes was subsequently confirmed by
on the reduction of the APC complexes, we prepared var- Jutand and co-workers who showed that, in the presence
ious alkali metal silanolates and compared their compe- of chloride ions, neutral complexes are formed from cat-
tence in this reduction. Having already demonstrated the ionic Pd species.19 The structure of the neutral bidentate
rapid reduction of the dppb complex using the potassium L2Pd(allyl)Cl complexes were determined to be h1 by X-
silanolate K+1a–, it was not surprising that both the rubid- ray crystal structure analysis which clearly show the s-
ium Rb+1a– and cesium silanolates Cs+1a– also reduced bonding (Pd–C) of the allyl group.25 From the current
Synlett 2006, No. 18, 2921–2928 © Thieme Stuttgart · New York

2924
S. E. Denmark, R. C. Smith
LETTER
Table 2 Effect of Ligands on the Reduction of APC Using Silanolates
Me
Me
Me
Me
O
P
P
Cl
Si
CH2Cl2 (0.01 M)
0 °C
Si
O–K+
Pd
+
+ 1a–
3–6
2a
K
Entry
Ligand
Complex
Time (min)
Conversion (%)a
1
2
no ligandb
5
1
NRc
86
3
4
5
6
PPh2
Ph2P
Ph2P
3
4
5
2
1
1
96
100
100
PPh2
PPh2
Ph2P
Fe
PPh2
PPh2
a Yields based on GC conversion using an internal standard.
b [C3H5PdCl]2.
c NR = no reaction; reaction was performed at r.t.
knowledge of the reactivity as well as solution and solid would not constitute reduction to Pd(0) and therefore
state structures of L2Pd(allyl)Cl complexes, a mechanistic could not contribute to the catalytic cycle.
picture for the reduction can be formulated.
In summary, we have unambiguously established that the
There are four limiting pathways by which the allyloxysi- reduction of APC in the presence of alkali metal silano-
lane may be formed by considering the permutations of lates proceeds via the formation of allyloxysilanes. This
peripheral or internal attack of the silanolate on the neutral conclusion has preparative significance because an active
or cationic complexes (Scheme 3). Attack by the silano- Pd(0) species can be generated independent of substrate,
late at the palladium center on the neutral complexes (path ligand, and counter ion of the silanolate. Further studies to
PdN) or the cationic complex (path PdC) generates the pal- elaborate the full mechanistic picture of the fluoride-free
ladium(II) silanolate complex i. This complex must un- cross-coupling of organosilanols are currently underway
dergo reductive elimination to form the observed allyl in these laboratories and the results will be reported in due
silyl ether 2 and Pd(0). Alternatively, attack of K+1– on the course.
s-bound allyl group (path allylN) or the p-bound allyl
group (path allylC) leads directly to the formation of 2
with concomitant generation of Pd(0). Although we can-
not unambiguously rule out any of these pathways, we can
bring some previous insights to bear on the question. First,
studies on the reaction of palladium aryloxides have
shown that the reductive elimination to form C–O bonds
is slow and generally requires elevated temperatures
therefore disfavoring the paths involving attack at palladi-
um.26 Second, although it has been established that the
cationic complex is a more reactive electrophile than the
neutral complex,19,21 the use of fairly non-polar solvents
would favor the neutral form. These findings would lead
to suggest that path allylN to be the course of reduction.
General Procedure for the Preparation of Allyloxysilanes
(Allyloxy)dimethylphenylsilane (2a)27,28
Et3N (1.02 mL, 7.33 mmol, 2.0 equiv) was added to a solution of al-
lyl alcohol (0.25 mL, 3.67 mmol, 1.0 equiv) in dry THF (7.3 mL) in
a flame-dried, 3-neck, 25-mL round-bottomed flask equipped with
an Ar inlet, septum, and glass stopper. The solution was cooled to
0 °C over 10 min. Then, phenyldimethylchlorosilane (0.61 mL,
3.67 mmol) was added dropwise over 1 min. Upon addition of the
chlorosilane, a white precipitate formed. The resulting mixture was
stirred at 0 °C for 1 h whereupon Et2O (15 mL) was added. The mix-
ture was then filtered through a medium-porosity fritted funnel and
the collected solids were washed with Et2O (10 mL). The filtrate
was concentrated in vacuo and the residue was purified by column
chromatography (silica gel, hexanes–EtOAc, 30:1) to afford 645
mg (91%) of 2a as a colorless oil. The data for 2a matched those re-
ported in the literature.27
However, any mechanism for the reduction of APC must
also account for the formation of disiloxane from K+1d–.
The simplest explanation, that the disiloxane is generated
by attack of K+1d– on the allyl silyl ether 2d, could be dis-
counted by a control experiment, which clearly demon-
strated that this pathway is not operative (in the absence
of Pd; Scheme 4). Alternatively, attack of silanolate
K+1d– on the palladium silanolate i could also generate 2d
with formation of a palladium alkoxide. This pathway
(Allyloxy)dimethyl(1-naphthyl)silane (2b)
1H NMR (500 MHz, CDCl3): d = 8.32 (d, 1 H, J = 8.1 Hz), 7.86–
7.91 (m, 2 H), 7.74 (dd, 1 H, J1 = 1.0 Hz, J2 = 6.6 Hz), 7.45–7.54
(m, 3 H), 5.92 (dddd, 1 H, J1 = J2 = 4.9 Hz, J3 = 10.3 Hz, J4 = 17.1
Hz), 5.26 (dddd, 1 H, J1 = J2 = J3 = 1.7 Hz, J4 = 17.1 Hz), 5.08
(dddd, 1 H, J1 = J2 = J3 = 1.5 Hz, J4 = 10.5 Hz), 4.15 (ddd, 2 H,
J1 = J2 = 1.2 Hz, J3 = 4.9 Hz), 0.56 (s, 6 H). 13C NMR (125 MHz,
Synlett 2006, No. 18, 2921–2928 © Thieme Stuttgart · New York

LETTER
Silanolate Redution of APC
2925
Table 3 Counter-Ion Effect on the Reduction of the Neutral dppb–h1-Allyl Complex 5
Ph
Ph
Me
Me
Me
Me
P
Cl
Si
Si
O–M+
CH2Cl2 (0.01 M)
0 °C
+
O
Pd
P
Ph Ph
M+ 1a–
5
2a
Entry
Counter ion (M+)
Na
Time (min)
Conversion (%)a
1
1
10
61
94
2
3
4
K
1
1
1
100
99
Rb
Cs
99
a Yields based on GC conversion using an internal standard.
attack at allyl
allylN
attack at Pd
PdN
ArylSiMe2O–K+
Ph2P
PPh2
Cl
Pd
reductive
OSiMe2Aryl
Ph2P
PPh2
elimination
Pd
2
OSiMe2Aryl
i
Cl–
+
allylC
PdC
ArylSiMe2O–K+
Ph2P
PPh2
Pd
Scheme 3 Possible reductive pathways to form allyloxysilanes
Me
Me
Me
Me
O
Si
Si
toluene, r.t.
O–K+
NR
+
F3C
F3C
K+ 1d–
2d
Scheme 4 Control experiment to probe formation of disiloxane from 2d
CDCl3): d = 137.0, 136.9, 135.6, 133.9, 133.3, 130.5, 128.9, 128.2,
126.1, 125.6, 125.0, 114.8, 64.1, –0.5. HRMS: m/z calcd for
C15H18OSi [M+]: 242.1127; found: 242.1123. Rf = 1.293; tR = 6.15
min (HP-1, 150 °C, 3.5 min, 25 °C/min, 220 °C, 1 min, 16 psi).29
(Allyloxy)dimethyl(4-trifluoromethyl)phenylsilane (2d)
1H NMR (500 MHz, CDCl3): d = 7.70 (d, 2 H, J = 7.6 Hz), 7.62 (d,
2 H, J = 7.8 Hz), 5.91 (dddd, 1 H, J1 = J2 = 4.9 Hz, J3 = 10.0 Hz,
J4 = 17.1 Hz), 5.26 (dddd, 1 H, J1 = J2 = J3 = 1.5 Hz, J4 = 10.3 Hz),
5.11 (dddd, 1 H, J1 = J2 = J3 = 1.7 Hz, J4 = 17.1 Hz), 4.16 (ddd, 2 H,
J1 = J2 = 1.7 Hz, J3 = 4.9 Hz). 13C NMR (125 MHz, CDCl3):
d = 142.6, 136.7, 133.8, 131.5 (q, JCF = 32.2 Hz), 124.4 (q,
JCF = 3.7 Hz), 124.2 (q, JCF = 272.5 Hz), 114.9, 64.2, –1.8. 19F
NMR (470 MHz, CDCl3, ref. CFCl3): d = –63.4. HRMS: m/z calcd
for C15H15OSiF3 [M+]: 260.0844; found: 260.0843. Rf = 0.8422;
tR = 3.35 min (HP-1, 115 °C, 2.0 min, 25 °C/min, 150 °C, 2.0 min,
16 psi).
(Allyloxy)dimethyl(4-methoxy)phenylsilane (2c)
1H NMR (500 MHz, CDCl3): d = 7.54 (d, 2 H, J = 8.6 Hz), 6.95 (d,
2 H, J = 8.6 Hz), 5.92 (dddd, 1 H, J1 = J2 = 4.9 Hz, J3 = 10.0 Hz,
J4 = 17.2 Hz), 5.27 (dddd, 1 H,, J1 = J2 = J3 = 1.7 Hz, J4 = 17.2 Hz),
5.10 (dddd, 1 H, J1 = J2 = J3 = 1.5 Hz, J4 = 10.3 Hz), 4.14 (ddd, 2 H,
J1 = J2 = 1.5 Hz, J3 = 4.9 Hz), 3.83 (s, 3 H), 0.40 (s, 6 H). 13C NMR
(125 MHz, CDCl3): d = 160.9, 137.1, 135.1, 128.6, 114.6, 113.5,
63.9, 55.0, –1.6; HRMS: m/z calcd for C12H18O2Si [M+]: 222.1076;
found: 222.1077; Rf = 0.8775; tR = 3.35 min (HP-1, 150 °C, 3.5
min, 25 °C/min, 175 °C, 0.5 min, 16 psi).
Synlett 2006, No. 18, 2921–2928 © Thieme Stuttgart · New York

2926
S. E. Denmark, R. C. Smith
LETTER
Neutral dppe h1-Allyl Complex [(Ph2P(CH2)2PPh2)Pd(h1-
(E)-(Allyloxy)dimethyl(1-heptenyl)silane (2e)
1H NMR (500 MHz, CDCl3): d = 6.18 (dt, 1 H, J1 = 6.3 Hz,
J2 = 18.8 Hz), 5.92 (dddd, 1 H, J1 = J2 = 4.9 Hz, J3 = 10.0 Hz,
J4 = 17.1 Hz), 5.62 (dt, 1 H, J1 = 1.7 Hz, J2 = 18.8 Hz), 5.24 (dddd,
1 H, J1 = J2 = J3 = 1.7 Hz, J4 = 17.1 Hz), 5.08 (dddd, 1 H,
J1 = J2 = J3 = 1.5 Hz, J4 = 10.3 Hz), 4.13 (ddd, 2 H, J1 = J2 = 1.5
Hz, J3 = 4.9 Hz), 2.12 (qd, 2 H, J1 = 1.5 Hz, J2 = 7.3 Hz), 1.40 (qn,
2 H, J = 7.6 Hz), 1.25–1.33 (m, 4 H), 0.89 (t, 3 H, J = 6.8 Hz), 0.17
(s, 6 H). 13C NMR (125 MHz, CDCl3): d = 150.4, 137.5, 127.0,
114.8, 64.0, 36.8, 31.6, 28.4, 22.7, 14.2, –1.6; HRMS: m/z calcd for
C12H24OSi [M+]: 212.1597; found: 212.1596. Rf = 0.9121; tR = 1.85
min (HP-1, 150 °C, 3.5 min, 16 psi).
C3H5)Cl] (2)
1H NMR (500 MHz, CDCl3): d& nbsp;= 7.57–7.63 (m, 8 H), 7.47–
7.50 (m, 12 H), 5.80 (qn, 1 H, J = 10.7 Hz), 4.15 (br s, 4 H), 2.86 (d,
4 H, J = 18.6 Hz). 13C (125 MHz, CDCl3): d = 132.6 (t, JCP = 6.4
Hz), 131.8, 129.6 (t, JCP = 5.5 Hz), 122.9 (t, JCP = 5.5 Hz), 70.9 (t,
J
CP = 17.5 Hz), 27.5 (t, JCP = 23.0 Hz). 31P (202 MHz, CDCl3,
H3PO4): d = 51.8.
Neutral dppp h1-Allyl Complex [(Ph2P(CH2)3PPh2)Pd(h1-
C3H5)Cl·CH2Cl2] (3)
The complex was recrystallized from CH2Cl2–hexane. Data for 3:
1H NMR (500 MHz, CDCl3): d = 7.51 (br s, 8 H), 7.28–7.35 (m, 12
H), 5.58 (qn, 1 H, J = 10.8 Hz), 5.26 (s, 2H), 3.73 (br s, 4 H), 3.11
(br s, 4 H), 2.03 (br s, 2 H). 13C (125 MHz, CDCl3): d = 132.8 (br),
130.2, 128.5 (t, JCP = 4.6 Hz), 119.3 (t, JCP = 6.4 Hz), 71.3 (t,
JCP = 16.6 Hz), 53.4, 24.9 (t, JCP = 13.8 Hz), 19.4. 31P (202 MHz,
CDCl3, H3PO4): d = 6.4.
General Procedure for the Reduction of APC with Silanolates
(Table 1)
(Allyloxy)dimethylphenylsilane (2a, Table 1, entry 1)
Bis(1,4-diphenylphosphino)butane (43 mg, 0.10 mmol, 1.0 equiv)
was added to a solution of (C3H5PdCl)2 (18 mg, 0.049 mmol, 0.5
equiv) and biphenyl (6.7 mg, 0.043 mmol) in dry toluene (0.2 mL)
in a flame-dried, 2-neck round-bottomed flask equipped with an Ar
inlet and septum. Upon addition of the dppb, a white precipitate
formed. The mixture was stirred for 5 min at r.t. Thereafter, potas-
sium dimethylphenylsilanolate was added as a solid (19 mg, 0.010
mmol) or as a solution in toluene (396 mL, 0.252 M, 0.10 mmol).
The reaction mixture was maintained at r.t. and the reaction
progress was monitored by GC at certain time intervals. Sampling
of the reaction was performed by removing a 30 mL aliquot of the
mixture by syringe. The aliquot was quenched onto a 10% aq 2-
(dimethylamino)ethanethiol hydrochloride solution (0.2 mL) and
was diluted with Et2O (0.5 mL). The organic portion was filtered
through a small plug of silica gel and then was eluted with 1.0 mL
of Et2O. The aliquot was analyzed by GC (HP-1, 150 °C, 3.5 min,
25 °C/min, 225 °C, 0.5 min, 16 psi).
Neutral dppf h1-Allyl Complex [(Fe(C5H5PPh2)2)Pd(h1-
C3H5)Cl] (6)
Data for 6 matched those reported in the literature.19
General Procedure for the Metal Counter Ion Study (Table 3)
Potassium Dimethylphenylsilanolate (Table 3, entry 2)
A solution of [(Ph2P(CH2)4PPh2)Pd(h1-C3H5)Cl] (12.4 mg, 0.020
mmol, 1.0 equiv) and biphenyl (5.0 mg, 0.324 mmol) in dry CH2Cl2
(1.9 mL, 0.01 M) was charged into a flame-dried, 2-neck, round-
bottomed flask equipped with an Ar inlet and septum. The solution
was cooled to 0 °C. Thereafter, potassium dimethylphenylsilanolate
was added as a solution in CH2Cl2 (70 mL, 0.284 M, 0.020 mmol).
The mixture was maintained at 0 °C and the reaction progress was
monitored by GC at certain time intervals. Sampling of the reaction
was performed by removing a 80 mL aliquot of the mixture by sy-
ringe. The aliquot was quenched onto a 10% aq 2-(dimethylami-
no)ethanethiol hydrochloride solution (0.2 mL) and was diluted
with Et2O (0.5 mL). The organic portion was filtered through a
small plug of silica gel and then was eluted with 1.0 mL of Et2O.
The aliquot was analyzed by GC (HP-1, 150 °C, 3.5 min, 25 °C/
min, 225 °C, 0.5 min, 16 psi).
General Procedure for the Complex Study (Table 2)
Neutral dppb h1-Allyl Complex [(Ph2P(CH2)4PPh2)Pd(h1-
C3H5)Cl] (5, Table 2, entry 4)
A solution of [(Ph2P(CH2)4PPh2)Pd(h1-C3H5)Cl] (12.4 mg, 0.020
mmol, 1.0 equiv) and biphenyl (5.0 mg, 0.324 mmol) in dry CH2Cl2
(1.9 mL, 0.01 M) was charged into a flame-dried, 2-neck, round-
bottomed flask equipped with an Ar inlet and septum. The reaction
mixture was cooled to 0 °C. Thereafter, potassium dimethylphenyl-
silanolate was added as a solution in CH2Cl2 (70 mL, 0.284 M, 0.020
mmol). The reaction was maintained at 0 °C and the reaction
progress was monitored by GC at certain time intervals. Sampling
of the reaction was performed by removing a 80 mL aliquot of the
mixture by syringe. The aliquot was quenched onto a 10% aq 2-
(dimethylamino)ethanethiol hydrochloride solution (0.2 mL) and
was diluted with Et2O (0.5 mL). The organic portion was filtered
through a small plug of silica gel and was then eluted with 1.0 mL
of Et2O. The aliquot was analyzed by GC (HP-1, 150 °C, 3.5 min,
25 °C/min, 225 °C, 0.5 min, 16 psi).
Preparation of Sodium Dimethylphenylsilanolate (Na+1a–)
To a suspension of NaH (81 mg, 3.31 mmol) in THF (4.5 mL) in a
flame-dried, 50-mL scintillation flask equipped with a 3-way Ar
adapter was added dimethylphenylsilanol (503.4 mg, 3.31 mmol) as
a solution in THF (2.0 mL) dropwise over 5 min. The resultant so-
lution was stirred for 30 min further. The volatiles were removed in
vacuo to give a semi-solid. The material was washed with dry hex-
ane (2.0 mL) and the volatiles were subsequently removed in vacuo
to afford 500 mg (87%) of Na+1a– as a white, powder solid. Data for
Na+1a–: 1H NMR (500 MHz, THF-d8): d = 7.58 (d, 2 H, J = 6.3 Hz),
7.15–7.24 (m, 3 H), 0.16 (s, 6 H).
General Procedure for the Preparation of the Neutral h1-allyl
Complexes19
Preparation of Potassium Dimethylphenylsilanolate (K+1a–)14
To a suspension of KH (356 mg, 8.87 mmol, 1.3 equiv) in THF (9.1
mL) in a flame-dried, 50-mL scintillation flask equipped with a 3-
way Ar adapter was added dimethylphenylsilanol (1.039 g, 6.821
mmol) as a solution in THF (2.0 mL) dropwise over 5 min. The re-
sultant mixture was stirred for 30 min further, and then was added
by syringe into two oven-dried vials previously purged with argon.
The vials were then centrifuged (1100 rpm) for 10 min, and the su-
pernatant was removed by syringe, with care taken not to disturb the
excess KH and KOH precipitate at the bottom of the vials. The su-
pernatant was then placed in a preweighed, flame dried, argon-
purged one-neck round-bottomed flask. The volatiles were removed
in vacuo to give a semi-solid. The material was washed with dry
hexane (2.0 mL) and the volatiles were subsequently removed in
Neutral dppb h1-Allyl Complex [(Ph2P(CH2)4PPh2)Pd(h1-
C3H5)Cl] (5)
Bis(1,4-diphenylphosphino)butane (291.2 mg, 0.6828 mmol, 2.0
equiv) was added to a solution of (C3H5PdCl)2 (124.9 mg, 0.341
mmol, 1.0 equiv) in dry toluene (15.0 mL) in a flame-dried, 2-neck,
25-mL, round-bottomed flask equipped with an Ar inlet and
septum. The resulting mixture was stirred for 30 min at r.t. at
which point the mixture was filtered through a medium-porosity
fritted funnel. The solids were further dried using high vacuum
(0.05 mmHg) overnight to afford 401 mg (96%) of 5 as a white sol-
id. Data for 5 matched those reported in the literature.19
Synlett 2006, No. 18, 2921–2928 © Thieme Stuttgart · New York

LETTER
Silanolate Redution of APC
2927
vacuo two times to give 1.150 g (89%) of K+1a– as an off-white sol-
id. Data for K+1a–: 1H NMR (500 MHz, THF-d8): d = 7.55 (d, 2 H,
J = 7.8 Hz), 7.21 (t, 2 H, J = 7.6 Hz), 7.13–7.16 (m, 1 H), 0.07 (s, 6
H).
(6) For representative examples, see: (a) Hiyama, T.; Hatanaka,
Y. Pure Appl. Chem. 1994, 66, 1471. (b) Wallow, T. I.;
Novak, B. M. J. Org. Chem. 1994, 59, 5034. (c) Feuerstein,
M.; Doucet, H.; Santelli, M. Tetrahedron Lett. 2004, 45,
8443.
Potassium Dimethyl(1-naphthyl)silanolate (K+1b–)
1H NMR (500 MHz, THF-d8): d = 8.67–8.69 (m, 1 H), 7.69–7.78
(m, 3 H), 7.31–7.38 (m, 3 H), 0.22 (s, 6 H). 13C NMR (125 MHz,
THF-d8): d = 147.7, 138.7, 134.8, 132.7, 129.9, 129.5, 128.5, 126.0,
125.7, 125.4, 5.1.
(7) Trzeciak, A. M.; Ziólkowski, J. J. Coord. Chem. Rev. 2005,
249, 2308.
(8) (a) Mason, M. R.; Verkade, J. G. Organometallics 1992, 11,
2212. (b) Amatore, C.; Jutand, A.; Barki’, M. A. M.
Organometallics 1992, 11, 3009. (c) Ozawa, F.; Kubo, A.;
Hayashi, T. Chem. Lett. 1992, 2177. (d)Amatore, C.; Carre,
E.; Jutand, A.; Barki’, M. A. M. Organometallics 1995, 14,
1818. (e) Amatore, C.; Jutand, A. J. Organomet. Chem.
1999, 576, 254.
Potassium Dimethyl(4-methoxy)phenylsilanolate (K+1c–)
1H NMR (500 MHz, THF-d8): d = 7.45 (d, 2 H, J = 8.5 Hz), 6.79 (d,
2 H, J = 8.3 Hz), 3.72 (s, 3 H), 0.05 (s, 6 H).
(9) Treciak, A. M.; Ciunik, Z.; Ziólkowski, J. J.
Organometallics 2002, 21, 132.
(10) Beletskaya, I. P.; Cheprakov, A. V. Chem. Rev. 2000, 100,
3009; and references therein.
(11) (a) Viciu, M. S.; Germaneau, R. F.; Navarro-Fernandez, O.;
Stevens, E. D.; Nolan, S. P. Organometallics 2002, 21,
5470. (b) Viciu, M. S.; Navarro, O.; Germaneau, R. F.;
Kelly, R. A.; Sommer, W.; Marion, N.; Stevens, E. D.;
Cavallo, L.; Nolan, S. P. Organometallics 2004, 23, 1629.
(12) Denmark, S. E.; Kobayashi, T. J. Org. Chem. 2003, 68,
5153.
(13) Trost, B. M.; Fullerton, T. J. J. Am. Chem. Soc. 1973, 95,
292.
(14) Denmark, S. E.; Sweis, R. F. In Metal-Catalyzed Cross-
Coupling Reactions, Vol. 1; de Meijere, A.; Diederich, F.,
Eds.; Wiley-VCH: Weinheim, 2004, Chap. 4.
(15) (a) Denmark, S. E.; Wehrli, D. Org. Lett. 2000, 2, 565.
(b) Denmark, S. E.; Neuville, L. Org. Lett. 2000, 2, 3221.
(c) Denmark, S. E.; Ober, M. H. Org. Lett. 2003, 5, 1357.
(d) Denmark, S. E.; Ober, M. H. Adv. Synth. Catal. 2004,
346, 1703. (e) Denmark, S. E.; Yang, S.-M. Tetrahedron
2004, 60, 9695. (f) Denmark, S. E.; Yang, S.-M. J. Am.
Chem. Soc. 2004, 126, 12432.
Potassium Dimethyl(4-trifluoromethyl)phenylsilanolate
(K+1d–)
1H NMR (500 MHz, THF-d8): d = 7.73 (d, 2 H, J = 7.3 Hz), 7.51 (d,
2 H, J = 7.6 Hz), 0.12 (s, 6 H).
Preparation of Rubidium Dimethylphenylsilanolate (Rb+1a–)
In a dry box, a one-neck flask was charged with dry, degassed ben-
zene (17 mL) followed by rubidium metal (296 mg, 3.46 mmol). To
this suspension was added dimethylphenylsilanol (554 mg, 3.64
mmol, 1.05 equiv) dropwise over 5 min. The resulting mixture was
stirred for 30 min further, and then was filtered through a medium-
porosity fritted funnel into a preweighed one-neck flask fitted with
a vacuum stopcock adaptor. The flask was removed from the dry
box and the solvent evaporated in vacuo to give a semi-solid. The
residue was transferred back into the dry box and was washed with
dry hexane (5 mL) and filtered through a medium-porosity fritted
funnel. The collected solids were further washed with dry hexane
(3 × 5 mL). The solids were placed in a flame-dried, 15-mL recov-
ery flask equipped with a vacuum stopcock adaptor and any excess
volatiles were removed in vacuo to give 302 mg (37%) of Rb+1a– as
1
a white, powdered solid. Data for Rb+1a–: H NMR (500 MHz,
THF-d8): d = 7.55 (dd, 2 H, J1 = 1.2 Hz, J2 = 7.8 Hz), 7.12–7.21 (m,
3 H), 0.07 (s, 6 H).
(16) Denmark, S. E.; Baird, J. D. Chem. Eur. J. 2006, 12, 4954.
(17) Denmark, S. E.; Ober, M. H., unpublished results.
(18) The formation of Pd(0) from this process was confirmed by
the identification of dppbPd(dba) in the reduction mixture
from K+1a–, dpppPd(allyl)Cl, and dba. In this experiment, no
palladium black was observed. The 31P NMR spectrum
(CDCl3) of this reaction mixture showed two broad peaks at
d = 23.3 and 17.2 ppm which are in good agreement with
those reported by Jutand for this compound (in THF) at d =
21.3 and 17.5 ppm. No free dppb was observed. See:
Amatore, C.; Broeker, G.; Jutand, A.; Khalil, F. J. Am.
Chem. Soc. 1997, 119, 5176.
Preparation of Cesium Dimethylphenylsilanolate (Cs+1a–)
Following the same procedure for the preparation of the rubidium
silanolate gave 272 mg (90%) of Cs+1a– as a white, powdered solid.
Data for Cs+1a–: 1H NMR (500 MHz, THF-d8): d = 7.56 (dd, 2 H,
J1 = 1.2 Hz, J2 = 7.8 Hz), 7.18–7.22 (m, 2 H), 7.12–7.15 (m, 1 H),
0.09 (s, 6 H).
Acknowledgment
We are grateful to the NIH RO1 GM63167. R.C.S. acknowledges
the University of Illinois for the Carl S. Marvel Graduate Fel-
lowship.
(19) Cantat, T.; Génin, E.; Giroud, C.; Meyer, G.; Jutand, A. J.
Organomet. Chem. 2003, 687, 365.
(20) The presence of the disiloxane was confirmed by GC-MS
analysis. This peak was the major constituent in the GC
chromatogram as determined by relative area percents.
(21) (a) Åkermark, B.; Zetterberg, K.; Hansson, S.;
Krakenberger, B.; Vitagliano, A. J. Organomet. Chem.
1987, 335, 133. (b) Oslob, J. D.; Åkermark, B.; Helquist, P.;
Norrby, P. Organometallics 1997, 16, 3015.
(c) Kranenburg, M.; Kamer, P. C. J.; van Leeuwen, P. W. N.
M. Eur. J. Inorg. Chem. 1998, 25. (d) van Haaren, R. J.;
Goubitz, K.; Fraanje, J.; van Strijdonck, G. P. F.; Oevering,
H.; Coussens, B.; Reek, J. N. H.; Kamer, P. C. J.; van
Leeuwen, P. W. N. M. Inorg. Chem. 2001, 40, 3363.
(22) The stability of the silanolate was tested using 1H NMR
spectroscopy. The 1H NMR spectrum of a 0.075 M solution
of K+1a– in CH2Cl2 was unchanged after standing at r.t. for
5 h.
References and Notes
(1) (a) Smidt, J.; Hafner, W. Angew. Chem. 1959, 71, 284.
(b) Hüttel, R.; Kratzer, J. Angew. Chem. 1959, 71, 456.
(2) Tsuji, J.; Takahashi, H.; Morikawa, M. Tetrahedron Lett.
1965, 49, 4387.
(3) Trost, B. M. Chem. Rev. 1996, 96, 395; and references
therein.
(4) Hayashi, T.; Uozumi, Y. J. Am. Chem. Soc. 1991, 113, 9887.
(5) Shirakawa, E.; Yoshida, H.; Kurahashi, T.; Nakao, Y.;
Hiyama, T. J. Am. Chem. Soc. 1998, 120, 2975.
Synlett 2006, No. 18, 2921–2928 © Thieme Stuttgart · New York

2928
S. E. Denmark, R. C. Smith
LETTER
(23) Kurts, A. K.; Sakembaeva, S. M.; Beletskaya, I. P.; Reutov,
O. A. Zh. Org. Chim. 1974, 10, 1572.
(24) (a) Powell, J.; Shaw, B. L. J. Chem. Soc. A 1967, 1839.
(b) Åkermark, B.; Åkermark, G.; Hegedus, L. S.; Zetterberg,
K. J. Am. Chem. Soc. 1981, 103, 3037.
(25) (a) Braunstein, P.; Naud, F.; Dedieu, A.; Rohmer, M.-M.;
DeCain, A.; Rettig, S. J. Organometallics 2001, 20, 2966.
(b) Kollmar, M.; Helmchen, G. Organometallics 2002, 21,
4711.
(26) However, this claim has not been established for
allylpalladium complexes. See: Mann, G.; Hartwig, J. F. J.
Am. Chem. Soc. 1996, 118, 13109.
(27) Gregg, B. T.; Cutler, A. R. Organometallics 1994, 13, 1039.
(28) The procedure was modified from: Alonso, E.; Guijarro, D.;
Yus, M. Tetrahedron 1995, 51, 11457.
(29) Response Factor (Rf) calculated using: Rf = (mmol of
internal standard*area of x)/(mmol of x*area of internal
standard).
Synlett 2006, No. 18, 2921–2928 © Thieme Stuttgart · New York