Mechanistic studies on the SCS-pincer palladium(II)-catalyzed tandem stannylation/electrophilic allylic substitution of allyl chlorides with hexamethylditin and benzaldehydes

Article abstract of DOI:10.1002/chem.201203049

This paper describes a mechanistic study of the SCS-pincer Pd ^II-catalyzed auto-tandem reaction consisting of the stannylation of cinnamyl chloride with hexamethylditin, followed by an electrophilic allylic substitution of the primary tandem-reaction product with 4-nitrobenzaldehyde to yield homoallylic alcohols as the secondary tandem products. As it turned out, the anticipated stannylation product, cinnamyl trimethylstannane, is not a substrate for the second part of the tandem reaction. These studies have provided insight in the catalytic behavior of SCS-pincer Pd^II complexes in the auto-tandem reaction and on the formation and possible involvement of Pd⁰ species during prolonged reaction times. This has led to optimized reaction conditions in which the overall tandem reaction proceeds through SCS-pincer Pd^II-mediated catalysis, that is, true auto-tandem catalysis. Accordingly, this study has provided the appropriate reaction conditions that allow the pincer catalysts to be recycled and reused. Minor isomer, major importance: A mechanistic study of the SCS-pincer Pd-catalyzed auto-tandem reaction is described consisting of the stannylation of cinnamyl chloride, followed by an electrophilic allylic substitution of the primary tandem reaction product to yield homoallylic alcohols (see scheme; SCS=[2,6-(CH₂SPh)₂C₆H₃] ^-). This study pointed out that the anticipated stannylation product, cinnamyl trimethylstannane, is not a substrate for the second part of the tandem reaction. Copyright

Full text of DOI:10.1002/chem.201203049

DOI: 10.1002/chem.201203049
Mechanistic Studies on the SCS-Pincer Palladium(II)-Catalyzed Tandem
Stannylation/Electrophilic Allylic Substitution of Allyl Chlorides with
Hexamethylditin and Benzaldehydes**
Niels J. M. Pijnenburg, Yves H. M. Cabon, Gerard van Koten, and
Robertus J. M. Klein Gebbink*[a]
Abstract: This paper describes a mech-
anistic study of the SCS-pincer PdII-cat-
alyzed auto-tandem reaction consisting
of the stannylation of cinnamyl chlo-
namyl trimethylstannane, is not a sub-
strate for the second part of the
tandem reaction. These studies have
provided insight in the catalytic behav-
ior of SCS-pincer PdII complexes in the
auto-tandem reaction and on the for-
mation and possible involvement of
Pd0 species during prolonged reaction
times. This has led to optimized reac-
tion conditions in which the overall
tandem reaction proceeds through
SCS-pincer PdII-mediated catalysis, that
is, true auto-tandem catalysis. Accord-
ingly, this study has provided the ap-
propriate reaction conditions that allow
the pincer catalysts to be recycled and
reused.
ACHTUNGTRENNUNGride with hexamethylditin, followed by
an electrophilic allylic substitution of
the primary tandem-reaction product
with 4-nitrobenzaldehyde to yield ho-
moallylic alcohols as the secondary
tandem products. As it turned out, the
anticipated stannylation product, cin-
Keywords: homogeneous catalysis ·
palladium · pincers · reaction mech-
anisms · substitution reactions
Introduction
pincer palladium complexes[7] or in an auto-tandem system
by using a single SCS-pincer palladium complex catalyzing
ECE-pincer palladium complexes have proven to be versa-
tile catalysts that show unique catalytic properties due to
both
reactions
(SCS=[2,6-(CH2SPh)2C6H3]ꢀ;
see
Scheme 1).[11]
ꢀ
1) the strong Pd C s bond that provides a stable ligand–
metal manifold for catalysis and 2) the limitation of availa-
ble coordination sites due to the tridentate nature of the
ECE-pincer ligand. These properties modify the catalytic
applications of palladium significantly.[1–6] Palladium pincer
complexes serve as excellent catalysts for, among others,
the stannylation of allylic chlorides, propargylic chlorides,
allylic alcohols, phosphonates and epoxides by using hexaal-
kylditin, as was published recently by Szabꢀ and co-work-
ers.[7,8] The same group also reported that palladium pincer
complexes can act as catalysts for the electrophilic allylic
substitution of allylstannanes with imines, sulfonimines, and
aldehydes.[9,10] Since some of the products of the stannyla-
tion reactions (e.g., allyl trialkylstannanes) can act as a
starting material for the electrophilic addition, studies have
been performed to combine these two reactions in a one-
pot procedure by using two different orthogonal ECE-
Scheme 1. Auto-tandem catalytic formation of 1-(4-nitrophenyl)-2-
phenyl-3-buten-1-ol catalyzed by the SCS-pincer Pd complex 1MeCN
[a] Dr. N. J. M. Pijnenburg, Dr. Y. H. M. Cabon, Prof. Dr. G. van Koten,
Prof. Dr. R. J. M. Klein Gebbink
.
Organic Chemistry and Catalysis
Debye Institute of Nanomaterials Science
Faculty of Science, Utrecht University
Universiteitsweg 99
3584 CH Utrecht (The Netherlands)
Fax : (+31)30-252-3615
A mechanism consisting of two catalytic cycles was pro-
posed for this auto-tandem reaction (see Scheme 2). This
mechanism is a combination of the mechanisms of the two
separate reactions.[7,9] First the SCS-pincer PdII complex
1MeCN reacts with hexamethylditin to give the SCS-pincer
E-mail: r.j.m.kleingebbink@uu.nl
[**] SCS=[2,6-(CH2SPh)2C6H3]ꢀ
PdII-SnMe3 complex 1SnMe (see Scheme 2, A). In the initial
3
4858
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 4858 – 4868

FULL PAPER
leaching of 27.5% Pd after four
runs, similar to the observed
amount of free ligands as ana-
lyzed by NMR spectroscopy.
This palladium leaching indi-
cates two important drawbacks
in the compartmentalized auto-
tandem catalysis setup. Firstly,
leaching decreases the recycling
efficiency of the dendritic cata-
lyst, because the amount of cat-
alytic centers in the dendritic
catalyst decreases with each
reuse. Secondly, if leaching
takes place, it is unclear wheth-
er the Pd0 species plays a role
as catalyst and thus contributes
to the product formation. In
case these species are catalyti-
cally active, this might have im-
plications for several reaction
parameters, including the over-
all activity and product selectiv-
Scheme 2. Proposed mechanism for the SCS-pincer Pd-catalyzed tandem stannylation/electrophilic allylic sub-
stitution reaction between allyl chlorides, hexamethylditin and benzaldehydes (X=Cl, MeCN).[11]
report [Pd
lyst, whereas recently we have reported on the use of [Pd-
(SCS)Cl] (1Cl) as catalyst.[12] Complex 1SnMe undergoes nu-
(SCS)
E
ity, as was described by Jones et al. in their review on active
species in palladium-catalyzed Heck and Suzuki–Miyaura
reactions.[13] They demonstrated, for example, that all the
catalytic activity in the Heck coupling of n-butyl acrylate
with iodobenzene catalyzed by soluble polymer-supported
SCS-pincer Pd complexes was caused by the leached Pd0
species and that the starting SCS-pincer PdII complexes
were inactive.[14]
Here, we report a more detailed study of the auto-tandem
reaction to obtain an insight in the catalytic behavior of
SCS-pincer PdII complexes and on the formation and possi-
ble involvement of Pd0 species in catalysis. The aim of this
study was to find optimized reaction conditions in which the
formation of potentially active Pd0 species can be avoided,
to create a catalytic system in which SCS-pincer PdII com-
plexes are the only catalytically active species. Under such
optimized conditions, the use of dendritic SCS-pincer palla-
dium complexes in a compartmentalized setup would allow
for their consecutive catalytic application without catalyst
decomposition and, accordingly, under constant operation
conditions. In our present studies, we have used monomeric
ACHTUNGTRENNUNG
cleophilic substitution with allyl chloride yielding allyltrime-
thylstannane and recovering the SCS-pincer Pd complex 1X
(Scheme 2, step B). In the second cycle, the regained cata-
lyst 1X undergoes a second transmetalation reaction in
which it reacts with the formed allyl stannane leading to the
formation of the reactive h1-allylpalladium complex
(Scheme 2, C). This complex then acts as a nucleophile to-
wards 4-nitrobenzaldehyde (Scheme 2, D). As the highest
nucleophilicity resides on the g-carbon atom of the allyl
fragment of 1h1-allyl, this reaction leads to the formation of a
mixture of syn- and anti homo-allylic alcohol products.
After product dissociation, complex 1X is regenerated in the
last step (Scheme 2, E).
Recent investigations on the recycling of dendritic SCS-
pincer Pd-catalysts have raised questions concerning this
PdII-only mechanism. In these investigations the catalytic
behavior of dendritic SCS-pincer palladium complexes in
the tandem reaction was studied both in homogeneous solu-
tion and in a membrane dialysis bag.[12] In particular, in the
compartmentalized dialysis bag experiments considerable
palladium leaching was observed by inductively coupled
ꢀ
SCS-pincer Pd Cl complexes in a series of kinetic experi-
ments in combination with DFT calculations. These results
have led to a more detailed mechanistic insight, which led
to the development of improved conditions for running the
tandem reaction of cinnamyl chloride with hexamethylditin
and 4-nitrobenzaldehyde by using dendritic SCS-pincer PdII
complexes in a compartmentalized reaction setup.
1
plasma mass spectrometry (ICP-MS) and H NMR analysis.
It was found that after four consecutive runs approximately
30–40% of the pincer moieties of the dendritic catalyst did
no longer contain a palladium center, indicating that Pd-
leaching had occurred through release of palladium from
the SCS-pincer ligands. These individual palladium centers
will agglomerate to palladium nanoparticles, and penetra-
tion of the palladium nanoparticle through the dialysis bag
may or may not occur, depending on its size. ICP-MS analy-
sis of the outer membrane solutions showed a cumulative
Results and Discussion
Kinetic reaction profile: An ideal tandem reaction consist-
ing of two subsequent reactions deals with two rate con-
Chem. Eur. J. 2013, 19, 4858 – 4868
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4859

J. M. Klein Gebbink et al.
stants: k1 for reaction A+B!C, and k2 for the subsequent
reaction C+D!E (see Scheme 3). In this simplified model
it is assumed that all reagents are orthogonal, that is, they
do not react with each other in the absence of a catalyst.
For the stannylation/electro-
philic substitution tandem reac-
tion, this model can be used by
assigning cinnamyl chloride as
A, hexamethylditin as B, pri-
Scheme 3. Simplified
model
mary product cinnamyl tri
ylstannane as C, 4-nitrobenz
dehyde as and secondary
product trimethyl(1-(4-nitro-
phenyl)-2-phenylbut-3-enyl-
ACHTUNGTRENNUNGoxy)ACHTUNGTRENNUNGstanACHTUNGTRENNUNGnane (or its aqueous workup reaction product 1-(4-
ACHTUNGTRENNUNG
for a tandem reaction with re-
agents A, B, and D and two
subsequent reactions with rate
constants k1 and k2.
A
R
ACHTUNGTRENNUNG
A
R
D
A
R
ACHTUNGTRENNUNG
nitrophenyl)-2-phenyl-3-buten-1-ol) as E.
Figure 2. Experimental kinetic profile for the SCS-pincer Pd-catalyzed
The kinetic profile of this tandem reaction will vary with
the values of the rate constants k1 and k2. With the input of
the experimental data that was collected for the SCS-pincer
palladium-catalyzed tandem reaction earlier by our group,[12]
we estimated k1 =8ꢂ10ꢀ3 molsꢀ1 and k2 =4ꢂ10ꢀ4 molsꢀ1. By
using these estimated values, a theoretical kinetic profile for
this tandem reaction (see Figure 1) was calculated by
using:[15]
tandem reaction. Conditions: cinnamyl chloride (0.80 mmol), hexa
ACHTUNGTRENNUNGmeth-
AHCTUNGTRENNUNG
^
(2 mol%) in THF (6 mL; ambient temperature, N2 atmosphere). : cin-
~
&
namyl chloride; : cinnamyl trimethylstannane; : trimethyl(1-(4-nitro-
phenyl)-2-phenylbut-3-enyloxy)stannane.
plexes,[12] thereby showing the single-site nature of each cat-
alytic center in these dendritic complexes. For this reason,
the present studies were carried out with monomeric SCS-
pincer Pd complex 1Cl.
The experimental kinetic profile shows that in the first 5 h
of the reaction cinnamyl chloride is consumed relatively fast
(84% conversion) and both the primary product cinnamyl
trimethylstannane and the secondary product 1-(4-nitro-
phenyl)-2-phenylbut-3-en-1-ol are formed in 50 and 34%
yields, respectively (see Part I in Figure 2). The formation of
1-(4-nitrophenyl)-2-phenylbut-3-en-1-ol did not show the si-
nusoidal kinetic curve as predicted for the formation of a
secondary product in a tandem reaction (see Figure 1). In-
stead, the instant formation of both the primary and sec
ACHTUNGTRENNUNGond-
AHCTUNGTRENNUNG
being formed in substantial amounts after as early as 15 min
(typically around 20–30% yields). In the second part of the
reaction (Part II in Figure 2), the reaction progression sud-
denly stops, resulting in a period with minor changes in the
concentration of the starting material, or the primary and
secondary products. After 24 h, in the third and last part of
the reaction (Part III in Figure 2), the reaction catalysis
starts again leading to a slow but complete conversion of
Figure 1. Theoretical kinetic profile for the tandem reaction with k1 =8ꢂ
10ꢀ3 molsꢀ1 and k2 =4ꢂ10ꢀ4 molsꢀ1. A: cinnamyl chloride; C: cinnamyl
trimethylstannane; E: trimethyl(1-(4-nitrophenyl)-2-phenylbut-3-en
oxy)stannane.
ACHTUNGERTNyNUNG l-
N
ACHTUNGTRENNUNG
*
*
½Aꢁ ¼ ½Aꢁ0 expðꢀk1 tÞ,
the primary product, cinnamyl trimethylstannane, to sec
ary product 1-(4-nitrophenyl)-2-phenylbut-3-en-1-ol in
days.
ACHTUNGTRENNUNGond-
G
5
*
*
*
*
½Bꢁ ¼ ½Aꢁ0 ½k1=ðk2ꢀk1Þꢁ ½expðꢀk1 tÞꢀexpðꢀk2 tÞꢁ
*
*
*
*
*
In conclusion, the observed reaction profile shows two
distinct sections. One section (Part I, Figure 2) that leads to
close to complete consumption of starting material and
rapid formation of both primary and approximately 35% of
the secondary product trimethyl(1-(4-nitrophenyl)-2-phenyl-
but-3-enyloxy)stannane. The second section involves a com-
bination of parts II and III (Figure 2), in which consumption
of the primary product to form the secondary product goes
½Cꢁ ¼ ½Aꢁ0 ½1 þ ð½k1 expðꢀk2 tÞꢀk2 expðꢀk1 tÞꢁ=½k2ꢀk1ꢁÞꢁ
The experimental kinetic profile of the tandem reaction
catalyzed by SCS-pincer Pd complex 1Cl reveals several dif-
ferences between the theoretical and experimental kinetic
profile for this reaction (see Figure 2). Similar reaction pro-
files were also observed for other monomeric SCS-pincer Pd
complexes as well as for dendritic SCS-pincer Pd com-
4860
www.chemeurj.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 4858 – 4868

SCS-Pincer Palladium-Catalyzed Substitution
FULL PAPER
through a significant lag phase and ultimately leads to
almost quantitative formation of trimethyl(1-(4-nitrophen-
yl)-2-phenylbut-3-enyloxy)stannane. These observations
strongly suggest that instead of a single catalytic species, the
involvement of at least two catalytically active species seems
to be required to describe the experimental kinetic profile
of the tandem reaction.
tored at regular intervals (see Figure 3). We found that the
kinetic order of the catalyst is 1.7 for this reaction (deter-
mined at 10% substrate conversion; see the inset in
To investigate these findings in more detail, a series of ki-
netic experiments were performed. Firstly, the two individu-
al reactions that form the tandem reaction were investigated
independently to gain more insight in the behavior of cata-
lyst 1Cl in the separate reaction steps. Secondly, control ex-
periments were carried out in which the reactivity of the
substrates with respect to each other and/or the catalyst was
investigated. The results of these experiments have allowed
us to propose a new mechanism that explains the full experi-
mental kinetic reaction profile as shown in Figure 2. This
mechanism is corroborated by additional DFT calculations.
Figure 3. The stannylation of cinnamyl chloride over time using different
amounts of catalyst 1Cl. Inset: ln ([1Cl]) versus time at 10% cinnamyl
chloride conversion.
First reaction step (stannylation):The conversion of cinna-
ACHTUNGTRENNUNGmyl chloride in the first step of the tandem reaction, that is,
the stannylation with hexamethylditin, was investigated by
using THF and CH2Cl2 as solvents and in the absence and
presence of 4-nitrobenzaldehyde. In these experiments the
loading of catalyst 1Cl was fixed at 2 mol%. The results of
these experiments are depicted in Table 1. This reaction ex-
Figure 3). Even at a catalyst loading of 0.0625 mol%, the re-
action progressed at a significant rate. The product was not
observed in a blank experiment without the pincer palladi-
um complex.
Poisoning experiments with mercury and polyvinylpyri-
dine (PVPy) were performed to probe the involvement of
Pd0 species in this reaction. Mercury (Hg0) is known to in-
tercept Pd0 by amalgamation of palladium colloids or by ad-
sorption of Pd0 particles to the mercury surface.[15,16] PVPy
binds free, unbound Pd atoms to the many pyridine ligands
that are present in the polymer.[17,18] Upon addition of either
of these poisons, no change in reactivity was observed.
These observations rule out the involvement of Pd0 in cycle
I (Scheme 2).
Table 1. Conversion of cinnamyl chloride in the SCS-pincer Pd-catalyzed
stannylation with hexamethylditin in the absence or presence of 4-nitro-
benzaldehyde in THF or CH2Cl2.[a]
4-Nitrobenzaldehyde Solvent
Conversion of cinnamyl chloride [%]
1 h
CH2Cl2 48
THF 72
CH2Cl2 46
THF 53
5 h
24 h
absent
absent
present
present
100
100
80
100
100
100
100
84
A closer examination of the separate stannylation reac-
1
[a] Conditions: cinnamyl chloride (0.80 mmol), hexamethylditin
(0.80 mmol), 1Cl (2 mol%) and, if present, 4-nitrobenzaldehyde
(0.80 mmol) in solvent (6 mL; ambient temperature, N2 atmosphere).
tion in the presence of 1Cl by using H NMR spectroscopic
analysis showed that small but significant amounts of 1-
phenyl-2-propenyl trimethylstannane, the branched isomer
of cinnamyl trimethylstannane, were formed in this reaction.
Interestingly, the maximum amount of this species reached
about 10% after 3 h, whereupon its concentration decreased
again (see Figure 4). Identification of 1-phenyl-2-propenyl
trimethylstannane is based on the comparison with reported
hibits a small solvent effect: reactions carried out in THF
tend to be slightly faster than those in CH2Cl2. Furthermore,
the stannylation of cinnamyl chloride in the absence of the
benzaldehyde electrophile was remarkably faster than in the
case in which the electrophile was present. For the separate
reaction, a complete conversion was found in both solvents
after 10 h, whereas after this time in the tandem setup, ap-
proximately 20% cinnamyl chloride remained present. The
stannylations in the tandem reaction were completed only
after a reaction time of 16 h.
1
NMR data of the isolated compound.[19] The H NMR spec-
trum recorded after 3 h of reaction time clearly shows typi-
cal signals for cinnamyl chloride, cinnamyl trimethylstan-
nane, as well as 1-phenyl-2-propenyl trimethylstannane (see
Figure 5).
An NMR experiment was performed to obtain insight
into the order of the catalyst in the stannylation reaction. A
solution containing equimolar amounts of cinnamyl chloride
and hexamethylditin (0.67 mmol) in CD2Cl2 was placed in a
series of NMR tubes, and solutions of a given concentration
of SCS-pincer Pd complex 1Cl were added to reach a con-
stant volume of 0.5 mL. The reaction progression was moni-
Second reaction step (electrophilic allylic substitution): In a
separate experiment, the isolated primary tandem product
cinnamyl trimethylstannane was treated with 4-nitrobenzal-
dehyde and the palladium pincer catalyst 1Cl in CH2Cl2 or
THF to determine the kinetics of cycle II (Scheme 2) of the
tandem in an independent manner. In the literature this re-
action has mainly been performed with PCP-pincer Pd com-
Chem. Eur. J. 2013, 19, 4858 – 4868
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4861

J. M. Klein Gebbink et al.
of 1Cl. Finally, a stoichiometric amount of catalyst 1Cl was
used, but even under these conditions no carbon–carbon
coupling occurred.
Under these conditions (i.e., by using SCS-pincer Pd com-
plex 1Cl in the presence or absence of trimethyltin chloride)
the reaction mixtures became darker after one day, probably
caused by slow decomposition of the pincer complex and
formation of palladium colloids in the solution. After a long
period of inactivity, eventually product formation was ob-
served in all of these dark reaction mixtures. When this
same reaction was performed by using the same loading of
PdACHTUNGRTENN(UG dba)2 or PdACHTUNGTRNEN(UGN PPh3)4 as catalyst, the reaction proceeded
Figure 4. The reaction profile of the stannylation of cinnamyl chloride in-
cluding the presence of 1-phenyl-2-propenyl trimethylstannane.
within 1 h showing a full conversion towards secondary
product 1-(4-nitrophenyl)-2-phenylbut-3-en-1-ol in a anti/syn
product ratio of 1.5:1.
The reactivity of 1-phenyl-2-propenyl trimethylstannane,
the branched isomer of cinnamyl trimethylstannane, was in-
vestigated in a similar way. To a solution of independently
synthesized 1-phenyl-2-propenyl trimethylstannane[19,21] stoi-
chiometric amounts of 4-nitrobenzaldehyde and catalytic
amounts (2 mol%) of SCS-pincer Pd complex 1Cl were
added and the reaction was followed by NMR spectroscopic
analysis. Immediate formation of the allylation product 1-(4-
nitrophenyl)-2-phenylbut-3-en-1-ol was observed. After 3 h
the reaction was complete and did not only yield the sec-
ACHUTNGERNoNUG ndAHCTUNGTRNENaUGN ry tandem product but also cinnamyl trimethylstan-
nane. The ratio between the tandem anti product, the
tandem syn product, and cinnamyl trimethylstannane was
3.5:1:1.
Next, 1-phenyl-2-propenyl trimethylstannane was treated
with SCS-pincer Pd-catalyst 1Cl only (i.e., no aldehyde was
added) to determine whether the isomerization of 1-phenyl-
2-propenyl trimethylstannane to cinnamyl trimethylstannane
is catalyzed by the SCS-pincer Pd complex. After 16 h, large
amounts of cinnamyl trimethylstannane were formed,
whereas there was no 1-phenyl-2-propenyl trimethylstan-
nane left. Because of the unstable character of both isomers,
decomposition products (amongst others allylbenzene) were
also formed. Indeed, when 1-phenyl-2-propenyl trimethyl-
stannane was stirred in CH2Cl2 without additives for 16 h,
mainly decomposition products were detected. In addition,
small amounts of cinnamyl trimethylstannane were ob-
served, whereas no starting material was recovered. This
result is remarkably different from the experiment in which
the SCS-pincer Pd-complex 1Cl was present. Apparently, 1Cl
catalyzes the isomerization from 1-phenyl-2-propenyl tri-
Figure 5. 1H spectrum of a stannylation reaction of cinnamyl chloride
with hexamethylditin in the presence of 1Cl at 50% substrate conversion
(S=solvent peak).
plexes.[9,20] In collaboration with the group of Szabꢀ, we
have reported that this particular reaction also works with
cationic SCS- and PCS-pincer Pd complexes as the second
reaction of the same tandem reaction that is investigated
here.[11] Surprisingly, when this second reaction step was in-
vestigated with the neutral SCS-pincer Pd complex 1Cl, there
was no product formation at all after 16 h. The only reaction
observed was the partial decomposition of cinnamyl triACHTUNTRGNEUNGmeth-
ꢀ
G
Even when this experiment was performed at reflux temper-
atures, homoallylic alcohol product formation was still not
observed after 16 h.
In another experiment, the effect of adding additional tri-
methyltin chloride to the reaction mixture was investigated.
Trimethyltin chloride is present in the tandem reaction mix-
ture as a byproduct of the first reaction step (see Scheme 1).
Because the electrophilic substitution reaction was success-
ful in the tandem reaction and not in the separate reaction,
the presence of trimethyltin chloride in the reaction mixture
might play a role. This appeared not to be the case, since
also after addition of trimethyltin chloride no reaction be-
tween cinnamyl trimethylstannane and 4-nitrobenzaldehyde
was observed after 16 h in the presence of catalytic amounts
ACHUTGTNRENmNUG ethHCATUNGTRENNyUGN lstannane to its linear isomer cinnamyl trimethylstan-
nane.
Control experiments: To get more insight in the reactivity
and orthogonality of the used reagents in the tandem reac-
tion, several experiments were performed in which the
added equivalents of substrates and the presence or absence
of the catalyst were varied. The outcome of these tests is
summarized in Table 2. Entries 1–3 show that Pd pincer
complex 1Cl is required for the stannylation reaction: no
blank reaction was observed. In the absence of 1Cl, also no
4862
www.chemeurj.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 4858 – 4868

SCS-Pincer Palladium-Catalyzed Substitution
FULL PAPER
Table 2. Control experiments on the reactivity and the orthogonality of
the different reagents in the tandem reaction.[a]
In the presence of an excess of both cinnamyl chloride
and hexamethylditin the tandem reaction went to comple-
tion in just a few hours (Table 2, entries 11–14). Figure 6
Entry
Substrate [equiv]
HMDT
Cat [%]
Products
CC
4NBA
1
2
3
4
5
6
7
8
9
10
11
12
13
14
1
1
1
0
1
1
0
1
1
1
10
5
3
1.5
1
1
1
1
1
0
1
1
10
10
10
5
3
1.5
1
0
1
1
0
1
1
10
1
10
1
2
–
–
–
2
2
2
2
2
2
2
2
2
2
product 2[b]
–
–
–
product 1
–
[c]
–
product 2[c]
product 1
product 1
product 2[d]
product 2[d]
product 2[d]
product 2[e]
1
1
1
[a] CC=cinnamyl chloride; HMDT=hexamethylditin; 4NBA=4-nitro-
benzaldehyde, product 1=cinnamyl trimethylstannane, product 2=tri-
methyl(1-(4-nitrophenyl)-2-phenylbut-3-enyloxy)stannane. All experi-
ments were performed at 258C and followed in time by 1H NMR analy-
sis. [b] The reaction profile of this entry was similar to the one described
in Figure 2. [c] Decomposition of 4-nitrobenzaldehyde was observed.
[d] Due to the excess of CC and HMDT, also 9 (entry 11), 4 (entry 12) or
2 (entry 13) equivalents of product 1 were formed, respectively. Complete
substrate conversion towards tandem product was observed in the first
5 h. [e] Reaction profile of this entry shows 70% product formation in
the first 5 h, a plateau at 70% of product formation and finally a gradual
completion of the reaction after 120 h.
Figure 6. Secondary tandem product formation using 4-nitrobenzaldehyde
&
^
and one ( ) or three equivalents ( ) of cinnamyl chloride and hexa
ylditin, respectively.
ACHTUNGTRENNUNGmeth-
AHCTUNGTRENNUNG
shows the reaction profile when 3 equivalents of cinnamyl
chloride and hexamethylditin were used (Table 2, entry 13).
A threefold excess of these substrates with respect to 4-ni-
trobenzaldehyde therefore resulted in a remarkable 60-fold
increase in the reaction rate (120 h vs. 2 h to reach comple-
tion) and an entirely different reaction profile. In fact, the
observed reaction profile lacks a lag phase and looks much
like the theoretical kinetic profile shown in Figure 1. Only
upon lowering the amounts of cinnamyl chloride and hexa-
reaction occurred between hexamethylditin and 4-nitro
aldehyde within 16 h (Table 2, entry 4). The presence of
hexamethylditin is also essential for the formation of the
secondary product: no direct palladium-catalyzed coupling
ACHTUNGTNERbNUNG enz-
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
A
ACHTUNGTRENNUNG
occurred between cinnamyl chloride and 4-nitrobenzalde-
hyde (Table 2, entry 6). Addition of the Pd-pincer complex
1Cl to a mixture of hexamethylditin and 4-nitrobenzaldehyde
led to decomposition of 4-nitrobenzaldehyde (Table 2,
entry 7). When this experiment was repeated with another
electrophile, that is, 4-cyanobenzaldehyde, no decomposition
was found implying the involvement of the nitro-group in
the degradation. The use of 4-cyanobenzaldehyde as the
substrate in the palladium-catalyzed tandem reaction led to
very similar reaction kinetics as described here for 4-nitro-
benzaldehyde and the formation of the respective secondary
tandem product 1-(4-cyanophenyl)-2-phenyl-3-buten-1-ol.
The use of a large excess of 4-nitrobenzaldehyde with re-
spect to cinnamyl chloride and hexamethylditin led to the
complete formation of secondary tandem product, but
hardly speeded up the reaction (Table 2, entry 8). Again, a
temporary state of reaction inactivity was observed, where-
upon product formation proceeded gradually to completion
after two days. When hexamethylditin was added in large
excess, the stannylation was speeded up enormously (full re-
action within 15 min), but the reaction did not proceed
beyond the primary product cinnamyl trimethylstannane
ACHTUNGTRENmNUNG ethylditin to 1.5 equiv with respect to 4-nitrobenzaldehyde
was a reaction plateau again observed. In this case, product
formation stopped after 5 h at 70% of secondary tandem
product (compare 40% secondary tandem product forma-
tion at equimolar substrate ratio, Figure 2) and was com-
plete after 120 h.
Discussion
Our investigation of the tandem reaction consisting of the
SCS-pincer palladium-catalyzed stannylation of cinnamyl
chloride by hexamethylditin and subsequent electrophilic
substitution by 4-nitrobenzaldehyde to form homo-allylic al-
cohols shows that the mechanism of this catalytic auto-
tandem reaction is more complex than was anticipated so
far. Based on our experiments, we propose an alternative
mechanism for the tandem reaction catalyzed by SCS-pincer
Pd complex 1Cl that explains the unusual reaction kinetic
profile of the reaction, including the high initial 1-(4-nitro-
phenyl)-2-phenylbut-3-en-1-ol formation and the sudden
halt of the formation of this secondary tandem product (see
Scheme 4). These observations could not be explained with
(Table 2, entry 9), even when a similar excess of 4-nitro
aldehyde was used (entry 10). Interestingly, no formation of
the secondary tandem product was observed in these cases.
ACHTUNGTNERbNUNG enz-
G
the catalytic cycle described by Gagliardo et al. for cationic
[11]
SCS-pincer Pd complexes 1MeCN
.
Chem. Eur. J. 2013, 19, 4858 – 4868
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4863

J. M. Klein Gebbink et al.
ments in which only the stan-
A
ACHTUNGTRENNUNG
a
amount of the branched SN2’
substitution product is formed.
This SN2’ product disappears
again before the stannylation
reaction goes to completion. It
either isomerizes in situ to the
more stable conjugated cinna-
AHCTUNGTERGmNNUN yl trimethylstannane (step C)
or enters the second catalytic
cycle. The SN2 product cinna-
A
trimethylstannane
ap-
peared unreactive and, there-
fore, does not enter the second
catalytic cycle toward the ben-
zaldehyde electrophile in sepa-
rate reactions. We believe that
the formation of 1-phenyl-2-
propenyl trimethylstannane is
the key to the unusual reaction
profile that was observed for
the tandem reaction.
As soon as the first reaction
step is complete, that is, all the
cinnamyl chloride and hexa-
A
ACHTUNGTRENNUNG
stannane is present, because it
has either reacted onward to
the secondary tandem product
or it has been converted to the
more stable cinnamyl trimethyl-
stannane. At this point, cycle 2
(Scheme 4) of the tandem reac-
tion cannot take place, since
the remaining cinnamyl tri-
ACHUTNGRENNUmG ethACHTUNTGNERyNUGN lstannane has proven to
be unreactive in this reaction.
This explains the sudden drop
in activity at the end of Part I
of
the
reaction
profile
(Figure 2). Furthermore, this
explains why higher amounts of
cinnamyl chloride and hexa-
Scheme 4. Proposed reaction mechanism for the tandem reaction between cinnamyl chloride, hexamethylditin,
and 4-nitrobenzaldehyde by using SCS-pincer palladium complexes.
A
ACHTUNGTNERyNUNG lditin relative to 4-nitro-
A
ACHTUNGTRENNUNG
Cycle 1: In the mechanism shown in Scheme 4, SCS-Pd com-
plex 1Cl is transmetalated in the first catalytic cycle by hexa-
ACHTUNGTRENNUNGmethylditin leading to a palladium-tin species (step A),
plateau at a higher percentage of secondary product (see
Table 2, entry 14), or to a completed tandem reaction with a
normal kinetic profile (entries 11–13).
which can undergo substitution with allyl chlorides like cin-
namyl chloride (step B). This reaction can either take place
through an SN2-type substitution leading to the linear prod-
uct cinnamyl trimethylstannane (step B1), or through an
SN2’-type substitution leading to the branched product 1-
phenyl-2-propenyl trimethylstannane (step B2). Experi-
An order of 1.7 in palladium was found for this stannyla-
tion reaction (see Figure 3), which suggests a certain degree
of positive cooperativity of SCS-pincer Pd fragments in the
rate-limiting step of the reaction. Therefore, catalysts show-
ing a high local concentration of catalytic centers, for exam-
ple, dendritic catalysts, might show a higher reaction rate
4864
www.chemeurj.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 4858 – 4868

SCS-Pincer Palladium-Catalyzed Substitution
FULL PAPER
than their monomeric analogues. Indeed, earlier investiga-
tions by us showed that the dendritic SCS-pincer Pd-com-
plexes are more active catalysts for the stannylation of cin-
namyl chloride.[12] Here, the monomeric catalyst showed
63% conversion of cinnamyl chloride after 1 h reaction,
whereas the G0 and the G1 dendritic catalysts are significant-
ly more active showing 71 and 85% conversion, respective-
ly.
In conclusion, only when the first reaction step in the
tandem is progressing, that is, when distinct amounts of 1-
phenyl-2-propenyl trimethylstannane are present, can the
second tandem step take place, since this species is required
to feed the second reaction step. As soon as the starting ma-
terials cinnamyl chloride and hexamethylditin become de-
cinnamyl chloride forming the primary reaction products
cinnamyl trimethylstannane and 1-phenyl-2-propenyl tri-
ACHUTNGERNmNUG ethAHCTUNGTRNENyUGN lstannane (step B in catalytic cycle 1, Scheme 4), but a
small portion is proposed to be reduced to release a free Pd0
atom (step H). The Pd0 atoms that are formed in this way
provide a starting point for a Pd0/PdII-based third catalytic
cycle.
In this cycle, the Pd0 center is proposed to undergo oxida-
tive addition with cinnamyl trimethylstannane forming a
palladium(II) h3-allyl intermediate similar to Pd0-catalyzed
allylic substitution reactions of allyl halides[24] and Stille cou-
plings.[25,26] It is widely accepted that this palladium(II) h3-
allyl intermediate has an electrophilic nature and reacts with
nucleophiles.[27] Nevertheless it is well-known that h3-allyl
groups show exchange between (eventual) syn and anti sub-
stituents and that the intermediate of this process is a h1-
allyl metal complex.[28] Recently, Nakamura and Yamamoto
found that bis-h3-allylpalladium complexes can behave as
nucleophiles and are able to react with electrophiles, like al-
dehydes.[29–31] Szabꢀ and co-workers have performed exten-
sive mechanistic studies on such palladium-catalyzed elec-
trophilic allylation reactions that start from bis-h3-allylpalla-
dium complexes.[32–35] In these cases, the catalytic reaction
proceeds because the aldehyde is able to coordinate to this
bis-h3-allylpalladium complex forming a h3-allyl-h1-allylpalla-
dium intermediate. Subsequent nucleophilic attack of the h1-
allyl group on the carbon–oxygen double bond of the alde-
hyde produces a h3-allyl-homoallyloxypalladium species. In
this mechanism, one of the allyl ligands is a spectator ligand
and therefore does not actively take part in the catalytic
cycle. Its role is important in blocking one half of the coor-
dination sphere and fine-tuning the activity of this reac-
tion.[33]
pleted, no further formation of 1-phenyl-2-propenyl tri
ylstannane occurs that leads to a status quo in the kinetic re-
action profile.
ACHTUNGTNERmNUNG eth-
ACHTUNGTRENNUNG
Cycle 2: Whereas in our previous investigation we assumed
that the second catalytic cycle starts with the formation of a
Pd-h1-allyl intermediate upon reaction of linear cinnamyl tri-
methylstannane with the (recovered) SCS-pincer Pd com-
plex 1Cl,[11] we now propose that this catalytic cycle starts
with a transmetalation between branched product 1-phenyl-
2-propenyl trimethylstannane and the SCS-pincer Pd-halide
complex. This leads to the formation of the h1-allyl palladi-
um species 1h1-allyl (Scheme 4, D). The attack of this species
on the carbonyl group of the benzaldehyde subsequently
takes place through the nucleophilic g-carbon, creating a ter-
tiary homo-allylic alcohol that is coordinated to the Pd
center in
a
h2-allyl fashion through the allyl moiety
(Scheme 4, E). In this step, the two prochiral reaction part-
ners react to form a mixture of secondary reaction products
favoring the anti products (RS, SR) above the syn products
(RR, SS). The anti products are formed in excess due to the
more advantageous spatial distribution of the substituents,
which reduces steric hindrance in the transition state in step
E (Scheme 4). In the absence of available protons, the prod-
uct is initially stabilized by the earlier released trimethyltin
cation. Finally, the formed product is liberated from Pd
(Scheme 4, step F), whereupon protonation in the workup
leads to the secondary tandem product 1-(4-nitrophenyl)-2-
phenyl-3-buten-1-ol (Scheme 4, step G).
In our case, the nucleophilic reactivity might be explained
by a similar mechanism that takes place on the surface of a
Pd nanoparticle, a situation in which also one half of the co-
ordination sphere is blocked. A surface Pd0 center might un-
dergo oxidative addition with cinnamyl trimethylstannane
forming a palladium(II) h3-allyl intermediate (Scheme 4,
step I), whereupon coordination of 4-nitrobenzaldehyde to
this intermediate generates a nucleophilic h1-allylpalladium
intermediate (Scheme 4, step J). This interaction likely is
part of an equilibrium in which the palladium(II) h3-allyl in-
termediate is favored. The h1-allylpalladium intermediate
can escape from the equilibrium by attacking the carbonyl
center of the aldehyde, leading to an alkoxypalladium spe-
cies (Scheme 4, step K). Through reductive elimination, the
catalytic cycle closes (Scheme 4, step L) and the released
product leads, after protonation in the workup of this reac-
tion, to the secondary tandem product (Scheme 4, step G).
The anti/syn product ratio of this secondary product that has
been formed through this Pd0/PdII mechanism was found to
be lower (1.5:1) than the product that has been exclusively
formed through a PdII-only mechanism (3.5–4.0:1).
Cycle 3: In the third cycle all further reactivity is proposed
to be mediated by the in situ formation of Pd0 species and
not by SCS-PdII complexes. The latter complexes only act as
precursors for the formation of Pd nanoparticles. Although
ECE-Pd complexes are known to be stable as long as the
temperature is not raised to too high values in the presence
of strong bases,[22,23] a dramatic color change of the tandem
reaction mixtures was observed after one day at room tem-
perature and visible Pd-black formation occurred at room
temperature after approximately two days.
We propose that Pd0 formation is caused by an auto-de-
composition of the SCS-pincer palladium-tin species 1SnMe
in the first catalytic cycle. This species mainly reacts with
DFT calculations: In the previous part, we demonstrated
that the SCS-pincer palladium complex 1Cl does not catalyze
3
Chem. Eur. J. 2013, 19, 4858 – 4868
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4865

J. M. Klein Gebbink et al.
the electrophilic allylic substitution of cinnamyl trimethyl-
stannane with 4-nitrobenzaldehyde, whereas it successfully
performs this reaction in the case of the isomeric substrate
1-phenyl-2-propenyl trimethylstannane. The most likely ex-
planation for this behavior is that complex 1Cl is able to
react with the branched 1-phenyl-2-propenyl trimethylstan-
nane to form the (transient) key reactive h1-allylpalladium
intermediate, whereas this transformation cannot be ach-
ieved in the case of the cinnamyl trimethylstannane sub-
strate.
A plausible, dissociative pathway for the reaction of com-
plex 1Cl with an allylstannane moiety leading to a h1-allylpal-
ladium complex 1h1-allyl is described in Scheme 5. It involves
Figure 7. Transition state of the formation of an h1-allylpalladium com-
plex by nucleophilic attack of a chloride anion on the tin atom of an al-
lylstannane coordinated to a palladium-pincer cation through an SN2-type
mechanism.
non-bulky linear h1-allyl fragment, which is further stabi-
lized by conjugation of the double bond with the phenyl
group.
This qualitative reasoning is reflected in the overall reac-
tion energy as computed by using DFT for both isomers. In
the case of the cinnamyl trimethylstannane substrate, the
transformation depicted in Scheme 5 is strongly endothermic
(DG0 =16.4 kcalmolꢀ1) and therefore would be difficult to
achieve using the current reaction conditions. For 1-phenyl-
2-propenyl trimethylstannane, however, the overall reaction
is almost athermic (DG0 =2.6 kcalmolꢀ1) and should conse-
quently be able to proceed smoothly. Therefore, we con-
clude that our DFT results are in agreement with the hy-
pothesis and our experimental results that 1-phenyl-2-pro-
penyl trimethylstannane is able to react with SCS-pincer Pd-
catalyst 1Cl to form a Pd-allyl intermediate, whereas cinnam-
yl trimethylstannane is unreactive towards this complex.
Scheme 5. Proposed dissociative pathway for the formation of a h1-allyl-
palladium complex from an allylstannane moiety and complex 1Cl.
1) dissociation of the chloride ligand from 1Cl, 2) coordina-
tion of the double bond of the allylstannane substrate to the
vacant site of the palladium center, and 3) SN2-type substitu-
tion of the chloride anion on the tin center, liberating the
corresponding chlorostannane and forming 1h1-allyl
.
The viability of such a reaction pathway was tested by
using DFT calculations on a model system in which all
methyl and phenyl groups of the pincer complex and the
stannane substrates were replaced by hydrogen atoms (R=
R’=R’’=R’’’=H). Preliminary results from these calcula-
tions indicate that the reaction is unlikely to proceed as it is
strongly endothermic (DG0 = ++12.6 kcalmolꢀ1), but that the
required transition state appears fairly accessible at room
temperature (DG† =20.6 kcalmolꢀ1, see Figure 7).
In the real system, the position of the phenyl substituent
of the allylstannane moiety appeared to have a crucial influ-
ence on the reactivity. This can be qualitatively rationalized
by using Scheme 5. For the cinnamyl trimethylstannane sub-
strate (R’=Ph and R’’=H), the starting material is stabi-
lized by conjugation of the allyl double bond with the
phenyl ring, whereas the final product presents a hindered
branched h1-allyl fragment without conjugation between the
double bond and the phenyl substituent. On the contrary,
for the 1-phenyl-2-propenyl trimethylstannane substrate
(R’=H, R’’=Ph), the starting material is branched and not
stabilized by conjugation, whereas the product exhibits a
Conclusion
The SCS-pincer Pd-catalyzed auto-tandem reaction consist-
ing of the stannylation of cinnamyl chloride followed by the
electrophilic allylic substitution with 4-nitrobenzaldehyde to
form 1-(4-nitrophenyl)-2-phenylbut-3-en-1-ol was studied. It
was found that when palladium pincer complex 1Cl is used
as the catalyst both cinnamyl trimethylstannane and its
branched isomer 1-phenyl-2-propenyl trimethylstannane are
formed in the first reaction step. The branched isomer
turned out to be the active substrate in the second pincer-
catalyzed reaction step, the electrophilic substitution with 4-
nitrobenzaldehyde towards the secondary tandem product.
These findings were corroborated by DFT calculations.
The consequence of this behavior is that as soon as the
first catalytic cycle has finished because all substrates for the
stannylation reaction have reacted, no branched product 1-
phenyl-2-propenyl trimethylstannane is present in the reac-
tion mixture. Since the linear isomer cinnamyl trimethyl-
4866
www.chemeurj.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 4858 – 4868

SCS-Pincer Palladium-Catalyzed Substitution
FULL PAPER
stannane is not reacting through the PdII-catalyzed second
catalytic cycle, but through another mechanism (catalytic
cycle 3, Scheme 4), this second cycle is no longer fed, and as
a consequence the electrophilic substitution with 4-nitro-
ACHTUNGTRENNUNGbenzACHTUNGTRENNUNGaldehyde can no longer take place. This means that the
total reaction progress stops and a plateau in the kinetic re-
action profile is observed.
ACHUTNGRENmNUG ethACHUTNGTRENyNGUN lstannane in 46% yield. Analytical data were in accordance with
the data published by Fong et al.[36]
Synthesis of 1-phenyl-2-propenyl trimethylstannane: A solution of cin-
namyl trimethylstannane (approximately 0.1m) in CDCl3 was placed into
a NMR tube. Irradiation of this tube with a high pressure mercury UV
lamp for 2 h at a distance of 2 cm to the lamp in a cooled water bath (to
compensate for the heat caused by the lamp), leads to complete conver-
sion of cinnamyl trimethylstannane to 1-phenyl-2-propenyl trimethylstan-
nane. This NMR tube was directly used for further investigations. Analyt-
ical data were in accordance with the data published by Takuwa.[19]
Addition of more equivalents of cinnamyl chloride and
hexamethylditin with respect to 4-nitrobenzaldehyde leads
to higher amounts of branched isomer 1-phenyl-2-propenyl
trimethylstannane. Formation of the secondary tandem
product 1-(4-nitrophenyl)-2-phenylbut-3-en-1-ol goes to
completion using three, or more equivalents of cinnamyl
chloride and hexamethylditin, because under these condi-
tions the second catalytic cycle is continuously fed. The re-
action rate enormously improves from one week, using one
equivalent of cinnamyl chloride and hexamethylditin, to
only 2 h by using three equivalents of these substrates,
thereby showing normal reaction kinetics for a tandem reac-
tion. Consequently, the tandem reaction takes place rapidly
through a PdII-only mechanism. Under these conditions, a
mechanism that makes use of the cinnamyl trimethylstan-
nane product that is also formed in cycle I and of Pd0 parti-
cles leached from the palladium pincer (cycle III) does not
take part in this tandem reaction, which improves reaction
rate, product selectivity and prevents Pd-leaching from the
SCS-pincer Pd complexes.
General protocol for the tandem reaction: SCS-pincer palladium complex
(2 mol%, 0.016 mmol, 7.4 mg), was added to a solution of cinnamyl
chloride (0.80 mmol, 122.1 mg, 113 mL), hexamethylditin (0.80 mmol,
275 mg, 174 mL), 4-nitrobenzaldehyde (0.80 mmol, 126.9 mg), and hexa-
ACHUTNGRENUmNG ethAHCUTNGTRENyNNGU lbenzene (internal standard, 0.088 mmol, 14.4 mg) in dry THF
(6 mL). The reaction was stirred at room temperature in a nitrogen envi-
ronment. Aliquots of 50 mL for NMR/GC analysis were regularly taken
with an airtight syringe.
Protocol for the first reaction step of the tandem reaction (stannylation):
The general protocol for the tandem reaction was followed, but no 4-ni-
trobenzaldehyde was added. The reaction was performed in dry THF
(6 mL) or dry CH2Cl2 (6 mL) and analyzed by NMR spectroscopy and
GC.
Variation on the concentration of catalyst in the first reaction step of the
tandem reaction (stannylation):
A solution of cinnamyl chloride
(67 mmol, 10.2 mg, 9.4 mL), hexamethylditin (67 mmol, 21.8 mg, 13.8 mL)
and hexamethylbenzene (internal standard, 7.4 mmol, 1.2 mg) in of
CD2Cl2 (0.5 mL) was added into a NMR tube. Subsequently a solution of
SCS-pincer palladium complex 1Cl in CD2Cl2 (0.1 mL) was introduced to
the NMR tube. A series of solutions were made containing 2, 1, 0.5, 0.25,
0.125, or 0.0625% of catalyst 1Cl. The experiments were performed inside
a Varian 300 MHz spectrometer at 258C.
For further catalytic application of this reaction, it is ad-
vised to use an excess of cinnamyl chloride and hexamethyl-
ditin relative to 4-nitrobenzaldehyde to guarantee a PdII-
only mechanism. In this way, this auto-tandem reaction was
successfully performed in a compartmentalized way by using
a dendritic SCS-pincer PdII-catalyst for four runs in a high
catalytic rate (89, 96, 96, and 92% for the respective runs)
showing no palladium leaching for the first two runs.[12]
Poisoning experiments for the first reaction step of the tandem reaction
(electrophilic allylic substitution): The general protocol for the stannyla-
tion was followed. Polyvinylpyridine (2% cross linked, 100 equiv,
80 mmol, 8.4 g) or mercury (2 drops) were added to the reaction mixture.
Analyses were performed by NMR spectroscopy and GC analysis.
Protocol for the second reaction step of the tandem(electrophilic allylic
substitution): SCS-pincer palladium complex 1Cl (2 mol%, 0.016 mmol,
7.4 mg), was added to
a solution of cinnamyl trimethylstannane
(0.80 mmol, 225 mg), 4-nitrobenzaldehyde (0.80 mmol, 126.9 mg) and
hexamethylbenzene (internal standard, 0.088 mmol, 14.4 mg) in dry THF
(6 mL). The reaction was stirred at room temperature in a nitrogen envi-
ronment. Aliquots of 50 mL for NMR/GC analysis were regularly taken
with an airtight syringe.
Experimental Section
Catalysis using 1-phenyl-2-propenyl trimethylstannane: A solution of 1-
phenyl-2-propenyl trimethylstannane (0.25 mmol, 70 mg) in CDCl3
(0.5 mL) was prepared, and 4-nitrobenzaldehyde (0.25 mmol, 38 mg) and
SCS-pincer palladium complex (1Cl, ꢂ5%, few mg) were added to this
solution. Limitations in the accuracy of the analytical balances used
meant that the catalyst loading could not exactly be determined. The ex-
periments were performed inside a Varian 300 MHz spectrometer at
258C.
General: All reactions were carried out using standard Schlenk techni-
ques under an inert dinitrogen atmosphere unless stated otherwise. All
solvents were carefully dried and distilled prior to use. All standard re-
agents were purchased commercially and used without further purifica-
tion. 1H (300 MHz), 13C (100 MHz) and 29Si (60 MHz) NMR spectra
were recorded on a Varian 400 MHz spectrometer at 258C, chemical
shifts (d) are given in ppm referenced to residual solvent resonances.
ICP-MS analyses were carried out by Kolbe Mikroanalytisches Laborato-
rium (Mꢃlheim a.d. Ruhr, Germany). GC analysis was carried out using
a Perkin–Elmer Clarus 500 GC equipped with an Alltech Econo-Cap
EC-5 column.
DFT calculations: All calculations were run in gas phase at the DFT
level on the software G03W by using the B3LYP functional and the basis
set H/C/S/Cl 6–31G*, Pd/Sn LANL2DZ.[37] All geometries were opti-
mized using the regular convergence criteria (keywords opt for inter-
mediates and opt=qst3 for transition states). Intermediates were charac-
terized by the absence of imaginary vibrations in a frequency calculation.
Transition states were characterized by the presence of a single imaginary
vibration in a frequency calculation. The following simplifications were
applied: 1) the phenyl groups of the pincer moieties were replaced by H
atoms and 2) the trimethyltin group was replaced by a stannyl group
(SnH3).
Synthesis of cinnamyl trimethylstannane: Cinnamyl chloride (8.0 mmol,
1.2 g, 1.1 mL), hexamethylditin (8.0 mmol, 2.8 g, 1.7 mL) and SCS-pincer
Pd complex 1Cl (1 mol%, 37 mg) were combined in CH2Cl2 (30 mL).
After a few hours, quantitative conversion towards cinnamyl trimethyl-
stannane was observed. The solution was separated from the catalyst by
using flash chromatography on neutral alumina using n-hexane as eluent.
The fractions that contained product were collected and PVPy
(100 equiv) was added. After stirring for 1 h, the colorless solution was
filtered and concentrated under vacuum leading to pure cinnamyl tri-
Chem. Eur. J. 2013, 19, 4858 – 4868
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4867

J. M. Klein Gebbink et al.
[22] A. Takuwa, T. Kanaue, Y. Nishigaichi, H. Iwamoto, Tetrahedron
Lett. 1995, 36, 575–578.
Acknowledgements
[23] D. E. Bergbreiter, P. L. Osburn, J. D. Frels, Adv. Synth. Catal. 2005,
347, 172–184.
The authors acknowledge the National Research School Combination
Catalysis (NRSC-C) for financial support.
[24] W. J. Sommer, K. Q. Yu, J. S. Sears, Y. Y. Ji, X. L. Zheng, R. J.
Davis, C. D. Sherrill, C. W. Jones, M. Weck, Organometallics 2005,
24, 4351–4361.
[1] M. Albrecht, G. van Koten, Angew. Chem. 2001, 113, 3866–3898;
Angew. Chem. Int. Ed. 2001, 40, 3750–3781.
[2] J. T. Singleton, Tetrahedron 2003, 59, 1837–1857.
[3] M. Q. Slagt, G. Rodriguez, M. M. P. Grutters, R. J. M. Klein Geb-
bink, W. Klopper, L. W. Jenneskens, M. Lutz, A. L. Spek, G. van Ko-
ten, Chem. Eur. J. 2004, 10, 1331–1344.
[25] A. de Meijere, S. J. Brase, Organomet. Chem. 1999, 576, 88–110.
[26] J. K. Stille, Angew. Chem. 1986, 98, 504–519; Angew. Chem. Int. Ed.
Engl. 1986, 25, 508–523.
[27] A. L. Casado, P. Espinet, J. Am. Chem. Soc. 1998, 120, 8978–8985.
[28] J. Tsujiin, Handbook of Organopalladium Chemistry for Organic
Synthesis (Ed.: E. Negishi), Wiley, New York, 2000.
[4] K. J. Szabꢀ, Synlett 2006, 811–824.
[29] R. H. Crabtree, The Organometallic Chemistry of the Transition
Metals, 4th ed., Wiley-Interscience, 2005, pp. 131–136.
[30] H. Nakamura, H. Iwama, Y. Yamamoto, Chem. Commun. 1996,
1459–1460.
[31] H. Nakamura, M. Bao, Y. Yamamoto, Angew. Chem. 2001, 113,
3308–3310; Angew. Chem. Int. Ed. 2001, 40, 3208–3214.
[32] Y. Yamamoto, H. Nakamura, Top. Organomet. Chem. 2005, 14, 211–
239.
[5] N. Selander, K. J. Szabꢀ, Chem. Rev. 2011, 111, 2048–2072.
[6] N. J. M. Pijnenburg, T. J. Korstanje, G. van Koten, R. J. M. Klein
Gebbinkin, Palladacycles on Dendrimers and Star-Shaped Molecules
(Eds.: J. Dupont, M. Pfeffer), Wiley-VCH, Weinheim, 2008,
pp. 361–398.
[7] O. A. Wallner, K. J. Szabꢀ, Org. Lett. 2004, 6, 1829–1831.
[8] N. Selander, K. J. Szabꢀ, Dalton Trans. 2009, 6267–6279.
[9] N. Solin, J. Kjellgren, K. J. Szabꢀ, J. Am. Chem. Soc. 2004, 126,
7026–7033.
[33] O. A. Wallner, K. J. Szabꢀ, J. Org. Chem. 2003, 68, 2934–2943.
[34] K. J. Szabꢀ, Chem. Eur. J. 2004, 10, 5269–5275.
[10] J. Aydin, K. S. Kumar, M. J. Sayah, O. A. Wallner, K. J. Szabꢀ, J.
Org. Chem. 2007, 72, 4689–4697.
[35] N. Solin, J. Kjellgren, K. J. Szabꢀ, Angew. Chem. 2003, 115, 3784–
3786; Angew. Chem. Int. Ed. 2003, 42, 3656–3658.
[11] M. Gagliardo, N. Selander, N. C. Mehendale, G. van Koten, R. J. M.
Klein Gebbink, K. J. Szabꢀ, Chem. Eur. J. 2008, 14, 4800–4809.
[12] N. J. M. Pijnenburg, H. P. Dijkstra, G. van Koten, R. J. M. Klein
Gebbink, Dalton Trans. 2011, 40, 8896–8905.
[13] N. T. S. Phan, M. van der Sluys, C. W. Jones, Adv. Synth. Catal. 2006,
348, 609–679.
[14] M. Weck, C. W. Jones, Inorg. Chem. 2007, 46, 1865–1875.
[15] K. J. Laidler, Reaction Kinetics, 3rd ed., Harper & Row, New York,
1987.
[16] J. A. Widegren, R. G. Finke, J. Mol. Catal. A: Chem. 2003, 198, 317–
341.
[17] C. Paal, W. Hartmann, Ber. Dtsch. Chem. Ges. 1918, 51, 711–737.
[18] K. Q. Yu, W. Sommer, M. Weck, C. W. Jones, J. Catal. 2004, 226,
101–110.
[19] K. Q. Yu, W. Sommer, J. M. Richardson, M. Weck, C. W. Jones, Adv.
Synth. Catal. 2005, 347, 161–171.
[36] C. W. Fong, W. Kitching, J. Organomet. Chem. 1970, 22, 89–94.
[37] Gaussian 09, Revision A.02, M. J. Frisch, G. W. Trucks, H. B. Schle-
gel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V.
Barone, B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X.
Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Son-
nenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa,
M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven,
J. A. Montgomery, Jr., J. E. Peralta, F. Ogliaro, M. Bearpark, J. J.
Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, R. Kobayashi, J.
Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar,
J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox,
J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E.
Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W.
Ochterski, R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A.
Voth, P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels, ꢄ.
Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski, D. J. Fox, Gaussian,
Inc., Wallingford CT, 2009.
[20] A. Takuwa, T. Kanaue, K. Yamashita, Y. Nishigaichi, J. Chem. Soc.
Perkin Trans. 1 1998, 1309–1314.
Received: August 28, 2012
[21] O. A. Wallner, K. J. Szabꢀ, Chem. Eur. J. 2006, 12, 6976–6983.
Published online: February 21, 2013
4868
www.chemeurj.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 4858 – 4868