A structure-activity relationship for pincer palladium(II) complexes - Influence of ring-size of metallacycles on the activity in allylic alkylation

Article abstract of DOI:10.1016/S0022-328X(03)00604-1

A series of new palladium(II) complexes derived from the well-known pincer complex [PdCl{(C₆H₃)(OPⁱ Pr₂)₂-2,6}] were synthesised: [PdX{(C₆ H₃)(OPⁱPr₂)₂-2,6}] (X = Br^-, I^-, OAc^-, OTf^-). A novel PCP′ pincer ligand 1-(ⁱPr₂ PO)-3-(ⁱPr₂POCH₂)(C₆ H₄) was prepared and complexed to palladium(II) to give [PdX{(C₆H₃)(OPⁱPr₂)-2- (CH₂OPⁱPr₂)-6}] (X=Cl^-, I^-, OAc^-, OTf^-, BF₄ ^-). The X-ray structure of [PdCl{(C₆H₃) (OPⁱPr₂)-2-(CH₂OPⁱ Pr₂)-6}] was solved and is discussed. These complexes were applied to the catalytic reaction of cinnamyl acetate with sodium dimethyl malonate in order to evaluate the influence of the ligand structure and co-ordinating or non-co-ordinating anions on the regioselectivity. A detailed analysis shows that palladium(II) complexes of the unsymmetrical PCP′ bis(phosphinito) ligand are much more active when compared to related complexes of the symmetrical PCP bis(phosphinito) ligand. The origin of this difference in activity is discussed.

Full text of DOI:10.1016/S0022-328X(03)00604-1

Journal of Organometallic Chemistry 681 (2003) 189ꢀ195
/
www.elsevier.com/locate/jorganchem
A structureꢀ
complexes * influence of ring-size of metallacycles on the activity in
allylic alkylation
/
activity relationship for pincer palladium(II)
/
Zhaohui Wang a, Michael R. Eberhard a,*, Craig M. Jensen a, Shiro Matsukawa b,
Yohsuke Yamamoto b,1
a Department of Chemistry, University of Hawaii, 2545 The Mall, Honolulu, HI 96822-2275, USA
b Department of Chemistry, Graduate School of Science, Hiroshima University 1-3-1 Kagamiyama, Higashi-Hiroshima 739-8526, Japan
Received 25 April 2003; received in revised form 22 June 2003; accepted 25 June 2003
Abstract
A series of new palladium(II) complexes derived from the well-known pincer complex [PdCl{(C6H3)(OPi Pr2)2-2,6}] were
synthesised: [PdX{(C6H3)(OPi Pr2)2-2,6}] (XꢁBrꢂ, Iꢂ, OAcꢂ, OTfꢂ). A novel PCP? pincer ligand 1-(i Pr2PO)-3-(i Pr2-
POCH2)(C6H4) was prepared and complexed to palladium(II) to give [PdX{(C6H3)(OPi Pr2)-2-(CH2OPiPr2)-6}] (XꢁClꢂ, Iꢂ
/
/
,
OAcꢂ, OTfꢂ, BF4ꢂ). The X-ray structure of [PdCl{(C6H3)(OPi Pr2)-2-(CH2OPi Pr2)-6}] was solved and is discussed. These
complexes were applied to the catalytic reaction of cinnamyl acetate with sodium dimethyl malonate in order to evaluate the
influence of the ligand structure and co-ordinating or non-co-ordinating anions on the regioselectivity. A detailed analysis shows
that palladium(II) complexes of the unsymmetrical PCP? bis(phosphinito) ligand are much more active when compared to related
complexes of the symmetrical PCP bis(phosphinito) ligand. The origin of this difference in activity is discussed.
# 2003 Elsevier Science B.V. All rights reserved.
Keywords: Phosphorus; Pincer; Palladium; PMP angle; Allylic alkylation
1. Introduction
which is the reason for their great air-, temperature- and
moisture-stability. Most pincer ligands are based on
one-atom linkages between the donor atom P and the
aryl-carbon. Thus, upon metal complexation the ligand
forms two five-membered metallacycles. Venanzi and
co-workers recently reported the synthesis of
PCP pincer ligands, monoanionic terdentate ligands
with the general formula [(C6H3)(APR2)2]ꢂ (Aꢁ
CH2,
/
O), have received great attention since the first example
was reported by Moulton and Shaw [1]. Their chemistry
has been reviewed recently [2]. PCP pincer complexes
were found to be active catalysts for a wide range of
reactions including dehydrogenation [3], transfer dehy-
drogenation [4], Heck reaction [5], Suzuki-coupling [6],
Negishi-type coupling [7] and enantioselective sharpless
epoxidations [8]. Pincer ligands bond most commonly in
the meridional h3-PCP co-ordination mode to metals,
[MBr{(C6H3)((CH2)2PPh2)2-2,6}] (MꢁPd, Pt), which
/
is the only example for a PCP pincer ligand forming two
six-membered metallacycles [9].
We previously reported on the synthesis and applica-
tions of [PdCl{(C6H3)(OPiPr2)2-2,6}] (1) [5c,5d,7]. Li-
gand exchange furnished the following new complexes:
[PdX{(C6H3)(OPiPr2)2-2,6}] (XꢁBrꢂ 2, Iꢂ 3, OTfꢂ 4,
/
OAcꢂ 5). We also wish to report the synthesis of the
first unsymmetrical PCP? pincer bis(phosphinite) 1-
(iPr2PO)-3-(iPr2POCH2)(C6H4) (7) and its palladium(II)
* Corresponding author. Tel.: ꢃ
/
1-808-534-7064; fax: ꢃ1-808-543-
/
complexes:
(Xꢁ
Clꢂ 8, Iꢂ 9, OAcꢂ 10). The X-ray structure of 8
has been solved and is compared to that of 1, which has
[PdX{(C6H3)(OPiPr2)-2-(CH2OPiPr2)-6}]
8182.
E-mail addresses: chmre@gold.chem.hawaii.edu (M.R. Eberhard),
yyama@sci.hiroshima-u.ac.jp (Y. Yamamoto).
/
1
To whom correspondence pertaining to crystallographic studies
should be addressed.
been published [5d].
0022-328X/03/$ - see front matter # 2003 Elsevier Science B.V. All rights reserved.
doi:10.1016/S0022-328X(03)00604-1

190
Z. Wang et al. / Journal of Organometallic Chemistry 681 (2003) 189ꢀ195
/
These complexes were applied to the allylic alkylation
of cinnamyl acetate with sodium dimethyl malonate.
Catalytic allylic alkylation is one of the most important
reactions in synthetic chemistry. Many advances regard-
ing the regioselectivity and enantioselectivity of this CÃ
/
C bond forming reaction have been made and several
reviews have appeared [10]. Surprisingly, pincer com-
plexes have not been studied for this reaction yet [11].
The activity and selectivity of these new complexes are
compared in order to evaluate the influence of the ligand
structure and co-ordinating or non-co-ordinating anions
on this reaction.
Fig. 1. Displacement ellipsoid plot (50% probability level) of 8.
Hydrogen atoms have been omitted for clarity. Pertinent bond lengths
2. Results and discussion
˚
(A) and angles (8): Pd(1)Ã
P(2) 2.2848(9); Pd(1)Ã
/
Cl(1) 2.370(1); Pd(1)Ã
C(1) 2.030(4); Cl(1)ÃPd(1)Ã
P(2) 95.16(2); Cl(1)ÃPd(1)ÃC(1) 176.0(1); P(1)Ã
P(2) 168.31(5); P(1)ÃPd(1)ÃC(1) 81.3(1); P(2)ÃPd(1)ÃC(1) 91.6(1).
/
P(1) 2.2590(9); Pd(1)Ã
P(1) 101.17(2);
Pd(1)Ã
/
/
/
/
Cl(1)ÃPd(1)Ã
/
/
/
/
/
/
2.1. Ligand synthesis and complexes
/
/
/
/
[PdCl{(C6H3)(OPiPr2)2-2,6}] (1) was reacted with
NaBr or NaI in acetone to give [PdX{(C6H3)(O-
All bond lengths are within the typical range. Note
that Pd(1)ÃP(2) is slightly longer than Pd(1)ÃP(1)
PiPr2)2-2,6}] (XꢁBrꢂ 2, Iꢂ 3), with AgOTf in THF
and AgOAc in diethyl ether to give [PdX{(C6H3)(O-
/
/
/
˚
˚
(2.2848(9) A vs 2.2590(9) A). The five-membered
metallacycle shows the expected distorted envelope
conformation whereas the six-membered metallacycle
reveals a boat-conformation. A comparison of the
structure of 8 with the structure of 1 [5d] reveals the
effect of increasing the ring-size of one of the metalla-
PiPr2)2-2,6}] (XꢁOTfꢂ 4, OAcꢂ 5).
/
Commercially available 3-hydroxybenzylalcohol 6 is
the starting material for the synthesis of ligand 7
(Scheme 1). Addition of two equivalents ClPiPr2 to a
solution of 6 in THF in the presence of DMAP furnishes
the moisture- and air-sensitive ligand after work-up in
84% yield. The reaction of 7 with [PdCl2(cod)] in
refluxing toluene gave the cyclometallated complex 8
as a colourless solid after recrystallisation or column
chromatography (75% yield). Magnetically inequivalent
phosphorus nuclei are observed for 8 in the 31P{1H}-
NMR spectrum with a coupling constant of 2J(PP?) 429
Hz clearly indicating the trans-disposition of the phos-
phorus donors. Metathesis reaction with NaI in acetone
and AgOAc in diethyl ether gave 9 (2J(PP?) 422 Hz) and
10 (2J(PP?) 446 Hz), respectively.
cycles. The PÃ
different: 160.380(6)8 for 1 and 168.31(5)8 for 8 (P(1)Ã
P(2)). Whilst the bond lengths are comparable in
/
PdÃ
/
P bond angles are significantly
/
Pd(1)Ã
/
both structures, it is clear that introducing a larger
metallacycle leads to a relaxation of the bond angles.
2.2. Allylic alkylation of cinnamyl acetate with sodium
dimethyl malonate
Allylic alkylation with soft nucleophiles has been
achieved with many metals e.g. Pd [10], Pt [12], Ir [13],
Rh [14], W [15] and Mo [15b,16]. Palladium is by far the
most popular choice for catalysts, which are usually
stabilized with monodentate or bidentate ligands. More-
over, it has been found that anions can affect regio- and
enantioselectivity e.g. Clꢂ [17] and Iꢂ [18] affect the
regioselectivity whereas Fꢂ affects the enantioselectivity
[19]. Tridentate monoanionic pincer complexes have not
been applied to this reaction yet [11]. Our objective was
Complexes 8ꢀ10 are unusual for pincer complexes in
/
that they are composed of five- and six-membered
metallacycles, respectively. To the best of our knowledge
they represent the first examples of pincer complexes
with different metallacycle ring-sizes. Single crystals of
8, suitable for X-ray diffraction, were obtained by slow
diffusion of pentane into a saturated solution of the
complex in dichloromethane. Fig. 1 shows its molecular
structure.
to study the regioselectivity of this CÃ/C bond forming
Scheme 1. Synthesis of ligand 7 and complexes 8ꢀ
/
10. Reagents and conditions: (i) Two equivalents ClPi Pr2, DMAP, THF, r.t.; (ii) [PdCl2(cod)],
THF, reflux; (iii) NaI, acetone, r.t.; (iv) AgOAc, Et2O, r.t., ultrasound.

Z. Wang et al. / Journal of Organometallic Chemistry 681 (2003) 189ꢀ
/195
191
reaction with regard to the structure of the ligand and
the nature of the anion X i.e. non-co-ordinating or co-
ordinating. We chose the standard reaction between
cinnamyl acetate with sodium dimethyl malonate for
this purpose. Catalytic experiments were conducted in
THF at either 25 or 75 8C using 5 mol.% catalyst. A
mixture containing cinnamyl acetate and the catalyst in
THF was added to sodium dimethyl malonate (gener-
ated from sodium hydride and dimethyl malonate in
THF) and analysed after the prescribed time. Cationic
complexes derived from 8 were generated in situ. The
results are summarised in Table 1.
In general the activity of all catalysts are high leading
to complete conversions of cinnamyl acetate and
isolated yields of about 70%. The formation of the
linear isomer clearly dominates (exclusively trans). This
is not unexpected for palladium catalysts, which mostly
lead to reaction at the less hindered terminus of the
substrate [20]. Whereas a slightly greater influence of X
on the regioselectivity is seen in the case of 1 and its
derivatives, no influence is apparent in case of 8 and its
derivatives. Cationic catalyst 4 gives the largest amount
of isomerisation (entry 5; 83:17). This is in contrast to a
system of 8/AgOTf, which under the same conditions
leads to very little isomerisation (entry 12; 95:5). Whilst
these results do not constitute an advance in allylic
alkylation chemistry, a detailed look into the activity of
these complexes revealed interesting trends regarding
these pincer catalysts. These results are shown in Fig. 2.
Indeed the activity of these catalysts are very much
dependent on the structure of the ligand and the nature
of X. Catalyst 1 and derivatives thereof show only
reasonable activity at higher temperatures, whereas 8
Fig. 2. Plot of conversion vs time for various catalyst systems.
and derivatives are highly active at room temperature.
At room temperature, as well as 75 8C, the activity for 1
and its derivatives increases with decreasing co-ordina-
tion strength of X i.e. Iꢂ B
/
Clꢂ BOTfꢂ. At 75 8C only
/
cationic complex 4 achieves comparable activity to 8/
AgOTf, 8/AgBF4 and 10, with ca. 90% conversion after
6 h. Catalytic systems 8/AgOTf, 8/AgBF4 and 10 behave
identically and are most active giving complete conver-
sion after 6 h. Catalysts 8 and 9 give 90 and 71%
conversion, respectively after 6 h; 100% conversion is
achieved after 12 h with 8, whereas catalyst 9 achieves
only 82% conversion after 12 h. In comparison the
activity of 1 is rather low at 25 8C with only 9%
conversion after 6 h. Even at 75 8C catalysts 1 and 3
only give 70 and 56% conversion after 6 h (80 and 71%,
respectively after 10 h) showing that the activity of 1 and
derivatives thereof is significantly lower than that of 8
and its derivatives. Interestingly the activity of catalysts
1 and 3 is always lower than that of catalysts 8 and 9,
respectively. The activity of catalysts 8 and derivatives
increases with the metal centre being more easily
Table 1
Regioselectivity in the allylic alkylation of cinnamyl acetate with
sodium dimethyl malonate
accessible e.g. in the order 9B
/
8B
/
10:
/
8/AgOTf:8/
/
AgBF4.
In order to test if the mechanism of this reaction
involves an allyl intermediate or a discreet enyl-species
Entry
Catalyst
T (8C)
t (h) a
Linear:branched
(sꢃp) the reaction of allylic deuterated acetate 11 with
/
1
1
25
75
75
75
75
75
25
25
25
25
25
75
72
16
16
16
16
16
12
12 b
6
96:4
95:5
90:10
97:3
83:17
94:6
94:6
93:7
94:6
94:6
94:6
95:5
sodium dimethyl malonate was examined (Fig. 3). A
similar labeling experiment has been conducted by
Evans and Nelson who observed for the first time enyl
intermediates in acyclic rhodium-catalysed allylic alky-
lation [14a].
2
1
3
2
4
3
5
4
6
5
7
8
8
9
10
9
10
11
12
8/AgBF4
8/AgOTf
8/AgOTf
6
6
16
a
100% conversion.
82% conversion.
Fig. 3. Deuterium-labeling experiment in order to test if an allyl-
mechanism is present.
b

192
Z. Wang et al. / Journal of Organometallic Chemistry 681 (2003) 189ꢀ195
/
1H-NMR analysis of the product after column
tion of observed rates and selectivities of catalytic
reactions [22]. One important observation is that if the
ligand bite angle is close to that expected for the
transition state of a reaction, the activation energy for
this step should be lower and the reaction step is
accelerated. If in fact this step is rate determining, the
catalytic cycle runs more smoothly and with higher
frequency. In the case of the ligand bite angle being close
to that of the reactant or product complex the energy of
the transition state will be higher and the energy of the
starting material is less affected. Owing to increased
activation energy the reaction is slower. van Leeuwen
and co-workers have studied the effect of the bite angle
of diphosphine ligands on activity in palladium-cata-
lysed allylic alkylation [23]. They observed an increase in
activity with increasing bite angles. It has to be noted,
however, that the ligands studied co-ordinate in a cis-
fashion to the metal whereas the ligands in this study are
monoanionic terdentate ligands with the phosphorus co-
ordinated trans to each other. Nevertheless, we also
chromatography shows that 1:1 ratios of 12:13 are
always obtained independent of the catalyst used. This
indicates that allyl intermediates are most likely formed
in this reaction with cationic and neutral catalyst
precursors. As is common in allylic alkylation the allylic
acetate is oxidatively added to the Pd(II) centre to form
a palladium allyl species and consequently reacts with
the nucleophile to yield the product. The presence or
absence of other ligands besides the PCP or PCP? ligand
influences the activity of the catalysts directly. Weaker
co-ordinating anions lead to more reactive catalyst
species. This is obvious as the first step must be co-
ordination of the allylic acetate to the metal centre upon
which it is then oxidatively added. It is not clear yet if
co-ordinating anions such as halides dissociate from the
metal centre or remain co-ordinated during the catalytic
cycle. Thus, a question which still needs to be answered
is if either intimate ion-pairs such as [PdPCP(allyl)-
X]ꢃ[OAcꢂ] or [PdPCP(allyl)]2ꢃ[Xꢂ][OAcꢂ] or solvent
separated ion-pairs such as [PdPCP(allyl)X]ꢃjj[OAcꢂ]
or [PdPCP(allyl)]2ꢃjj[Xꢂ]jj[OAcꢂ] are formed when X
is a halide. Ongoing experiments in our laboratories
with the platinum analogues of 1 and 8 are addressing
this question. Preliminary results show that residues of
experiments with 1, 3, 8 and 10, which were examined
spectroscopically by 31P{1H}-NMR after the reaction
had finished, only had the original catalysts present.
Halides were not displaced by acetate although a large
excess of acetate is present in the reaction mixture.
Furthermore, an in situ 31P{1H}-NMR study of the
allylic alkylation reaction using 1 and 4 at 75 8C and 8,
8/AgBF4, 8/AgOTf and 10 at room temperature as
catalysts only showed species to be present of which
one is the starting material complex and the others can
be tentatively assigned to alkene co-ordination product
and oxidative addition product. Especially for the
experiments with 8, 8/AgBF4, 8/AgOTf and 10 only
trans-co-ordinated species are observed (typical cou-
observe an increase in activity with an increase in the PÃ
/
MÃP angle. The larger PÃMÃP angle and the more
/
/
/
flexible structure of 8 and its analogues is more suited
for oxidative addition of allyl acetates leading to better
conversions and a more efficient reaction than that of 1
and its analogues.
In fact, one may expect that palladium pincer com-
plexes of 1,3-(iPr2POCH2)2(C6H4) with two six-mem-
bered metallacycles will show even higher activity in
allylic alkylation than any of the complexes in this
study. Ligand 1,3-(iPr2POCH2)2(C6H4) is prepared in a
straightforward manner from 1,3-(HOCH2)2(C6H4).
However, the reaction of this ligand with either
[PdCl2(cod)] or [PdCl2(NCPh)2] did not give the desired
pincer complex but only polymeric species. We are in the
process of developing a different approach towards
[PdX{(C6H3)(CH2OPiPr2)2-2,6}] in order to complete
this study.
pling constants of 400ꢀ500 Hz). In none of the
/
experiments do we observe phosphorus ligand dissocia-
tion or ring-opening [21].
3. Conclusion
Interestingly, 8 and derivatives thereof prove to be
more active than complex 1 and its derivatives. We think
this can be attributed to two facts: Complexes of 7 are
electronically different to complexes of 1,3-(iPr2-
PO)2(C6H4) simply due to the fact that one donor is
an aryl dialkylphosphinite, the other an alkyl dialkyl-
phosphinite. But more importantly, comparing bond
angles of complexes of 1 and 8 shows that the PMP
angle for 8 is significantly larger by 88. Ligand bite
In conclusion we have prepared the novel unsymme-
trical pincer ligand 7. A series of palladium(II) com-
plexes derived from this ligand were compared to
complexes derived from the symmetrical pincer analo-
gue 1,3-(iPr2PO)2(C6H4). All complexes were applied
and found to be active catalysts for the allylic alkylation
of cinnamyl acetate with sodium dimethyl malonate.
They show little isomerisation activity giving only linear
product. However, palladium complexes of 7 are
significantly more active catalysts than palladium com-
plexes of 1,3-(iPr2PO)2(C6H4). We attribute this increase
in activity to the introduction of a six-membered
metallacycle into the pincer complex structure. This in
angles (or PÃ
/
MÃ
/
P angles) for metal complexes with
bidentate phosphorus ligands have been studied exten-
sively over the last 25 years. Dierkes and van Leeuwen
stated that they are a useful parameter for the explana-

Z. Wang et al. / Journal of Organometallic Chemistry 681 (2003) 189ꢀ
/195
193
turn leads to a more flexible complex structure and an
P angle. Moreover, the catalytic
activity depends strongly on the presence or absence of
concentrated in vacuum to give 7 (0.299 g, 0.84 mmol,
1
84%) as a colourless oil. H-NMR (300 MHz, CD2Cl2):
increase in the PÃ
/
MÃ
/
dꢁ
CH(CH3)2), 1.71ꢀ
2H, CH2O), 6.82ꢀ
NMR (75 MHz, CD2Cl2): dꢁ
17.9, 18.1, 28.2, 28.4, 28.6, 74.2 (d, Jꢁ
114.7, 118.0, 117.8, 117.7, 120.7, 129.3, 141.5 (d, Jꢁ
Hz), 159.7 (d, Jꢁ8 Hz) *
31P{1H}-NMR (121 MHz,
CD2Cl2): dꢁ
150.8 (CArOPiPr2), 156.0 (CH2OPiPr2).
/
0.89ꢀ
/
1.10 (m, 12H, CH(CH3)2), 1.58ꢀ
1.98 (m, 1H, CH(CH3)2), 4.61 (ABX,
7.14 (m, 4H, ArH) *
13C{1H}-
16.96, 17.04, 17.6, 17.8,
34 Hz, CH2O),
/
1.70 (m, 1H,
co-ordinating anions.
/
/
/
/
4. Experimental
/
/
8
4.1. General procedure
/
/
/
All manipulations were carried out using standard
Schlenk procedures under nitrogen. Solvents (CH2Cl2,
toluene, THF, diethyl ether) were freshly distilled from
either CaH2 or K/benzophenone under nitrogen. 3-
4.1.1.2. [PdCl{(C6H3)(OPiPr2)-2-(CH2OPiPr2)-6}]
(8). To a suspension of [PdCl2(cod)] (0.285 g, 1 mmol)
in toluene (10 ml) was added dropwise with stirring a
solution of 7 (0.356 g, 1 mmol) in toluene (10 ml). The
suspension was refluxed for 4 h during which time a
clear yellow solution formed. The solvent was removed
in vacuum and the solid was dissolved in CH2Cl2 and
filtered over a short SiO2 pad to give 8 (0.373 g, 0.75
mmol, 75%) as a pale yellow solid, which can be
recrystallised from diethyl ether. Single crystals were
grown from CH2Cl2/pentane. 1H-NMR (300 MHz,
Hydroxybenzylalcohol,
4-dimethylaminopyridine
DMAP, chlorodiisopropylphosphine, pyridine, acetic
anhydride, dimethyl malonate, cinnamyl alcohol, silver
triflate, silver tetrafluoroborate and sodium hydride
were purchased from Aldrich and used without further
purification. Silver acetate was purchased from Mal-
linckrodt. Silica gel (100ꢀ200 mesh), alumina, celite 545
/
and magnesium sulfate were purchased from Fisher
Scientific. [PdCl2(cod)] was prepared according to a
previously published method [24]. Complex 1 was
prepared according to the published procedure [5d].
1H-, 13C{1H}- and 31P{1H}-NMR spectra were re-
corded on a Varian Unity Inova 300 spectrometer at
ambient temperature of the probe using the deuterated
solvent to provide the field/frequency lock. Chemical
shifts are reported in parts per million relative to high
1
2
CD2Cl2): dꢁ
CH(CH3)2), 1.29 (dd, 6H, 1Jꢁ
CH(CH3)2), 1.36 (dd, 6H, 1Jꢁ
CH(CH3)2), 1.42 (dd, 6H, 1Jꢁ
CH(CH3)2), 2.37ꢀ
18.0 Hz, 2H,
ArH) *
13C{1H}-NMR (75 MHz, CD2Cl2): dꢁ
16.9, 17.3, 17.4, 18.3, 18.4, 27.6, 27.7, 27.9, 28.0, 28.6,
28.7, 28.86, 28.94, 78.1 (d, Jꢁ3 Hz, CH2O), 112.2 (d,
Jꢁ14 Hz), 121.3, 126.5, 132.9, 139.1 (d, Jꢁ11 Hz),
168.1 (d, Jꢁ14.9 Hz) *
31P{1H}-NMR (121 MHz,
CD2Cl2): dꢁ189.6 (d, Jꢁ
429 Hz, CArOPiPr2), 151.4
(d, 2Jꢁ429 Hz, CH2OPiPr2) *
C: 46.06 (45.89) H:
6.69 (6.69).
/
1.14 (dd, 6H, Jꢁ
14.4 Hz, 2Jꢁ
7.2 Hz, 2Jꢁ
7.2 Hz, 2Jꢁ
/
13.8 Hz, Jꢁ
/
6.9 Hz,
/
/6.9 Hz,
/
/
2.4 Hz,
3.0 Hz,
/
/
/
2.60 (m, 2H, CH(CH3)2), 4.78 (d, Jꢁ
CH2O), 6.69ꢀ7.4 (m, 3H,
16.7,
/
/
/
/
1
frequency of TMS for H and 13C{1H} and relative to
/
high frequency of 85% H3PO4 for 31P{1H} *
/
elemen-
/
/
tal analysis were performed by Oneida Research Ser-
vices.
/
/
2
/
/
Identity and purity of the products in the catalysis
experiments were determined by gas chromatography
with a gas chromatograph GC HP 5980A with flame
ionization detector (FID), equipped with a HP-1
capillary column (25 m) from Hewlett Packard, by gas
chromatography with a gas chromatograph GC HP
5890 Series II with a mass selective detector HP 5971
/
/
4.1.1.3. [PdX{(C6H3)(OPiPr2)2-2,6}] (XꢁBrꢂ 2, Iꢂ
/
3) [PdI{(C6H3)(OPiPr2)-2-(CH2OPiPr2)-6}] (9). A
solution of either complex 1 or 8 in acetone was treated
with excess NaX and stirred for several hours. Filtration
and evaporation of the solvent yielded the respective
(GCꢀMS), equipped with a HP-1MS capillary column
/
1
(30 m) from Hewlett Packard, and by NMR spectro-
scopy on a Varian Unity Inova 300 spectrometer.
complexes. H-NMR spectra of the products are iden-
tical to the starting material 1H-NMR spec-
tra *
/
31P{1H}-NMR (121 MHz, CD2Cl2): dꢁ
/
190.3
422 Hz, CArOPiPr2), 154.1
2
4.1.1. Ligand and complex synthesis
(2); 194.2 (3); 195.1 (d, Jꢁ
/
2
(d, Jꢁ
422 Hz, CH2OPiPr2) (9).
/
4.1.1.1. 1-(iPr2PO)-3-(iPr2POCH2)(C6H4) (7). To a
solution of 3-hydroxybenzylalcohol 6 (0.124 g, 1 mmol)
and DMAP (0.244 g, 2 mmol) in THF (20 ml) was
added dropwise with stirring ClPiPr2 (0.305 g, 0.318 ml,
2 mmol) at room temperature (r.t.). A precipitate
formed immediately and the mixture was stirred over-
night. The solvent was removed in vacuum and the
residue extracted repeatedly with toluene. The toluene
extracts were filtered over celite and the filtrate was
4.1.1.4. [PdOTf{(C6H3)(OPiPr2)2-2,6}] (4). A solu-
tion of complex 1 in THF was treated with one
equivalent of AgOTf and stirred for 1 h. Filtration
through celite and evaporation of the solvent yielded the
product. 1H-NMR spectrum are of the product is
identical to the starting material 1H-NMR spec-
trum *
/
31P{1H}-NMR (121 MHz, CD2Cl2): dꢁ
/
189.9.

194
Z. Wang et al. / Journal of Organometallic Chemistry 681 (2003) 189ꢀ195
/
4.1.1.5. [PdOAc{(C6H3)(OPiPr2)2-2,6}]
(5);
1H), 7.25ꢀ
MHz, CDCl3): dꢁ
126.8, 110.0, 76.3, 21.3 (ꢃ
aromatic region).
/
7.4 (m, 10H, ArH) *
139.7, 128.8, 128.3, 127.8, 127.7,
overlapping signals in the
/
13C{1H}-NMR (75
[PdOAc{(C6H3)(OPiPr2)-2-(CH2OPiPr2)-6}] (10).
A solution of either complex 1 or 8 in diethyl ether
was sonicated in the presence of AgOAc for 30 min.
Filtration and evaporation of the solvent yielded the
/
/
1
respective complexes. H-NMR spectra of the products
1
are identical to the starting material H-NMR spectra.
4.3. X-ray data collection and structure determination
NMR data (5) *
1.18ꢀ1.27 (m, 24H, CH(CH3)2), 1.90 (s, 3H,
CH3COOꢂ), 2.41 (m, 4H, CH(CH3)2), 6.42 (d, Jꢁ
Hz, 2H, ArH), 6.87 (t, Jꢁ Hz, 1H,
ArH) *
31P{1H}-NMR (121 MHz, CDCl3): dꢁ
187.5 * (10) *
31P{1H}-NMR (121 MHz, THF,
C6D6): dꢁ
184.4 (d, 2Jꢁ446 Hz, CArOPiPr2), 151.5
446 Hz, CH2OPiPr2).
/
1H-NMR (300 MHz, CDCl3): dꢁ
/
Crystals of 8 suitable for X-ray analysis were obtained
from slow diffusion of pentane into a saturated solution
of 8 in CH2Cl2. Intensity data were collected at 200 K on
a Mac Science DIP2030 imaging plate equipped with
/
/
8
/8
/
/
graphite-monochromated Moꢀ
/
Ka radiation (lꢁ
/
˚
/
/
0.71073 A). Unit cell parameters were determined by
autoindexing several images in each data set separately
with program DENZO. For each data set, rotation
images were collected in 38 increments with a total
rotation of 1808 about f. Data were processed by using
SCALEPACK. The structure was solved using the teXsan
system and refined by full-matrix least-squares (Table
2).
/
/
2
(d, Jꢁ
/
4.2. Allylic alkylation experiments
To a suspension of NaH (10 mg, 0.44 mmol) in THF
(1 ml) was added dimethyl malonate (46 ml, 0.4 mmol).
The mixture was stirred for 5 min and then transferred
to a Schlenk tube with a stir bar. To this was added the
Pd catalyst (0.01 mmol) (and one equivalent of a silver
salt) and cinnamyl acetate (34 ml, 0.2 mmol) or trans-
1,3-diphenyl-3-deuterium-2-propenyl-1-acetate 11 (50.3
mg, 0.2 mmol) in THF (1 ml). Undecane (10 ml, 0.047
mmol) was used as internal standard. The tube was then
sealed and the mixture was stirred for the prescribed
time at r.t. or immersed in a silicon oil bath at 75 8C,
5. Supplementary material
Crystallographic data for the structural analysis have
been deposited with the Cambridge Crystallographic
Data Centre, CCDC no. 205746 for compound 8.
Copies of this information may be obtained free of
charge from The Director, CCDC, 12 Union Road,
respectively. The reaction mixture was analysed by GCꢀ
MS. The products were isolated via column chromato-
graphy.
/
Cambridge CB2 1EZ, UK (Fax: ꢃ44-1223-336033; e-
/
Table 2
Crystallographic data
4.2.1. PhCDÄ
/
CHCH(OAc)Ph (11)
Compound
8
PhCÅCCH(OH)Ph was prepared from phenylacety-
/
lene (deprotonated with BuLi) and benzaldehyde in
THF and obtained as a colourless oil after column
chromatography. 1H-NMR (300 MHz, CD2Cl2): dꢁ
Empirical formula
M
C19H33ClO2P2Pd
497.27
200
Temperature (K)
Colour
/
yellow
triclinic
¯
P1
5.37ꢀ
/
5.40 (m, 1H), 6.38ꢀ
/
6.42 (m, 1H), 7.23ꢀ7.47 (m,
/
Crystal system
Space group
10H, ArH) *
/
13C{1H}-NMR (75 MHz, CD2Cl2): dꢁ
/
/
˚
143.3, 136.8, 131.9, 128.8, 128.0, 127.9, 126.7, 126.5,
89.3, 86.4, 75.1, 75.0.
a (A)
8.2150(1)
11.0300(2)
13.0140(3)
102.455(1)
95.322(1)
90.691(1)
1145.88(4)
2
˚
b (A)
˚
c (A)
PhCDÄ
/
CHCH(OH)Ph was prepared according to a
1
similar procedure by Grant and Djerassi [25]. H-NMR
a (8)
b (8)
g (8)
3
(300 MHz, CDCl3): dꢁ
(m, 1H), 7.23ꢀ7.47 (m, 10H, ArH) *
(75 MHz, CDCl3): dꢁ141.2, 131.9, 128.9, 128.63,
128.55, 127.5, 126.9, 126.2, 122.7, 68.0, 65.0.
PhCDÄCHCH(OH)Ph was dissolved in diethyl ether.
/
5.39 (d, 1H, Jꢁ
/
6.6 Hz), 6.40
/
/
13C{1H}-NMR
3
˚
U (A )
Z
/
m Moꢀ
/
Ka (cmꢂ1
Crystal dimensions (mm)
)
1.076
0.350ꢄ/0.200ꢄ0.150
/
/
u Range (8)
12.6ꢀ27.9
/
Acetic anhydride, triethylamine and a catalytic amount
of DMAP were added and the reaction mixture was
stirred for 3 h. Typical work-up gave 11 as a colourless
No. of data/restraints/parameters
No. of reflections observed
Final R2 and wR2 indices a
3841/0/226
3950
0.0344, 0.0992
0.0344
3841
Conventional R index [F2ꢀ
/
3s(F2)] a
solid after column chromatography. 1H-NMR of 11 and
1
Reflections with F2ꢀ3s(F2)
Goodness-of-fit
/
comparison to the H-NMR of PhCHÄ
/
CHCH(OAc)Ph
shows the absence of desired proton. 1H-NMR (300
MHz, CDCl3): dꢁ2.14 (s, 3H), 6.33 (m, 1H), 6.44 (m,
1.069
a
wꢁ
/
1/[s2(Fo)ꢃ0.0856(Fo)2].
/
/

Z. Wang et al. / Journal of Organometallic Chemistry 681 (2003) 189ꢀ
/195
195
situ by mixing the ligand and the metal salt. Thus, it is not a well-
defined species and most likely consists of co-ordination com-
pounds in which the ligand actually behaves as a bidentate ligand
and not as a tridentate pincer ligand.
mail: deposit@ccdc.cam.ac.uk or www: http://
www.ccdc.cam.ac.uk).
[12] (a) A.J. Blacker, M.L. Clark, M.S. Loft, M.F. Mahon, M.E.
Humphries, J.M.J. Williams, Chem. Eur. J. 6 (2000) 353;
(b) J.M. Brown, J.E. MacIntyre, J. Chem. Soc. Perkin Trans. 1
(1985) 961;
Acknowledgements
This work was supported by the US Department of
Energy Hydrogen Program.
(c) H. Kurosawa, J. Chem. Soc. Dalton Trans. (1979) 939.
[13] K. Fuji, N. Kinoshita, K. Tanaka, T. Kawahata, Chem. Com-
mun. (1999) 2289.
[14] (a) P.A. Evans, J.D. Nelson, J. Am. Chem. Soc. 120 (1998) 5581;
(b) P.A. Evans, J.D. Nelson, Tetrahedron Lett. 38 (1998) 1725.
[15] (a) B.M. Trost, M.-H. Hong, J. Am. Chem. Soc. 106 (1984) 6837;
(b) A.V. Malkov, I.R. Baxendale, D.J. Mansfield, P. Kocˇovsky`, J.
Chem. Soc. Perkin Trans. 1 (2001) 1234.
References
[1] C.J. Moulton, B.L. Shaw, J. Chem. Soc. Dalton Trans. (1976)
1020.
[2] (a) M. Albrecht, G. van Koten, Angew. Chem. Int. Ed. 40 (2001)
3750;
[16] F. Glorius, M. Neuburger, A. Pfaltz, Helv. Chim. Acta 84 (2001)
3178.
[17] (a) G.C. Lloyd-Jones, S.C. Stephen, Chem. Eur. J. 4 (1998) 2539;
(b) J.T. Singleton, Tetrahedron 59 (2003) 1837;
(c) M.E. van der Boom, D. Milstein, Chem. Rev. 103 (2003) 1759.
[3] C.M. Jensen, Chem. Commun. (2000) 2443.
[4] P. Dani, T. Karlen, R.A. Gossage, S. Gladiali, G. van Koten,
Angew. Chem. Int. Ed. 39 (2000) 743.
˚
(b) S. Hansson, P.-O. Norrby, M.P.T. Sjogren, B. .Ackermark,
¨
M.E. Cucciolito, F. Giordano, A. Vitagliano, Organometallics 12
(1993) 4940;
¨
(c) A. Gogoll, J. .Ornebro, H. Grennberg, J.E. Backvall, J. Am.
¨
[5] (a) M. Ohff, A. Ohff, M.E. van der Boom, D. Milstein, J. Am.
Chem. Soc. 119 (1997) 11687;
Chem. Soc. 116 (1994) 3631.
[18] M. Kawatsura, Y. Uozumi, T. Hayashi, Chem. Commun. (1998)
217.
(b) F. Miyazaki, K. Yamaguchi, M. Shibasaki, Tetrahedron Lett.
40 (1999) 7378;
[19] U. Burckhardt, M. Baumann, A. Togni, Tetrahedron: Asymmetry
8 (1997) 155.
(c) D. Morales-Morales, R. Redon, C. Yung, C.M. Jensen, Chem.
Commun. (2000) 1619;
(d) D. Morales-Morales, C. Grause, K. Kasaoka, R. Redon, R.E.
[20] (a) B.M. Trost, T.R. Verhoeven, J. Am. Chem. Soc. 98 (1976) 630;
(b) E. Keinan, M. Sahai, Chem. Commun. (1984) 648;
Cremer, C.M. Jensen, Inorg. Chim. Acta 300ꢀ302 (2000) 958;
/
˚
(c) B. .Ackermark, S. Hansson, B. Krakenberger, A. Vitagliano,
(e) D. Morales-Morales, R.E. Cremer, C.M. Jensen, J. Organo-
met. Chem. 654 (2002) 44.
K. Zetterverg, Organometallics 3 (1984) 679.
[21] Nevertheless at this point we cannot completely exclude the
possibility of one donor-atom being hemilabile. Such a species
may only be present in very low concentrations and thus may not
be observable in the 31P{1H}-NMR spectrum.
[6] R.B. Bedford, S.M. Draper, P.N. Scully, S.L. Welch, New J.
Chem. 24 (2000) 745.
[7] M.R. Eberhard, Z. Wang, C.M. Jensen, Chem. Commun. (2002)
818.
[22] P. Dierkes, P.W.N.M. van Leeuwen, J. Chem. Soc. Dalton Trans.
(1999) 1519.
[8] F. Gorla, A. Togni, L.M. Venanzi, A. Albinati, F. Lianza,
Organometallics 13 (1994) 1607.
[23] (a) M. Kranenburg, P.C.J. Kamer, P.W.N.M. van Leeuwen, Eur.
J. Inorg. Chem. (1998) 25;
[9] Q.-N. Lu, S. Chaloupka, L.M. Venanzi, Wuji Huaxue 17 (2001)
227.
(b) R.J. van Haaren, H. Oevering, B.B. Coussens, G.P.F. van
Strijdonck, J.N.H. Reek, P.C.J. Kamer, P.W.N.M. van Leeuwen,
Eur. J. Inorg. Chem. (1999) 1237.
[10] (a) B.M. Trost, D.L. van Vranken, Chem. Rev. 96 (1996) 395;
(b) G. Consiglio, R.M. Waymouth, Chem. Rev. 89 (1989) 257.
[11] (a) Longmire, Zhang, Tetrahedron Lett. 38 (1997) 1725;
(b) The authors reported the application of a chiral PCP pincer
ligand in allylic alkylation. However, the catalyst was prepared in
[24] D. Drew, J.R. Doyle, Inorg. Syn. 28 (1990) 348.
[25] B. Grant, C. Djerassi, J. Org. Chem. 39 (1974) 968.