Hybrid NCS palladium pincer complexes of thiophosphorylated benzaldimines and their ketimine analogs

Article abstract of DOI:10.1016/j.jorganchem.2012.03.029

A series of novel hybrid pincer ligands bearing thiophosphoryl and imino groups as donating arms was obtained by the condensation of 3- diphenylthiophosphorylbenzaldehyde with RNH₂ (R = Ph, ^tBu) to afford aldimine derivatives 2a, b o

Full text of DOI:10.1016/j.jorganchem.2012.03.029

Journal of Organometallic Chemistry 711 (2012) 52e61
Contents lists available at SciVerse ScienceDirect
Journal of Organometallic Chemistry
journal homepage: www.elsevier.com/locate/jorganchem
Hybrid NCS palladium pincer complexes of thiophosphorylated benzaldimines
and their ketimine analogs
Diana V. Aleksanyan a, Vladimir A. Kozlov a, Nikolay E. Shevchenko b, Valentine G. Nenajdenko b,
,
Andrei A. Vasil’ev c, Yulia V. Nelyubina a, Ivan V. Ananyev a, Pavel V. Petrovskii a, Irina L. Odinets a
*
a A. N. Nesmeyanov Institute of Organoelement Compounds, Russian Academy of Sciences, 28 Vavilova Street, Moscow 119991, Russia
b Moscow State University, Leninskie Gory 1-3, 119991 Moscow, Russia
c N. D. Zelinsky Institute of Organic Chemistry, Russian Academy of Sciences, 47 Leninsky Prosp., 119991 Moscow, Russia
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of novel hybrid pincer ligands bearing thiophosphoryl and imino groups as donating arms was
obtained by the condensation of 3-diphenylthiophosphorylbenzaldehyde with RNH2 (R ¼ Ph, tBu) to
afford aldimine derivatives 2a, b or by the reaction of lithiated 3-(thiophosphoryl)bromobenzene with an
appropriate substituted lactam yielding their ketimine analogs 2cee with 5e7-membered azacy-
cloalkene moieties. The direct cyclopalladation of the ligands with (PhCN)2PdCl2 in MeCN under reflux
Received 13 January 2012
Received in revised form
22 March 2012
Accepted 23 March 2012
led to k
3-NCS pincer complexes 3bee with two five-membered fused metallacycles, isolated in low to
Keywords:
moderate yields (12e53%); their structures were confirmed by multinuclear NMR and X-ray diffraction
study. The palladacycles demonstrated high activity as (pre)catalysts for the Suzuki cross-coupling of
phenylboronic acid with aryl bromides, which was found to increase in the series 3e w 3d <3c< 3b, i.e.
passing from ketimine derivatives 3d, e with larger azacycloalkene to their analog with smaller five-
membered cyclic imine moiety and further to benzaldimine complex 3b. The tendency observed was
explained by the controlled release of Pd(0) catalytically active species in the case of less sterically
hindered complexes.
Organothiophosphorus ligand
Cyclopalladation
Palladium
Pincer complex
Suzuki reaction
Ó 2012 Elsevier B.V. All rights reserved.
1. Introduction
palladium complexes, the systematic investigations in this area are
rather scarce and optimal routes for desymmetrization are still
Over the last two decades pincer complexes bearing a specific
monoanionic terdentate backbone have emerged as a unique class
of organometallic compounds mostly owing to their high catalytic
activity in various chemical processes. A vast majority of pincer
complexes is based on symmetrical derivatives (bis(phosphines),
bis(thioethers), bis(amines) etc.), and their structural features as
well as physical and chemical properties have been extensively
reviewed [1,2]. But recently the interest in this area has been
switched to hybrid derivatives with two different coordination
arms, especially those combining “hard” and “soft” donor atoms
[3]. The unsymmetrical pincer palladacycles with such hemilabile
ligands demonstrated high catalytic activity, e.g., for the Suzuki
cross-coupling, often surpassing in the efficiency their symmetric
analogs [4,5]. However, despite an increasing number of publica-
tions devoted to the catalytic performance of hybrid pincer
unclear.
As a part of our ongoing studies on pincer-type palladacycles
with organothiophosphorus ligands, we have developed a series
of 5,6-membered Pd(II) pincer complexes based on
3-diphenylthiophosphoryloxybenzaldimines
I (Fig. 1), which
were proved to be reasonably good (pre)catalysts for the Suzuki
coupling between phenylboronic acid and aryl bromides [6].
Among the related NCS palladacycles IIeIV formed by thio-
phosphorylated benzothiazoles, bearing as coordination arms
along with the P]S group the imine moiety of a benzothiazole
ring, complexes II, III with two fused metallacycles of different
sizes were much more active than their 5,5-membered coun-
terpart IV [7].
According to the kinetic experiments, 31P NMR monitoring of
a catalyst destiny after the catalytic cycle, Hg poisoning tests, and
observation of palladium black formation, complexes IeIII repre-
sented only precatalysts, liberating catalyticallyactive Pd(0) particles.
In other words, these data support the Pd(0)/Pd(II) catalytic cycle. At
the same time, in the case of 5,5-membered NCS palladacycle IV and
* Corresponding author. Tel.: þ7 499 135 9356.
E-mail address: odinets@ineos.ac.ru (I.L. Odinets).
0022-328X/$ e see front matter Ó 2012 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2012.03.029

D.V. Aleksanyan et al. / Journal of Organometallic Chemistry 711 (2012) 52e61
53
S
S
Ph2P
S
O
O
X
Ph2P
S
Pd
Cl
Ph2P
Pd
Cl
N
Ph2P
Pd
Cl
N
Pd
N
S
S
S
N
R
Cl
IV
t
R=Ph, OMe, Bu
X=O, NH
II
5,5-membered
complex
I
III
the least active
in Suzuki reaction
5,6-membered palladacycles
Fig. 1. NCS hemilabile pincer complexes with organothiophosphorus ligands.
related 5,5-membered SCS’ pincer complexes [8] bearing thio-
phosphoryl and thiocarbamoyl donating arms, which retain their
structure over the catalytic process, we cannot disclaim the alterna-
tive Pd(II)/Pd(IV) mechanism. Therefore, it seems reasonable to
perform the synthesis of NCS pincer ligands and corresponding
5,5-membered palladacycles having another type of imine moiety to
get more insight into this phenomenon.
Herein, we report on the synthesis of new NCS unsymmetrical
pincer ligands, namely, thiophosphorylated benzaldimines and
their ketimine analogs, bearing as coordination arms along with the
phosphine sulfide group either cyclic or acyclic imine moiety,
which are capable of forming NCS pincer complexes with two five-
membered fused metallacycles. Of particular interest is the
preparation of ketimine derivatives which could afford hybrid
pincer ligands with 5e7-cyclic ketimine fragment, unknown for the
symmetric NCN bis(imine) pincer prototypes [9]. The introduction
of a bulky imine moiety would allow additional estimation of the
effect of steric hindrances on the catalytic performance of Pd(II)
pincer complexes derived in the model Suzuki reaction.
imines on silica gel, and were used in the subsequent synthesis of
palladium complexes without purification.
A convenient synthetic route to the thiophosphoryl-ketimino
analogs with a cyclic imine moiety was developed via metalation
of thiophosphorylated bromobenzene with butyllithium followed
by the treatment with an equivalent of appropriate substituted
lactam according to the earlier developed synthetic strategy [11].
The process was devised starting from commercially available 1,3-
dibromobenzene which was converted into the key precursor e 3-
(bromophenyl)diphenylphosphine
sulfide
via
mono-
phosphorylation with Ph2PCl followed by in situ addition of
elemental sulfur to the corresponding phosphine (Scheme 2) [12].
The sequential treatment of monothiophosphorylated bromide
with n-BuLi and N-vinyl pyrrolidone, N-(diethoxymethyl)valer-
olactam or N-vinyl caprolactam led to the formation of the target
ligands with five-, six-, and seven-membered azacycloalkene
moiety, isolated as viscous yellowish oils in moderate yields
(38e61%) after chromatography purification.
The structures of ligands 2aee were confirmed by multinuclear
NMR (31P, 1H, and 13C) and IR spectroscopy data. Thus, in the IR
spectra the characteristic
1618e1646 cmꢀ1, while the band corresponding to the thio-
phosphoryl group stretching vibrations appeared at ca. 640 cmꢀ1
n(C]N) band was observed at
2. Results and discussion
.
The singlet signals of the phosphorus atoms in the 31P NMR spectra
of 2aee were detected at 43e44 ppm e the region typical for tri-
arylphosphine sulfides. The signals of aldimine protons in the 1H
NMR spectra of acyclic imines 2a and 2b appeared as singlets at
8.24 and 8.44 ppm, respectively. The characteristic resonances of
(CH2)n units in the 1H NMR spectra of their ketimine analogs 2cee
were observed in the expected regions (see Experimental). The 13C
NMR spectroscopy data is also consistent with the proposed
2.1. Synthesis of ligands and complexes
Thiophosphorylated benzaldimines 2a, b with an acyclic imine
moiety were readily derived from the condensation of 3-
diphenylthiophosphorylbenzaldehyde 1 with aniline or tert-butyl-
amine, respectively, as viscous yellowish oils in almost quantitative
yields (Scheme 1). The starting thiophosphorylated benzaldehyde 1
was synthesized, in turn, by the reaction of the Grignard reagent of
2-(3-bromophenyl)-1,3-dioxolane with diphenylchlorophosphine
followed by the addition of sulfur and acidic deprotection according
to the literature procedure [10]. Note that ligands 2a, b were
obtained with w5% admixture of the starting benzaldehyde 1,
which we failed to separate due to the decomposition of the target
structures of the final ligands (e.g., dC
]
in the range of
N
165e174 ppm, etc.).
The direct cyclopalladation of these ligands at the C2 position of
a central phenyl ring to afford the corresponding NCS pincer
complexes 3aee was found to proceed most efficiently while
O
O
NR
HC
CHO
CHO
i-iv
HO
RNH2, MgSO4
OH
TosH
S
PPh2
S
PPh2
CH2Cl2, r.t.
Br
Br
2a,b
1
i Mg, THF,
ii Ph2PCl,
t
R=Ph (a), Bu (b)
iii [S],
iv HCl
Scheme 1. Synthesis of 3-thiophosphorylbenzaldimines 2a, b.

54
D.V. Aleksanyan et al. / Journal of Organometallic Chemistry 711 (2012) 52e61
i-iii
i-ii
Ph2P
S
Ph2P
S
Br
Br
Br
( )n
i BuLi
i BuLi
ii Ph2PCl
iii [S]
N
O
ii
2c-e
RN
( )n
n=1 (c), 2 (d), 3 (e)
R = vinyl (c,d),
CH(OEt)2 (e)
Scheme 2. Synthesis of thiophosphorylated ketimines 2cee.
refluxing the equimolar amounts of 2aee and (PhCN)2PdCl2 in
acetonitrile for 1e8 h depending on the ligand structure (Scheme 3).
Although the yields of the complexes isolated in pure form were not
high, ranging from 12 to 54%, the harnessing of other conditions was
unsuccessful. Thus, the reaction in dichloromethane at room
temperature (i.e., under milder conditions) was too sluggish,
providing only w15% of complex 3b in a month (according to the 31P
NMR monitoring data) with the predominance of some coor-
dinatively bonded complexes of PdCl2 [13]. The application of
a nonpolar solvent (benzene) under reflux did not afford a pincer
product at all. Moreover, elevation of the reaction temperature to
120 ꢁC (replacement of MeCN for benzonitrile) did not result in any
significant increase of the yields of the target products. Note that we
failed to isolate complex 3awith a phenyl substituent at the nitrogen
atom of the aldimine group due to its decomposition on silica gel.
Complexes 3bee are pale-yellow or yellow crystalline solids
that are air and moisture stable both in the solid state and in
solution as well as thermally stable up to T > 240 ꢁC.
P]S group and sp2-nitrogen atom e causes the distortion of the
coordination polyhedron of the palladium atom, the bending along
the N.S line. In 3b and 3s, the dihedral angles between the CPdSN
and ClPdSN planes are as high as w4ꢁ (3.6 and 4.5ꢁ, respectively),
the angles C(2)Pd(1)Cl(1) are 175.4(1) and 175.6(1)ꢁ, and the
chlorine atoms deviate by 0.15e0.19 Å from the mean square plane
of S(1), N(1), C(2), and Rd(1) atoms. In the case of 3d with six-
membered azacycloalkene moiety, this distortion is much less
pronounced. Thus, in the molecules of 3d_nosolv and 3d_solv the
angles C(2)Pd(1)Cl(1) are 178.0(1) and 176.7(1)ꢁ, the angles
between the planes C(2)Pd(1)S(1)N(1) and Cl(1)Pd(1)S(1)N(1) are
1.1 and 0.9ꢁ, and the deviation of the atom Cl(1) is only 0.01 and
0.03 Å, respectively. Moreover, both 5SN and 5PS metallacycles in
complexes 3b, 3s, and 3d_nosolv are planar, while in the case of
3d_solv the first one is also planar, but the second has a “twist”
conformation with the atoms P(1) and S(1) deviating by 0.22 and
0.38 Å, respectively. Note that this study revealed the unexpected
crystallization of palladacycle 3b in a chiral space group P43 as
a result of cross-linking supramolecular aggregation, which is a rare
case for pincer complexes that do not include a stereogenic center
in their molecule (see, e.g., ref. [14]). This is, apparently, a result of
noncenrosymmetric supramolecular assembling formed by rather
The structures of the complexes were assigned based on the 31P,
1H NMR and IR spectroscopy data as well as X-ray diffraction anal-
ysis. Thus, the bands at ca. 600 and 1540 cmꢀ1 in the IR spectra of
3bee are diagnostic of the coordinated P]S and C]N bonds, being
shifted to the lower frequencies relative to those of the free ligands
by 41e44 and 76e96 cmꢀ1, respectively. The strongly deshielded
signals of phosphorus resonances for 3bee (DdP ¼ 19e25 ppm)
unambiguously indicate the coordination of thiophosphoryl group,
while the upfiled shift of the imine proton signal in the 1H NMR
spectra of 3b (DdH ¼ 0.17 ppm) evidences the complexation by the
aldimine moiety. Moreover, the disappearance of low-field doublets
of the H(C2)-protons with an overall reduction of aromatic proton
intensity by1H confirms the occurrence of metalation. Of note is also
a general tendency of strong concomitant displacement of the
signals of remaining benzene core protons to lower frequencies
upon complex formation (see Experimental). The elemental analysis
data is also in a good agreement with the suggested compositions of
complexes 3bee. The structures of the palladacycles derived were
also confirmed by the X-ray diffraction analysis for complexes 3b, 3c,
and 3d, the latter crystallizing with a solvent chloroform molecule
(3d_solv) and without it (3d_nosolv).
weak CeH.p contacts; CeH.Cl and CeH.S interactions
complete the formation of 3D cross-linked network (Fig. S1 in
Supplementary material). The centrosymmetric crystal structures
of the other complexes are assembled by numerous but also weak
CeH.Cl, CeH.p, CeH.S, and H.H contacts together with p.p
interactions in both solvated and non-solvated 3d complexes as
well as CeH.Cl contacts with participation of chloroform mole-
cules and Cl.Cl interactions between them in 3d_solv only.
It should be mentioned that the yields of the pincer-type
complexes obtained via direct cyclopalladation in the case of the
According to the X-ray diffraction data for these complexes
(Figs. 2e4), the presence of different donor atoms e sulfur atom of
3a 20%*
PdCl2(PhCN)2
3b 34%
3c 54%
3d 12%
3e 34%
Ph
Ph
PPh2
S
P
S
CH3CN, reflux
N
N
Pd
2a-e
31P NMR,
R
R
*
Cl
3a-e
not isolated
t
C=NR: R=Ph (a), Bu (b) or
(d),
(e)
(c),
N
N
N
Scheme 3. Direct cyclopalladation of 3-thiophosphorylated benzaldimines 2a, b and
their ketimine analogs 2cee.
Fig. 2. General view of complex 3b in representation of atoms by thermal ellipsoids
(p ¼ 50%).

D.V. Aleksanyan et al. / Journal of Organometallic Chemistry 711 (2012) 52e61
55
to 53e66% upon elongation of a thiophosphoryl coordination arm
[6]. Therefore, desymmetrization of a pincer ligand structure via
introduction of the second thiophosphoryl coordination arm
instead of imine group facilitates the process of direct cyclo-
metallation which may start either from the first coordination of
the thiophilic palladium atom with the soft sulfur donor or from the
simultaneous coordination of a metal with both donor centers. In
other words, the desymmetrization of a pincer ligand obviously
serves as a factor increasing the efficiency of direct cyclopalladation
of the above ligands to produce pincer-type complexes.
2.2. Catalytic studies
It should be noted that before the beginning of our investigation,
the data concerning the catalytic activity of pincer palladacycles
with organothiophosphorus ligands were limited to a single publi-
cation of Le Floch [17] dealing with the palladium-catalyzed elec-
trophilic allylation of aldehydes with allylstannanes using
Fig. 3. General view of complex 3c in representation of atoms by thermal ellipsoids
(p ¼ 50%).
symmetric SPS pincer complex featuring a central
l
4-phosphinine
other SCY (Y]S0, N) hybrid pincer ligands bearing thiophosphoryl
function [6e8] are much higher than those observed for
thiophosphoryl-imino ligands 2aee. However, it is well known that
the direct cyclopalladation of symmetric bis(imine) derivatives is
often failed due to the kinetically preferred metalation at
4,6-positions of the central phenyl ring to give bis(monopallada-
cyclic) species [15,16]. This was observed for dialdiminobenzenes
with Et [15a,b], CH2Ph, n-Bu, Oct [15a], and tetrahydrofurfuryl [15c]
substituents at the imine nitrogen atoms upon metalation with
Pd(OAc)2 in chloroform [15a] or toluene [15c] or with Li2[PdCl4] in
methanol [15b]. Only in the case of bis(N-cyclohexylimine) deriv-
atives the pincer-type product was obtained in less than 15% yield
under reflux with Pd(OAc)2 in glacial acetic acid [9a]. Taking into
account the strong dependence of the reaction time on the nature
of an imine donating moiety for thiophosphoryl-imino ligands
2aee, it can be concluded that in our case the formation of pincer
complexes 3aee is also complicated by the processes involving
imine group (e.g., metalation at the C6-position of the benzene
core, hydrolysis of an aldimine bond under the action of liberating
HCl and so on). Although we did not manage to isolate and analyze
the side products in any case, the comparison of the results on the
direct cyclopalladation of NCN bis(imine) derivatives and hybrid
pincer ligands with thiophosphoryl(oxy) and imine donating
groups allowed to conclude that the yield of pincer complexes
increases from <15% in the case of symmetric bis(imine) deriva-
tives [9a] to 12e54% upon the replacement of one of the N-donor
groups for the soft S-donor phosphine sulfide moiety, and further
unit and two lateral phosphine sulfide groups. As mentioned above,
in our study we used the SuzukieMiyaura cross-coupling as a model
reaction to estimate the catalytic activity of SCS0 and NCS pincer
palladacycles and elucidate if they could serve as true catalysts.
Therefore, complexes 3bee were tested in the cross-coupling
between aryl bromides and phenylboronic acid similar to the study
performed for the related complexes. Note that all experiments were
carried out under the conditions previously successfully adopted for
the other NCS pincer complexes: heating in DMF at 120 ꢁC using
K3PO4 as a base, which allowed us to make a correct comparison
between the results both for the present series and complexes IeIV.
Therefore, this test is an appropriate activity evaluation benchmark.
Fig. 5 outlines the results as kinetic evolution of the conversions of
two electronically varied substrates, namely, 4-bromoacetophenone
and 4-bromoanisole in time at the catalyst loading of 1 mol%. As is
obvious, in the reaction of 4-bromoacetophenone (Fig. 5A) complex
3b based on aldimine derivative appeared to be more active than its
ketimino analogs 3cee, providing almost quantitative conversion of
the substrate (97%) in 1 h. Among the ketimino analogs, complex 3c
with a five-membered azacycloalkene moiety was slightly more
active than the other ones for the first hour of the reaction, while
subsequently palladacycles 3cee equalized in their activity towards
this easy-to-couple substrate. While passing to the electronically
deactivated 4-bromoanisole (Fig. 5B), the structureeactivity depen-
dences for the complexes of the present series became more
pronounced. Again, complex 3b was the most active one, providing
over90% ofthecorrespondingbiaryl product in2 h, whileits ketimino
counterpart 3c with a five-membered azacycloalkene moietyled only
to 67% conversion in 5 h. More sterically hindered complexes 3d, e
based on six- and seven-membered cyclic imine derivatives were
almost inactive in the reaction with 4-bromoanisole at this concen-
tration of a (pre)catalyst.
Using the most active complex of this study e aldimine deriv-
ative 3b as a representative example, it was shown that the high
conversion level of 4-bromoacetophenone could be reached within
1 h even when the amount of a catalyst precursor was reduced to
0.1 mol%. Although further reduction in the catalyst loading to
0.01 mol% significantly decelerates the coupling, the overall yield of
the biaryl product in 5 h still remains rather good (84%) (Fig. 6A).
However, further decrease of the catalyst loading up to 0.001 mol%
led to a drop in the conversion to several percent only. For the less
active substrate, 4-bromoanisole (Fig. 6B), unlike ketimino deriva-
tives 3cee being inactive already at the 1 mol% concentration,
complex 3b showed rather high activity even at 0.1 mol% loading,
although no coupling was observed upon further decrease (to
0.01 mol%) of the complex amount.
Fig. 4. General view of complex 3d_nosolv (3d_solv is practically the same) in
representation of atoms by thermal ellipsoids (p ¼ 50%).

56
D.V. Aleksanyan et al. / Journal of Organometallic Chemistry 711 (2012) 52e61
Fig. 5. Plots of conversion vs time for the couplings of 4-bromoacetophenone (A) and 4-bromoanisole (B) with phenylboronic acid catalyzed by 1 mol% of complexes 3bee.
The 31P NMR monitoring of a complex’s destiny after the cata-
lytic cycle (the representative reaction was the coupling of 4-
bromoacetophenone and phenylboronic acid at 1 mol% catalyst
loading) revealed that the least active palladacycles 3d, e decom-
pose with the formation of only one phosphorus-containing
product with dP at ca. 38 ppm, being situated in the region of free
P]S derivatives, thereby, suggesting the rupture of the PdeS and
probably PdeC bonds. The 31P NMR spectrum for the reaction
catalyzed by complex 3c, having a five-membered azacycloalkene
moiety, displayed along with the signal at w38 ppm two singlets at
w26 and w30 ppm. These species were tentatively assigned, as
before in the case of complex IV [7], to the metalated and non-
metalated phosphoryl-containing derivatives, presumably result-
ing from the transformation of the thiophosphoryl group into P]O
moiety and thermally induced demetalation of the starting
complex under the reaction conditions. At the same time, the 31P
NMR spectrum of the reaction mixture catalyzed by the most active
complex 3b showed an intensive signal at 31.2 ppm and a set of
resonances in the range of 58.3e62.36 ppm e the region typical for
metalated P¼S-coordinated species. Thus, none of the pincer
complexes 3bee serves as a catalyst in its starting form.
of the stability and activity in catalysis of symmetrical SCS palla-
dium pincer complexes formed by bis(thioether) ligands revealed
that these compounds are not actual catalysts in the Heck cross-
coupling and leached Pd(0) species following Pd(0)/Pd(II) mecha-
nism [19] similar to 5,6-membered hybrid SCS0 and NCS analogs in
the Suzuki reaction [6,7]. At the same time, the stability of hybrid
SCS0 Pd complexes tested in the Suzuki reaction [8] gave evidence of
Pd(II)/Pd(IV) mechanism.
The kinetic studies performed for the conversion of 4-
bromoacetophenone (see Fig. 5A) revealed the sigmoidal kinetics
and distinct induction period required for the generation of cata-
lytically active species at least for ketimino derivatives 3cee. This
fact substantiates our conclusion that these complexes serve only
as precatalysts in contrast to the above mentioned 5,5-membered
SCS0 complexes [8] and NCS complex IV [7]. Moreover, the lack of
activity of complexes 3bee under conditions of the so-called
mercury drop test [20] indicates that the true catalysts in these
systems have Pd(0) nature. Thus, addition of a few drops of Hg(0) to
the reaction mixtures consisting of 4-bromoacetophenone, phe-
nylboronic acid, K3PO4, and 0.1 mol% of complex 3b or 1 mol% of
complexes 3cee prior to heating led only to the trace amounts of
a biaryl product in each case. Furthermore, when excess of Hg(0)
was added at the reaction time of 30 min (3b) or 1.5 h (3cee), the
coupling terminated at the conversion level of 75e78%, 72%,
51e52%, and 33e35% in the case of complexes 3b, 3c, 3d, and 3e,
The mechanism of the cross-coupling reactions catalyzed by
pincer complexes is a matter of considerable debate and both
Pd(0)/Pd(II) and Pd(II)/Pd(IV) catalytic cycles were proposed
[2e,18]. Motivated by conflicting reports, the detailed investigations
Fig. 6. Plots of conversion vs time for the couplings of 4-bromoacetophenone (A) and 4-bromoanisole (B) with phenylboronic acid catalyzed by different amounts of complex 3b.

D.V. Aleksanyan et al. / Journal of Organometallic Chemistry 711 (2012) 52e61
57
Table 1
NCS Pd(II) pincer complexes with organothiophosphorus ligands catalyzing the
Suzuki cross-coupling of 4-bromoanisole and phenylboronic acid. Reaction condi-
tions: DMF, 120 ꢁC, 5 h, K3PO4.
Cat. load (mol%) Yield of 4-methoxybiphenyl, %*
I
II
III IV 3b 3c 3d 3e
R ¼ OMe R ¼ tBu X ¼ O X ¼ NH
1
0.1
0.01
86
99
0
91
84
3
98.5 99
100 99.5
92 98
61 97 93 67
3
e
2
e
81 92 78
92 77
e
5
e
e
e
*Estimated by GC.
passing from the aldimine-based derivative to its ketimine
analogs arranged in order of increasing azacycloalkene ring sizes,
corresponds to the increase of readiness of the reductive elimi-
nation of LePdePh species 5 to generate Pd(0) particles and
biaryls 6. If that is the case, the more efficient generation of Pd(0)
particles by sterically crowded complexes 3d, e can explain the
lower activity of these precatalysts in the Suzuki reaction, since
higher concentrations of the released Pd(0) particles would
facilitate their agglomeration which, in turn, would lead to the
formation of inactive species. Therewith, the generation of these
catalytically active species in the systems with five-membered
azacycloalkene complex 3c and especially aldimine derivative
3b occurs in a controlled manner, which is responsible for their
higher activity. The controlled (constant and reluctant) in situ
release of Pd(0) catalytically active species from their inactive
precursors is a new concept in Pd(0)/Pd(II)-catalysis [22], which
Fig. 7. Conversion of 4-bromoacetophenone as a function of time for the Suzuki
coupling with 0.1 mol% of complex 3b in the presence of Hg(0).
respectively (3e4 points up to t ¼ 4 h). Fig. 7 illustrates the
representative reaction catalyzed by complex 3b.
In general, the formation of Pd(0) particles can occur via two
pathways: homolytic cleavage (of the compound itself) or by
intervention of one of the aryl reagents in the catalytic reaction
via biaryl coupling. Previously, we have shown that NCS pincer
complexes IIeIV based on thiophosphorylated benzothiazoles
can produce Pd(0) catalytically active species just by the reduc-
tive elimination of ortho-metalated species with an aryl group
introduced by phenylboronic acid [21]. So we decided to check if
a similar mechanism of formation of catalytically active Pd(0)
particles is operative in the case of complexes 3bee (Scheme 4).
Indeed, the heating of complex 3e with 10 eq. of phenylboronic
acid and 15 eq. of K3PO4 in DMF at 120 ꢁC for 30 min led to the
formation of Pd black and two phosphorus-containing
compounds with dP at 71 ppm, diagnostic of the starting pincer
complex, and 38 ppm - the signal characteristic of the free P]S
derivatives which can be tentatively assigned to the biaryl
product 6. Note that the latter signal has already been detected in
the 31P NMR spectra of the reaction mixture after catalytic cycle
(vide supra) confirming the formation of biaryl product over the
Suzuki reaction. Note that for the related LePdePh species of
type 5 derived from bis(phosphine) PCP pincer complexes [18a,b]
no reductive elimination with the liberation of Pd(0) and
formation of biaryls was detected, and therefore, the Pd(II)/Pd(IV)
catalytic cycle was suggested. However, in the case of NCS pincer
complexes 3bee, the hemilabile coordination may facilitate such
a reductive elimination by decoordination of one of the ancillary
arms in order to achieve the ortho-mutual disposition of two
leaving phenyl fragments. In this respect, apparently the increase
of steric hindrances in the series 3b < 3c < 3d < 3e, i.e., while
may become
a powerful alternative to the conventional
approaches that imply addition of various stabilizers for Pd(0)
species (phosphine or carbene ligands, tetrabutylammonium
bromide [23] and so on).
Finally, to compare the catalytic activity of NCS pincer Pd(II)
complexes with organothiophoshorus ligands of different nature,
Table 1 outlines the results obtained for the coupling of a repre-
sentative substrate e 4-bromoanisole.
Independently of the nature of the true catalyst, a general
tendency is obvious: the catalytic activity of NCS pincer-type pal-
ladacycles based on hemilabile thiophosphoryl-imino ligands
increases while passing from the complexes with two fused five-
member
ed metallacycles IV, 3bee to 5,6-membered complexes
IeIII, and from ketimine derivatives to their aldimine analogs, and
further to the complexes having as a coordination arm the flexible
heterocyclic imine moiety.
Finally, efforts in application of hybrid pincer-type palladacycles
with organothiophosphorus ligands, including those of NCS-type,
in different catalytic reactions where they could serve as true
catalysts [24] are ongoing in our laboratory.
3. Conclusions
To summarize the results presented, wehavedevised the original
synthetic approaches to novel hybrid pincer ligands bearing as
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
P
S
P
S
P
S
P
S
reductive
PhB(OH)2
K3PO4
elimination
Pd Ph
N
Pd(0)
Pd Cl
N
Ph
N
+
Pd
Ph
N
5
3b-e
6
Scheme 4. Suppositional way of generation of Pd(0) species.

58
D.V. Aleksanyan et al. / Journal of Organometallic Chemistry 711 (2012) 52e61
coordination arms either cyclic or acyclic imine moiety along with
the thiophosphoryl group. The direct cyclopalladation of these
ligands afforded NCS pincer palladium complexes with two five-
membered fused metallacycles. The catalytic activity of the
complexes obtained in the Suzuki coupling of phenylboronic acid
with aryl bromides was shown to increase while passing from the
ketimine derivatives 3d, e to their analog 3c with a smaller five-
membered cyclic imine moiety, and further to complex 3b formed
by thiophosphorylated benzaldimine. This tendency was explained
by the reduction of steric hindrances with reducing azacycloalkene
ring sizes and, more profoundly, upon a transition to the aldimine
derivative that led to the controlled release of Pd(0) particles.
1278(w), 1310(w), 1436(m), 1481(m), 1499(m), 1513(w), 1601(m),
1621(m) (C]N), 3050(w).
4.2.2. N-{[3-Diphenylthiophosphoryl]phenyl]methylidene}-tert-
butylamine, 2b
31P{1H} NMR (161.97 MHz, CDCl3):
d
43.25 ppm. 1H NMR
(400.13 MHz, CDCl3): d 1.25 (s, 9H, C(CH3)3), 7.42e7.53 (m, 7H, HAr),
7.60 (dd, 1H, H(C4), 3JHH ¼ 7.6 Hz, 3JHP ¼ 12.6 Hz), 7.72 (dd, 4H, o-H
3
3
in P(S)Ph2, JHH ¼ 7.0 Hz, JHP ¼ 13.4 Hz), 8.06 (d, 1H, H(C2),
3JHP ¼ 13.7 Hz), 8.07 (d, 1H, H(C6), JHH ¼ 6.9 Hz), 8.24 (s, 1H,
3
CH]N). 13C{1H} NMR (100.61 MHz, CDCl3):
d 29.33 (s, CH3), 57.34
(s, C(CH3)3), 128.30 (d, m-C in P(S)Ph2, 3JCP ¼ 12.3 Hz), 128.37 (d, C5,
3JCP ¼ 12.6 Hz),129.51 (d, C6, 4JCP ¼ 2.8 Hz),131.39 (d, p-C in P(S)Ph2,
4JCP ¼ 2.8 Hz), 131.94 (d, o-C in P(S)Ph2, 2JCP ¼ 10.8 Hz), 132.17 (d, C3,
1JCP ¼ 68.3 Hz), 132.37 (d, ipso-C in P(S)Ph2, 1JCP ¼ 87.4 Hz), 132.62
4. Experimental
2
2
(d, C2, JCP ¼ 11.4 Hz), 133.32 (d, C4, JCP ¼ 10.5 Hz), 137.36 (d, C1,
4.1. General remarks
3JCP ¼ 12.0 Hz), 153.81 (s, C]N). IR (KBr,
n
/cmꢀ1): 500(w), 517(m),
615(w), 642(s) (P]S), 691(s), 716(s), 750(w), 1103(m), 1203(w),
1367(w), 1436(m), 1480(w), 1573(vw), 1587(vw), 1642(m) (C]N),
2867(w), 2901(w), 2928(w), 2966(m), 3053(w).
All manipulations with organolithium compounds were
carried out using flame-dried glass equipment and anhydrous
solvents under argon, otherwise no precautions were taken to
exclude air and moisture. Tetrahydrofuran was distilled from
sodium/benzophenone, dichloromethane and acetonitrile were
distilled over P2O5. 2-(3-Bromophenyl)-1,3-dioxolane [25] and
3-(diphenylthiophosphoryl)benzaldehyde 1 [10] were obtained
according to the literature procedures. All other chemicals and
solvents were used as purchased.
4.3. Synthesis of (3-bromophenyl)(diphenyl)phosphine sulfide
To a solution of (3-bromophenyl)diphenylphosphine, prepared
according to the literature procedure [12] from 1,3-
dibromobenzene (5.4 g, 23 mmol), n-BuLi (1.6 M in hexane, 15 mL,
24 mmol), and Ph2PCl (4.8 g, 22 mmol), in 30 mL of THF at ꢀ50 ꢁC,
sulfur (0.8 g, 25 mmol) was added portionwise. Then the reaction
mixture was allowed to warm to room temperature under stirring
and left overnight. The resulting solution was quenched with water
and diluted with Et2O. The organic layer was separated, dried over
anhydrous MgSO4, and evaporated to dryness. The resulting brown
semisolid residue was crystallized from EtOH followed byadditional
purification by silica gel column chromatography (EtOAc) to give the
desired product as a white crystalline solid. Yield: 2.5 g (30%). Mp:
92e94 ꢁS (benzene-hexane (1:3)). 31P{1H} NMR (161.97 MHz,
NMR spectra were recorded on Bruker Avance-300 and Bruker
Avance-400 spectrometers, and the chemical shifts (d) were inter-
nally referenced by the residual solvent signals relative to tetra-
methylsilane (1H and 13C) or externally to H3PO4 (31P). The 13C NMR
spectra were registered using the JMODECHO mode; the signals for
the C atoms bearing odd and even numbers of protons have
opposite polarities. The numeration for carbon atoms of the central
benzene ring in the descriptions of the 1H and 13C NMR spectral
data of ligands 2aee and complexes 3bee is in agreement with
IUPAC nomenclature used for ligands 2aee. The same principle of
numbering was used to describe the solid-state molecular struc-
tures characterized by X-ray crystallography.
Column chromatography was carried out using Merck silica gel
60 (230e400 mesh ASTM). IR spectra were recorded on a Magna-
IR750 Fourier spectrometer (Nicolet) in KBr pellets, resolution
2 cmꢀ1, 128 scans. The assignment of absorption bands in the IR
spectra was made according to ref. [26]. HRMS were obtained on
a Varian MAT CH7A instrument with an ESI source at 70 eV. Melting
points were determined with a MPA 120 EZ-Melt Automated
Melting Point Apparatus and were uncorrected.
CDCl3):
d d 7.30 (dt, 1H,
42.76 ppm. 1H NMR (400.13 MHz, CDCl3):
H(C5), 3JHH ¼ 7.8 Hz, 4JHP ¼ 3.3 Hz), 7.44e7.48 (m, 4H, HAr), 7.51e7.57
(m, 2H, HAr), 7.59e7.64 (m, 2H, HAr), 7.70 (dd, 4H, o-H in P(S)Ph2,
3JHH ¼ 7.3 Hz, 3JHP ¼ 13.5 Hz), 7.86 (d,1H, H(C2), 3JHP ¼ 13.2 Hz). Anal.
Calcd. for C18H14BrPS: C, 57.92; H, 3.78; P, 8.30. Found: C, 57.74; H,
3.68; N, 8.23%.
4.4. General procedure for the synthesis of ketimine ligands 2cee
A
solution of (3-bromophenyl)(diphenyl)phosphine sulfide
(500 mg,1.34 mmol) in 5 mL of THF was added dropwise to a stirred
mixture of 2.5 M solution of n-butyllithium in hexane (0.59 mL,
1.47 mmol) and tetrahydrofuran (20 mL), keeping the temperature
in the range from ꢀ80 to ꢀ90 ꢁC. After stirring for 30 min at ꢀ80 ꢁC,
the reaction mixture was cooled to ꢀ90 ꢁC and then a solution of
the corresponding N-protected lactam (1.34 mmol: 149 mg of N-
vinyl pyrrolidone, 270 mg of N-(diethoxymethyl)valerolactam or
187 mg of N-vinyl caprolactam) in 5 mL of THF was added dropwise.
The cooling bath was removed and the temperature was allowed to
rise to r.t. The resulting reaction mixture was quenched with
saturated aqueous solution of NH4Cl (5 mL), the organic phase was
separated and slowly added (over 15 min) to aq. 3 M HCl (20 mL)
under stirring at 40 ꢁC. During the addition, the organic solvent was
collected under reduced pressure (120 torr) using a short-path
distilling head. Then the reaction mixture was stirred for 20 min
at 40 ꢁC and carefully made alkaline with aq. 10 M NaOH (15 mL)
under cooling (ice bath). The water solution obtained was extracted
with CH2Cl2 (2 ꢂ 15 mL), the combined organic layers were dried
over K2CO3 and concentrated under reduced pressure. The
4.2. General procedureforthe synthesisof benzaldimineligands 2a, b
A
mixture of 3-(diphenylthiophosphoryl)benzaldehyde
1
(0.85 g, 2.6 mmol), the corresponding amine (2.6 mmol), and
MgSO4 (3.20 g, 2.7 mmol) in 9 mL of CH2Cl2 was refluxed for 5 h and
left under ambient conditions for 3 days. The resulting mixture was
filtered and evaporated to dryness to give quantitatively the desired
ligands 2a, b as viscous yellowish oils (95% pure according to the 31P
and 1H NMR data), which were used in the synthesis of metal
complexes without further purification.
4.2.1. N-{[3-(Diphenylthiophosphoryl)phenyl]methylidene}aniline, 2a
31P{1H} NMR (161.97 MHz, CDCl3):
d
43.14 ppm. 1H NMR
(400.13 MHz, CDCl3):
d 7.70e7.80 (m, 4O, HAr), 7.47e7.62 (m, 13O,
3
HAr), 8.17 (d, 1H, H(C6), JHH ¼ 7.6 Hz), 8.20 (d, 1H, H(C2),
3JHP ¼ 14.2 Hz), 8.44 (s, 1H, CH]N). IR (KBr,
n
/cmꢀ1): 512(m),
614(w), 639(m) (P]S), 691(s), 715(s), 750(m), 1101(m), 1173(w),

D.V. Aleksanyan et al. / Journal of Organometallic Chemistry 711 (2012) 52e61
59
resulting residue was purified by column chromatography to give
2cee as viscous yellowish oils (eluent: hexaneeEtOAc (from 5/1 to
1/1)).
997(w), 1058(w), 1102(s), 1253(w), 1343(w), 1398(w), 1436(s),
1480(w), 1572(vw), 1586(vw), 1630(m) (C]N), 2850(w), 2922(m),
3052(w). HRMS (ESI): calcd. for C24H25NPS (M þ H) 390.1440, found
390.1440.
4.4.1. 5-[3-(Diphenylthiophosphoryl)phenyl]-3,4-dihydro-2H-
pyrrole, 2c
4.5. General procedure for the synthesis of pincer complexes 3aee
Yield: 300 mg (61%). 31P{1H} NMR (161.97 MHz, CDCl3):
d
43.43 ppm. 1H NMR (400.13 MHz, CDCl3):
d
1.98e2.06 (m, 2H,
A solution of (PhCN)2PdCl2 (50 mg, 0.130 mmol) and the cor-
responding ligand (0.137 mmol) in 7 mL of MeCN was refluxed for 1
(in the case of ligand 2e), 3 (2a, d), 5 (2b) or 8 h (2c). To isolate
palladacylces 3a, b, d, the reaction mixture was evaporated to
dryness, the residue was washed with Et2O and purified by column
chromatography (eluent: CH2Cl2). Note that due to the decompo-
sition on silica gel, complex 3a having a phenyl substituent at the
nitrogen atom of the imine moiety was not isolated. Analytically
pure samples of complexes 3b, d were obtained by recrystallization
from CHCl3eEtOH (1:5) mixture. In the case of complexes 3c, e,
insoluble in acetonitrile, the resulting precipitate of pincer products
were filtered off, washed with CH3CN and Et2O (3c) or DMSOeEt2O
(1:3) mixture (3e) and dried in vacuo.
NCH2CH2), 2.89 (t, 2H, N¼CeCH2, 3JHH ¼ 7.0 Hz), 4.05 (t, 2H, NCH2,
3JHH ¼ 7.3 Hz), 7.40e7.55 (m, 7H, HAr), 7.58e7.81 (m, 5H, o-H in P(S)
3
Ph2 þ HAr), 8.08 (d, 1H, H(C6), JHH ¼ 7.8 Hz), 8.19 (d, 1H, H(C2),
3JHP ¼ 13.9 Hz). 13C{1H} NMR (100.61 MHz, CDCl3):
d 22.75
(NCH2CH2), 35.09 (N]CeCH2), 61.76 (CH2N), 128.69 (d, C5þm-C in
P(S)Ph2, 3JCP ¼ 12.1 Hz), 130.61 (d, C6, JCP ¼ 2.9 Hz), 131.53 (d, C2,
4
2JCP ¼ 11.1 Hz), 131.80 (d, p-C in P(S)Ph2, 4JCP ¼ 2.9 Hz), 132.3 (d, o-C
in P(S)Ph2, 2JCP ¼ 10.5 Hz), 133.45 (d, C3, 1JCP ¼ 85.0 Hz), 132.76 (d,
1
2
ipso-C in P(S)Ph2, JCP ¼ 83.4 Hz), 133.91 (d, C4, JCP ¼ 11.1 Hz),
3
135.15 (d, C1, JCP ¼ 13.1 Hz), 172.53 (s, C]N). IR (KBr,
n
/cmꢀ1):
511(m), 519(s), 613(w), 640(s) (P]S), 691(m), 698(m), 716(vs),
746(m), 753(m), 996(w), 1088(sh,w), 1100(s), 1174(w), 1325(m),
1416(w), 1436(s), 1479(w), 1568(vw), 1618(m) (C]N), 2856(w),
2921(w), 2968(w), 3051(vw). HRMS (ESI): calcd. for C22H20NNaPS
(M þ Na) 384.0946, found 384.0953.
4.5.1. [2-(tert-Butyliminomethyl)-6-(diphenylthiophosphoryl)
phenyl]palladium chloride, 3b
Yellow crystalline solid. Yield 34%. Mp: >240 ꢁC (decomp.). 31P
4.4.2. 6-[3-(Diphenylthiophosphoryl)phenyl]-2,3,4,5-
tetrahydropyridine, 2d
{1H} NMR (161.98 MHz, CDCl3):
d
63.58 ppm. 1H NMR (400.13 MHz,
3
CDCl3):
d
1.64 (s, 9H, C(CH3)3), 6.96 (dd, 1H, H(C4), JHH ¼ 7.7 Hz,
Yield: 195 mg (39%). 31P{1H} NMR (161.97 MHz, CDCl3):
3JHP ¼ 8.9 Hz), 7.12 (dt, 1H, H(C5), 3JHH ¼ 7.7 Hz, 4JHP ¼ 4.3 Hz), 7.43
3
d
44.18 ppm. 1H NMR (400.13 MHz, CDCl3):
d
1.60e1.66 and
(d, 1H, H(C6), JHH ¼ 7.6 Hz), 7.50 (dt, 4H, m-H in P(S)Ph2,
4
1.75e1.81 (both m, 2H þ 2H, CH2CH2), 2.55 (t, 2H, N]CeCH2,
3JHH ¼ 7.9 Hz, JHP ¼ 4.6 Hz), 7.61 (t, 2H, p-H in P(S)Ph2,
3
3
3JHH ¼ 6.3 Hz), 3.79 (t, 2H, CH2N, JHH ¼ 5.7 Hz), 7.41e7.52 (m,
3JHH ¼ 7.4 Hz), 7.78 (dd, 4H, o-H in P(S)Ph2, JHH ¼ 7.2 Hz,
7H, HAr), 7.60 (dd, 1H, H(C4), 3JHH ¼ 7.9, 3JHP ¼ 12.6 Hz), 7.70 (dd, 4H,
3JHP ¼ 13.6 Hz), 8.07 (s,1H, CH]N). IR (KBr,
n
/cmꢀ1): 529(m), 598(s)
3
3
o-H in P(S)Ph2, JHH ¼ 7.8, JHP ¼ 13.1 Hz), 7.97 (d, 1H, H(C6),
(R]S), 690(s), 705(s), 721(s), 751(m), 806(w), 969(m), 997(w),
1103(s), 1107(s), 1184(m), 1197(m), 1241(w), 1368(m), 1436(s),
1481(w), 1546(m) (S]N), 1597(m), 2919(w), 2962(w), 3021(w).
Anal. Calcd. for C23H23ClNPPdS$0.1 CHCl3: C, 52.32; H, 4.39; N, 2.64;
S, 6.05. Found: C, 52.39; H, 4.51; N, 2.71; S, 5.88%.
3JHH ¼ 7.6 Hz), 8.17 (d, 1H, H(C2), JHP ¼ 14.0 Hz). 13C NMR
3
(100.61 MHz, CDCl3):
d 19.61 (s, CH2), 21.74 (s, CH2), 27.17 (s, CH2),
3
50.07 (s, CH2N), 128.44 (d, C5, JCP ¼ 12.5 Hz), 128.74 (d, m-C in
P(S)Ph2, 3JCP ¼ 12.5 Hz), 129.36 (d, C6, 4JCP ¼ 2.9 Hz), 130.08 (d, C2,
2JCP ¼ 11.5 Hz),131.72 (d, p-C in P(S)Ph2, 4JCP ¼ 2.9 Hz),132.34 (d, o-C
in P(S)Ph2, 2JCP ¼ 10.5 Hz), 132.72 (d, C3, 1JCP ¼ 83.4 Hz), 132.93 (d,
4.5.2. [2-(3,4-Dihydro-2H-pyrrol-5-yl)-6-(diphenylthiophosphoryl)
phenyl]palladium chloride, 3c
1
2
ipso-C in P(S)Ph2, JCP ¼ 85.3 Hz), 133.25 (d, C4, JCP ¼ 10.6 Hz),
3
140.66 (d, C1, JCP ¼ 12.5 Hz), 165.34 (s, C]N). IR (KBr,
n
/cmꢀ1):
Pale-yellow crystalline solid. Yield 54%. Mp: >270 ꢁC (decomp.).
507(m), 517(s), 615(w), 643(s) (P]S), 692(s), 697(s), 716(vs),
748(m), 756(w), 997(w), 1026(w), 1074(w), 1100(s), 1352(w),
1400(m), 1437(s), 1480(m), 1572(vw), 1586(vw), 1636(m) and
1646(w) (both C]N), 2857(w), 2939 (m), 3052(w). HRMS (ESI):
calcd. for C23H23NPS (M þ H) 376.1283, found 376.1270.
31P{1H} NMR (161.98 MHz, DMSO-d6): 64.39 ppm. 1H NMR
d
(400.13 MHz, DMSO-d6): d 2.13e2.20 (m, 2H, NCH2CH2), 2.97 (t, 2H,
3
N]CeCH2, JHH ¼ 8.0 Hz), 4.11 (m, 2H, NCH2), 7.04e7.25 (m, 6H,
HAr), 7.32e7.56 (m, 3H, HAr), 7.70 (dd, 4H, o-H in P(S)Ph2,
3
3JHH ¼ 8.1 Hz, JHP ¼ 13.5 Hz). IR (KBr,
n
/cmꢀ1): 516(m), 533(s),
597(s) (P]S), 620(w), 689(m), 704(m), 721(m), 751(m), 793(w),
998(w), 1103(vs), 1199(w), 1303(w), 1357(m), 1436(s), 1482(w),
1542(m) (C]N), 1568(w), 1602(w), 2864(w), 2925(w), 3052(w).
Anal. Calcd. for C22H19ClNPPdS: C, 52.61; H, 3.81; N, 2.79. Found: C,
52.63; H, 3.71; N, 2.88%.
4.4.3. 7-[3-(Diphenylthiophosphoryl)phenyl]-3,4,5,6-tertrahydro-
2H-azepine, 2e
Prior to purification by preparative chromatography (according
to the general procedure) the residue was dried with toluene using
a DeaneStark trap. Yield: 187 mg (38%). 31P{1H} NMR (161.97 MHz,
CDCl3):
d
43.54 ppm. 1H NMR (400.13 MHz, CDCl3):
d
1.47e1.53 (m,
4.5.3. [2-(Diphenylthiophosphoryl)-6-(3,4,5,6-tetrahydro-2-
pyridinyl)phenyl]palladium chloride, 3d
2H, CH2), 1.56e1.61 (m, 2H, CH2), 1.79e1.85 (m, 2H, CH2), 2.74e2.77
(m, 2H, N]CeCH2), 3.77e3.79 (m, 2H, CH2N), 7.39e7.50 (m, 7H,
HAr), 7.57 (dd, 1H, H(C4), 3JHH ¼ 7.7 Hz, 3JHP ¼ 12.5 Hz), 7.70 (dd, 4H,
Yellow crystalline solid. Yield 12%. Mp: >265 ꢁC (decomp.). 31P
{1H} NMR (161.98 MHz, CDCl3): 63.37 ppm. 1H NMR (400.13 MHz,
d
3
3
o-H in P(S)Ph2, JHH ¼ 7.3 Hz, JHP ¼ 13.3 Hz), 7.88 (d, 1H, H(C6),
CDCl3): d 1.78e1.87 (m, 4H, CH2CH2), 2.69e2.72 (m, 2H, N]CeCH2),
3JHH ¼ 7.8 Hz), 8.12 (d, 1H, H(C2), JHP ¼ 14.2 Hz). 13C NMR
4.18e4.21 (m, 2H, CH2N), 6.96e7.00 (m, 1H, H(C4)), 7.09 (dt, 1H,
H(C5), 3JHH ¼ 7.6 Hz, 4JHP ¼ 4.5 Hz), 7.26 (d, 1H, H(C6), 3JHH ¼ 7.6 Hz),
7.50 (dt, 4H, m-H in P(S)Ph2, 3JHH ¼ 7.7 Hz, 4JHP ¼ 3.2 Hz), 7.60 (d, 2H,
3
(100.61 MHz, CDCl3): d 23.55 (s, CH2), 25.99 (s, CH2), 31.22 (s, CH2),
31.49 (s, CH2), 52.71 (s, CH2N), 128.50 (d, C5, 3JCP ¼ 12.5 Hz), 128.63
3
4
3
(d, m-C in P(S)Ph2, JCP ¼ 12.1 Hz), 129.98 (d, C6, JCP ¼ 2.9 Hz),
130.47 (d, C2, 2JCP ¼ 11.5 Hz), 131.72 (d, p-C in P(S)Ph2, 4JCP ¼ 2.9 Hz),
132.32 (d, o-C in P(S)Ph2, 2JCP ¼ 11.1 Hz),132.83 (d, ipso-C in P(S)Ph2,
p-H in P(S)Ph2, JHH ¼ 7.4 Hz), 7.77 (dd, 4H, o-H in P(S)Ph2,
3
3JHH ¼ 7.8 Hz, JHP ¼ 13.5 Hz). IR (KBr,
n
/cmꢀ1): 514(m), 527(s),
554(w), 600(s) (P]S), 620(w), 691(s), 705(s), 721(m), 742(m),
753(m), 781(w), 997(w), 1111(s), 1158(w), 1185(w), 1261(w),
1329(w), 1365(m), 1399(m), 1436(s), 1480(w), 1546(m) (C]N),
1568(w), 1600(w), 2857(w), 2942 (m), 3051(w). Anal. Calcd. for
1JCP ¼ 92.5 Hz), 132.87 (d, C3, JCP ¼ 85.5 Hz), 132.91 (d, C4,
1
2JCP ¼ 10.5 Hz), 142.14 (d, C1, JCP ¼ 12.5 Hz), 173.80 (s, C]N). IR
3
(KBr, n
/cmꢀ1): 516(m), 614(w), 640(s) (P]S), 691(s), 717(s), 748(m),

60
D.V. Aleksanyan et al. / Journal of Organometallic Chemistry 711 (2012) 52e61
Table 2
corresponding H atoms are bonded. Crystal data and structure
refinement parameters for 3b, 3c, 3d_nosolv, and 3d_solv are given
in Table 2. All calculations were performed using the SHELXTL
software [27].
Crystal data and structure refinement parameters for 3b, 3c, 3d_nosolv, and
3d_solv.
3b
3c
3d_nosolv
23H21
ClNPPdS
516.29
100
3d_solv
C24H22Cl4
NPPdS
635.66
100
Empirical formula
C23H23
ClNPPdS
518.30
100
C22H19
C
ClNPPdS
502.26
100
4.7. Catalytic studies
Formula weight
T, K
In a typical experiment a solution of 0.25 mmol of aryl bromide,
0.375 mmol of PhB(OH)2, 0.5 mmol of K3PO4, and the mentioned
amount of the corresponding palladium complex (used as titrated
solutions in DMF) in 1 mL of DMF was heated at 120 ꢁC from 30 min
up to 5 h. To determine the conversion of aryl bromide, aliquots of
the reaction mixture, taken after the specified time of heating, were
treated with water (3e4 ml), extracted with benzene, and analyzed
by GC. Hg(0) poisoning experiments were performed according to
ref. [28] with 300 eq. of Hg relative to the metal complex. Hg was
added either at t ¼ 0 min or 30 min to the reaction mixture.
Crystal system
Space group
Z
a, Å
b, Å
Tetragonal
P43
4
9.0862(5)
9.0862(5)
26.6352(12) 14.0838(11) 8.944(5)
90.00
90.00
90.00
2199.0(2)
1.566
Triclinic
P-1
2
8.4842(5)
9.1409(5)
Monoclinic
P21/c
4
9.012(5)
27.864(16)
Triclinic
P-1
2
9.9915(4)
11.1299(4)
11.9419(5)
78.3312(6)
76.9215(6)
89.4350(6)
1265.91(9)
1.668
c, Å
ꢁ
ꢁ
ꢁ
a
,
96.903(1)
98.314(1)
111.679(1)
986.26(11)
1.691
90.00
110.215(9)
90.00
2108(2)
1.627
11.91
b
,
g
,
V, Å3
Dcalc (g cmꢀ1
)
Linear absorption,
(cmꢀ1
F(000)
11.42
12.7
13.15
m
)
1048
57
504
58
1040
58
636
58
Appendix A. Supplementary material
ꢁ
2
qmax
,
Reflections measured 12807
11807
5235
16628
5600
15295
6727
Complete crystallographic data for the structures in this paper
have been deposited at the Cambridge Crystallographic Data Centre
CCDC 862466 (3b), 862467 (3c), 862468 (3d_nosolv), and 862469
(3d_solv) contain the supplementary crystallographic data for this
paper. These data can be obtained free of charge via www.ccdc.cam.
ac.uk/data_request/cif.
Independent
4995
reflections
Observed reflections 4829
4908
4691
6014
[with I > 2
Parameters
R1
s(I)]
253
244
253
289
0.0291
0.0666
1.009
0.0197
0.0508
1.009
0.0364
0.0803
1.009
0.0279
0.0660
1.009
wR2
GOF
Drmax
/
Drmin (e Åꢀ3
)
0.729/ꢀ0.480 0.566/ꢀ0.506 1.167/ꢀ0.803 1.041/ꢀ0.576
Appendix B. Supplementary material
Supplementary data related to this article can be found online at
doi:10.1016/j.jorganchem.2012.03.029.
C23H21ClNPPdS$0.5 CHCl3: C, 49.00; H, 3.76; N, 2.43. Found: C, 48.81;
H, 3.67; N, 2.34%.
References
4.5.4. [2-(Diphenylthiophosphoryl)-6-(3,4,5,6-tetrahydro-2H-
azepin-7-yl)phenyl]palladium chloride, 3e
[1] D. Morales-Morales, C. Jensen (Eds.), The Chemistry of Pincer Compounds,
Elsevier, New York, 2007.
[2] (a) M. Albrecht, G. van Koten, Angew. Chem. Int. Ed. 40 (2001) 3750;
(b) J.T. Singleton, Tetrahedron 59 (2003) 1837;
Pale-yellow crystalline solid. Yield: 34%. Mp: >305 ꢁC
(decomp.). 31P{1H} NMR (121.49 MHz, CDCl3): 68.53 ppm. 1H NMR
d
(300.13 MHz, CDCl3): 1.60e1.70 (m, 4H, CH2CH2), 1.87e1.89 (m, 2H,
CH2), 2.83e2.87 (m, 2H, N]CeCH2), 4.26e4.29 (m, 2H, CH2N), 7.03
(c) M.E. van der Boom, D. Milstein, Chem. Rev. 103 (2003) 1759;
(d) D. Morales-Morales, Rev. Soc. Quim. Mex. 48 (2004) 338;
(e) J. Dupont, C.S. Consorti, J. Spencer, Chem. Rev. 105 (2005) 2527;
(f) I. Moreno, R. SanMartin, B.R. Ines, F. Churruca, E. Dominguez, Inorg. Chim.
Acta 363 (2009) 1903;
3
3
(dd, 1H, H(S4), JHH ¼ 7.5 Hz, JHP ¼ 8.5 Hz), 7.14 (dt, 1H, H(C5),
4
3
3JHH ¼ 7.4 Hz, JHP ¼ 4.8 Hz), 7.33 (d, 1H, H(C6), JHH ¼ 7.5 Hz),
7.49e7.54 (m, 4H, m-H in P(S)Ph2), 7.60e7.65 (m, 2H, p-H in
P(S)Ph2), 7.73 (dd, 4H, o-H in P(S)Ph2, 3JHH ¼ 7.2 Hz, 3JHP ¼ 13.6 Hz).
(g) D. Morales-Morales, Curr. Org. Synth. 6 (2009) 169.
[3] For the first review on unsymmetrical pincer complexes, see I. Moreno,
R. SanMartin, B. Ines, M.T. Herrero, E. Dominguez, Curr. Org. Chem. 13 (2009)
878.
IR (KBr,
n
/cmꢀ1): 527(s), 547(w), 599(s) (P]S), 619(w), 691(s),
[4] (a) Zh. Wang, M.R. Eberhard, C.M. Jensen, Sh. Matsukawa, Y. Yamamoto.,
J. Organomet. Chem. 681 (2003) 189;
704(m), 720(m), 751(m), 790(w), 998(w), 1076(w), 1108(vs),
1197(w), 1276(w), 1354(m), 1390(m), 1436(vs), 1482(w), 1543(m)
(C]N), 1588(m), 2857(w), 2931(m), 2983(w), 3048(w). Anal. Calcd.
for C24H13ClNPPdS$0.25 DMSO: C, 53.51; H, 4.49; N, 2.55. Found: C,
53.60; H, 4.28; N, 2.55%.
(b) O.A. Wallner, K.J. Szabo, Org. Lett. 6 (2004) 1829;
(c) J.-F. Gong, Y.-H. Zhang, M.-P. Song, C. Xu, Organometallics 26 (2007) 6487;
(d) M. Gagliardo, N. Selander, N.C. Mehendale, G. van Koten, R.J.M. Klein
Gebbink, K.J. Szabo, Chem. Eur. J. 14 (2008) 4800;
(e) B.-Sh. Zhang, Ch. Wang, J.-F. Gong, M.-P. Song, J. Organomet. Chem. 694
(2009) 2555;
(f) J.M. Serrano-Becerra, S. Hernandez-Ortega, D. Morales-Morales, Inorg.
Chim. Acta 363 (2010) 1306;
(g) M.A. Solano-Prado, F. Estudiante-Negrete, D. Morales-Morales, Polyhedron
29 (2010) 592.
4.6. Crystal structure determination and data collection
Crystals of complexes 3bed suitable for X-ray diffraction study
were grown by recrystallization from CH2Cl2eEt2O (1:4) (3b),
CH2Cl2eCH3CN (1:1) (3c), and CHCl3eEtOH (1:5) (3d). The X-ray
diffraction experiments were carried out with a SMART APEX2 CCD
[5] For the recent review encompassing comparative data on catalytic perfor-
mance of symmetric and unsymmetrical pincer complexes, see J.-L. Niu, X.-
Q. Hao, J.-F. Gong, M.-P. Song, Dalton Trans. 40 (2011) 5135.
[6] V.A. Kozlov, D.V. Aleksanyan, Yu.V. Nelyubina, K.A. Lyssenko, P.V. Petrovskii,
A.A. Vasil’ev, I.L. Odinets, Organometallics 29 (2010) 2054.
[7] V.A. Kozlov, D.V. Aleksanyan, Yu.V. Nelyubina, K.A. Lyssenko, P.V. Petrovskii,
A.A. Vasil’ev, I.L. Odinets, Organometallics 30 (2011) 2920.
[8] V.A. Kozlov, D.V. Aleksanyan, Yu.V. Nelyubina, K.A. Lyssenko, E.I. Gutsul,
L.N. Puntus, A.A. Vasil’ev, P.V. Petrovskii, I.L. Odinets, Organometallics 27
(2008) 4062.
diffractometer, using graphite monochromated Mo-K
a radiation
(
l
¼ 0.71073 Å, -scans) at 100 K. The structures were solved by
u
direct methods and refined by the fullmatrix least-squares against
F2 in anisotropic approximation for non-hydrogen atoms. The H(C)
atom positions were calculated. All hydrogen atoms were refined in
isotropic approximation within the riding model with the Uiso(H)
parameters equal to 1.2 Ueq(Ci), where U(Ci) are the equivalent
thermal parameters of the carbon atoms to which the
[9] (a) J.M. Vila, M. Gayoso, M.T. Pereira, M. Lopez-Torres, J.J. Fernandez,
A. Fernandez, J.M. Ortigueira, J. Organomet. Chem. 506 (1996) 165;
(b) W.J. Hoogervorst, C.J. Elsevier, M. Lutz, A.L. Spek, Organometallics 20
(2001) 4437;
(c) J.S. Fossey, Ch.J. Richards, Organometallics 21 (2002) 5259;

D.V. Aleksanyan et al. / Journal of Organometallic Chemistry 711 (2012) 52e61
61
(d) W.J. Hoogervorst, K. Goubitz, J. Fraanje, M. Lutz, A.L. Spek, J.M. Ernsting,
C.J. Elsevier, Organometallics 23 (2004) 4550;
(via Schiff-base condensation) on a preliminary palladated isophthalaldehyde
core, see ref. [9e].
(e) K. Takenaka, M. Minakawa, Y. Uozumi, J. Am. Chem. Soc. 127 (2005) 12273;
(f) S.H. Oakley, M.P. Coogan, R.J. Arthur, Organometallics 26 (2007) 2285;
(g) M. Stol, D.J.M. Snelders, M.D. Godbole, R.W.A. Havenith, D. Haddleton,
G. Clarkson, M. Lutz, A.L. Spek, G.P.M. van Klink, G. van Koten, Organometallics
26 (2007) 3985;
[17] C. Piecgaczyk, T. Cantat, N. Mézailles, P. Le Floch, J. Org. Chem. 72 (2007) 4228.
[18] (a) R. Gerber, O. Blacque, Ch.M. Frech, ChemCatChem 1 (2009) 393;
(b) D. Olsson, O.F. Wendt, J. Organomet. Chem. 694 (2009) 3112;
(c) J.L. Bolliger, O. Blacque, M. Ch. Frech, Angew. Chem. Int. Ed. 46 (2007)
6514.
(h) J.S. Fossey, M.L. Russell, K.M. Abdul Malik, Ch.J. Richards, J. Organomet.
Chem. 692 (2007) 4843;
[19] (a) K. Yu, W. Sommer, M. Weck, C.W. Jones, Adv. Synth. Catal. 347 (2004) 161;
(b) K. Yu, W. Sommer, M. Weck, C.W. Jones, J. Catal. 226 (2004) 101;
(c) D.E. Bergbreiter, P.L. Osburn, J. Frels, Adv. Synth. Catal. 347 (2004) 172.
[20] (a) J.A. Widegren, R.G. Finke, J. Mol. Catal. A: Chem. 198 (2003) 317;
(b) N.T.S. Phan, M. Van Der Sluys, C.W. Jones, Adv. Synth. Catal. 348 (2006)
609;
(i) D.H. Lee, S.J. Hong, S. Park, Bull. Korean Chem. Soc. 29 (2008) 187.
[10] G.P. Schiemenz, K. Hermann, Liebigs Ann. Chem. (1973) 1480.
[11] (a) A. Giovanni, D. Savoia, A. Umani-Ronchi, J. Org. Chem. 54 (1989) 228;
(b) M. Haslego, C. Maryanoff, L. Scott, K. Sorgi, Heterocycles 35 (1993) 643.
[12] J.-Ch. Poupon, A.A. Boezio, A.B. Charette, Angew. Chem. Int. Ed. 45 (2006) 1415.
[13] The coordination nature of these complexes answering to a set of signals at
40.8e46.7 ppm in the 31P NMR spectrum of the reaction mixture was
confirmed by the collapse of these signals in a singlet signal of the free ligand
after addition of DMSO.
[14] W.B. Cross, Ch.G. Daly, R.L. Ackerman, I.R. George, K. Singh, Dalton Trans. 40
(2011) 495.
[15] (a) S. Chakladar, P. Paul, K. Venkatsubramanian, K. Nag, J. Chem. Soc. Dalton
Trans. (1991) 2669;
(b) S. Chakladar, P. Paul, K. Nag, Polyhedron 10 (1991) 1513;
(c) A. Fernandez, E. Pereira, J.J. Fernandez, M. Lopez-Torres, A. Suarez,
R. Mosteiro, M.T. Pereira, J.M. Vila, New J. Chem. 26 (2002) 895.
[16] Several efficient synthetic approaches that have been suggested to produce
Pd(II) pincer complexes of bis(imines) require the application of C2-
functionalized starting compounds. For the synthesis of NCN bis(imine)
pincer complexes by the oxidative addition of the CeBr bond of 2-bromo-
substituted ligands across Pd(0) precursor, see ref. [9h]; for the very efficient
ligand introduction route consisting in the construction of a ligand framework
(c) D. Astruc, Inorg. Chem. 46 (2007) 1884;
(d) Y. Tsuji, T. Fujihara, Inorg. Chem. 46 (2007) 1895;
(e) K. Yu, W. Sommer, J.M. Richardson, M. Weck, C.W. Jones, Adv. Synth. Catal.
347 (2005) 161.
[21] Such a mechanism of the formation of catalytically active species in the Suzuki
cross-coupling promoted by palladium complexes was also proposed for
monopalladacycles: (a) R.B. Bedford, C.S.J. Cazin, Chem. Commun. (2001)
1540;
(b) R.B. Bedford, S.L. Welch, Chem. Commun. (2001) 129.
[22] Ch. Rohlich, K. Kohler, Adv. Synth. Catal. 352 (2010) 2263.
[23] (a) T. Jeffery, M. David, Tetr. Lett. 39 (1998) 5751;
(b) T. Jeffery, Tetrahedron 52 (1996) 10113.
[24] N. Selander, K.J. Szabó, Chem. Rev. 111 (2011) 2048.
[25] R.A.W. Johnstone, P.M. Quan, J. Chem. Soc. (1963) 5706.
[26] L.J. Bellamy, The Infrared Spectra of Complex Molecules, Wiley, New York,
1975.
[27] G.M. Sheldrick, Acta Cryst. A 64 (2008) 112.
[28] K.S. Weddle, J.D. Aiken, R.G. Finke, J. Am. Chem. Soc. 120 (1998) 5653.