Cyclopalladation of meta -(diphenylthiophosphoryloxy)benzaldimines: NCS and unexpected NCO 5,6-membered pincer palladium complexes

Article abstract of DOI:10.1021/om9009894

Unsymmetrical NCS-pincer ligands of a new type, namely, m-(diphenylthiophosphoryloxy)benzaldimines 3 (1-[Ph₂P(S)O]-3-[CH - NR]-C₆H₄, R = OMe (3a), Ph (3b), ^tBu (3c)), were obtained in two steps starting from commercially available 3-hydroxybenzaldehyde. These ligands easily underwent cyclopalladation at the C-2 position of the central benzene ring in the reaction with PdCl ₂(PhCN)₂ in benzene or benzene-methanol solutions to afford the corresponding hybrid pincer complexes 4a-c with five- and six-membered fused metallacycles in moderate to good yields. The same reaction in dichloromethane followed by treatment with alcohol resulted in unexpected formation of the related NCO-palladacycles 5a,b, along with the above NCS-complexes. Complexes 5 present the products of formal oxidation of the P - S group in the starting ligand, which apparently proceeds in the metal ion coordination sphere. Realization of κ³-NCS and κ³-NCO coordination in 4a-c and 5a,b, respectively, was unambiguously confirmed by X-ray diffraction analysis as well as multinuclear (¹H, ¹³C, ³¹P) NMR, IR, and Raman spectroscopy. The NCS-pincer complexes 4a-c demonstrated excellent catalytic activity for the Suzuki cross-coupling reactions of aryl halides with phenylboronic acid.

Full text of DOI:10.1021/om9009894

2054 Organometallics 2010, 29, 2054–2062
DOI: 10.1021/om9009894
Cyclopalladation of meta-(Diphenylthiophosphoryloxy)benzaldimines:
NCS and Unexpected NCO 5,6-Membered Pincer Palladium Complexes
V. A. Kozlov,† D. V. Aleksanyan,† Yu. V. Nelyubina,† K. A. Lyssenko,† A. A. Vasil’ev,‡
P. V. Petrovskii,† and I. L. Odinets*,†
†A.N. Nesmeyanov Institute of Organoelement Compounds, Russian Academy of Sciences,
28 Vavilova Street, Moscow 119991, Russian Federation, and ‡N.D. Zelinsky Institute of Organic Chemistry,
Russian Academy of Sciences, 47 Leninsky Prospect, 117913 Moscow, Russian Federation
Received November 13, 2009
Unsymmetrical NCS-pincer ligands of a new type, namely, m-(diphenylthiophosphoryloxy)-
benzaldimines 3 (1-[Ph2P(S)O]-3-[CHdNR]-C6H4, R = OMe (3a), Ph (3b), tBu (3c)), were obtained
in two steps starting from commercially available 3-hydroxybenzaldehyde. These ligands easily
underwent cyclopalladation at the C-2 position of the central benzene ring in the reaction with
PdCl2(PhCN)2 in benzene or benzene-methanol solutions to afford the corresponding hybrid pincer
complexes 4a-c with five- and six-membered fused metallacycles in moderate to good yields. The
same reaction in dichloromethane followed by treatment with alcohol resulted in unexpected
formation of the related NCO-palladacycles 5a,b, along with the above NCS-complexes. Complexes
5 present the products of formal oxidation of the PdS group in the starting ligand, which apparently
proceeds in the metal ion coordination sphere. Realization of κ3-NCS and κ3-NCO coordination in
4a-c and 5a,b, respectively, was unambiguously confirmed by X-ray diffraction analysis as well as
multinuclear (1H, 13C, 31P) NMR, IR, and Raman spectroscopy. The NCS-pincer complexes 4a-c
demonstrated excellent catalytic activity for the Suzuki cross-coupling reactions of aryl halides with
phenylboronic acid.
Introduction
physical properties (including catalytic activity, photolumi-
nescence, etc.) compared to their symmetric analogues.2,3
As a part of our research on unsymmetrical pincer com-
plexes with organothiophosphorus ligands, recently we re-
ported the syntheses of pincer-type palladacycles with
coordinated thiophosphoryl and thiocarbamoyl (I)4 and
thiophosphoryl and thiophosphoryloxy donor moieties
(II).5 Note that complexes II appeared to be the first example
of cyclometalation involving a thiophosphoryloxy group.
While the palladacycles of both series may be attributed to
the κ3-SCS0 unsymmetrical type, compounds II present the
rare examples of pincer complexes containing two fused
metallacycles of different size. In continuation of this investi-
gation, it was of interest to design unsymmetrical pincer
complexes containing five- and six-membered metallacycles
formed by coordination sites of completely different natures.
Azomethynes represent one of the most popular types of
ligands in organometallic and coordination chemistry. A
literature survey has revealed a few monocyclic and pincer
palladium complexes with participation of the imine func-
tionality. As for monopalladacycle species, particular atten-
tion should be drawn to Milstein’s imine palladacycle III,6
which exhibited exceptional activity in the Heck reaction of
Pincer complexes with a terdentate monoanionic back-
bone are of particular interest due to the readiness of fine-
tuning their steric and electronic properties via variation of
the metal center nature and its immediate and distant
surroundings. An extending scope of applications of pincer
complexes, particularly during the last decades, has led to
more and more investigations in this area.1 Among the
strategies used to improve the chemical and physical proper-
ties of these compounds, desymmetrization of a pincer struc-
ture holds a special position. Unsymmetrical pincer
complexes were shown to exhibit unique reactivity and
*Corresponding author. E-mail: odinets@ineos.ac.ru.
(1) (a) Albrecht, M.; van Koten, G. Angew. Chem., Int. Ed. 2001, 40,
3750. (b) Singleton, J. T. Tetrahedron 2003, 59, 1837. (c) van der Boom,
M. E.; Milstein, D. Chem. Rev. 2003, 103, 1759. (d) Peris, E.; Crabtree, R. H.
Coord. Chem. Rev. 2004, 248, 2239. (e) Morales-Morales, D. Rev. Soc.
Quım. Mex. 2004, 48, 338. (f) Dupont, J.; Consorti, C. S.; Spencer, J. Chem.
Rev. 2005, 105, 2527. (g) Pughl, D.; Danopoulos, A. A. Coord. Chem. Rev.
2007, 251, 610. (h) Morales-Morales, D.; Jensen, C. M. Eds. The Chemistry
of Pincer Compounds; Elsevier: New York, 2007.
(2) (a) Wang, Z.; Eberhard, M. R.; Jensen, C. M.; Matsukawa, S.;
Yamamoto, Y. J. Organomet. Chem. 2003, 681, 189. (b) Poverenov, E.;
Gandelman, M.; Schimon, L. J. W.; Rozenberg, H.; Ben-David, Y.; Milstein,
D. Organometallics 2005, 24, 1082. (c) Poverenov, E.; Leitus, G.; Schimon,
L. J. W.; Milstein, D. Organometallics. 2005, 24, 5937. (d) Gagliardo, M.;
Selander, N.; Mehendale, N. C.; van Koten, G.; Klein Gebbink, R. J. M.;
Szabo, K. J. Chem.;Eur. J. 2008, 14, 4800.
´
ꢀ
(4) Kozlov, V. A.; Aleksanyan, D. V.; Nelyubina, Yu. V.; Lyssenko,
K. A.; Gutsul, E. I.; Puntus, L. N.; Vasil’ev, A. A.; Petrovskii, P. V.;
Odinets, I. L. Organometallics 2008, 27, 4062.
(3) For the first review on unsymmetrical palladium pincer complexes
(5) Kozlov, V. A.; Aleksanyan, D. V.; Nelyubina, Yu. V.; Lyssenko,
K. A.; Gutsul, E. I.; Vasil’ev, A. A.; Petrovskii, P. V.; Odinets, I. L.
Dalton Trans. 2009, 8657.
see: Moreno, I.; SanMartin, R.; Ines, B.; Herrero, M. T.; Domı
´
Curr. Org. Chem. 2009, 13, 878.
nguez, E.
r
pubs.acs.org/Organometallics
Published on Web 04/12/2010
2010 American Chemical Society

Article
Organometallics, Vol. 29, No. 9, 2010 2055
Scheme 1. Unsymmetrical Pincer Complexes with Organothio-
phosphorus Ligands
Scheme 3. Synthesis of meta-Thiophosphoryloxybenzaldimines
and Their NCS-Palladacycles
Scheme 2. Palladacycles with an Imine Functionality
obtained by two reciprocal routes differing by the sequence
of thiophosphorylation and Schiff base formation depend-
ing on the nature of the amino component. Thus, in
the case of O-methylhydroxylamine the synthesis of the
oxime 1 was accomplished first followed by thiophosphor-
ylation to afford the corresponding product 3a. In contrast,
N-phenyl- and tert-butyl-substituted m-(thiophosphoryloxy)-
benzaldimines 3b,c were prepared from thiophosphorylated
benzaldehyde 2. The synthesis of 3c was readily carried out in
dichloromethane solution at room temperature, while 3b was
obtained only upon prolonged heating. In all cases thiophos-
phorylation with Ph2P(S)Cl was carried out under phase
transfer catalysis conditions using triethylbenzylammonium
chloride (TEBA) as a catalyst.
bromobenzene and methyl acrylate, providing turnover
numbers up to 106. This pioneering work promoted numer-
ous investigations in this area, and a series of imine- and
oxime-based palladacycles were further developed and tested
as catalytic precursors for various chemical processes.7,1f
Among symmetric pincer complexes, only a few pallada-
cycles of general formula IV based on bis(imino)benzenes
have been described.8,9 These complexes were obtained
through direct cyclometalation,8 a “ligand-introduction”
route,9 or oxidative addition.10 It should be noted that
unsymmetrical 5,5-membered pincer complexes V contain-
ing imine and phosphinite moieties were recently developed
by two independent research groups11 in parallel with our
investigations.
Structures of ligands 3a-c were confirmed by IR and 31P,
1H, and 13C NMR spectra. Thus, the IR spectra of 3a-c
show the intensive absorption bands at 1605-1638 cm-1
relating to the valent vibrations of the CdN groups. The
singlet signals at 82-83 ppm in the 31P NMR spectra for
3a-c are typical for the phosphorus atom of the thiophos-
phinic acid aryl esters. Upon transformation of 2 into
corresponding aldimines 3b,c, the aldehyde proton signal
observed at 9.87 ppm in the 1H NMR spectrum of 2
disappeared and the signals assigned to the protons of the
CHdN moieties were observed at ca. 8 ppm. Moreover, the
structures of 3b and 3c were confirmed by a single-crystal
X-ray diffraction analysis. In line with the molecular geometry
examination, the bond lengths and angles of 3b and 3c
(Figure 1) fall into the range typical for compounds of this type.
The ligands 3a-c underwent direct cyclopalladation in the
reaction with (PhCN)2PdCl2 in benzene solution to give the
corresponding NCS-pincer complexes in moderate yields
(reflux, 2 h, yield ca. 30% for 4a and ca. 50% for 4b,c).
More prolonged heating of the reaction mixture did not
provide significant increase of the palladacycle yield, as
shown using ligand 3c as a representative example. The yield
of 4a was increased to 66% by performing the reaction in the
presence of methanol as a cosolvent. Such an effect can be
explained by homogenization of the reaction mixture, which
is not achieved in neat benzene even under heating. Despite
the close nature of the complexes 4a-c, their isolation
Therefore, the elaboration of a new hybrid pincer system
containing a thiophosphoryloxy group along with an imine
moiety anchored on the same benzene backbone and inves-
tigation of its ability to undergo cyclopalladation seems
reasonable. Herein, we report the convenient synthesis of
m-(thiophosphoryloxy)benzaldimines bearing thiophosphi-
nite and imine coordination sited and their direct cyclo-
palladation to form unsymmetrical 5,6-membered pallada-
cycles.
Results and Discussion
A convenient synthetic approach to m-(thiophosphoryloxy)-
benzaldimines was devised starting from commercially avail-
able 3-hydroxybenzaldehyde (Scheme 3). The ligands were
(6) Ohff, M.; Ohff, A.; Milstein, D. Chem. Commun. 1999, 357.
(7) Beletskaya, I. P.; Cheprakov, A. V. J. Organomet. Chem. 2004,
689, 4055.
(8) Vila, J. M.; Gayoso, M.; Pereira, M. T.; Torres, M. L.; Fernandez,
J. J.; Fernandez, A.; Ortigueira, J. M. J. Organomet. Chem. 1996, 506,
165.
(9) Takenaka, K.; Minekawa, M.; Uozumi, Y. J. Am. Chem. Soc.
2005, 127, 12273.
(10) Fossey, J. S.; Russell, M. L.; Malik, K. M. A.; Richards, C. J.
J. Organomet. Chem. 2007, 692, 4843.
(11) (a) Zhang, B.-C.; Wang, C.; Gong, J.-F.; Song, M.-P. J. Organo-
met. Chem. 2009, 694, 2555. (b) Yorke, J.; Sanford, J.; Decken, A.; Xia, A.
Inorg. Chim. Acta 2010, ASAP. doi: 10.1016/j.ica.2009.12.031

2056 Organometallics, Vol. 29, No. 9, 2010
Kozlov et al.
Figure 1. General view of 3b and one of the independent molecules of 3c in a representation of atoms by thermal ellipsoids (p = 50%).
The disorder in 3b is omitted for clarity.
direction of the shift for 4a is inverse to that for 4b,c. Finally,
the absence of the C2-H proton resonance gives unam-
biguous evidence for C2-metalation. In the 13C NMR spec-
trum of 4b, the most significant changes compared to that
of the corresponding ligand 3b comprise the downfield shifts
of the resonances corresponding to C2 (Δδ = ca. 14 ppm) and
C3 (Δδ = 10 ppm) carbon atoms. The almost 2-fold increase
of the 3JPC(2) coupling constant was observed. Nevertheless,
the chemical shifts of the C1 carbon atoms are practically the
same for both 3b and NCS-pincer complex 4b obtained from
this ligand. The CdN absorption bands in the IR spectra of
4b,c were found to be shifted to the low-frequency region by
∼30 cm-1 relative to the band in the corresponding free
ligands. Comparing the Raman and IR spectra of 3 and 4,
the lines and bands at ca. 640cm-1 in the spectra of the ligands
Figure 2. General view of 4a with representation of atoms by
thermal ellipsoids (p = 50%).
and ca. 620 cm-1 in the spectra of the corresponding com-
plexes were assigned to free and coordinated PdS groups,
respectively. Furthermore, X-ray diffraction analysis per-
differed depending on the substituent at the nitrogen atom of
the imine functionality (see Experimental Section). The rate
formed for 4a-c (Figure 2-4) unambiguously confirmed
of cyclopalladation of 3a-c was found to be significantly
formation of the palladium pincer complexes in all cases.
slower in dichloromethane solution (room temperature), and
The 31P NMR monitoring of the cyclopalladation in di-
chloromethane solution demonstrated a set of signals at 85-86
ppm of different intensities downfield shifted relative to the
signals of free ligands along with the upfield shifted signals of
it took 5 days to achieve about 30-40% yield of 4b and 3
weeks to produce 4a in a yield of only ca. 20%.
The complexes 4a-c are yellow crystalline solids that are
air and moisture stable both in the solid state and in solu-
the final NCS-pincer complexes. These downfield shifted
tions. Complexes 4a,c are characterized with rather high
signals may be assigned to the nonmetalated palladium(II)
thermal stability and do not undergo decomposition up to
complexes with thiophosphoryloxybenzaldimines 3a-c and
230 and 265 °C, respectively, while the decomposition point
that with thiophosphoryloxybenzaldehyde 2. The latter is
for 4b is substantially lower (135 °C).
apparently formed via the partial hydrolysis of an imine
moiety occurring under the action of HCl generated in the
course of cyclopalladation, in the presence of the solvent
residual moisture. Indeed, when 3b and (PhCN)2PdCl2 were
reacted in wet dichloromethane, bis(aniline)palladium dichloride,
The 31P NMR spectra of the complexes 4a-c displayed
singlets at 75-76 ppm, upfield shifted (Δδ = 6.2-7.7 ppm)
relative to the signals of the corresponding free ligand. The
shifting of phosphorus resonances indicates the coordination
by the metal center at the sulfur atom of the PdS group in the
(PhNH2)2PdCl2, was isolated and characterized, in particular, by
ligand.4,5,13 In turn, the shift of proton signals of the imine
1
means of X-ray diffraction analysis (see Figure S1).
group in the H NMR spectra of 4a-c by 0.2-0.6 ppm
confirms the coordination of the aldimine moiety. The
Note that application of nonpolar solvents such as diethyl
ether or hexane, commonly used for isolation of pincer
complexes via precipitation, did not result in pure samples
of 4a,b formed in the corresponding CH2Cl2 reaction mix-
tures. At the same time, surprising results were obtained when
reaction mixtures were treated with alcohol (either methanol
or ethanol). Thus, addition of methanol led to the decom-
position of the above-mentioned intermediate nonmetalated
palladium complexes with signals in the 85-86 ppm region
followed by precipitation of PdCl2. Correspondingly, the
(12) Poverenov, E.; Efremenko, I.; Frenkel, A. I.; Ben-David, Y.; W.
Shimon, L. J.; Leitus, G.; Konstantinovski, L.; Martin, J. M. L.;
Milstein, D. Nature 2008, 455, 1093.
(13) (a) Kanbara, T.; Yamamoto, T. J. Organomet. Chem. 2003, 688,
15. (b) Meguro, H.; Koizumi, T.; Yamamoto, T.; Kanbara, T. J. Organomet.
Chem. 2008, 693, 1109. (c) Doux, M.; Bouet, C.; Mezailles, N.; Ricard, L.;
Le Floch, P. Organometallics 2002, 21, 2785. (d) Doux, M.; Le Floch, P.; Jan,
Y. J. Mol. Struct. (THEOCHEM) 2005, 724, 73. (e) Doux, M.; Piechaczyk,
O.; Cantat, T.; Mezailles, N.; Le Floch, P. C. R. Chim. 2007, 10, 1.

Article
Organometallics, Vol. 29, No. 9, 2010 2057
Figure 3. General view of 4b with representation of atoms by thermal ellipsoids (p = 50%). The disorder of the phenyl ring is omitted
for clarity.
Figure 5. General view of 5b with representation of atoms by
thermal ellipsoids (p = 50%).
Figure 4. General view of 4c with representation of atoms by
thermal ellipsoids (p = 50%).
Scheme 4. Synthesis of NCS- and NCO-Pincer Complexes upon
MeOH Treatment of 3a,b and (PhCN)2PdCl2 Dichloromethane
Reaction Mixtures
disappearanceof thesesignals and concomitantappearance of
the resonances of the NCS-complexes 4a,b along with the
signal of thiophosphorylated aldehyde 2 were observed in the
31P NMR spectra. Besides the above signals, the 31P NMR
spectra of dichloromethane reaction mixtures treated with
alcohol showed new resonances at ca. 43.5 ppm, the region
typical for compounds bearing a coordinated PdO moiety.
Moreover, after evaporation of the reaction mixtures and
recrystallization of the resulting residues from CH2Cl2-Et2O
(1:2), the 31P NMR spectra of the crystalline samples obtained
also revealed signals at ∼43.5 ppm along with the signals of
the NCS-pincer complexes 4a,b. Fortunately, the crystalline
sample isolated via workup of the reaction mixture in the case
of the N-phenyl-substituted ligand 3b was found to be suitable
for X-ray analysis, which revealed a superposition of two
pincer complexes with NCS (complex 4b) and NCO (complex
5b) coordination in a 7:3 ratio (see Figure 2S in the Supporting
Information). In other words, the alcohol treatment of the
dichloromethane reaction mixtures resulted in unexpected
side formation of a palladacycle bearing a phosphoryloxy
moiety coordinated to the Pd(II) ion instead of the thiophos-
phoryloxy one (Scheme 4). It should be noted that according
to the 31P NMR data the ratio of NCS- and NCO-complexes
in the crude reaction mixtures (approximately 7:3) was the
same as was observed in the crystalline sample isolated.
The pairs of complexes bearing coordinated thiophos-
phoryl- and phosphoryloxy moieties 4a,band 5a,b, respectively,
were further successfully separated using thin-layer chromato-
graphy; however the yield of the pure NCO-complexes 5a,b
isolated in such a manner did not exceed 5%. Such low yields
may be explained by partial decomposition of NCO-complexes
on a solid support. The structure of complex 5b after purifica-
tion was elucidated by X-ray diffraction technique (Figure 5).
Concerning the spectral data for NCO-complexes 5a,b,
the resonances of coordinated OP(O) groups in the 31P
NMR spectra are observed at ca. 44 ppm. Their 1H NMR
spectra are similar to those for the complexes 4a,b bearing a
thiophosphoryloxy moiety and unambiguously confirm
the occurrence of C-metalation and coordination of an
aldimine group by the palladium ion. The IR spectra of
the NCO-complex 5b demonstrates the intensive absorption
bands of PdO and CdN groups at 1210 and 1587 cm-1
,
respectively. Although the NCO-complex 5a, with an oxime
moiety, was characterized only by means of 31P and 1H NMR
spectroscopy, the similarity of spectral pattern for 5a and 5b
allows suggesting for 5a the same structure with κ3-NCO
coordination.

2058 Organometallics, Vol. 29, No. 9, 2010
Kozlov et al.
˚
Table 1. Selected Bond Lengths(A) and Angles (deg) in 4a-c and 5b
Scheme 5. First Example of a PdO-Containing Platinum Pincer
Complex
4a
4b
4c
5b
Pd(1)-Cl(1)
Pd(1)-C(2)
Pd(1)-N(1)
Pd(1)-X(1)
(X = S(1), O(1))
P(1)-S(1)a
2.39289(17) 2.3585(5) 2.3974(6) 2.3647(10)
2.0160(6)
2.0432(6)
2.29535(18) 2.3013(6) 2.3031(6) 2.059(3)
2.006(2)
1.9988(16) 1.984(4)
2.0560(18) 2.0949(15) 2.018(3)
1.9786(2)
1.2902(9)
1.9746(7) 1.9547(7) 1.501(3)
1.291(3)
N(1)-C(19)b
1.279(2)
1.298(5)
C(2)Pd(1)N(1)
C(2)Pd(1)X(1)
(X = S(1), O(1))
N(1)Pd(1)Cl(1)
X(1)Pd(1)Cl(1)
(X = S(1), O(1))
80.09(2) 81.10(7)
98.767(17) 99.56(6)
81.11(6)
96.76(5)
80.87(15)
93.57(14)
also did not afford the formation of the NCO-complexes 5. The
reaction of 3a with (PhCN)2PdCl2 in methanol as a sole solvent
resulted in NCS-complex 4a along with the PdO analogue of
the starting ligand, namely, N-methoxy-(m-diphenylphosphor-
yloxy)benzaldimine (6) [see Supporting Information] in a 9:1
ratio rather than NCO-complex 5a. Furthermore, unlike the
ligand 3a, the above-mentioned PdO ligand 6 did not form the
corresponding NCO-pincer complex 5b upon interaction with
(PhCN)2PdCl2 either in dichloromethane solution or under
heating in benzene. It should be emphasized that such an
oxidation of the thiophosphoryloxy group in the presence of
alcohol was not observed in the cyclopalladation of 1-thiophos-
phoryloxy-3-thiophosphorylbenzenes.5 On the basis of these
results, we may suggest that the unexpected NCO-complexes 5
are formed in the oxidation of the PdS group in the metal ion
coordination sphere in the intermediate noncyclopalladated
species formed due to the starting PdS-containing pincer
ligand. However, the mechanism of such oxidation requires
additional detailed investigation.
An important point to emphasize is that the complexes 5a,
b present only the second example of the noble metal pincer
complexes bearing a coordinated phosphoryl group. The
first one, recently reported by Milstein’s group,12 represents
the platinum unsymmetrical NCO-pincer complex resulting
from insertion of the terminal oxo-ligand into the Pt-P bond
of the Pt(IV) PCN-pincer complex containing a phosphine
moiety (Scheme 5). Such an intramolecular oxygen transfer
occurred in the absence of an external oxygen acceptor due to
the instability of the initial oxo-complex at room tempera-
ture. However, no close analogy can be drawn with the
formation of the complexes 5a,b, as the starting platinum
oxo-complex was preformed under action of a strong oxi-
dant, namely, dioxirane.
Catalytic Studies. The application of symmetrical pincer
complexes as catalysts is well studied and presents the most
important sphere of their usage. At the same time, far less
information about the catalytic activity of their unsymme-
trical analogues is available. These data mostly concern the
palladium-catalyzed cross-coupling reactions, in particular
the Suzuki-Miyaura cross-coupling of aryl halides with
arylboronic acids,1h,4,5,11,14 being one of the most powerful
methods for CAr-CAr bond formation. Indeed, due to
commercial availability of the starting materials, the rela-
tively mild reaction conditions, tolerance of a broad range of
functionalities, easy handling, and removal of the nontoxic
boron-containing byproduct, as well as the possibility of
using water as a solvent (or cosolvent) and solid-support
materials, this cross-coupling reaction has gained promi-
nence in recent years at an industrial level, mainly in the
synthesis of pharmaceuticals and fine chemicals. The recent
advances involving also nonconventional methodologies
97.046(17) 94.56(5) 101.76(4) 97.72(9)
84.125(6) 84.732(18) 80.777(18) 87.91(8)
a PdS bonds in 3b and 3c are 1.9289(8) and 1.9351(7) A, respectively.
b NdC bonds in 3b and 3c are 1.253(5) and 1.269(2) A, respectively.
Table 1 summarizes the main geometrical parameters for
NCS- and NCO-pincer complexes 4a-c and 5b, respectively.
As one can see, the principal bond lengths in these complexes
are almost the same independent from the coordination
pattern. Furthermore, the conformation of five- and six-
membered palladacycles along with distortions of the co-
ordinated palladium polyhedron (with the only exception of
the most sterically hindered 4c) are also rather close to each
other. Thus, according to the X-ray data, complex 5b is
almost isostructural to the related 4b, which in turn can
explain the formation of the above-mentioned solid solution
(see Figure S2). In 4a, 4b, and 5b the five-membered rings are
flat, while the six-membered ones are characterized by a sofa
conformation with the deviation of the P(1) atom by
0.63-0.78 A. In 4c the introduction of the bulky tBu
substituent leads to the increase of cycle puckering for both
five-membered (envelope with the deviation of the C(3) atom
by 0.16 A) and six-memebred (distorted sofa with the devia-
tion of P(1) and O(1) atoms by 1.06 and 0.32 A, respectively)
rings. The steric effects are also reflected in the distortion
of the square-planar configuration of the Pd(1) atom. The
latter is characterized by a tetrahedral configuration with
the folding along the S(1) N(1) line in 4a-4c and the
3 3 3
O(2) N(1) line in 5b. The maximum folding that is reflected
3 3 3
in the dihedral angle between the Pd(1)C(2)S(1)N(1) and
Pd(1)Cl(1)S(1)N(1) planes was also observed for sterically
hindered 4c (10.2°), while in the other structures the corres-
ponding angle was almost the same, at ca. 4°. Furthermore,
the significant puckering of the above palladacycles in 4c leads
to the rather high value (17.7°) of the dihedral angle between
the PdONClC square and the central aromatic cycle.
It should be noted that Pd-C and Pd-Cl bond lengths in
4a-c and 5b are almost equal to the corresponding ones
(2.008(2) and 2.4013(6) A) in the relative 5,6-membered
SCS0-pincer complexes II.5 Thus, we may conclude that
Pd-C and Pd-Cl bond lengths and consequently the energy
of these bonds are almost independent of the nature of the
coordinated heteroatoms in the ligand as well as the puck-
ering of the palladacycles and dihedral angles between the
PdX2ClC square and the central aromatic cycle.
The mechanism of formation of these PdO palladacycles
is still obscure since the preformed NCS-complexes 4a,b are
stable in CH2Cl2-MeOH solution and do not transform into
the corresponding phosphoryl analogues 5a,b in time. More-
over, forced blowing of air and addition of water to the
dichloromethane reaction mixtures at different time points
(14) Serrano-Becerra J. M., Hernandez-Ortega S., Morales-Morales D.
Inorg. Chim. Acta, 2010, ASAP. doi: 10.1016/j.ica.2010.02.003.

Article
Table 2. Palladacycles 4-Catalyzed Suzuki Cross-Coupling
Organometallics, Vol. 29, No. 9, 2010 2059
Table 3. Influence of Different Solvents on the Conversion in the
Suzuki Cross-Coupling Catalyzed by 4b
reaction
temperature yield, %a
entry
X
cat. mol %
solvent
toluene
dioxane
EtOHb
toluene
dioxane
dioxanec
90% EtOH
1
2
3
4
5
6
7
Ac
0.1
0.1
0.1
1
1
1
105
105
80
91
86
82
79
67
55
79
X/yield, %a
OMe
catalyst
mol %
R
C(O)Me
NMe2
OMe
105
105
105
80
4a
4a
4a
4b
4b
4b
4c
4c
4c
1
OMe
OMe
OMe
Ph
100
100
93
100
100
100
100
99
86
99
0
79
95
3
91
84
3
35
23
0.1
0.01
1
0.1
0.01
1
1
25
30
a Reaction conditions: K3PO4, Bu4NBr, 5 h. b Without Bu4NBr.
c Cs2CO3 as a base.
Ph
Ph
tBu
tBu
tBu
40
40
nitrogen atom of the thiocarbamide group provided 80-84%
yield when used in the amount of 3 mol % even at more
prolonged reaction times.4 The best catalyst of type II having
5,6-membered palladacycles similar to compounds 4a-c gave
93% and 68% of 4-methoxybiphenyl when used in 1 and
0.1 mol % amounts, respectively. Therefore, the higher asym-
metry for a pincer palladium complex with at least one PdS
coordinating group may serve as a factor of its higher catalytic
activity in the Suzuki-Miyaura cross-coupling.
Furthermore, using N-phenyl-substituted complex 4b as a
representative example, we investigated the influence of a
few common solvents for the above cross-coupling reaction
on the conversion. Independently from the solvent in use, the
yields varied in a relatively narrow range for the same
starting substrate and were slightly lower compared with
those obtained in DMF. Such a decrease of the yields can be
explained by the lower reaction temperature. Nevertheless,
promoting the coupling of 4-chloroacetophenone, the cata-
lyst 4c demonstrates yields of the product equal to 22% and
8% at 120 °C in DMF and xylene, respectively. In contrast
to iminophosphite PCN palladium pincer complexes V,11
replacement of K3PO4 by Cs2CO3 resulted in a decrease of
the product yield. However, comparing the catalytic activity
of PCN-complexes V with those for NCS-complex 4b,
we may mention that the latter provided similar yields of
4-methoxybiphenyl already after 5 h (vs 18 h11) with other
conditions being equal.
0.1
0.01
91
a Reaction conditions: K3PO4, Bu4NBr, DMF, 120 °C, 5 h.
that have been applied to the Suzuki-Miyaura reaction are
well documented in the literature.15 The main challenges in
this cross-coupling pertain to the use of less reactive chloro-
arenes and bromoarenes bearing electron-donating groups,
and in these cases only a few pincer complexes displayed
catalytic activity. Noteworthy, the best results for coupling of
chloroarenes were achieved just with phosphorus-containing
pincer complexes or with palladacycles modified by car-
benes. That was related to the extra stability provided by
these ligands toward the low-ligated catalytically active
Pd(0) species involved in the catalytic cycle.1h
Hence, the catalytic activity of complexes 4a-c was
examined in a standard Suzuki-Miyaura reaction as a
benchmark evaluation to compare with previously reported
data for the other above-mentioned pincer complexes having
at least one thiophosphoryl coordinating group.
First, we performed the experiments with aryl bromides
differing in the electronic properties of substituents in the
para-position in DMF solution at 120 °C for 5 h similar to the
procedure used for testing of pincer palladium complexes
formed by 3-thiophosphorylbenzoic acid thioamides (I)4 and
1-thiophosphoryloxy-3-thiophosphorylbenzenes(II).5Table2
summarizes the results obtained.
In all experiments the normal dependence on the electronic
properties of the substituent X at the aryl bromide was ob-
served. For 4-bromoacetophenone activated by an electron-
withdrawing group, complete conversions were reached even
using 0.01 mol % of 4a-c. For the least active 4-bromo-N,N-
dimethylaniline the yields decreased to ca. 25-40% using both
1 mol % and 0.1 mol % of catalyst. In the case of intermediate
4-bromoanisole, the yields of the corresponding biphenyl pro-
duct were in the range 79-91% and 84-99% with 1 mol % and
0.1 mol % of a catalyst, respectively. However, none of the
catalysts 4a-c were active in the reaction with this substrate
when used at 0.01 mol %.
In general, all the catalysts 4a-c demonstrated similar
activity under these conditions; that is, no pronounced effect
was observed depending on the substituent at the nitrogen
atom of the CdN group. Nevertheless, they were more effec-
tive in this reaction compared with reference catalysts I and II.
Thus, the reaction with bromoanisole complexes I having 5,5-
membered palladacycles and differing in substituents at the
The higher catalytic efficiency of the complexes described
may be attributed to their hemilabile properties, as they
contain rather weakly bonded thiophosphoryloxy groups that
may facilitate substrate interaction with the metal center.
According to Hermann’s hypothesis,16 palladacycles may
serve as a reservoir providing the catalytic cycle with highly
catalytically active Pd0 particles. Furthermore, in the absence
of tetrabutylammonium bromide;a salt known to stabilize
palladium nanoparticles17;only approximately 10% decrease
of the yield of the final biaryl product was observed under
catalysis by 4b. Therefore, we cannot exclude the catalytic
mechanism comprising the cleavage of the palladium-carbon
bond and/or formation of palladium nanoparticles as the
active form of the catalysts.
Conclusions
To conclude the results presented, we developed a convenient
approach to meta-diphenylthiophosphoryloxybenzaldimines
€
(16) Hermann, W. A.; Bohm, V. P. W.; Reisinger, C.-P. J. Organo-
met. Chem. 1999, 576, 23.
(17) Bolliger, J. L.; Blacque, O.; Frech, C. M. Angew. Chem. 2007, 46,
6514.
(15) (a) Alonso, F.; Beletskaya, I. P.; Yus, M. Tetrahedron 2008, 64,
3047 (and referencies therein). (b) Schneider, F.; Stolle, A.; Ondruschka, B.;
Hopf, H. Org. Process Res. Dev. 2009, 13, 44. (c) Barder, T. E.;Walker, S. D.;
Martinelli, J. R.; Buchwald, S. L. J. Am. Chem. Soc. 2005, 127, 4685.

2060 Organometallics, Vol. 29, No. 9, 2010
Kozlov et al.
as NCS-pincer ligands bearing a thiophosphoryloxy group and
imine moieties as coordination sites. These ligands undergo
direct cyclopalladation, affording unsymmetrical NCS-pincer
complexes with fused five- and six-membered metallacycles
under reaction with (PhCN)2PdCl2 in benzene at elevated
temperature. When the same reaction was performed in di-
chloromethane, addition of an alcohol surprisingly resulted in
oxidation of a PdS group in the metal ion coordination sphere
to yield NCO-pincer complexes along with the expected NCS
ones. These 5,6-membered palladium NCS-pincer complexes
demonstrated the highest catalytic activity in the Suzuki cross-
coupling reaction among known ECS-analogues.
1223, 1239, 1437, 1700 (CdO), 2744, 2847, 3066. Anal. Calcd
for C19H15O2PS: C, 67.44; H, 4.47. Found: C, 67.58; H, 4.55.
O-{3-[(Methoxyimino)methyl]phenyl} Diphenylthiophosphinate,
3a. A solution of Na2CO3 (0.77 g, 7.3 mmol) and TEBA (4 mol %)
in 8 mL of H2O was slowly dropwise added to a stirred solution of
1 (1.10 g, 7.3 mmol) and Ph2P(S)Cl (1.84 g, 7.3 mmol) in 14 mL of
benzene at 0-5 °C. The resulting mixture was stirred for 2.5 h at
50-55 °C and, after cooling to room temperature, diluted with
20 mL of water and benzene. After separation of the organic layer,
the water phase was additionally washed with benzene (25 mL).
The combined benzene solution was dried over anhydrous
Na2SO4 and evaporated to dryness. The resulting residue was
crystallized from hexane-Et2O (2:1, 15 mL) and recrystallized
from hexane (15 mL) to give 1.70 g of 3a as a white solid. Yield:
64%. Mp: 79-81 °C (hexane). 31P{1H} NMR (161.98 MHz,
CDCl3): δ 82.49 ppm. 1H NMR (400.13 MHz, CDCl3): δ 3.93 (s,
3H, OCH3), 7.04 (d, 1H, H-C4, 3JHH = 8.3 Hz), 7.17-7.25 (m,
2H, HAr), 7.33 (d, 1H, H-C6, 3JHH = 7.7 Hz), 7.45-7.55 (m, 6H,
Experimental Section
If not noted otherwise, all manipulations were carried out
without taking precautions to exclude air and moisture. Benzene
was distilled over sodium benzophenone ketyl. Dichloromethane
was distilled from P2O5. The starting 3-hydroxybenzaldehyde
was recrystallized from water prior to use. 3-Hydroxy-N-meth-
oxybenzaldimine 118 and Ph2P(S)Cl19 were obtained according to
the literature procedures. All other chemicals and solvents were
used as purchased without further purification.
HAr), 7.92 (s, 1H, CHdN), 7.97 (dd, 4H, o-H in P(S)Ph2, 3JPH
=
13.8 Hz, 3JHH = 7.1 Hz). IR (KBr, ν/cm-1): 643 (PdS), 691, 721,
745, 792, 845, 978, 1054, 1117, 1142, 1233, 1437, 1603 (CdN),
2816, 2933. Raman (solid, Δν/ cm-1): 239, 644 (PdS), 998, 1232,
1597, 1605 (CdN). Anal. Calcd for C20H18NO2PS: C, 65.38; H,
4.94, N 3.81. Found: C, 65.60; H, 4.84, N, 3.67.
O-[3-[(Phenylimino)methyl]phenyl] Diphenylthiophosphinate,
3b. A mixture of 2 (1.90 g, 5.6 mmol), aniline (0.52 g, 5.6 mmol),
and MgSO4 (0.34 g, 2.8 mmol) in 15 mL of dichloromethane was
refluxed for 6 h under stirring and left for 2 days. After
evaporation to dryness, the resulting residue was crystallized
from hexane and recrystallized from EtOH to yield 1.5 g (65%)
of 3b as a white solid. Mp: 87-88 °C (EtOH). 31P{1H} NMR
NMR spectra were recorded on Bruker Avance-300 and
Bruker Avance-400 spectrometers, and the chemical shifts (δ)
were internally referenced by the residual solvent signals relative
to tetramethylsilane (1H and 13C) or externally to H3PO4 (31P).
The 13C NMR spectra were registered using the JMODECHO
mode; the signals for the C atom 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 spectral data is in agreement with IUPAC nomenclature
used for the ligands. The same principle of numbering was used
for the description of solid-state molecular structures character-
ized by X-ray crystallography.
Column chromatography was carried out using Merck silica
gel 60 (230-400 mesh ASTM). Analytical TLCs were performed
with Merck silica gel 60 F254 plates. IR spectra were recorded
on a “Magna-IR750” Fourier spectrometer (Nicolet), resolu-
tion 2 cm-1, 128 scans. The assignment of the absorption bands
in the IR spectra was made according to ref 20. Raman spectra
were recorded with a LabRAM Jobin-Yvon Raman spectro-
meter with an exciting He-Ne laser line of 632.8 nm. Melting
points were determined with an Electrothermal IA9100 digital
melting point apparatus and are uncorrected.
1
(121.49 MHz, CDCl3): δ 82.80 ppm. H NMR (300.13 MHz,
CDCl3): δ 7.15-7.25 (m, 4H, HAr), 7.30-7.41 (m, 3H, HAr),
7.46-7.68 (m, 8H, HAr), 7.96-8.04 (m, 4H, HAr), 8.33 (s, 1H,
CHdN). 13C{1H} NMR (100. 61 MHz, CDCl3): δ, 120.68 (s,
o-C in NPh), 121.69 (d, C6, 3JCP = 5.1 Hz), 124.30 (d, C2, 3JCP
=
4.8 Hz), 125.05 (d, C4, 4JCP = 1.3 Hz), 125.95 (s, p-C in NPh),
128.42 (d, m-C in P(S)Ph2, JCP = 13.5 Hz), 128.96 (s, m-C in
3
NPh), 129.45 (d, C5, 4JCP = 1.0 Hz), 131.23 (d, o-C in P(S)Ph2,
4
2JCP = 11.5 Hz), 132.05 (d, p-C in P(S)Ph2, JCP = 2.9 Hz),
1
133.98 (d, ipso-C in P(S)Ph2, JCP = 110.9 Hz), 137.60 (s, C3),
150.87(d, C1, 2JCP =8.1 Hz), 151.51 (s, ipso-CinNPh), 159.00 (s,
CHdN). IR (KBr, ν/cm-1): 641 (PdS), 692, 721, 728, 840, 875,
960, 1104, 1115, 1197, 1253, 1436, 1579, 1630 (CdN), 2863, 3050.
Anal. Calcd for C25H20NOPS: C, 72.62; H, 4.88: N, 3.39. Found:
C, 72.74; H, 4.81; N, 3.31.
O-[3-[(tert-Butylimino)methyl]phenyl] Diphenylthiophosphinate,
3c. 3c was obtained analogously to 3b, except that the reaction
was carried out at room temperature. Yield: 86%. Mp: 75-77 °C
(hexane). 31P{1H} NMR (121.49 MHz, CDCl3): δ 82.26 ppm. 1H
NMR (300.13 MHz, CDCl3): δ 1.25 (s, 9H, C(CH3)3), 7.04-7.08
(m, 1H, H-C4), 7.23 (t, 1H, H-C5, 3JHH = 7.9 Hz), 7.43-7.56 (m,
8H, HAr), 7.95-8.03 (m, 4H, HAr), 8.15 (s, 1H, CHdN). 13C{1H}
NMR (75.47 MHz, CDCl3): δ 29.46 (s, CH3), 57.21 (s, C(CH3)3),
3-(Diphenylthiophosphoryl)oxybenzaldehyde, 2. A solution
of Na2CO3 (0.90 g, 8.2 mmol) and TEBA (4 mol %) in 8 mL of
water was slowly dropwise added to a stirred solution of
3-hydroxybenzaldehyde (1.00g, 8.2mmol) andPh2P(S)Cl(2.07 g,
8.2 mmol) in 14 mL of benzene at 0-5 °C. The resulting reaction
mixture was stirred for 2.5 h at 50-55 °C and after cooling to
roomtemperaturediluted with 20 mL of water andbenzene. After
separation of the organic layer, the water phase was additionally
washed with benzene (25 mL). The combined benzene solution
was dried over anhydrous Na2SO4 and evaporated to dryness.
The resulting residue was purified by silica gel column chromato-
graphy (eluent: hexane-EtOAc (5:1)) to give 1.90 g of 2 as a
white solid. Yield: 72%. Mp: 70-71 °C (Et2O). 31P{1H} NMR
120.92 (s, C2), 122.95 (s, C6), 124.15 (s, C4), 128.38 (d, 3JCP
=
13.2, m-C in P(S)Ph2), 129.16 (s, C5), 131.23 (d, o-C in P(S)Ph2,
2JCP = 11.5,), 131.98 (s, p-C in P(S)Ph2), 134.05 (d, ipso-C in
P(S)Ph2, 1JCP = 113.1 Hz), 138.64 (s, C3), 150.73 (d, C1, 2JCP
=
7.1 Hz), 153.89 (s, CHdN). IR (KBr, ν/cm-1): 644 (PdS), 688,
719, 736, 791, 846, 981, 1107, 1117, 1135, 1206, 1245, 1437, 1478,
1580, 1638 (CdN), 2961, 3072. Raman (solid, Δν/ cm-1): 239, 644
(PdS), 999, 1589, 1605, 1638 (CdN), 3059. Anal. Calcd for
C23H24NOPS: C, 70.21; H, 6.15: N, 3.56. Found: C, 70.35; H,
6.21; N, 3.51.
1
(161.98 MHz, CDCl3): δ 83.75 ppm. H NMR (400.13 MHz,
CDCl3): δ 7.32-7.40 (m, 2H, HAr), 7.45-7.53 (m, 7H, HAr),
7.62 (d, 1H, HAr, 3JHH = 7.3 Hz), 7.98 (dd, 4H, o-H in P(S)Ph2,
3JHH = 7.5 Hz, 3JPH = 13.8 Hz), 9.87 (s, 1H, CHO). IR (KBr,
ν/cm-1): 637 (PdS), 692, 725, 737, 796, 849, 1107, 1118, 1135,
Preparation of the Palladium Complexes 4a-c in Benzene/
Benzene-Methanol. [2-[(Diphenylthiophosphoryl)oxy]-6-[(methoxy-
imino)methyl]phenyl]palladium Chloride, 4a. A solution of PdCl2-
(PhCN)2 (42.2 mg, 0.110 mmol) in 5 mL of benzene was slowly
dropwise added to a solution of 3a (40.4 mg, 0.110 mmol) in 6 mL
of C6H6 under stirring. Then, the reaction mixture was diluted with
(18) Boehringer, C. H. Patent DE 1934443 (A1); Chem. Abstr. 1971,
74, 99676.
(19) Ko, E. C. F.; Robertson, P. E. Can. J. Chem. 1973, 51, 597.
(20) Bellamy, L. J. The Infrared Spectra of Complex Molecules; Wiley:
New York, 1975.

Article
Organometallics, Vol. 29, No. 9, 2010 2061
Table 4. Crystal Data and Structure Refinement Parameters
3b
3c
4a
4b
4c
5b
formula
MW
T, K
C
413.45
100
25H20NOPS
C23H24NOPS
393.46
120
C20H17ClNO2PPdS
508.23
100
C26H21Cl3NOPPdS
639.22
100
C23H23ClNOPPdS
534.30
100
C26H21Cl3NO2PPd
623.16
100
cryst syst
space group
Z(Z0)
monoclinic
P21/n
4
monoclinic
P21/c
8(2)
monoclinic
P21/c
4(1)
monoclinic
C2/c
8(1)
orthorhombic
Pbca
8(1)
monoclinic
C2/c
8(1)
a, A
b, A
10.108(2)
16.831(4)
12.130(3)
91.406(8)
2063.1(8)
1.331
2.51
864
56
15 717
5136
3687
0.0486
0.1271
1.017
26.116(2)
9.5744(7)
17.7198(14)
107.978(2)
4214.4(6)
1.240
2.42
1664
58
45 054
11 184
7297
0.0538
0.1379
1.005
0.841/-0.271
13.3324(3)
10.8818(3)
14.5024(4)
114.223(1)
1918.77(9)
1.759
13.14
1016
100
15 2546
20 127
17 438
0.0231
0.0700
31.0174(14)
9.9746(5)
17.4994(8)
109.130(1)
5115.1(4)
1.660
12.04
2560
58
30 322
6788
5937
0.0290
0.1000
1.001
1.289/-0.740
15.550(3)
16.818(4)
16.818(4)
90.00
4398.2(17)
1.614
11.48
2160
58
51 336
5857
30.418(7)
9.9067(18)
17.049(4)
104.959(9)
4963.5(18)
1.668
11.61
2496
58
19 719
6597
5000
0.0440
0.1329
1.008
0.898/-1.319
c, A
β, deg
V, A3
dcalc, g cm-3
3
μ, cm-1
F(000)
2θmax, deg
reflns collected
indep reflns (Rint
obsd reflns [I > 2σ(I)]
R1
)
5041
0.0212
0.0673
1.001
wR2
GOF
1.004
1.398/-1.705
ΔFmax/ΔFmin, e A-3
0.414/-0.341
0.535/-0.388
4.5 mL of MeOH and heated at 50-60 °C for 1.5 h. After cooling to
room temperature the resulting mixture was filtered and evaporated
to dryness to give 55.9 mg of 4a as a yellow crystalline solid. Yield:
66%. Mp: dec > 230 °C. 31P{1H} NMR (161.98 MHz, CDCl3): δ
76.26 ppm. 1H NMR (400.13 MHz, CDCl3): δ 4.00 (s, 3H, OCH3),
7.09 (d, 1H, H-C4, 3JHH = 8.1 Hz), 7.14-7.18 (m, 1H, H-C5),
7.27 (d, 1H, H-C6, 3JHH = 7.0 Hz), 7.55-7.68 (m, 6H, HAr), 7.94
(dd, 4H, o-H in P(S)Ph2, 3JHH = 7.7 Hz, 3JPH = 13.9 Hz), 8.49 (s,
1H, CHdN). IR (KBr, ν/cm-1): 626 (PdS), 694, 715, 727, 790,
912, 1030, 1108, 1118, 1199, 1439, 1582 (CdN), 2927. Raman (solid,
Δν/ cm-1): 170, 239, 276 (Pd-Cl), 391, 624 (PdS), 994, 1216, 1548,
1584 (CdN),1598, 3051. Anal. Calcd for C20H17ClNO2PPdS: C,
47.26; H, 3.37; N, 2.76. Found: C, 47.11; H, 3.31; N, 2.65.
give 4c as a yellow crystalline solid. Yield: 55.8 mg (53%). Mp:
dec > 265 °C. 31P{1H} NMR (161.97 MHz, CDCl3): δ 74.61
ppm. 1H NMR (400.13 MHz, CDCl3): δ 1.60 (s, 9H, C(CH3)3),
6.88-6.90 (m, 1H, H-C4), 7.07-7.15 (m, 2H, H-C5,6),
7.47-7.52 (m, 5H, HAr), 7.56-7.60 (m, 2H, HAr), 7.93 (s, 1H,
CHdN), 7.95-7.98 (m, 4H, HAr). IR (KBr, ν/cm-1): 632
(PdS), 696, 719, 786, 933, 1108, 1120, 1200, 1438 (P-Ph),
1610 (CdN), 2965, 3055. Anal. Calcd for C23H23ClNOPPdS:
C, 51.70; H, 4.34; N, 2.62. Found: C, 51.87; H, 4.44; N, 2.64.
Preparation of the Complexes 4a/5a and 4b/5b in Dichloro-
methane-Methanol Solutions. A solution of PdCl2(PhCN)2
(0.222 mmol) in 4 mL of dichloromethane was slowly dropwise
added to a solution of 3a,b (0.222 mmol) in 2 mL of CH2Cl2. The
reaction mixture was left under ambient conditions for 5 days in
the case of 3b and 3 weeks in the case of 3a. Then, 3 mL of MeOH
was added to the half-evaporated reaction mixture to give a 1:1
CH2Cl2-MeOH solution. In 5-6 h the resulting mixture was
filtered and evaporated to dryness. The obtained residue was
washed with Et2O (15 mL) and purified by thin-layer chromato-
graphy (eluent CH2Cl2-EtOAc (4:1)) to give the complexes 4a,b
and 5a,b.
[2-[(Diphenylthiophosphoryl)oxy]-6-[(phenylimino)methyl]-
phenyl]palladium Chloride, 4b. A benzene (8 mL) solution of
PdCl2(PhCN)2 (72.2 mg, 0.188mmol) was slowlydropwise added
to a solution of 3b (77.8 mg, 0.188 mmol) in 8 mL of C6H63 The
resulting reaction mixture was refluxed for 2 h under stirring.
After cooling to room temperature, the precipitate was filtered
off and recrystallized from CH2Cl2-Et2O (1:2, 12 mL) to give
4b as a yellow crystalline solid. Yield: 53.1 mg (51%). Mp: dec >
135 °C. 31P{1H} (161.98 MHz, CDCl3): δ 76.48 ppm. 1H NMR
Complexes 4a and 5a (R = OMe). Yields: 22% (4a), 2% (5a).
Spectral data for 5a: 31P{1H} NMR (161.98 MHz, CDCl3): δ
43.83 ppm. 1H NMR (400.13 MHz, CDCl3): δ 4.00 (s, 3H,
3
(400.13 MHz, CDCl3): δ 7.03 (dd, 1H, H-C4, JHH = 8.0 Hz,
4JHP = 1.2 Hz), 7.13-7.17 (m, 1H, H-C5), 7.25-7.38 (m, 6H,
3
4
H
Ar), 7.49-7.54 (m, 4H, HAr), 7.58-7.62 (m, 2H, HAr), 7.96 (dd,
OCH3), 6.87 (dd, 1H, H-C4, JHH = 6.0 Hz, JPH = 2.9 Hz),
7.11-7.12 (m, 2H, H-C5,C6), 7.49-7.64 (m, 6H, HAr), 7.96 (dd,
4H, o-H in P(S)Ph2, 3JHH = 7.5 Hz, 3JPH = 13.1 Hz), 8.01 (s,
1H, CHdN).
4H, o-H in P(S)Ph2, 3JHH = 7.2 Hz, 3JPH = 14.0 Hz), 8.03 (s, 1H,
CHdN). 13C{1H} NMR (100. 61 MHz, CDCl3/DMSO-d6): δ
121.63 (d, C6, JCP = 8.1 Hz), 122.58 (s, o-C in NPh), 125.61
3
(s, C4), 125.79 (s, C5), 126.28 (s, p-C in NPh), 126.84 (s, m-C
in NPh), 126.93 (d, ipso-C in P(S)Ph2, 1JCP = 108.6 Hz), 128.02
(d, m-C in P(S)Ph2, 3JCP = 13.9 Hz), 130.67 (d, o-C in P(S)Ph2,
Complexes 4b and 5b (R = Ph). Yields: 38% (4b), 5% (5b). 5b:
Mp: dec > 180 °C. 31P{1H} NMR (121.97 MHz, CDCl3): δ
43.98 ppm. 1H NMR (300.13 MHz, CDCl3): δ 6.94 (d, 1H, H-C4,
4
2JCP = 11.7 Hz), 132.67 (d, p-C in P(S)Ph2, JCP = 2.2 Hz),
3JHH = 7.8 Hz), 7.14 (m, 1H, H-C5), 7.22 (d, 1H, H-C6, 3JHH
=
138.17 (d, C2, 3JCP = 10.3 Hz), 147.21 (s, C3), 148.51 (s, i-NPh),
150.82 (d, C1, 2JCP = 8.1 Hz), 175.31 (d, CHdN, 4JCP = 5.1 Hz).
IR (KBr, ν/cm-1): 628 (PdS), 690, 716, 744, 924, 1108, 1117,
1204, 1437, 1587, 1603 (CdN), 3052. Anal. Calcd for C25H19-
7.1 Hz), 7.28-7.37 (m, 5H, HAr), 7.48-7.65 (m, 6H, HAr), 7.85
(s, 1H, CHdN), 7.97 (dd, 4H, o-H in P(S)Ph2, 3JHP = 13.1 Hz,
3JHH = 7.4 Hz). IR (KBr, ν/cm-1): 691, 734, 926, 1134, 1152,
1210 (P-O-Ar), 1439 (P-Ph), 1587 (CdN), 3050. Anal. Calcd
ClNOPPdS 0.20CH2Cl2: C, 52.98; H, 3.42; N, 2.45. Found: C,
3
for C25H19ClNO2PPd 0.33CH2Cl2: C, 53.70; H, 3.50: N, 2.47.
3
53.03; H, 3.35; N, 2.39.
Found: C, 53.68; H, 3.54; N, 2.41.
[2-[(Diphenylthiophosphoryl)oxy]-6-[(tert-butylimino)methyl]-
phenyl]palladium Chloride, 4c. A benzene (8 mL) solution of
PdCl2(PhCN)2 (76.2 mg, 0.199 mmol) was slowly dropwise
added to a solution of 3c (78.2 mg, 0.199 mmol) in 8 mL of
C6H6. The resulting reaction mixture was refluxed for 2 h under
stirring. After cooling to room temperature the resulting mix-
ture was filtered and evaporated to dryness. The obtained
residue was recrystallized from CH2Cl2-Et2O (1:2, 12 mL) to
Catalytic Experiments. In a typical experiment a solution of
0.25 mmol of aryl bromide, 0.375 mmol of PhB(OH)2, 2 mmol
of K3PO4, 0.5 mmol of Bu4NBr, and the mentioned amount of
the corresponding palladium complex (used as titrated solu-
tions in DMF) in 1 mL of DMF was heated at 120 °C over 5 h.
After cooling the reaction mixture was immediately filtered,
treated with water, extracted with benzene, and analyzed by
GC and 31P NMR.

2062 Organometallics, Vol. 29, No. 9, 2010
Kozlov et al.
X-ray Crystallography. Single crystals were grown by recrys-
tallization from EtOH (3b) and hexane (3c) or by slow evapora-
tion from CH2Cl2-hexane (1:3), (4b, 5b), CHCl3 (4a), and
CH2Cl2 (4c). All diffraction data were collected on a Bruker
SMART APEX II CCD diffractometer [λ(Mo KR) = 0.71072 A,
ω-scans] at 100 K. The substantial redundancy in data allows
empirical absorption correction to be performed with SA-
DABS,21 using multiple measurements of equivalent reflections.
The structures were solved by direct methods and refined by the
full-matrix least-squares technique against F2 in the anisotropic-
isotropic approximation. All calculations were performed with
the SHELXTL software package.22 Crystal data and structure
refinement parameters are listed in Table 2.
Acknowledgment. The authors are grateful to the Russian
Basic Research Foundation (grant 08-03-00508) for financial
support.
Supporting Information Available: Synthesis of the compara-
tive ligand 6, i.e., O-[3-[(methoxyimino)methyl]phenyl] diphenyl-
phosphonate, bearing an O-PdO group, as well as crystal data
and structure refinement parameters for crystals (PhNH2)2PdCl2
andsolidsolution formedby4band 5b are available free of charge
via the Internet at http://pubs.acs.org.
€
(21) Sheldrick, G. M. SADABS; University of Gottingen, 1996.
(22) SHELXTL, version 6.1; Bruker AXS Inc.: Madison, WI, 2005.