Palladium(II) ion promoted hydroamination of di(phenylethynyl)phenylphosphine and aniline: A facile synthesis of a six-membered P-N heterocycle

Article abstract of DOI:10.1016/S0022-328X(01)01074-9

The six-membered aromatic heterocycle 1,4-2H-1,2,4,6-tetraphenyl-1,4-azaphosphabenzene was generated efficiently via a [PdCl₂(NCMe)₂] promoted hydroamination between PhP(C=CPh)₂ and aniline. The heterocyclization reaction

Full text of DOI:10.1016/S0022-328X(01)01074-9

Journal of Organometallic Chemistry 643–644 (2002) 4–11
www.elsevier.com/locate/jorganchem
Palladium(II) ion promoted hydroamination of
di(phenylethynyl)phenylphosphine and aniline: a facile synthesis of
a six-membered P–N heterocycle
Xueming Liu a, Terence K.W. Ong a, S. Selvaratnam a, Jagadese J. Vittal a,
Andrew J.P. White b, David J. Williams b, Pak-Hing Leung a,*
a Department of Chemistry, Faculty of Science, National Uni6ersity of Singapore, Kent Ridge Crescent, Singapore 119260, Singapore
b Department of Chemistry, Imperial College, London SW 7b2 AY, UK
Received 3 May 2001; accepted 21 June 2001
Dedicated to Professor F. Mathey on the occasion of his 60th birthday, with our best wishes and most sincere congratulations
Abstract
The six-membered aromatic heterocycle 1,4-2H-1,2,4,6-tetraphenyl-1,4-azaphosphabenzene was generated efficiently via a
[PdCl2(NCMe)2] promoted hydroamination between PhP(CꢀCPh)2 and aniline. The heterocyclization reaction adopts a stepwise
mechanism during which the unstable intermediate imino complex cis-dichloro{1,2-diphenyl-3-phenyl(phenylethynyl)phosphino-1-
aza-1-propene}palladium(II) is formed. The end product of the cyclization reaction is a palladium complex in which the
six-membered aromatic P–N heterocyclic ring is coordinated to the metal center as a typical monodentate ligand via phosphorus.
The azaphospha heterocycle can subsequently be liberated from the palladium reaction activator as an air-stable yellow solid by
treating the palladium complex with KCN. In contrast to the intermediate imino-complex, where the CꢁN is readily hydrolyzed
by HCl, the aromatic heterocycle is stable in strongly acidic environments. © 2002 Elsevier Science B.V. All rights reserved.
Keywords: Azaphospha heterocycle; Phosphinoalkyne; Hydroamination reaction; Imino-phosphine; Acid hydrolysis; Crystal structures
cal relationships with analogous pyridine derivatives,
and on their phosphorusꢂcarbon double bond charac-
1. Introduction
ters in the conjugative interaction [4]. More recently,
studies have shown that successive replacement of the
aromatic-carbons in phosphinine by phosphorus [5] or
nitrogen [6] significantly affects the aromaticity of the
six-membered ring, enhancing the ring reactivity
markedly. However, the syntheses of these interesting
six-membered heterocycles generally involve inefficient
multi-step processes. The literature methods generally
involved the initial oxidation and hydrolysis of bis-
alkynylphosphines into the corresponding diketo phos-
phine oxides followed by the silane compound
promoted reductive couplings between the diketo-phos-
phine oxides and the selected amines under prolonged
and strong heating conditions [7]. In some cases, it was
necessary to conduct the reductive coupling reaction in
boiling toluene for 3 days. In this article, we report a
simple and efficient approach for the preparation of a
The syntheses and the reactions of heterocyclic phos-
phorus compounds have been extensively studied over
many years by Mathey and his co-workers [1]. These
heterocycles have been shown to have rich metal–lig-
and chemistry. For example, they are able to form the
classic P–M bonds and participate in the p-accepting
‘‘ring to metal’’ bonding [2]. Among the various
families of phospha-heterocycles, the heteroaromatic
five-membered phosphole core is perhaps the best-stud-
ied system, both by the synthetic and theoretical
chemists. Similarly, the six-membered phosphinine core
has also received considerable attention and has been
the subject of a number of theoretical investigations [3].
In particular, studies have been focused on their chemi-
* Corresponding author. Tel.: +65-772-2682; fax: +65-779-1691.
E-mail address: chmlph@nus.sg (P.-H. Leung).
0022-328X/02/$ - see front matter © 2002 Elsevier Science B.V. All rights reserved.
PII: S0022-328X(01)01074-9

X. Liu et al. / Journal of Organometallic Chemistry 643–644 (2002) 4–11
5
ion promoted heterocyclization reaction between
di(phenylethynyl)phenylphosphine and aniline.
six-membered P–N heterocycle 1,4-2H-1,2,4,6-te-
traphenyl-1,4-azaphosphinine from the palladium(II)
2. Results and discussion
2.1. Formation of a 1,4-azaphosphabenzene heterocycle
No reaction between di(phenylethynyl)phenyl-
phosphine and aniline was observed in the absence of a
transition metal ion. However, in the presence of
PdCl2(NCMe)2 as the reaction promoter, the cycliza-
tion
reaction
between
di(phenylethynyl)phenyl-
phosphine and excess aniline proceeded smoothly in
acetonitrile at 75 °C (Scheme 1). The reaction was
monitored by 31P-NMR spectroscopy and was found to
be complete within 3 h. The 31P-NMR spectrum of the
crude reaction mixture in CDCl3 exhibited only one
singlet at l −10.8, thus indicating the generation of a
sole product in this reaction. The product was purified
by column chromatography and re-crystallized from
dichloromethane–diethyl ether and was obtained as
orange prisms in 59% yield, m.p.\225 °C. Fig. 1
shows the X-ray structure of the product palladium(II)
complex 1 which contains the monodentate P–N hete-
rocycle, a coordinated aniline from the excess reagent
used, and two trans-orientated chloro ligands. Selected
bond distances and angles are given in Table 1. The
X-ray structural analysis thus establishes unambigu-
ously that the palladium(II) ion promoted hydroamina-
tion reaction between di(phenylethynyl)phenylphos-
phine and aniline indeed occurred and the desired
six-membered P–N heterocycle was generated. The ge-
ometry at palladium is slightly distorted square plane
with angles in the ranges of 88.0(1)–91.1(1)° and
176.8(1)–179.0(1)° and a mean deviation from planar-
Scheme 1.
Fig. 1. Molecular structure of complex 1.
,
,
ity of 0.025 A. The Pd(1)ꢂP(1) [2.231(1) A], Pd(1)ꢂN(2)
,
,
[2.147(2) A], Pd(1)ꢂCl(1) [2.296(1) A], and Pd(1)ꢂCl(2)
Table 1
,
Selected bond distances (A) and angles (°) of complex 1
,
[2.302(1) A] bond distances are typical. The geometry at
phosphorus is distorted tetrahedral with angles in the
range of 98.8(1)–118.2(1)°. The distortions involve the
contraction of C(1)ꢂP(1)ꢂC(21) to 98.8(1)° and the
Bond lengths
Pd(1)ꢂP(1)
Pd(1)ꢂCl(1)
P(1)ꢂC(15)
P(1)ꢂC(1)
C(21)ꢂC(22)
N(1)ꢂC(22)
2.231(1)
2.296(1)
1.821(2)
1.770(2)
1.342(3)
1.398(3)
Pd(1)ꢂN(2)
Pd(1)ꢂCl(2)
P(1)ꢂC(21)
C(1)ꢂC(2)
N(1)ꢂC(2)
N(1)ꢂC(9)
2.147(2)
2.302(1)
1.766(2)
1.351(3)
1.392(3)
1.463(3)
drastic
enlargement
of
C(1)ꢂP(1)ꢂPd(1)
and
C(21)ꢂP(1)ꢂPd(1) to 118.2(1) and 115.3(1)°, respec-
tively. Most importantly, the six atoms in the hetero-
cyclic ring lie in one plane, with a maximum deviation
Bond angles
,
of only 0.03 A. Furthermore, the geometry at N(1) is
P(1)ꢂPd(1)ꢂCl(1)
P(1)ꢂPd(1)ꢂCl(2)
Cl(1)ꢂPd(1)ꢂCl(2)
C(15)ꢂP(1)ꢂPd(1)
C(1)ꢂP(1)ꢂC(21)
C(1)ꢂP(1)ꢂPd(1)
C(22)ꢂC(21)ꢂP(1)
C(2)ꢂN(1)ꢂC(22)
C(9)ꢂN(1)ꢂC(22)
C(1)ꢂN(2)ꢂN(1)
90.1(1)
91.1(1)
176.8(1)
107.5(1)
98.8(1)
118.2(1)
125.4(2)
122.0(2)
116.0(2)
124.2(2)
P(1)ꢂPd(1)ꢂN(2)
Cl(1)ꢂPd(1)ꢂN(2)
N(2)ꢂPd(1)ꢂCl(2)
C(1)ꢂP(1)ꢂC(15)
C(21)ꢂP(1)ꢂPd(1)
C(1)ꢂP(1)ꢂC(15)
C(2)ꢂC(1)ꢂP(1)
C(2)ꢂN(1)ꢂC(9)
Pd(1)ꢂN(2)ꢂC(29)
N(1)ꢂC(22)ꢂC(21)
179.0(1)
90.7(1)
88.0(1)
109.0(1)
115.3(1)
109.0(1)
124.9(2)
120.0(2)
112.5(1)
124.0(2)
trigonal-planar with angles in the range of 116.0(2)–
122.0(2)°, indicating that the nitrogen atom in the
six-membered ring adopts the classic sp2 hybridization.
These structural features are consistent with the aro-
matic nature of the P–N heterocycle. In agreement with
the solid state structural studies, the 1H-NMR spectrum
of 1 in CDCl3 recorded a doublet for the two identical
PCH protons in the low field region (l 7.73, 2JPH=13.7
Hz).

6
X. Liu et al. / Journal of Organometallic Chemistry 643–644 (2002) 4–11
sphere, however, is distorted slightly from the square-
planar with the angles in the ranges of 86.2(1)–93.9(1)
and 176.4(1)–178.4(1)°. Similar to that observed in the
monomeric complex 1, the six-membered heterocyclic
rings in 2 adopt a near-planar geometry, the maximum
,
deviation from planarity being 0.02 A. The nitrogen
center is trigonal-planar and adopts the sp2 hybridiza-
tion, with angles in the range of 119.7(5)–120.9(5)°. It
is important to note that the isolation of 2 from the
hydrochloric acid reaction with 1 unambiguously estab-
lished that the P–N heterocycle is stable in strong
acidic conditions.
The six-membered P–N heterocycle could be liber-
ated quantitatively both from the monomeric complex
1 and from the dimeric species 2 by treatment with
aqueous potassium cyanide. Thus the P–N heterocycle
3 was obtained as air-stable yellow solids. The 31P-
NMR spectrum of the liberated heterocycle in CDCl3
exhibits a singlet at l −41.6.
Fig. 2. Molecular structure of complex 2.
Table 2
,
Selected bond distances (A) and angles (°) of complex 2
Bond lengths
Pd(1)ꢂCl(1)
Pd(1)ꢂCl(1C)
Pd(1)ꢂCl(2)
Pd(1)ꢂP(1)
P(1)ꢂC(1)
P(1)ꢂC(21)
C(1)ꢂC(2)
2.332(2)
2.458(2)
2.271(2)
2.204(2)
1.767(6)
1.753(6)
1.330(8)
C(21)ꢂC(22)
C(2)ꢂN(1)
C(22)ꢂN(1)
C(2)ꢂC(3)
C(22)ꢂC(23)
N(1)ꢂC(9)
P(1)ꢂC(15)
1.337(8)
1.412(8)
1.398(8)
1.520(9)
1.507(9)
1.457(8)
1.804(7)
2.2. Mechanistic in6estigations
From a mechanistic standpoint, the formation of the
six-membered P–N heterocycle involves a hydroamina-
tion reaction between aniline and both alkyne groups in
di(phenylethynyl)phenylphosphine. It has been well es-
tablished that this class of hydroamination process can
be activated via metal complexation at the amine func-
tion or at the alkyne moiety [8]. In the above reaction,
however, the kinetic activation step may have involved
the coordination of both di(phenylethynyl)phenyl-
phosphine and aniline on the palladium(II) activator,
although the thermodynamic heterocyclic product was
isolated as a simple monodentate phosphorus ligand.
Furthermore, the two alkyne groups could adopt a
concerted hydroamination reaction mechanism to form
the P–N ring or proceeded the heterocyclization pro-
cess via a step-wise fashion. In order to establish the
reaction mechanism for this cyclization reaction, we
attempted to isolate and identify all possible intermedi-
ates generated in the reaction. To facilitate the isolation
of these reactive species, a similar palladium(II) pro-
moted cyclization reaction was repeated in which only
two molar equivalents of aniline were used and the
reaction mixture was heated in acetonitrile at 75 °C for
0.5 h (Scheme 2). At this stage, the 31P-NMR spectrum
of the reaction mixture in CDCl3 exhibited only one
new singlet at l 12.0. This signal had not been observed
previously and could be due to a key intermediate
generated at the early stage of the heterocyclization
reaction. Unfortunately, attempts to isolate this inter-
mediate were unsuccessful as it was found to be a
chemically reactive species that transformed slowly into
the heterocyclic complex 1, even at room temperature.
The reaction could however, be quenched with acid.
Thus when this reactive material was briefly treated
Bond angles
Cl(1)ꢂPd(1)ꢂCl(1C)
Cl(1)ꢂPd(1)ꢂCl(2)
Cl(1)ꢂPd(1)ꢂP(1)
Pd(1)ꢂCl(1)ꢂPd(1A)
P(1)ꢂPd(1)ꢂCl(1C)
P(1)ꢂPd(1)ꢂCl(2)
Cl(2)ꢂPd(1)ꢂCl(1C)
Pd(1)ꢂCl(1C)ꢂPd(1A)
Cl(1)ꢂPd(1A)ꢂCl(1C)
C(1)ꢂP(1)ꢂPd(1)
C(1)ꢂP(1)ꢂC(21)
C(21)ꢂP(1)ꢂC(15)
C(21)ꢂP(1)ꢂPd(1)
C(1)ꢂP(1)ꢂC(15)
86.2(1)
178.4(1)
93.9(1)
93.8(1)
176.4(1)
86.7(1)
93.3(1)
93.8(6)
86.2(1)
116.2(2)
98.8(3)
108.6(3)
117.4(2)
105.9(3)
C(15)ꢂP(1)ꢂPd(1)
P(1)ꢂC(1)ꢂC(2)
C(1)ꢂC(2)ꢂN(1)
C(1)ꢂC(2)ꢂC(3)
C(3)ꢂC(2)ꢂN(1)
C(2)ꢂN(1)ꢂC(22)
C(2)ꢂN(1)ꢂC(9)
C(9)ꢂN(1)ꢂC(22)
N(1)ꢂC(22)ꢂC(21)
N(1)ꢂC(22)ꢂC(23)
109.0(2)
126.1(5)
123.7(6)
118.8(6)
117.4(6)
120.9(5)
119.2(5)
119.7(5)
125.2(6)
119.1(6)
C(23)ꢂC(22)ꢂC(21) 115.6(6)
C(22)ꢂC(21)ꢂP(1)
C(1)ꢂP(1)ꢂC(15)
P(1)ꢂC(15)ꢂC(16)
125.1(5)
105.9(3)
118.5(5)
The coordinated aniline ligand in complex 1 could be
removed chemoselectively by treatment with concen-
trated hydrochloric acid to generate the dimeric com-
plex 2. The highly crystalline dimer
2 readily
crystallized from dichloromethane–diethyl ether as or-
ange prisms in 95% yield. The 31P-NMR spectrum of 2
in DMSO-d6 exhibits a sole singlet at l −9.1. A single
crystal X-ray structural analysis of 2 was achieved (Fig.
2). Selected bond distances and angles are given in
Table 2. The structural studies of 2 reveal that the
chloro-bridged dipalladium(II) complex has crystallo-
graphic inversion symmetry, with each palladium atom
coordinated to one P–N heterocycle and two bridging
and one terminal chloro ligands. The monodentate
P–N heterocycle ligands adopt a trans geometry in
solid state. The four-membered di-m-chlorobispalladium
bridging unit is planar. The palladium coordination

X. Liu et al. / Journal of Organometallic Chemistry 643–644 (2002) 4–11
7
Scheme 2.
unchanged after it had been heated in acetonitrile at 75
°C for 3 h. Addition of aniline to this reaction sample
also did not initiate any further transformation. How-
ever the imino-phosphine chelate in 5 is unstable in
strong acidic conditions. For example, further treat-
ment of 5 with hydrochloric acid resulted in the cleav-
age of the PdꢂN and the imino CꢁN bonds to form the
monodentate keto phosphine ligands in complex 6. The
dimeric keto-phosphine complex was obtained as or-
ange prisms in 85% yield. The molecular connectivity of
6 has been established unambiguously by X-ray crystal-
lography and shows the structure to be the centrosym-
metric dimer as illustrated in Fig. 4. The planar Pd2Cl2
ring has asymmetric Pd–Cl linkages with that trans to
with concentrated hydrochloric acid, the dimeric com-
plex 2 was obtained in 34% yield. In addition, the acid
treatment also produced a novel imino-phosphine com-
plex 5 as a minor product in 5% isolated yield. The
molecular structure of 5 is depicted in Fig. 3. Selected
bond distances and angles are given in Table 3. The
structural investigations reveal that 5 is a cis-dichloro
palladium complex containing a novel 5-membered
imino-phosphine chelate. The geometry at palladium is
slightly distorted square planar with angles at palla-
dium in the ranges of 81.2(1)–95.2(1) and 172.5(1)–
,
175.4(1)°. The N(1)–C(2) [1.294(6) A] and O(1)ꢂC(22)
,
[1.220(5) A] distances are typical for the nitro-
genꢂcarbon and oxygenꢂcarbon double bonds, respec-
tively. On the other hand, the C(1)ꢂC(2) distance
[1.496(6) A] is typical for a carbonꢂcarbon single bond.
,
Cl(2) [2.325(1) A] being significantly shorter than that
,
trans to phosphorus [2.412(1) A]; the PdꢂCl(2) distance
,
,
,
is 2.271(1) A. The PdꢂP bond length [2.220(1) A] is
marginally longer than the analogous distance in 2. The
palladium coordination plane is planar to within only
Consistent with the formation of a keto and a imino
functions, the IR spectrum (KBr) of 5 exhibited strong
absorption signals at 1670.1 cm−1 (wCꢁO) and 1591.8
cm−1 (wCꢁN).
,
0.053 A reflecting the non-linear PꢂPdꢂCl(1) angle of
In CDCl3 the 31P-NMR spectrum of 5 exhibited a
1
singlet at l 41.7. The H-NMR signals due to the four
non-equivalent PCH protons in 5 were observed as
individual doublet of doublet at l 4.12 (2JHH=18.5 Hz,
2
2JPH=14.4 Hz), 4.39 (2JHH=16.9 Hz, JPH=12.5 Hz),
4.91 (2JHH=16.9 Hz, 2JPH=10.8 Hz) and 5.13 (2JHH
=
18.5 Hz, 2JPH=12.9 Hz). The presence of these 1H-
NMR signals confirm that the imino-phosphine chelate
is stable in solution and does not undergo the imine–
enamine tautomeric transformation rapidly into its
enamino counterpart [9]. Furthermore, the complex is
found to be stable towards prolonged heating. The
31P-NMR studies confirmed that the complex remained
Fig. 3. Molecular structure of complex 5.

8
X. Liu et al. / Journal of Organometallic Chemistry 643–644 (2002) 4–11
Table 3
mechanism. Thus we believe that the sole reactive inter-
mediate that is initially generated is the imino-phos-
phine complex 4 or its enamino analogue 7, or an
equilibrium mixture of both tautomeric complexes 4
and 7 (Scheme 2). We have recently reported a similar
imine–enamine tautomerism with several chiral palladi-
um(II) complexes containing the bidentate ligand
Ph2PCH2C(Ph)ꢁNPh [9]. In principle, complex 7 can
undergo the heterocyclization reaction. Sterically, how-
ever, it is unlikely that the second intra-chelate hy-
droamination reaction can occur while both
phosphorus and nitrogen donors are coordinated to the
palladium activator. It is noteworthy, however, that
bidentate P–N ligands frequently form labile complexes
with palladium(II) [10]. We therefore also believe that
the PdꢂN bond in 7 is kinetically labile, and thus the
enamino-nitrogen can easily leave the metal activator
and undergo a further hydroamination reaction with
the second alkyne substituent on phosphorus to form
the P–monodentate heterocycle. The vacant coordina-
tion site generated from the original PdꢂN bond cleav-
age is then taken up by aniline to form complex 1. On
the other hand, when the intermediate 4 was briefly
quenched with hydrochloric acid, the remaining alkyne
substituent was hydrolyzed to form the ketone function
but the imino-phosphine chelate remained intact. Con-
sidering the structure of the hydrolyzed complex 5, it is
interesting to note that a combination of the double
bond rearrangement and a simple intramolecular con-
densation reaction between the keto group and the
imine function may also result in the formation of the
six-membered P–N heterocycle. However, attempts to
carry out this intra-molecular heterocyclization reaction
under a variety of viable conditions were not successful.
We therefore infer that complex 5 is not one of the
possible intermediates in the current heterocycle
synthesis.
,
Selected bond distances (A) and bond angles (°) of complex 5
Bond lengths
Pd(1)ꢂCl(1)
Pd(1)ꢂCl(2)
Pd(1)ꢂP(1)
Pd(1)ꢂN(1)
P(1)ꢂC(1)
C(1)ꢂC(2)
N(1)ꢂC(2)
2.295(1)
2.367(1)
2.190(1)
2.068(3)
1.817(4)
1.496(6)
1.294(6)
P(1)ꢂC(21)
C(21)ꢂC(22)
C(22)ꢂO(1)
N(1)ꢂC(9)
C(2)ꢂC(3)
P(1)ꢂC(15)
C(22)ꢂC(23)
1.818(4)
1.511(6)
1.220(5)
1.451(5)
1.485(6)
1.811(4)
1.485(6)
Bond angles
Cl(1)ꢂPd(1)ꢂCl(2)
Cl(1)ꢂPd(1)ꢂN(1)
Cl(1)ꢂPd(1)ꢂP(1)
Cl(2)ꢂPd(1)ꢂN(1)
Cl(2)ꢂPd(1)ꢂP(1)
N(1)ꢂPd(1)ꢂP(1)
Pd(1)ꢂN(1)ꢂC(9)
Pd(1)ꢂN(1)ꢂC(2)
C(2)ꢂN(1)ꢂC(9)
N(1)ꢂC(2)ꢂC(1)
N(1)ꢂC(2)ꢂC(3)
92.2(1)
172.5(1)
91.4(1)
95.2(1)
175.4(1)
81.2(1)
120.5(3)
120.0(3)
119.4(4)
118.2(4)
126.0(4)
C(1)ꢂC(2)ꢂC(3)
C(2)ꢂC(1)ꢂP(1)
C(1)ꢂP(1)ꢂPd(1)
C(1)ꢂP(1)ꢂC(15)
C(15)ꢂP(1)ꢂC(21)
Pd(1)ꢂP(1)ꢂC(21)
P(1)ꢂC(21)ꢂC(22)
C(21)ꢂC(22)ꢂO(1)
115.7(4)
106.6(3)
99.7(2)
107.8(2)
108.3(2)
120.4(2)
113.9(3)
120.3(4)
C(21)ꢂC(22)ꢂC(23) 118.3(4)
O(1)ꢂC(22)ꢂC(23)
C(1)ꢂP(1)ꢂC(21)
121.4(4)
111.3(2)
Fig. 4. Molecular structure of complex 6.
174.5(1)°. The angles at phosphorus range between
103.6(2) and 117.2(2)°, the largest departure from tetra-
hedral being that for PdꢂPꢂC(15) and reflecting the
steric interaction between Cl(2) and the C(15) methyl-
ene group. There are no packing interactions of note,
In conclusion, the present study provides a novel and
efficient synthetic route to the preparation of the six-
membered P–N heterocycles via the hydroamination of
the prochiral phosphinoalkyne. Currently, we are ap-
plying this methodology to the synthesis of a range of
substituted azaphosphinines.
,
the shortest intermolecular contact being 2.94 A be-
tween O(8) and its symmetry related counterpart; the
geometry of this approach, however, does not indicate
a carbonyl···carbonyl dipole···dipole interaction.
3. Experimental
The isolation of both imino-phosphine 5 and keto-
phosphine complex 6 unveiled several crucial mechanis-
tic issues regarding the formation of the P–N
heterocycle. For instance, the isolation of the imino-
phosphine chelate in 5 confirmed that only one of the
two alkyne substituents in di(phenylethynyl)phenyl-
phosphine was initially involved in the hyroamination
reaction with aniline. Thus the heterocyclization reac-
tion between coordinated di(phenylethynyl)phenyl-
phosphine and aniline proceeds via a stepwise reaction
Reactions involving air-sensitive compounds were
performed under a positive pressure of purified nitro-
gen. NMR spectra were recorded at 25 °C on Bruker
ACF300 and AMX500 spectrometers. Elemental analy-
ses were performed by the Elemental Analysis Labora-
tory of the Department of Chemistry at the National
University of Singapore. The phosphinoalkyne
phenyl(diphenylethynyl)phosphine was prepared by the
literature method [11].

X. Liu et al. / Journal of Organometallic Chemistry 643–644 (2002) 4–11
9
3.1. Heterocyclization reaction: synthesis of the
monomeric complex 1
min. The resulting yellow organic layer was separated,
washed with water, and dried over anhydrous MgSO4.
Upon removal of the solvent, the product was obtained
as air-stable pale yellow solids (72 mg, 72%). 31P-NMR
A mixture of PdCl2(NCMe)2 (1.0 g, 3.86 mmol),
PhP(CꢀCPh)2 (1.2 g, 3.86 mmol) and aniline (1.8 g,
19.30 mmol) in MeCN (250 ml) was refluxed at 75 °C
for 3 h. The solvent was the removed under reduced
pressure to give a black residue. The crude product was
purified by silica gel column chromatography (EtOAc–
hexane 1:1, v/v). The product was obtained as orange
crystals after crystallization from CH2Cl2–Et2O (1.5 g,
58.8%) 31P-NMR (CDCl3) l −10.8 (s). 1H-NMR
1
(CDCl3) l −41.6 (s). H-NMR (CDCl3) l 5.79 (d, 2H,
2JPH=5.6 Hz, 2×ꢁCH), 6.78–7.70 (m, 20H, aromat-
ics). Anal. Found: C, 83.8; H, 5.1; N, 3.1. Calc. for
C28H22NP: C, 83.4; H, 5.5; N, 3.5%. The heterocycle
could be liberated similarly from the dimer 2.
3.4. Isolation of the dichloro imino-phosphine complex
5
2
(CDCl3) l 7.73 (d, 2H, JPH=13.7 Hz, ꢁCH), 8.73–
3
9.50 (m, 23 H, aromatics), 10.08 (dd, 2H, JHH=7.4
A mixture of PdCl2(NCMe)2 (0.5 g, 1.93 mmol),
PhP(CꢀCPh)2 (0.6 g, 1.93 mmol) and aniline (0.4 g, 3.86
mmol) in MeCN (75 ml) was refluxed at 75 °C for 0.5
h. The reaction was allowed to cool to room tempera-
ture (r.t.) and the solvent was removed immediately
under reduced pressure to give a black residue. The
residue was redissolved in CH2Cl2 (30 ml) and treated
with concentrated HCl (10 ml) for 15 min. The solution
was washed with water, and dried over anhydrous
MgSO4. Both the dimeric complex 2 (0.25 g, 34%) and
the hydrolyzed imino-phosphine complex 5 were ob-
tained separately by the slow fractional crystallization
of the residue from CH2Cl2–Et2O. Complex 5 was
isolated as highly crystalline yellow prisms (15 mg, 5%).
31P-NMR (CD2Cl2): l 41.7(s). 1H-NMR (CD2Cl2) l
3
Hz, JPH=12.6 Hz, o-PPh). Anal. Found: C, 60.4; H,
4.1; N, 4.5. Calc. for C34H29N2Cl2PPd: C, 60.6; H, 4.3;
N, 4.1%.
3.2. Acid hydrolysis of monomer 1 and synthesis of the
dimer 2
The monomeric complex 1 (0.5 g, 0.74 mmol) was
treated with concentrated HCl (20 ml) in CH2Cl2 (50
ml) for 15 min. The solution was washed with water,
and dried over anhydrous MgSO4. Crystallization of
the crude product from CH2Cl2–Et2O gave the dimeric
complex 2 as orange prisms (0.42 g, 94.7%). 31P-NMR
1
(DMSO-d6) l −9.1 (s). H-NMR (DMSO-d7) l 5.66
(d, 4H, 2JPH=12.5 Hz, ꢁCH), 6.89–7.66 (m, 36H,
aromatics), 8.17 (dd, 4H, 3JHH=6.4 Hz, 3JPH=12.0
Hz, o-PPh). Anal. Found: C, 57.5; H, 4.0; N, 2.4. Calc.
for C56H44N2Cl4P2Pd2: C, 57.9; H, 3.8; N, 2.4%.
2
2
4.12 (dd, 1H, JHH=18.5 Hz, JPH=14.4 Hz, CaHH),
2
2
4.39 (dd, 1H, JHH=16.9 Hz, JPH=12.5 Hz, CbHH),
2
3
4.91 (dd, 1H, JHH=16.9 Hz, JPH=10.8 Hz, CbHH),
2
2
5.13 (dd, 1H, JHH=18.5 Hz, JPH=12.9 Hz, CaHH),
3
6.87–7.95 (m, 18H, aromatics); 8.02 (dd, 2H, JHH
=
6.0 Hz, 3JPH=12.8 Hz, o-PPh). IR (cm−1, KBr)
1670.1, 1591.8, 1445.4, 1288.0, 1186.1, 986.5, 743.4,
692.7. Anal. Found: C, 55.8; H, 4.1; N, 2.4. Calc. for
C28H24NCl2OPPd: C, 56.2; H, 4.0; N, 2.3%.
3.3. Liberation of the six-membered P–N heterocycle.
Isolation of 1,4-2H-1,2,4,6-tetraphenyl-1,4-
azaphosphabenzene (3)
A solution of complex 1 (0.12 g, 0.18 mmol) in
CH2Cl2 (20 ml) was stirred vigorously with a saturated
aqueous solution of KCN (0.5 g, 7.69 mmol) for 30
3.5. Acid hydrolysis of 5 and isolation of the dimeric
keto-complex 6
Table 4
A solution of the dichloro complex 5 (50 mg, 0.08
mmol) in CH2Cl2 (25 ml) was treated with concentrated
HCl (10 ml) at r.t. for 2 h. The organic layer was
separated, washed with water and then dried over
anhydrous MgSO4. Removal of the solvent gave a
yellow residue. The dimer 6 (Table 4) was obtained as
orange prisms by crystallization of the residue from
CH2Cl2–Et2O (37 mg, 85%). 31P-NMR (CDCl3) l 16.6
,
Selected bond distances (A) and bond angles (°) of the complex 6
Bond lengths
PdꢂP
PdꢂCl(1A)
PdꢂCl(1)
PdꢂCl(2)
2.220(1)
2.325(1)
2.412(1)
2.271(1)
PꢂC(15)
PꢂC(6)
PꢂC(7)
1.814(4)
1.813(2)
1.833(5)
1.514(6)
C(7)ꢂC(8)
Bond angles
PꢂPdꢂCl(2)
Cl(2)ꢂPdꢂCl(1A)
Cl(2)ꢂPdꢂCl(1)
PdꢂCl(1)ꢂPdA
C(6)ꢂPꢂC(7)
C(6)ꢂPꢂPd
92.6(1)
177.3(1)
91.9(1)
PꢂPdꢂCl(1A)
PꢂPdꢂCl(1)
Cl(1)ꢂPdꢂCl(1A)
C(6)ꢂPꢂC(15)
C(15)ꢂPꢂC(7)
C(15)ꢂPꢂPd
89.6(1)
174.5(1)
86.0(1)
103.6(2)
110.0(2)
117.2(2)
119.6(3)
1
2
(s). H-NMR (CDCl3) l 4.40 (ABX, 4H, JPH=16.5,
2
2JHH=13.7, 2×CH2), l 4.67 (ABX, 4H, JPH=16.7,
94.1(1)
2JHH=10.8, 2×CH2), 7.40–7.92 (m, 26H, aromatics),
109.4(2)
111.8(1)
104.8(2)
3
3
8.02 (dd, 4H, JPH=12.7, JHH=7.2, o-PPh×2) Anal.
Found: C, 50.2; H, 3.7. Calc. for C28H24NCl2O4PPd: C,
50.5; H, 3.7%.
C(7)ꢂPꢂPd
PꢂC(7)ꢂC(8)

10
X. Liu et al. / Journal of Organometallic Chemistry 643–644 (2002) 4–11
Table 5
Crystallographic data for complexes 1, 2, 5 and 6
1
2
5
6
Formula
Formula weight
Space group
Crystal system
Unit cell dimensions
C34H29Cl2N2PPd
673.86
P21/c
C
56H44Cl4N2P2Pd2·2CH2Cl2
C28H24Cl2NOPPd·CH2Cl2
683.68
C44H38O4P2Cl4Pd2
1047.28
1331.32
P21/c
Monoclinic
(
(
P1
P1
Monoclinic
Triclinic
Triclinic
,
a (A)
14.119(1)
17.495(1)
13.266(1)
–
111.410(1)
–
11.807(1)
16.211(1)
15.007(1)
–
91.038(1)
–
9.820(2)
12.586(2)
14.066(3)
113.843(3)
99.531(3)
103.087(3)
1482.8(5)
2
9.533(1)
10.273(1)
11.466(2)
77.019(9)
86.180(12)
80.614(10)
1079.1(3)
1
,
b (A)
,
c (A)
h (°)
i (°)
k (°)
V (A )
3
,
3050.7(3)
4
2872.0(2)
2
Z
T (K)
223
1.467
223
1.540
173
1.531
293
1.612
Dcalc (g cm−3
)
,
u (A)
0.71073
8.62
0.0298
0.0665
0.71073
10.93
0.0834
0.1407
0.71073
10.63
0.0432
1.54178
100.4
0.0389
v (cm−1
R1
)
Rw
0.1263
0.1025
3.6. Crystal structure determination of 1, 2, 5 and 6
Acknowledgements
Crystal data for all complexes and a summary of the
crystallographic analyses are given in Table 5. For
compounds 1, 2 and 5, diffraction data were collected
at the National University of Singapore on a Siemens
SMART OCD diffractometer with Mo–Ka radiation
(graphite monochromator) using v-scans. Compound 6
was analyzed at the Imperial College using a Siemens
P4/PC diffractometer with Cu–Ka radiation (graphite
monochromator) using v-scans. For all four com-
pounds, semiempirical absorption corrections were ap-
plied and refinements by full-matrix least-square were
based on SHELXL 93 [12]. All non-hydrogen atoms were
refined anisotropically. Hydrogen atoms were intro-
duced at fixed distance from carbon and nitrogen atoms
and were assigned fixed thermal parameters.
We are grateful to the National University of Singa-
pore for the financial support, a Research Fellowship to
S.S. and a PhD research scholarship to X.M.L. We
express our appreciation to Professor G. Bertrand for
his invitation to take part in this occasion.
References
[1] (a) K.B. Dillon, F. Mathey, J.F. Nixon, Phosphorus: The Car-
bon Copy, Wiley, Chichester, UK, 1998;
(b) S. Choua, H. Sidorenkova, T. Berclaz, M. Geoffroy, P. Rosa,
N. Mezailles, L. Ricard, F. Mathey, P. Le Floch, J. Am. Chem.
Soc. 122 (2000) 12227;
(c) E. Mattmann, D. Simonutti, L. Ricard, F. Mercier, F.
Mathey, J. Org. Chem. 66 (2001) 755;
(d) N. Mezailles, N. Maigrot, S. Hamon, L. Ricard, F. Mathey,
P. Le Floch, J. Org. Chem. 66 (2001) 1054;
4. Supplementary material
(e) U. Rhorig, N. Mezailles, N. Maigrot, L. Ricard, F. Mathey,
P. Le Floch, Eur. J. Inorg. Chem. 12 (2000) 2565.
[2] E. Deschamps, L. Ricard, F. Mathey, Organometallics 20 (2001)
1499.
[3] (a) L. Nyulaszi, T. Veszpre´mi, J. Phys. Chem. 100 (1996) 6456;
(b) K. Baldridge, M.S. Gordon, J. Am. Chem. Soc. 110 (1988)
4204;
(c) P.D. Burrow, A.J. Ashe III, D.J. Belville, K.D. Jvrdan, J.
Am. Chem. Soc. 104 (1982) 425.
[4] G. Frison, A. Sevin, N. Avarvari, F. Mathey, P. Le Floch, J.
Org. Chem. 64 (1999) 5524.
Crystallographic data for the structural analysis have
been deposited in cif format with the Cambridge Crys-
tallographic Data Centre, CCDC nos. 162699, 162700,
162701 and 162702 for compounds 1, 2, 5 and 6,
respectively. Copies of this data may be obtained free
of charge from The Director, CCDC, 12 Union Road,
Cambridge CB2 1EZ, UK (Fax: +44-1223-336033;
e-mail: deposit@ccdc.cam.ac.uk or www: http://
www.ccdc.cam.ac.uk).
[5] S.R. Bachrach, P. Magdalinos, Theochem 368 (1996) 1 and
references cited therein.

X. Liu et al. / Journal of Organometallic Chemistry 643–644 (2002) 4–11
11
[6] (a) A.A. Tolmachev, S.I. Dovgopoly, A.N. Kostyuk, E.S. Kozlov,
A.O. Pushechnikov, W. Holzer, Hetroatom Chem. 10 (1999)
391;
(b) E. Fluck, F. Rosche, G. Heckmann, F. Weller, Heteroatom
Chem. 6 (1995) 355.
[7] G. Markl, D. Matthes, Angew. Chem. Int. Ed. Engl. 11 (1972)
1019.
[9] X.M. Liu, K.F. Mok, P.-H. Leung, Organometallics 20 (2001)
3918.
[10] J.W. Martin, J.A.L. Palmer, S.B. Wild, Inorg. Chem. 23 (1984)
2664.
[11] A.J. Carty, N.K. Hota, H.A. Patel, T.J. O’Connor, Can. J. Chem.
49 (1971) 2707.
[12] G.M. Sheldrick, SHELXL 93, Program for Crystal Structure
Refinement, University of Go¨ttingen, Go¨ttingen, Germany, 1993.
[8] T.E. Mu¨ller, M. Beller, Chem. Rev. 98 (1998) 675.