Zirconium(IV) tris(phosphinoamide) complexes as a tripodal-type metalloligand: A route to Zr-M (M = Cu, Mo, Pt) heterodimetallic complexes

Article abstract of DOI:10.1002/ejic.200700007

Treatment of ZrCl₄ with 3 equiv. of Li(PPh₂NR) (R = tBu, iPr) gives tris(phosphinoamide)zirconium complexes, (PPh ₂NR)₃-ZrCl [R = tBu (1a), iPr (1b)], in high yield. Crystal structures of 1a,b show that three phosphorus atoms are coordinated to the zirconium center in the solid state, whereas variable temperature NMR studies indicate a reversible coordination/dissociation process of three phosphorus atoms in the solution state. Reaction of 1a,b with CuCl give rise to the formation of Zr-Cu heterodimetallic complexes, ClCu(Ph₂PNR) ₃-ZrCl [R = tBu (2a), iPr (2b)]. The molecular structures of 2a,b show that the Cu atom adopts a pseudotetrahedral coordination geometry with tripodal phosphorus moieties and a chlorine atom, whereas the ligand arrangement around the Zr atom is trigonal bipyramidal with the linear Cl-Zr-Cu axis at the apical site. A Zr-Mo heterodimetallic complex, (CO)₃-Mo(Ph ₂PNiPr)₃ZrCl (3), is synthesized from 1b and Mo(CO) ₃-(CH₃CN)₃, in which the ligands around the Mo center are arranged octahedrally, and three phosphorus moieties are coordinated in the fac-P₃ fashion. The reaction of 1b with a square planar Pt^II precursor, such as (COD)PtCl₂, is unique and gives (κ²-Ph₂PNiPr)Pt(Ph₂PNiPr) ₂ZrCl₃ (4), which is the first example of a Zr-Pt zwitterionic heterodimetallic complex. The reaction involves intramolecular migration of two chlorine atoms from Pt to Zr as well as that of a Ph ₂PNiPr moiety from Zr to Pt. Wiley-VCH Verlag GmbH & Co. KGaA, 2007.

Full text of DOI:10.1002/ejic.200700007

FULL PAPER
DOI: 10.1002/ejic.200700007
Zirconium(IV) Tris(phosphinoamide) Complexes as a Tripodal-Type
Metalloligand: A Route to Zr–M (M = Cu, Mo, Pt) Heterodimetallic
Complexes
Takashi Sue,[a] Yusuke Sunada,[a,b] and Hideo Nagashima*[a,b]
Keywords: Phosphinoamide / Heterodimetallic / Tripodal ligands
Treatment of ZrCl4 with 3 equiv. of Li(PPh2NR) (R = tBu, iPr)
gives tris(phosphinoamide)zirconium complexes, (PPh2NR)3-
ZrCl [R = tBu (1a), iPr (1b)], in high yield. Crystal structures
of 1a,b show that three phosphorus atoms are coordinated
to the zirconium center in the solid state, whereas variable
temperature NMR studies indicate a reversible coordination/
dissociation process of three phosphorus atoms in the solu-
the apical site. A Zr–Mo heterodimetallic complex, (CO)3-
Mo(Ph2PNiPr)3ZrCl (3), is synthesized from 1b and Mo(CO)3-
(CH3CN)3, in which the ligands around the Mo center are
arranged octahedrally, and three phosphorus moieties are co-
ordinated in the fac-P3 fashion. The reaction of 1b with a
square planar PtII precursor, such as (COD)PtCl2, is unique
and gives (κ2-Ph2PNiPr)Pt(Ph2PNiPr)2ZrCl3 (4), which is the
tion state. Reaction of 1a,b with CuCl give rise to the forma- first example of a Zr–Pt zwitterionic heterodimetallic com-
tion of Zr–Cu heterodimetallic complexes, ClCu(Ph2PNR)3-
ZrCl [R = tBu (2a), iPr (2b)]. The molecular structures of 2a,b
show that the Cu atom adopts a pseudotetrahedral coordina-
tion geometry with tripodal phosphorus moieties and a chlo-
rine atom, whereas the ligand arrangement around the Zr
atom is trigonal bipyramidal with the linear Cl–Zr–Cu axis at
plex. The reaction involves intramolecular migration of two
chlorine atoms from Pt to Zr as well as that of a Ph2PNiPr
moiety from Zr to Pt.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2007)
sult of their special structural and chemical properties.[3]
Tridentate tertiary phosphane ligands can be classified into
two types according to their preferred coordination geome-
tries: one is the fac-P3 (tripodal) coordination mode,
whereas the other is the mer-P3 coordination mode. The
phosphane 1,1,1-tris[(diphenylphosphanyl)methyl]ethane
(triphos) is the best-known fac-P3 tripodal-type phosphane
ligand for transition metals.[4] For instance, treatment of
triphos with CuCl or Mo(CO)3(CH3CN)3 reportedly gives
(triphos)CuCl[4a] or (triphos)Mo(CO)3, respectively.[4b]
These examples clearly demonstrate that the tripodal li-
gands rigidly coordinate to a single transition-metal center
in the fac-P3 fashion so as to stabilize the mononuclear
transition-metal complexes with tetrahedral and octahedral
coordination geometries. This feature is different from the
mer-P3 ligands that are often used to stabilize a multimet-
allic framework, which bridges over the dual metals by vir-
tue of its structural flexibility.
An interesting recent development used to control the
electronic properties of the phosphane ligand is the intro-
duction of a Lewis acid center in the ligand backbone.[5]
As shown in Figure 1, the “amphoteric ligand” reported by
Labinger is a typical example, in which both the Lewis
acidic R2Al moieties and Lewis basic R2P units are in-
cluded in one molecule; unique reactivity of the aluminum
phosphinoamide Ph2PNtBuAlEt2 was discovered in the re-
action with several transition-metal carbonyls.[6,7] The bi-
1. Introduction
It is well-known in coordination chemistry that ligand
design is important for the synthesis of transition-metal
complexes with unique structures and properties. Phospha-
nes are one of the most investigated ligands, especially in
their application to homogeneous catalysis. Numerous
monodentate and bidentate phosphane ligands have been
developed, and their ligation to appropriate transition met-
als has contributed to the opening of new aspects in cata-
lytic transformations of organic molecules and polymers.[1]
Electronic and steric effects of these phosphane ligands are
well-documented by using Tolman’s χ values and cone
angles for monodentate phosphanes and bite angles for bi-
dentate phosphane ligands.[2] Thus, the chemical properties
of most transition-metal complexes having monodentate
and/or bidentate phosphane ligands are predictable in mod-
ern organometallic chemistry. Although relatively little has
been studied compared with chemistry of mono- and biden-
tate phosphane ligands, tridentate phosphane ligands have
attracted the attention of organometallic chemists as a re-
[a] Graduate School of Engineering Sciences, Kyushu University
Kasuga, Fukuoka, 816-8580 Japan
[b] Institute for Materials Chemistry and Engineering,
Kasuga, Fukuoka, 816-8580 Japan
E-mail: nagasima@cm.kyushu-u.ac.jp
Supporting information for this article is available on the
WWW under http://www.eurjic.org or from the author.
Eur. J. Inorg. Chem. 2007, 2897–2908
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2897

T. Sue, Y. Sunada, H. Nagashima
FULL PAPER
Scheme 1.
dentate version of the amphoteric ligand was lately reported
We were interested in the fact that the cyclopentadienyl
by Bourissou and coworkers who reported the synthesis of group in previous examples of the fac-P3 tripodal-type am-
diphosphanylborane (o-iPr2P-C6H3)2B(Ph) and its ligation photeric ligands reduced the Lewis acidity of the zirconium
to the RhCl moiety in the reaction with [Rh(nbd)Cl]2.[8,9] center. As described above, we synthesized the titanium
We reported earlier a titanium version of bidentate ampho- phosphinoamides as a bidentate amphoteric ligand, which
teric ligands, (Ph2PNR)2TiCl2 (R = tBu, iPr), which are was successfully ligated to organoruthenium or platinum
good precursors to Ti–Pt and Ti–Ru heterodimetallic com- species to give the corresponding Ti–Ru and Ti–Pt com-
plexes, YYЈPt(Ph2PNtBu)2TiCl2 (Y, YЈ = Cl; Y = Me, YЈ plexes. Of interest are the structures of these heterodimetal-
= Cl; Y = p-Tol, YЈ = Cl; Y, YЈ = Me) and Cp*Ru(µ- lic complexes: the Ti–Ru compound has a bridging chlorine
Cl)(Ph2PNtBu)2Ti = O as shown in Scheme 1.[10] The Ti– atom, whereas the existence of a PtǞTi dative bond is sug-
Pt complexes have a trigonal bipyramidal titanium and a gested from the crystal structure of the Ti–Pt complexes.
square planar platinum center,[10a] whereas the Ti–Ru com- The Lewis acidity of the titanium center is probably the
plex has titanium and ruthenium atoms with trigonal bipy- reason for these structural features. In this context, we ex-
ramidal and tetrahedral geometries, respectively.[10b] Syn- amined the synthesis of (PPh2NR)3TiCl (R = iPr, tBu) by
thesis of a tridentate amphoteric ligand was examined by the reaction of TiCl4 with 3 equiv. of LiNRPPh2; however,
the design of the fac-P3 tripodal-type compounds contain- (PPh2NR)2TiCl2 was the only product at room temperature
ing a cyclopentadienyl or tris(pyrazolyl)hydroborate zirco- and only a small amount of the species assignable to the
nium moiety in the molecule; Cp*Zr(OCH2PPh2)3 or desired (PPh2NR)3TiCl (R = iPr, tBu) was detectable by
TpZr(OCH2PPh2)3 acts as
a tripodal-like ligand in NMR spectroscopy when the reaction mixture was heated
Cp*Zr(OCH2PPh2)3Ni(CO), Cp*Zr(OCH2PPh2)3RhMe, at 70 °C. In sharp contrast, the reaction of 3 equiv. of
and TpZr(OCH2PPh2)3Mo(CO)3.[11]
lithium phosphinoamide with ZrCl4 successfully
led to the formation of tris(phosphinoamide)zirconium,
(PPh2NR)3ZrCl [R = tBu (1a), iPr (1b)]. In this paper, we
describe the preparation, molecular structure, and solution
dynamics of 1a,b and their coordination behavior to several
late transition metals. The heterodimetallic complexes
ClCu(Ph2PNR)3ZrCl
and
(CO)3Mo(Ph2PNiPr)3ZrCl
showed that these zirconium phosphinoamides behave as a
fac-P3 tripodal ligand which effectively stabilizes the late-
transition-metal complexes with tetrahedral and octahedral
geometries. A unique reaction involving complicated inter-
metallic ligand rearrangement occurred in the reaction of
1b with (COD)PtCl2, which afforded a zwitterionic di-
metallic complex, (κ2-Ph2PNiPr)Pt(Ph2PNiPr)2ZrCl3 (4).
2. Results and Discussion
2.1 Synthesis and Characterization of Tris(phosphinoamide)-
zirconium Complexes (Ph2PNtBu)3ZrCl (1a) and
(Ph2PNiPr)3ZrCl (1b)
Treatment of 3 equiv. of Li(Ph2PNR) (R = tBu, iPr) with
zirconium tetrachloride in THF afforded tris(phosphino-
amide)zirconium complexes, (Ph2PNR)3ZrCl [R = tBu (1a),
iPr (1b)], in 86 and 84% yield, respectively, as colorless crys-
Figure 1. Monodentate, bidentate, and tridentate phosphorus li-
gands containing a Lewis acidic center.
2898
www.eurjic.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2007, 2897–2908

(PPh2NR)3ZrCl Complexes as a Tripodal-Type Metalloligand
FULL PAPER
tals (Scheme 2). A single 31P NMR resonance appeared for
1a,b at –11.8 and –12.0 ppm, respectively, which were 34.4
and 46.5 ppm shifted to higher field relative to
that of the phosphinoamide ligands Ph2PNHtBu and
Ph2PNHiPr.[12] Although dynamic behavior was observed,
1
as described later in detail, H and 13C NMR spectroscopy
clearly showed signals assignable to Ph2P and tBu or iPr
groups. The molecular structures of 1a,b were determined
by X-ray diffraction analysis and are shown in Figure 2;
representative bond lengths and angles for 1a,b are summa-
rized in Table 1. The asymmetric unit of 1a contains three
crystallographically independent but chemically equivalent
molecules. Ambiguous value of the Flack parameter (0.42)
suggests the structure of 1a is inversion twinned, which is
sometimes seen in a trigonal space group. There is a three-
fold axis along with the Zr–Cl bond in the solid-state struc-
ture of all three independent molecules of 1a. One of the
molecular structures of 1a is described in Figure 2. Three
phosphinoamide ligands are connected to the zirconium
center by a Zr–N bond and all of the three phosphorus
moieties are coordinated. Both 1a,b assume a distorted tet-
rahedral geometry with one chlorine atom and three centers
of the N–P bonds. In the solid-state structure, all three
phosphorus atoms are oriented away from the chlorine
atom, and no isomer was found in the crystal. The N–P
bond lengths are comparable to those seen in
Figure 2. Molecular structures of (Ph2PNtBu)3ZrCl (1a) (top)
(Ph2PNiPr)3ZrCl (1b) (bottom) with 50% probability ellipsoids.
Eisen’s zirconium tetrakis(phosphinoamide) complex, Hydrogen atoms are omitted for clarity.
Zr(Ph2PNPh)4,[13] our previously reported titanium com-
plexes, (Ph2PNR)2TiCl2 (R = tBu, iPr), and other known
phosphinoamide or related complexes.[14] The Zr–N bond
lengths around 2.15 Å are slightly longer than those found
in previously reported zirconium amide complexes,[15]
whereas the Zr–P bond lengths are somewhat shorter than
those of known zirconium phosphane complexes.[16]
Although all of the three phosphorus atoms are bonded
with the zirconium center in the solid state, NMR spectro-
scopic studies suggested a dynamic coordination/dissoci-
ation process of these three phosphorus atoms that is sim-
ilar to that of titanium phosphinoamide, (Ph2PNR)2-
TiCl2,[10a] and Zr(Ph2PNPh)4.[13] In a representative exam-
1
ple, H NMR spectra of 1a are dependent on the tempera-
ture measured: at room temperature, three broad signals at
δ = 6.99–7.17 ppm and one singlet at δ = 1.47 ppm were
seen, which are assignable to the protons of the Ph group
and the methyl portion of the tBu group. At –90 °C, the
spectrum consists of six signals at δ = 6.53 (m, 2 H), 6.66
(t, 2 H), 6.95 (t, 1 H), 7.20 (t, 2 H), 7.24 (m, 2 H), and 7.29
(t, 1 H) ppm and a slightly broadened singlet at δ =
1.37 ppm, which are assignable to two sets of phenyl signals
Scheme 2.
Table 1. Representative bond lengths and angles for 1a,b.
Bond lengths [Å]
Bond angles [°]
1a
1b
1a
1b
Zr–Cl(1)
Zr–N(1)
Zr–N(2)
Zr–N(3)
Zr–P(1)
Zr–P(2)
Zr–P(3)
N(1)–P(1)
N(2)–P(2)
N(3)–P(3)
2.505(2) [2.488(2), 2.525(2)]
2.183(6) [2.181(4), 2.180(3)]
2.4865(9)
2.150(2)
2.137(2)
2.101(2)
2.6659(9)
2.6577(9)
2.6516(9)
1.645(2)
1.648(3)
1.656(2)
Cl–Zr–N(1)
Cl–Zr–N(2)
Cl–Zr–N(3)
N(1)–Zr–N(2) 120.0(2) [119.9(2), 120.2(2)]
N(1)–Zr–N(3)
91.09(13) [91.07(12), 89.70(11)] 86.28(7)
86.18(8)
93.81(8)
133.29(10)
117.07(10)
109.40(10)
2.626(2) [2.617(2), 2.608(2)]
1.654(5) [1.662(5), 1.656(4)]
N(2)–Zr–N(3)
N(1)–Zr–P(1) 38.87(14) [39.22(13), 39.21(12)] 38.09(7)
N(2)–Zr–P(2)
N(3)–Zr–P(3)
38.30(8)
38.64(7)
Eur. J. Inorg. Chem. 2007, 2897–2908
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
2899

T. Sue, Y. Sunada, H. Nagashima
FULL PAPER
of the ortho-, meta-, para-, meta-, ortho-, and para protons, ing zirconium tetrakis(phosphinoamide). It should be noted
respectively, and a signal for the tBu group. 1H NMR spec- that we have never seen the formation of (Ph2PNR)2ZrCl2
troscopic resonances due to the phenyl moieties were un- (R = tBu, iPr) even in the reaction of ZrCl4 with less than
1
equivocally assigned by a H–1H selective decoupling mea- 2 equiv. of Li(Ph2PNR), though Zhang et al. recently re-
surement. The spectrum is explained by the existence of ported the synthesis of bis(phosphinoamide)zirconium
three magnetically equivalent phosphinoamide ligands, in complexes, (Ph2PNR)2ZrCl2 (R = tBu, Ph), by the reaction
which two phenyl groups bonded with each phosphorus of 2 equiv. of Li(Ph2PNR) (R = tBu, Ph) with ZrCl4; no
atom are inequivalent. Similar variable temperature NMR detailed spectroscopic data were described.[18] Repeated
studies of 1b also showed dynamic behavior; two signals attempts to reproduce the result of Zhang were unsuccess-
due to the Ph group [δ = 7.12–7.17 (m, 12 H), 7.22–7.26 ful. In a typical example, 2 equiv. of Li(Ph2PNtBu) was re-
(m, 18 H) ppm] appeared at room temperature, which were acted with ZrCl4 in THF from which only complex 1a was
broadened at –90 °C to show three broad signals. Similarly, obtained in 30~50% yield and unreacted ZrCl4 was reco-
the methyl and methine signals of the iPr group were grad- vered as a THF adduct. This is interesting in comparison
ually broadened at lower temperatures, and only a single set with our previous result that the reaction of TiCl4 with
of broad signals was seen even at –90 °C. Because 1b is not 2 equiv. of Li(Ph2PNR) resulted in the exclusive formation
very soluble in CD2Cl2 at low temperatures, the well-re- of (Ph2PNR)2TiCl2.
solved spectra below –90 °C, in which two doublets due to
the inequivalent methyl groups should appear, were unable
to be obtained because of technical problems. These spec-
tral changes are consistent with the fact that phosphorus
atoms are coordinated to the zirconium center at lower tem-
peratures, whereas reversible dissociation takes place at
2.2 Use of 1 as a Tripodal-Type Metalloligand:
Construction of Zr–Cu and Zr–Mo Heterodimetallic
Complexes ClCu(Ph2PNR)3ZrCl [R = tBu (2a), iPr (2b)]
and (CO)3Mo(Ph2PNiPr)3ZrCl (3)
higher temperatures in solution states; this is favorable for
the possible usage of 1 as a tripodal metalloligand as de-
As described above, the fluxional behavior of 1 in the
variable temperature H NMR spectroscopic studies sug-
1
scribed later.[17]
In close analogy to these results, Eisen and coworkers gested that 1 was a fac-P3 tripodal-type metalloligand. It is
reported the preparation and structure determination of known that the fac-P3 tripodal-type is suitable as the ligand
Zr(Ph2PNPh)4, which also showed dynamic behavior in of complexes with tetrahedral or octahedral geometry.[3] As
variable temperature NMR studies.[13] The four phosphi- a representative model of a tetrahedral complex, CuI com-
noamide ligands are bonded to the zirconium atom by Zr– plexes were prepared. Treatment of 1 with 1 equiv. of CuCl
N and ZrǟP bonds in the solid state. In solution, the coor- in CH3CN at room temperature afforded Zr–Cu heterodi-
dination of the phosphorus atoms is reversible. Although metallic complexes, ClCu(Ph2PNR)3ZrCl [R = tBu (2a), iPr
the synthetic method of Eisen’s complex is similar to that (2b)], as yellow crystals in 78 and 82% yield, respectively
of 1, neither Zr(Ph2PNtBu)4 nor Zr(Ph2PNiPr)4 was de- (Scheme 3). The products were unequivocally characterized
tected in our experiments. As a representative experiment, by NMR spectroscopy, in which no dynamic behavior was
ZrCl4 was treated with 4 equiv. of Li(Ph2PNiPr) in THF at visible in the temperature range from –90 to 30 °C. In a
1
1
room temperature overnight; H and 31P NMR spectra of typical example, the H NMR spectrum of 2a gave signals
the crude product showed the formation of 1b as a single due to the ortho-, meta-, and para protons of six magneti-
product; Zr(Ph2PNiPr)4 was not detected. This is presum- cally equivalent Ph groups in an integral ratio of 12:12:6,
ably attributable to the steric bulkiness of the tBu and iPr and one singlet at δ = 1.50 ppm, which is assignable to the
groups, which prevents the reaction of 1 to the correspond- methyl protons of the tBu group. In the 13C NMR spec-
Scheme 3.
2900
www.eurjic.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2007, 2897–2908

(PPh2NR)3ZrCl Complexes as a Tripodal-Type Metalloligand
FULL PAPER
trum, four signals at δ = 127.9 (meta-Ph), 129.5 (para-Ph),
The asymmetric unit of 2b contains two crystallographi-
132.8 (ortho-Ph), and 134.5 (ipso-Ph) ppm are observed, cally independent but chemically equivalent molecules.
which are assignable to the six magnetically equivalent Bond lengths and angles of the second molecule are given
phenyl groups, along with two signals at δ = 34.8 (CMe3) in brackets in Table 2. The molecular structures show that
and 61.8 (CMe3) ppm due to the three magnetically equiva- two metal centers are linked by three phosphinoamide li-
lent tBu groups. This clearly demonstrates the C3v sym- gands. The zirconium center adopts a trigonal bipyramidal
metry of the molecule. A singlet appeared at δ = –30.6 ppm coordination geometry with the three nitrogen atoms in the
in the 31P NMR spectrum of 2a; peak broadening due to basal position and the Cl(1)–Zr–Cu as the apical axis. The
the interaction with the quadrupolar copper center is good sum of the equatorial angles around the zirconium atom is
evidence of the coordination of three phosphorus moieties ca. 357 °. Alternatively, the geometry around the copper is
to the Cu atom. Spectral data of 2b is analogous to those distorted tetrahedrally because of the effective stabilization
of 2a in the points that three Ph2PNiPr ligands are magneti- of tetrahedral geometry given by tripodal-type P3 coordina-
cally equivalent (the integral ratio of methyl and methine tion. The Cl(1)–Zr–Cu–Cl(2) axis is arranged linearly, the
protons of iPr and three Ph protons is 18:3:12:12:6) and bent angles of Cl(1) and Cl(2) from the Zr–Cu axis are
broadening of the 31P NMR spectroscopic resonance (δ = 0.77(2)° and 3.12(4)°, respectively, in 2a. The N–P bond
–11.2 ppm) is ascribed to the coordination of the phospho- lengths of 2a,b are approximately 0.1 Å shorter than the
rus moiety to the copper center. The C3v symmetric struc- normal N–P single bonds, but comparable to those of start-
ture deduced from these spectroscopic features is confirmed ing zirconium complexes 1a,b, which reflect the partial N–
by X-ray diffraction analysis of 2a,b of which the ORTEP P double bond character. This was supported by the sum
drawings are shown in Figure 3, and selected bond lengths of the three angles at the N atom (359.3–359.9°), which is
and angles are summarized in Table 2.
close to the theoretical value of 360° for the planar sp2-
hybridized nitrogen atom. The shorter bond lengths of N–
P were also observed in our Ti–Pt and Ti–Ru phosphi-
noamide complexes,[10] Zr(Ph2PNPh)4,[13] and related com-
plexes.[14] The Cu–P [2.3126(12)–2.3626(11) Å] bond
lengths are slightly longer than those of (triphos)CuCl
(2.282–2.299 Å),[4e] but comparable to related copper phos-
phane complexes such as (PPh3)3CuCl (2.348 Å).[19]
Reaction of 1b with an octahedrally arranged molybde-
num precursor led to the formation of Zr–Mo heterodimet-
allic complex 3 (Scheme 3). In 3, ligands around the molyb-
denum center are arranged octahedrally, to which the zirco-
nium phosphinoamide moieties are coordinated in a facial
coordination mode. Treatment of 1b with Mo(CO)3-
(CH3CN)3 in CH2Cl2 at room temperature resulted in the
replacement of the CH3CN ligands by three phosphorus
atoms of 1b to give a Zr–Mo heterodimetallic complex,
(CO)3Mo(Ph2PNiPr)3ZrCl (3), in 70% yield as pale yellow
1
crystals. The H NMR spectrum of 3 showed signals for
three magnetically equivalent Ph2PNiPr moieties, which in-
dicates the C3v symmetric structure of 3 in solution. No
dynamic behavior was observed in variable temperature
NMR spectroscopic experiments. The 31P NMR spectrum
of 3 showed a singlet at δ = 36.3 ppm, which is low-field
shifted by ca. 20 ppm relative to (triphos)Mo(CO)3 (δ
=16.8 ppm). The IR spectrum of 3 shows characteristic
CϵO stretching bands at 1987 and 1913 cm–1, which is 50–
70 cm–1 shifted to higher frequencies relative to those of
(triphos)Mo(CO)3 (1937, 1844 cm–1).[4b] The back-donation
from molybdenum to the π* orbital of CO in 3 is apparently
weaker than that in (triphos)Mo(CO)3. There are two pos-
sible explanations for this: one is direct through-space inter-
action between Lewis acidic Zr and Mo atoms; the other is
through-bond interaction by way of the Zr–N–P–Mo
bond.[20]
The molecular structure determined by X-ray diffraction
analysis is depicted in Figure 4, and representative bond
lengths and angles are summarized in Table 2.The two
Figure 3. Molecular structures of ClCu(Ph2PNtBu)3ZrCl (2a) (top)
and ClCu(Ph2PNiPr)3ZrCl (2b) (bottom) with 50% probability el-
lipsoids. Hydrogen atoms are omitted for clarity.
Eur. J. Inorg. Chem. 2007, 2897–2908
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
2901

T. Sue, Y. Sunada, H. Nagashima
FULL PAPER
Table 2. Representative bond lengths and angles for dimetallic complexes.
Bond lengths [Å]
Bond angles [°]
2a
2b[a]
3
4
2a
2b[a]
3
4
Zr–M
2.6069(7)
2.6854(6)
[2.6790(6)]
2.9741(5)
2.964(4)
Cl(1)–Zr–M
Zr–M–Cl(2)
N(1)–Zr–N(2)
N(1)–Zr–N(3)
N(2)–Zr–N(3)
P(1)–M–P(2)
P(1)–M–P(3)
P(2)–M–P(3)
Zr–N(1)–P(1)
Zr–N(2)–P(2)
Zr–N(3)–P(3)
179.23(2)
178.93(3)
[176.01(3)]
174.99(3)
[174.43(3)]
180
163.01(4)
Zr–Cl(1)
Zr–Cl(2)
Zr–Cl(3)
Zr–N(1)
Zr–N(2)
Zr–N(3)
M–P(1)
2.4314(11) 2.4026(10) 2.4427(12) 2.390(2)
176.88(4)
[2.4048(13)]
2.506(2)
118.44(14) 116.79(13)
[116.00(13)]
120.34(14) 118.90(14)
[118.76(11)]
120.29(12) 121.04(14)
[121.65(12)]
85.94(3)
84.92(2)
83.22(3)
93.1(2)
95.0(2)
95.3(2)
119.66(9) 99.8(2)
119.72(10)
2.4936(14)
2.149(2)
2.156(4)
2.157(3)
2.100(3)
[2.085(3)]
2.108(4)
[2.114(3)]
2.107(3)
[2.108(2)]
2.114(2)
2.115(2)
2.113(2)
2.140(3)
2.181(4)
119.68(8)
83.48(3)
[82.94(3)]
81.59(2)
[82.50(3)]
80.85(3)
[80.51(2)]
98.5(2)
[99.0(2)]
99.8(2)
[99.4(2)]
99.6(2)
100.16(2) 101.03(4)
100.20(2) 106.94(4)
100.17(2) 145.57(12)
99.08(8) 102.2(2)
99.08(8) 105.0(2)
99.08(8)
2.3626(11) 2.3146(10) 2.5787(5)
[2.3126(12)]
2.3356(12) 2.3306(12) 2.5796(8)
[2.3265(9)]
2.3408(12) 2.3458(11) 2.5783(7)
[2.3576(10)]
1.665(3)
1.649(3)
1.659(3)
2.2439(11)
2.3270(11)
2.2548(11)
1.641(4)
M–P(2)
M–P(3)
N(1)–P(1)
N(2)–P(2)
N(3)–P(3)
1.666(3)
[1.667(3)]
1.671(3)
[1.674(3)]
1.668(3)
[1.668(3)]
1.671(2)
1.671(2)
1.671(2)
[99.0(2)]
1.641(4)
1.615(4)
[a] The asymmetric unit of 2b consists of two crystallographically independent but chemically equivalent molecules. Bond lengths and
angles of the second molecule are given in brackets.
metal centers are connected by three phosphinoamide brid- Mo(CO)3 [1.155(3) Å].[4b] The Mo–Zr and Mo–P distances
ges and three phosphorus atoms are coordinated to the mo- suggest that the above-described shift of the ν(CϵO) band
˜
lybdenum center in a fac-P3 fashion. The Mo–P bond to higher frequencies can be better explained by through-
lengths of 3 [2.5783(7)–2.5796(8) Å] are slightly longer rela- bond interactions.
tive to those found in previously reported (triphos)-
Mo(CO)3 [2.516(1) Å],[4b] whereas C–O bond lengths of
2.3 Construction of Zr–Pt Heterodimetallic Complexes with
1.140(3) Å are comparable to those of (triphos)-
the Use of PtII Precursors that Do Not Provide Suitable
Coordination Geometries for fac-P3 Ligands
In the above-described two types of heterodimetallic
complexes, the zirconium phosphinoamides act as the fac-
P3 tripodal-type ligand. The tetrahedral geometry of the
CuI and octahedral structure of the Mo0 are suitable for the
fac-P3 tripodal-type coordination. One question concerns
the fate of the zirconium phosphinoamides in the reaction
with the second metal complex fragment that does not have
the appropriate coordination geometry for the fac-P3 tripo-
dal-type coordination. To resolve this question, we chose a
square-planar arranged PtII precursor as the reactant in the
reaction with 1. Treatment of 1b with (COD)PtCl2 in
CH2Cl2 at room temperature for 1 h afforded a novel Zr–
Pt heterodimetallic complex, (κ2-Ph2PNiPr)Pt(Ph2PNiPr)2-
ZrCl3(4) in 75% yield (Scheme 4). The 31P NMR spectrum
of 4 showed the presence of three different phosphorus
atoms with a satellite signal due to the coupling with 195Pt,
δ = 12.9 (JP,P = 6.7, 286 Hz, JPt,P = 3153 Hz), 3.4 (JP,P
=
6.7, 13.9 Hz, JPt,P = 3403 Hz), and –32.5 ppm (JP,P = 13.9,
286 Hz, JPt,P = 1578 Hz). In contrast to the C3v symmetric
structures of the Zr–Cu and Zr–Mo complexes giving a sin-
Figure 4. Molecular structure of (CO)3Mo(Ph2PNiPr)3ZrCl (3)
with 50% probability ellipsoids. Hydrogen atoms are omitted for
clarity.
2902
www.eurjic.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2007, 2897–2908

(PPh2NR)3ZrCl Complexes as a Tripodal-Type Metalloligand
FULL PAPER
gle 31P NMR spectroscopic resonance, the 31P NMR spec- atom of the κ2-coordinated phosphinoamide and the other
trum of 4 clearly demonstrates that three Ph2PNiPr ligands two are trans to each other. The value of each coupling
in 4 are magnetically inequivalent and the complex has no constant is consistent with this interpretation.
symmetry. The nonsymmetric structure of 4 is also sup-
ported by the 1H NMR spectrum of 4; in a typical example,
there exists three doublets at δ = 0.71, 1.15, and 1.45 ppm,
which are due to the methyl protons of the iPr group. It is
noteworthy that one of them (δ =0.71 ppm) is significantly
shifted to a higher field (vide infra).
Scheme 4.
Figure 5. Molecular structure of (κ2-Ph2PNiPr)Pt(Ph2PNiPr)2-
The molecular structure of 4 was established by X-ray
ZrCl3(4) with 50% probability ellipsoids. Hydrogen atoms are
study. Figure 5 shows the ORTEP view of 4 and selected
bond lengths and angles of 4 are listed in Table 2. Two
metal centers are linked by two phosphinoamide bridges in
a boat form similar to Ti–Pt and Ti–Ru phosphinoamide
complexes. Ligand arrangement around the zirconium atom
is pseudooctahedral with two nitrogen atoms and two chlo-
rine atoms at the basal plane and Cl(1) and Pt at the apical
position. Geometry around the platinum atom is distorted
square planar with the Pt atom out of the plane defined by
P(1), P(2), and the center of N(3)–P(3) (ca. 0.077 Å). A
similar distorted structure was also observed in
YYЈPt(Ph2PNtBu)2TiCl2 complexes in which the Pt atom
is out of the plane by ca. 0.181–0.328 Å. The N(3)–P(3)
bond length of 1.615(4) Å is ca. 0.03 Å shorter than those
of the bridging phosphinoamide ligands, which indicates
that there is a slightly increased double bond character of
the N(3)–P(3) bond. The ligand arrangement indicates that
4 is formally considered to have a zwitterionic form with
ZrIV– and PtII+ centers. To the best of our knowledge, this
is the first example of a zwitterionic Zr–Pt heterodimetallic
complex. The molecular structure provided clues for under-
standing the NMR spectroscopic data, which included
three magnetically inequivalent methyl and methine signals
omitted for clarity.
This unusual ligand rearrangement in the reaction of 1b
with a square planar platinum complex fragment provided
some mechanistic insight. The reaction is likely to occur
intramolecularly and the ligand migration is internuclear.
The dimetallic structure of 4, in which the Zr and Pt exist
close enough to exchange their individual ligands, is the key
to accomplish the rearrangement. The possible mechanism
is illustrated in Scheme 5. Three phosphorus atoms of zirco-
1
of the iPr group that were observed in the H NMR spec-
trum and three 31P NMR spectroscopic resonances with
different chemical shifts and 31P–195Pt coupling constants.
The nonsymmetric structure of 4 is reasonable in giving
three doublets at δ = 0.71, 1.15, and 1.45 ppm that are due
to the methyl protons of the iPr group. An unusual upfield
shift of one of the three doublets due to the methyl group
in the iPr moieties can be ascribed to the methyl proton of
the iPr of κ2-coordinated phosphinoamide moiety, which is
effectively shielded by one of the phenyl rings of the PPh2
group of the bridging phosphinoamide ligand. The signal
at δ = 3.4 ppm in the 31P NMR spectrum is trans to the N Scheme 5.
Eur. J. Inorg. Chem. 2007, 2897–2908
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
2903

T. Sue, Y. Sunada, H. Nagashima
FULL PAPER
nium phosphinoamide 1b reversibly dissociate to their un- Zr–Pt distances of 2.9741(5) and 2.964(4) Å in complexes 3
coordinated forms in solution and two of them react with and 4, respectively, are nearly identical to the sum of the
a square-planar PtII species to form A. Coordination of the van der Waals radii of Zr–Mo and Zr–Pt (ca. 3.05 Å)[21] (98
remaining phosphorus atom to the PtII center results in in- and 97% of the sum of the radii, respectively). The Rh–B
ternuclear migration of a chlorine atom from Pt to Zr to interaction is proposed in Bourissou’s complex, {[(o-iPr2P-
give zwitterionic intermediate B. The tricyclic structure of C6H3)2B(Ph)]RhCl}2, from chemical shifts of the 11B NMR
B generates significant ring strain that triggers the cleavage spectroscopic resonance and the Rh–B distance found from
and recombination of the Zr–N and Pt–Cl bonds to form X-ray crystallography. This was supported by DFT calcula-
4. These results clearly demonstrate that zirconium phos- tions {BP86/[CEP-31G(Rh), 6-31G*(other elements)]}, in
phinoamides 1a,b behave as fac-P3 tripodal-type metalloli- which bonding interaction was observed at the HOMO of
gands in the reaction with transition metals that are able to the complex.[8] However, our preliminary theoretical calcu-
attain a tetrahedral or octahedral geometry while coordi- lations (single-point calculation of the structure of 2a/HF/
nated to the metal center with a square planar structure. LanL2DZ) did not provide clear evidence for the bonding
This geometry is unfavorable for the coordination of the interaction at the HOMO level. Thus, further theoretical
tripodal-type ligand and as such an internuclear ligand ar- studies with a higher functional level, including those of 3,
rangement is induced to release the ring strain of the pri- 4, previously reported Ti–Pt heterodimetallic complexes,
marily formed intermediates.
and related complexes, should be performed to discuss
metal–metal interactions.
In summary, 1 is useful as a metalloligand for the forma-
tion of Zr–M heterodimetallic compounds, though the co-
ordination mode of 1 was dependent on the preferred coor-
3. Concluding Remarks
As described above, we attained success in the prepara- dination environment of the second transition–metal cen-
tion of tris(phosphinoamide)zirconium complexes, ters. Current efforts are directed towards the synthesis of
(Ph2PNR)3ZrCl [R = tBu (1a), iPr (1b)], which reacted with novel heterodimetallic complexes with several second tran-
a second metal complex fragment to form heterodimetallic sition-metal complex fragments having other coordination
complexes, ClCu(Ph2PNR)3ZrCl (2) and (CO)3Mo- geometries. Their systematic results, including structural
(Ph2PNiPr)3ZrCl (3). In the Zr-Cu and Zr-Mo complexes, properties and dynamic behaviors, are expected to open the
the coordination geometry around the late-transition-metal way to construct novel heterodimetallic complexes with
center is defined as pseudotetrahedral and octahedral, unique structures, electronic properties, and reactivities.
respectively, owing to the effective coordination of they
P3-moiety of 1 in the tripodal fashion. The reaction of 1b
with square planar arranged (COD)PtCl2 is unique and
Experimental Section
gives a zwitterionic heterodimetallic complex (κ2-Ph2-
General: Manipulation of air- and moisture-sensitive organometal-
lic compounds was carried out under a dry argon atmosphere by
using standard Schlenk tube techniques associated with a high-vac-
uum line, or with the nitrogen filled glove box. All solvents were
distilled from appropriate drying reagents prior to use (THF, hex-
PNiPr)Pt(Ph2PNiPr)2ZrCl3 (4); this reaction involves intra-
molecular migration of two chlorine atoms from Pt to Zr
as well as that of a iPrNPPh2 moiety from Zr to Pt. In
heterodimetallic complexes 2–4, the Lewis acidic zirconium
and the late-transition-metal atoms are closely located; this
realized the internuclear ligand arrangement seen in the for-
mation of 4. The driving force for the rearrangement can
be attributed to the fact that the hard chlorine atom pre-
ferred the hard zirconium center, whereas the soft phosphi-
noamide ligand preferred the soft platinum atom. Of inter-
est is the possible dative interaction of the late transition
metal to the zirconium atom. As reported earlier, we pro-
posed the TiǟPt interaction in the YYЈPt(Ph2PNtBu)2-
TiCl2 complexes from the crystal structures. There is an oc-
cupied dz2 orbital in the platinum atom, and this possibly
interacted with the Lewis acidic titanium center. The TiǟPt
bond lengths decrease in the order Cl2Pt(Ph2PNtBu)2TiCl2
[2.8358(6) Å] Ͼ MeClPt(Ph2PNtBu)2TiCl2 [2.7547(10) Å]
Ͼ Me2Pt(Ph2PNtBu)2TiCl2 [2.6860(11) Å]; this is explained
by assuming that the higher electron dense platinum com-
plex moiety can make a stronger dative bond interaction.
The molecular structure of 2 revealed that the Zr–Cu bond
length of 2a [2.6069(7) Å] is slightly shorter than the sum
of the van der Waals radii of Zr and Cu (ca. 2.74 Å)[21] (ca.
1
ane; Ph2CO/Na, CH3CN, CH2Cl2; CaH2). H, 13C, and 31P NMR
spectra were recorded with a JEOL Lambda 600 or a Lambda 400
spectrometer at ambient temperature unless otherwise noted. 1H,
13C, and 31P NMR chemical shifts (δ values) were given in ppm
relative to the solvent signal (1H, 13C) or standard resonances (31P;
external 85% H3PO4). IR spectra were recorded with a JASCO FT/
IR-550 spectrometer. Melting points were measured with a Yanaco
SMP3 micro melting point apparatus. Elemental analyses were per-
formed by the Elemental Analysis Center, Faculty of Science, Kyu-
shu University, or with a Perkin–Elmer 2400/CHN analyzer. Start-
ing materials, phosphinoamide,[12] [Mo(CO)3(CH3CN)3],[22] and
[23]
(COD)PtCl2 were synthesized by the method reported in the lit-
erature.
(Ph2PNR)3ZrCl [R = tBu (1a), iPr (1b)]: In a 100-mL Schlenk tube,
Ph2PNHtBu (2589 mg, 10.06 mmol) was dissolved in THF
(30 mL). A hexane solution of nBuLi (1.56 , 6.5 mL, 10.14 mmol)
was added to the solution at –78 °C, and the solution was slowly
warmed to room temperature. After stirring at room temperature
for 3 h, the mixture was cooled to –78 °C and a THF (15 mL) solu-
tion of ZrCl4 (800 mg, 3.43 mmol) was added dropwise. The
slightly yellow solution turned off white in color and a white solid
95% of the sum of the radii). In contrast, the Zr–Mo and precipitated. After 12 h, all of the volatiles were removed under
2904 Eur. J. Inorg. Chem. 2007, 2897–2908
www.eurjic.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

(PPh2NR)3ZrCl Complexes as a Tripodal-Type Metalloligand
FULL PAPER
vacuum, and the formed crude product was dissolved in CH2Cl2
(20 mL). The insoluble materials were filtered off through a G3
filter, and the resulting filtrate was concentrated to a volume of
5 mL. Hexane (20 mL) was added to this supersaturated CH2Cl2
solution. The desired product was obtained as colorless crystals in
86% yield (2580 mg, 2.88 mmol). In a similar fashion, 1b was ob-
pellet): ν = 1987 (CϵO), 1913 (CϵO) cm–1.C H Cl MoN -
O3P3Zr (1033.49): calcd. C 55.78, H 4.97, N 4.07; found C 55.48,
H 4.91, N 4.00.
˜
48 51
2
3
(κ2-Ph2PNiPr)Pt(Ph2PNiPr)2ZrCl3 (4): A 20-mL Schlenk tube was
charged with 1b (50 mg, 0.059 mmol) and (COD)PtCl2 (21 mg,
0.056 mmol), and the atmosphere was replaced by argon. The mix-
ture was dissolved in CH2Cl2 (5 mL), and the solution was stirred
at room temperature for 1 h. The color of the reaction mixture
changed from yellow to red. The solution was cooled to –30 °C to
form pale yellow crystals of 4 in 75% yield (47 mg, 0.042 mmol).
1
tained in 84% yield as colorless crystals. 1a: M.p. 108 °C (dec). H
NMR (600 MHz, CD2Cl2): δ = 1.47 (s, 27 H, tBu), 6.99–7.17 (br.,
30 H, Ph) ppm. 13C NMR (150 MHz, CD2Cl2): δ = 34.1 (CMe3),
60.0 (m, CMe3), 127.5 (m, meta-Ph), 128.8 (para-Ph), 133.4 (m,
ortho-Ph), 136.75 (ipso-Ph) ppm. 31P{1H} NMR (243 MHz,
CD2Cl2): δ = –11.8 (s) ppm. C48H57ClN3P3Zr (895.59): calcd. C
64.37, H 6.42, N 4.69; found C 64.14, H 6.42, N 4.45. 1b: M.p.
1
M.p. 126–128 °C (dec). H NMR (600 MHz, CD2Cl2): δ = 0.71 [d,
JH,H = 6.8 Hz, 6 H, (CH3)2CH], 1.15 [d, JH,H = 6.0 Hz, 6 H,
(CH3)2CH], 1.45 [d, JH,H = 6.0 Hz, 6 H, (CH3)2CH], 2.97 [m, 1 H,
(CH3)2CH], 3.47 [m, 1 H, (CH3)2CH], 3.62 [m, 1 H, (CH3)2CH],
6.75–6.80 (m, 4 H, Ph), 6.90–6.97 (m, 4 H, Ph), 7.07–7.12 (m, 2 H,
Ph), 7.13–7.17 (m, 4 H, Ph), 7.22–7.27 (m, 2 H, Ph), 7.37–7.42
(m, 4 H, Ph), 7.51–7.60 (m, 10 H, Ph) ppm. 13C NMR (150 MHz,
CD2Cl2): δ = 25.2 [(CH3)2CH], 25.8 [(CH3)2CH], 27.0 [(CH3)2CH],
49.2 [(CH3)2CH], 54.2 [(CH3)2CH], 54.7 [(CH3)2CH], 127.5 (Ph),
128.2 (Ph), 129.0 (Ph), 130.5 (Ph), 130.7 (Ph), 133.0 (Ph), 133.1
(Ph), 133.4 (Ph), 134.1 (Ph) ppm. 31P{1H} NMR (243 MHz,
CD2Cl2): δ = 12.9 (dd, JP,P = 6.7, 286 Hz, JPt,P = 3153 Hz), 3.4 (dd,
JP,P = 6.7, 13.9 Hz, JPt,P = 3403 Hz), –32.5 (dd, JP,P = 13.9, 286 Hz,
JPt,P = 1578 Hz) ppm.C45H51Cl3N3P3PtZr (1119.51): calcd. C
48.28, H 4.59, N 3.75; found C 48.14, H 4.66, N 3.68.
1
107–108 °C (dec). H NMR (600 MHz, CD2Cl2): δ = 1.20 [d, JH,H
= 6.6 Hz, 18 H, (CH3)2CH], 3.89 [sept, JH,H = 6.6 Hz, 3 H, (CH3)2-
CH], 7.12–7.17 (m, 12 H, meta-Ph), 7.22–7.26 (m, 18 H, ortho and
para-Ph) ppm. 13C NMR (150 MHz, CD2Cl2): δ = 26.8 [(CH3)2-
CH], 53.2 [(CH3)2CH], 128.1 (meta-Ph), 129.3 (para-Ph), 133.4 (m,
ortho-Ph), 136.9 (m, ipso-Ph) ppm. 31P{1H} NMR (243 MHz,
CD2Cl2): δ = –12.0 (s) ppm.C45H51ClN3P3Zr (853.51): calcd. C
63.32, H 6.02, N 4.92; found C 63.34, H 6.52, N 5.40.[24]
ClCu(Ph2PNR)3ZrCl [R = tBu (2a), iPr (2b)]: A 20-mL Schlenk
tube was charged with 1a (44 mg, 0.050 mmol) and CuCl (5 mg,
0.051 mmol), and the atmosphere was replaced by argon. The mix-
ture was dissolved in CH3CN (5 mL). The color of the solution
changed immediately from colorless to yellow, from which a yellow
precipitate was separated. After removal of the solvent in vacuo,
the solid was washed with hexane (3ϫ2 mL). The crude product
was dissolved in warm CH2Cl2, and the insoluble materials (unre-
acted CuCl) were filtered off. The filtrate was cooled to –30 °C to
form yellow microcrystals of 2a in 78% yield (38 mg, 0.038 mmol).
In a similar fashion, 1b was obtained in 82% yield as yellow micro-
crystals. 2a: M.p. 116–117 °C (dec). 1H NMR (600 MHz, CD2Cl2):
δ = 1.50 (s, 27 H, tBu), 6.96 (t, J = 7.4 Hz, 12 H, meta-Ph), 7.15
(t, J = 7.4 Hz, 6 H, para-Ph), 7.42 (br., 12 H, ortho-Ph) ppm. 13C
NMR (150 MHz, CD2Cl2): δ = 34.8 (CMe3), 61.8 (m, CMe3), 127.9
(meta-Ph), 129.5 (para-Ph), 132.8 (ortho-Ph), 134.5 (m, ipso-Ph)
ppm. 31P{1H} NMR (243 MHz, CD2Cl2): δ = –30.6 (br) ppm.
C48H57Cl2CuN3P3Zr (994.59): calcd. C 57.97, H 5.78, N 4.22;
found C 57.51, H 5.94, N 4.32. 2b: M.p. 117–118 °C (dec). 1H
NMR (600 MHz, CD2Cl2): δ = 1.38 [d, JH,H = 6.6 Hz, 18 H,
(CH3)2-
Table 3. Crystallographic data for zirconium phosphinoamides.
1a
1b
Empirical formula
Formula weight
Crystal system
Lattice type
Space group
a [Å]
b [Å]
c [Å]
α [°]
C48H57ClN3P3Zr
895.59
C45H51ClN3P3Zr
853.51
trigonal
primitive
P3 (#143)
19.4576(14)
19.4576(14)
10.8143(10)
90
monoclinic
primitive
P21/n (#14)
11.293(2)
16.724(3)
22.575(4)
90
β [°]
γ [°]
90
120
3545.7(5)
3
1.258
1404.00
4.249
colorless, prism
93.930(2)
90
4253.6(12)
4
1.333
1776.00
Volume [Å3]
Z value
CH], 3.85 [m, 3 H, (CH3)2CH], 6.98 (t, JH,H = 7.7 Hz, 12 H, meta-
Ph), 7.19 (t, JH,H = 7.7 Hz, 6 H, para-Ph), 7.34 (br., 12 H, ortho-
Ph) ppm. 13C NMR (150 MHz, CD2Cl2): δ = 27.2 [(CH3)2CH],
53.2 [(CH3)2CH], 128.1 (meta-Ph), 130.0 (para-Ph), 133.7 (m, ortho-
Ph), 134.8 (m, ipso-Ph) ppm. 31P{1H} NMR (243 MHz, CD2Cl2):
δ = –11.2 (br) ppm. C45H51Cl2CuN3P3Zr (952.52): calcd. C 56.74,
H 5.40, N 4.41; found C 56.87, H 5.80, N 4.59.
Dcalcd. [gcm–3]
F(000)
µ(Mo-Kα) [cm–1]
Crystal color, habit
4.686
colorless, platelet
Crystal dimensions [mm] 0.15ϫ0.05ϫ0.02 0.25ϫ0.15ϫ0.08
No. observations
(all reflections)
No. variables
Reflection/parameter
ratio
10782
9605
506
21.31
529
18.16
(CO)3Mo(Ph2PNiPr)3ZrCl (3): A 20-mL Schlenk tube was charged
with 1b (428 mg, 0.501 mmol) and Mo(CO)3(CH3CN)3 (155 mg,
0.511 mmol), and the atmosphere was replaced by argon. The mix-
ture was dissolved in CH2Cl2 (5 mL), and the solution was stirred
at room temperature for 1 h. The color of the reaction mixture
changed from yellow to red. The solution was kept at –30 °C to
form pale yellow microcrystals of 3 in 70% yield (365 mg,
0.353 mmol). M.p. 227–228 °C (dec). 1H NMR (600 MHz,
CD2Cl2): δ = 1.11 [d, JH,H = 6.8 Hz, 18 H, (CH3)2CH], 3.67 [m, 3
H, (CH3)2CH], 7.25–7.28 (m, 12 H, meta-Ph), 7.36–7.41 (m, 18 H,
para and ortho-Ph) ppm. 13C NMR (150 MHz, CD2Cl2): δ = 26.0
[(CH3)2CH], 53.0 [(CH3)2CH], 128.9 (m, meta-Ph), 129.8 (para-Ph),
133.4 (m, ortho-Ph), 138.0 (m, ipso-Ph), 218.3 (m, CO) ppm.
R (all reflections)
R1 [IϾ2.00σ(I)][a]
wR2 (all reflections)[b]
GOF
Flack parameter
Max shift/error in final
cycle
0.0825
0.0757
0.1890
1.114
0.42(5)
0.000
0.0705
0.0466
0.1494
1.006
0.000
1.23
Maximum peak in final 4.30
diff. map [eÅ–3]
Minimum peak in final
–0.75
–0.76
diff. map [eÅ–3]
31P{1H} NMR (243 MHz, CD2Cl2): δ = 36.3 (s) ppm. IR (KBr [a] R1 = Σ|Fo| – |Fc|/Σ|Fo|. [b] wR2 = [Σ{w(Fo2 – Fc2)2}/Σ{w(Fo ) }]1/2
.
2 2
Eur. J. Inorg. Chem. 2007, 2897–2908
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
2905

T. Sue, Y. Sunada, H. Nagashima
FULL PAPER
Table 4. Crystallographic tables for the heterodimetallic complexes.
2a
2b
3
4
Empirical formula
C48H57Cl2N3P3CuZr C45H51ClN3P3CuZr·CH2Cl2 C48H57ClO3N3P3MoZr· C45H51Cl3N3P3PtZr
1/4C6H6
Formula weight
Crystal system
Lattice type
Space group
a [Å]
b [Å]
c [Å]
α [°]
β [°]
994.59
987.96
1033.49
trigonal
1119.51
monoclinic
primitive
P21/n (#14)
11.539(2)
23.527(4)
16.727(3)
90
92.090(2)
90
4538.1(13)
4
1.638
2224.00
36.137
yellow, chip
0.20ϫ0.09ϫ0.08
10368
triclinic
primitive
triclinic
primitive
R-centered
¯
¯
¯
P1 (#2)
P1 (#2)
R3 (#148)
10.968(2)
11.948(2)
19.980(4)
79.863(9)
84.577(9)
63.397(7)
2304.2(7)
2
1.433
1028.00
9.443
yellow, chip
0.20ϫ0.08ϫ0.05
10146
10.9996(12)
21.564(2)
21.801(2)
66.326(3)
83.226(4)
85.984(4)
4701.4(8)
4
1.396
2028.00
9.799
yellow, block
0.20ϫ0.18ϫ0.10
20685
14.516(3)
14.516(3)
39.098(7)
90
90
120
7135(2)
6
1.477
3240.00
6.839
yellow, block
0.20ϫ0.15ϫ0.05
3630
γ [°]
Volume [Å3]
Z value
Dcalcd. [gcm–3]
F(000)
µ(Mo-Kα) [cm–1]
Crystal color, habit
Crystal dimensions [mm]
No. observations (all reflections)
No. variables
580
1024
194
506
Reflection/parameter ratio
R (all reflections)[a]
R1 [IϾ2.00σ(I)][b]
wR2 (all reflections)
GOF
Max shift/error in final cycle
Maximum peak in final diff. map
[eÅ–3]
17.49
0.0765
0.0535
0.1675
1.005
0.000
1.62
20.20
0.0733
0.0572
0.1359
1.067
0.001
1.13
18.71
20.49
0.0515
0.0454
0.0855
1.223
0.002
3.73
0.0403
0.0379
0.0942
1.125
0.000
0.80
Minimum peak in final diff. map
–0.92
–1.58
–1.07
–1.43
[eÅ–3]
[a] R1 = Σ|Fo| – |Fc|/Σ|Fo|. [b] wR2 = [Σ{w(Fo – Fc ) }/Σ{w(Fo2)2}]1/2
.
2
2 2
X-ray Data Collection and Reduction: Single crystals of all com-
plexes were grown from CH2Cl2/pentane. X-ray crystallography
was performed with a Rigaku Saturn CCD area detector with
graphite monochromated Mo-Kα radiation (λ = 0.71070A). The
reflection data were collected at 123(2) K by using ω scan in the θ
obtained are reliable enough to discuss the atom connectivity, bond
lengths, and angles. CCDC-632038 (1a), -632039 (1b), -632040
(2a), -632041 (2b), -632042 (3), and -632043 (4) contain the supple-
mentary crystallographic data for this paper. These data can be
obtained free of charge from The Cambridge Crystallographic
range of 3.1 Յ θ Յ 27.5° (1a), 3.0 Յ θ Յ 27.5° (1b), 3.0 Յ θ Յ Data Centre via www.ccdc.cam.au.uk/data_request/cif.
27.5° (2a), 3.1 Յ θ Յ 27.5° (2b), 3.1 Յ θ Յ 27.5° (3), 3.1 Յ θ
Supporting Information (see footnote on the first page of this arti-
Յ 27.5° (4). The data obtained were processed with Crystal-Clear
1
cle): Variable temperature NMR spectroscopic data (1a,b) and H,
(Rigaku) on a Pentium computer, and were corrected for Lorentz
13C, and 31P NMR spectroscopic data (1a,b, 2a,b, 3, 4).
and polarization effects. The structures was solved by direct meth-
ods (SIR 97[25] for 1a,b, and 2b, SIR 92[26] for 2a, 3, and 4) and
expanded by using Fourier techniques.[27] The non-hydrogen atoms
were refined anisotropically, and the hydrogen atoms by using the
Acknowledgments
riding model. The final cycle of full-matrix least-squares refinement
on F2 was based on 10782 observed reflections and 563 variable
This work was supported by the Ministry of Education, Culture,
Sports, Science and Technology, Japan by Grant-in-Aid for Scien-
tific Research on Priority Areas (No. 15036253 and 16033246),
“Reaction Control of Dynamic Complexes”.
parameters for 1a, 9605 observed reflections and 529 variable pa-
rameters for 1b, 10146 observed reflections and 580 variable param-
eters for 2a, 20685 observed reflections and 1113 variable param-
eters for 2b, 10383 observed reflections and 599 variable parameters
for 3, and on 10368 observed reflections and 556 variable param-
eters for 4. Neutral atom scattering factors were taken from Cromer
and Waber.[28] All calculations were performed with the Crystal-
Structure[29,30] crystallographic software package. Details of the fi-
nal refinement are summarized in Tables 3 and 4, and the number-
ing scheme employed is shown in Figures 2, 3, 4, and 5 which was
drawn with ORTEP at 50% probability ellipsoids. Unsatisfactory
quality of the crystal in the crystallographic analysis of 1a provided
two large residual densities near the Zr atom. However, the data
[1] a) P. Dierkes, P. W. N. M. van Leeuwen, J. Chem. Soc., Dalton
Trans. 1999, 1519–1529; b) P. W. N. M. van Leeuwen, P. C. J.
Kamer, J. N. H. Reek, P. Dierkes, Chem. Rev. 2000, 100, 2741–
2770; c) Z. Freixa, P. W. N. M. van Leeuwen, Dalton Trans.
2003, 1890–1901; d) T. Appleby, J. D. Woollins, Coord. Chem.
Rev. 2002, 235, 121–140; e) D. M. A. Minahan, W. E. Hill,
C. A. McAuliffe, Coord. Chem. Rev. 1984, 55, 31–54; f) R.
2906
www.eurjic.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2007, 2897–2908

(PPh2NR)3ZrCl Complexes as a Tripodal-Type Metalloligand
FULL PAPER
Noyori, Angew. Chem. Int. Ed. 2002, 41, 2008–2022; g) R.
Noyori, T. Ohkuma, Angew. Chem. Int. Ed. 2001, 40, 40–73.
C. A. Tolman, Chem. Rev. 1977, 77, 313–348.
H. A. Mayer, W. C. Kaska, Chem. Rev. 1994, 94, 1239–
1272.
1994; b) Y. Sunada, T. Sue, T. Matsumoto, H. Nagashima, J.
Organomet. Chem. 2006, 691, 3176–3182.
a) G. S. Ferguson, P. T. Wolczanski, L. Párkányi, M. C. Zonne-
vylle, Organometallics 1988, 7, 1967–1979; b) G. S. Ferguson,
P. T. Wolczanski, Organometallics 1985, 4, 1601–1605; c) O.
Kühl, S. Joachim, J. Sieler, E. Hey-Hawkins, Polyhedron 2001,
20, 2171–2177.
H. H. Sisler, N. L. Smith, J. Org. Chem. 1961, 26, 611–613.
O. Kuhl, T. Koch, F. B. Somoza Jr, P. C. Junk, E. Hey-Hawk-
ins, D. Plat, M. S. Eisen, J. Organomet. Chem. 2000, 604, 116–
125.
[11]
[2]
[3]
[4]
a) S. Gambarotta, S. Strologo, C. Floriani, A. Chiesi-Villa, C.
Guastini, Organometallics 1984, 3, 1444–1445; b) O. Walter, T.
Klein, G. Huttner, L. Zsolnai, J. Organomet. Chem. 1993, 458,
63–81; c) A. Muth, O. Walter, G. Huttner, A. Asam, L. Zsolnai,
Ch. Emmerrich, J. Organomet. Chem. 1994, 468, 149–163; d)
L. Sacconi, S. Midollini, J. Chem. Soc., Dalton Trans. 1972,
1213–1216; e) D. J. Fife, H. J. Mueh, C. F. Campana, Acta
Crystallogr., Sect. C: Cryst. Struct. Commun. 1993, 49, 1714–
1716; f) A. F. Kowalski, M. T. Ashby, J. Am. Chem. Soc. 1995,
117, 12639–12640.
It is well-known that the reaction of a Lewis acidic boron cen-
ter with an appropriate anion gives the corresponding borate.
Peters et al. described the synthesis of a new type of bidentate
and fac-P3 tripodal phosphanes containing a borate anion, bis
[12]
[13]
[14]
a) M. T. Gamer, P. W. Roesky, Inorg. Chem. 2004, 43, 4903–
4906; b) P. W. Roesky, Inorg. Chem. 2006, 45, 798–802; c) N.
Poetschke, M. Nieger, M. A. Khan, E. Niecke, M. T. Ashby,
Inorg. Chem. 1997, 36, 4087–4093; d) P. W. Roesky, Heteroat.
Chem. 2002, 13, 514–520; e) M. T. Gamer, P. W. Roesky, Inorg.
Chem. 2005, 44, 5963–5965; f) C. Qi, S. Zhang, J. Sun, J. Or-
ganomet. Chem. 2005, 690, 2941–2946; g) T. K. Panda, M. T.
Gamer, P. W. Roesky, Inorg. Chem. 2006, 45, 910–916; h) P. W.
Roesky, M. T. Gamer, N. Marinos, Chem. Eur. J. 2004, 10,
3537–3542; i) P. W. Roesky, M. T. Gamer, M. Puchner, A.
Greiner, Chem. Eur. J. 2002, 8, 5265–5271; j) D. Fenske, B.
Maczek, K. Maczek, Z. Anorg. Allg. Chem. 1997, 623, 1113–
1120.
a) C. Airoldi, D. C. Bradley, H. Chudzynska, M. B. Hurst-
house, K. M. A. Malik, P. R. Raithby, J. Chem. Soc., Dalton
Trans. 1980, 2010–2015; b) X. Yu, S. Bi, I. A. Guzei, Z. Liu,
Z.-L. Xue, Inorg. Chem. 2004, 43, 7111–7119; c) P. Renner, C.
Galka, H. Memmler, U. Kauper, L. H. Gade, J. Organomet.
Chem. 1999, 591, 71–77; d) see also ref.[15a]
a) M. D. Fryzuk, H. D. Williams, S. J. Rettig, Inorg. Chem.
1983, 22, 863–868; b) F. A. Cotton, P. A. Kibala, Acta Crys-
tallogr., Sect. C: Cryst. Struct. Commun. 1991, 47, 270–272.
One of the reviewers suggested that the dynamic behavior ob-
served here could also be interpreted as the twisting of the
phosphinoamide ligand without dissociation of the phospho-
rus moiety from the Zr center. Although this possibility could
not be ruled out, we consider the coordination/dissociation
process of the phosphinoamide ligands to be reasonable for the
following reasons. First, a similar process was also proposed in
the solution dynamics of Zr(Ph2PNPh)4 and (Ph2PNR)2TiCl2.
Second, the facile reaction with CuCl and other complex frag-
ments suggests the existence of uncoordinated phosphorus
atoms in solution.
[5]
or tris[(diphenylphosphanyl)methyl]borate such as [PhnB-
–
(CH2PPh2)4–n
]
(n = 1, 2). They successfully ligate to various
transition metals such as Fe, Co, Ni, Pd and Pt complex frag-
ments. Seea) S. J. Thomas, J. C. Peters, J. Am. Chem. Soc. 2001,
123, 5100–5101; b) T. A. Betley, J. C. Peters, Angew. Chem. Int.
Ed. 2003, 42, 2385–2389; c) J. C. Thomas, J. C. Peters, Inorg.
Chem. 2003, 42, 5055–5073; d) T. A. Betley, J. C. Peters, Inorg.
Chem. 2003, 42, 5074–5084; e) J. C. Thomas, J. C. Peters, J.
Am. Chem. Soc. 2003, 125, 8870–8888; f) C. C. Lu, J. C. Peters,
J. Am. Chem. Soc. 2004, 126, 15818–15832; g) C. C. Lu, J. C.
Peters, J. Am. Chem. Soc. 2002, 124, 5272–5273; h) see ref.[9b]; i)
J. C. Peters, J. C. Thomas, J. A. Dupont, C. G. Riordan, Inorg.
Synth. 2004, 34, p., 8; j) S. D. Brown, M. P. Mehn, J. C. Peters,
J. Am. Chem. Soc. 2005, 127, 13146–13147; k) C. C. Lu, C. T.
Saouma, M. W. Day, J. C. Peters, J. Am. Chem. Soc., DOI:
10.1021/ja06043 58; l) S. D. Brown, J. C. Peters, J. Am. Chem.
Soc. 2005, 127, 1913–1923; m) C. C. Lu, J. C. V. Peters, Inorg.
Chem. 2006, 45, 8597–8607; n) J. Li, P. I. Djurovich, B. D. Al-
leyne, M. Yousufuddin, N. N. Ho, J. C. Peters, R. Bau, M. E.
Thompson, Inorg. Chem. 2005, 44, 1713–1727; o) D. M. Jenk-
ins, J. C. Peters, J. Am. Chem. Soc. 2005, 127, 7148–7165; p)
J. C. Peters, J. D. Feldman, T. D. Tilley, J. Am. Chem. Soc.
1999, 121, 9871–9872; q) C. M. Thomas, N. P. Mankad, J. C.
Peters, J. Am. Chem. Soc. 2006, 128, 4956–4957; r) C. M.
Thomas, J. C. Peters, Organometallics 2005, 24, 5858–5867; s)
J. Li, P. I. Djurovich, B. D. Alleyne, M. Yousufuddin, N. N. Ho,
J. C. Thomas, J. C. Peters, R. Bau, M. E. Thompson, Inorg.
Chem. 2005, 44, 1713–1727; t) M. T. Whited, E. Rivard, J. C.
Peters, Chem. Commun. 2006, 1613–1615; u) I. R. Shapiro,
D. M. Jenkins, J. C. Thomas, M. W. Day, J. C. Peters, Chem.
Commun. 2001, 2152–2153; v) A. A. Barney, A. F. Heyduk,
D. G. Nocera, Chem. Commun. 1999, 2379–2380.
a) J. A. Labinger, J. N. Bonfiglio, D. L. Grimmett, S. T. Masuo,
E. Shearin, J. S. Miller, Organometallics 1983, 2, 733–740; b)
H. H. Karsch, A. Appelt, F. H. Kohler, G. Muller, Organome-
tallics 1985, 4, 231–238; c) O. T. Beachley Jr, C. Tessier-Youngs,
Organometallics 1983, 2, 796–801; d) F. Thomas, S. Schluz, M.
Nieger, K. Nattinen, Chem. Eur. J. 2002, 8, 1915–1924; e) D. F.
Clemens, D. H. H. Sisler, W. S. Brey, Inorg. Chem. 1966, 5, 527–
533.
For other Lewis acid containing monodentate ligands, see: S.
Bontemps, G. Bouhadir, K. Miqueu, D. Bourissou, J. Am.
Chem. Soc. 2006, 128, 12056–12057.
S. Bontemps, G. Gornitzka, G. Bouhadir, K. Miqueu, D. Bour-
issou, Angew. Chem. Int. Ed. 2006, 45, 1611–1614.
[15]
[16]
[17]
[18]
[19]
[20]
C. Qi, S. Zhang, J. Sun, Appl. Organomet. Chem. 2006, 20, 138–
141.
J. T. Gill, J. J. Mayerle, P. S. Wekker, D. F. Lewis, D. A. Ucko,
D. J. Barton, Inorg. Chem. 1976, 15, 1155–1168.
One reviewer suggested that the CO stretching vibration in the
IR spectrum could also be shifted as a result of the geometric
constrains given by the tripodal phosphorus ligand. This can-
not be ruled out.
a) F. A. Cotton, G. Wilkinson, P. L. Gaus (Eds.), Basic Inor-
ganic Chemistry, 3rd ed., John Wiley & Sons, New York, 1995,
pp. 60–61; b) A. Bondi, J. Phys. Chem. 1964, 68, 411–451; c)
R. D. Shannon, Acta Crystallogr., Sect. A 1976, 32, 751–767.
D. P. Tate, W. R. Knipple, J. M. Augl, Inorg. Chem. 1962, 1,
433–434.
H. C. Clark, L. E. Manzer, J. Organomet. Chem. 1973, 59, 411–
428.
The extreme moisture sensitivity of the complex prevented its
exact elemental analysis even though we used our best tech-
niques, including sample preparation in a glove box, weight
measurement in a glove bag, and elemental analysis as soon as
the weight measurement was complete.
A. Altomare, M. Burla, M. Camalli, G. Cascarano, C. Giacov-
azzo, A. Guagliardi, A. Moliterni, G. Polidori, R. Spagna,
SIR97, 1999.
[6]
[7]
[21]
[22]
[23]
[24]
[8]
[9]
For other Lewis acid containing bidentate ligands, see: a) L.
Turculet, J. D. Feldman, T. D. Tilley, Organometallics 2004, 23,
2488–2502; b) C. Thomas, J. C. Peters, Polyhedron 2004, 23,
2901–2913.
a) H. Nagashima, T. Sue, T. Oda, A. Kanemitsu, T. Matsum-
oto, Y. Motoyama, Y. Sunada, Organometallics 2006, 25, 1987–
[25]
[26]
[10]
A. Altomare, G. Cascarano, C. Giacovazzo, A. Guagliardi, M.
Burla, G. Polidori, M. Camalli, SIR92, 1994.
Eur. J. Inorg. Chem. 2007, 2897–2908
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
2907

T. Sue, Y. Sunada, H. Nagashima
[27] P. T. Beurskens, G. Admiraal, G. Beurskens, P. W. Bosman, R. [30] J. R. Carruthers, J. S. Rollett, P. W. Betteridge, D. Kinna, L.
FULL PAPER
de Gelder, R. Israel, J. M. M. Smits, The DIRDIF-99 Program
System: Technical Report of the Crystallography Laboratory,
University of Nijmegen, Nijmegen, The Netherlands, 1999.
[28] D. T. Cromer, J. T. Waber, International Tables for X-ray Crys-
tallography, Kynoch Press, Birmingham, U. K., 1974, vol. 4.
[29] Crystal Structure 3.8: Crystal Structure Analysis Package, Ri-
gaku and Rigaku/MSC, 9009 New Trails Dr., The Woodlands,
Texas, 77381 USA, 2000–2006.
Pearce, A. Larsen, E. Gabe, CRYSTALS Issue 11, Chemical
Crystallography Laboratory, Oxford, U. K., 1999.
Received: January 3, 2007
Published Online: May 15, 2007
2908
www.eurjic.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2007, 2897–2908