Contrasting formation of a (phenylthio)phosphinimine and (phenylthio)phosphazide. Synthesis of metal complexes

Article abstract of DOI:10.1021/ic001303x

Much of what is known about phosphinimide and phosphinimine complexes of the transition and main-group metals arises from the numerous structural studies described by Dehnicke and co-workers. While some studies of the chemistry of the complexes of these ligands have appeared, in general, such systems are unexplored. Recently, we have demonstrated that early metal-phosphinimide species can act as highly active olefin polymerization catalysts. Although Reetz et al. have begun to employ chiral bidentate bisphosphinimines in asymmetric synthesis, in general, heteroatomic bidentate ligands containing phosphinimine and phosphazides donors have received little attention. In efforts to develop the new potentially bidentate ligand systems for use with late transition metals, we have investigated the synthesis of phosphinimine and phosphazide derivatives of trialkylphosphines that incorporate thioether units. The structures of Ni and Fe complexes of these ligands are presented, and the implications for the development of new bidentate phosphinimine-based ligand systems are considered.

Full text of DOI:10.1021/ic001303x

Inorg. Chem. 2001, 40, 3827-3829
Table 1. Crystallographic Parametersa
3827
Contrasting Formation of a
(Phenylthio)phosphinimine and
(Phenylthio)phosphazide. Synthesis of Metal
Complexes
3
4
5
formula
formula
weight
C16H28Cl2FeNPS C41H64Cl2N6NiP2S2 C38H67Cl2FeN6P2S2
424.17
896.65
860.79
a(Å)
b(Å)
c(Å)
â(deg)
cryst syst
9.32850(10)
15.2783(3)
15.0720(2)
102.1040(10)
monoclinic
21.5918(2)
10.60690(10)
22.79520(10)
106.3120(10)
monoclinic
P2(1)/c
13.917(5)
21.409(7)
15.513(5)
93.821(8)
monoclinic
P2(1)/n
Luc LePichon and Douglas W. Stephan*
School of Physical Sciences, Chemistry and Biochemistry,
University of Windsor, Windsor, Ontario, Canada N9B 3P4
space group P2(1)/c
Volume (Å3) 2100.36(5)
5010.45(7)
1.189
4612(3)
1.240
ReceiVed NoVember 17, 2000
Dcalcd
(g cm-1
1.341
Much of what is known about phosphinimide and phosphin-
imine complexes of the transition and main-group metals arises
from the numerous structural studies described by Dehnicke and
co-workers.1,2 While some studies of the chemistry of the
complexes of these ligands have appeared, in general, such
systems are unexplored. Recently, we have demonstrated that
early metal-phosphinimide species can act as highly active
olefin polymerization catalysts.3,4 Although Reetz et al.5 have
begun to employ chiral bidentate bisphosphinimines in asym-
metric synthesis, in general, heteroatomic bidentate ligands
containing phosphinimine2 and phosphazide6-8 donors have
received little attention.9,10 In efforts to develop the new
potentially bidentate ligand systems for use with late transition
metals, we have investigated the synthesis of phosphinimine
and phosphazide derivatives of trialkylphosphines that incor-
porate thioether units. The structures of Ni and Fe complexes
of these ligands are presented, and the implications for the
development of new bidentate phosphinimine-based ligand
systems are considered.
)
Z
4
4
2
abs coeff,
1.144
0.674
0.635
µ, mm-1
R (%)
Rw (%)
0. 0478
0.1202
0.0869
0.1957
0.0534
0.1554
a All data collected at 24 °C with Mo KR radiation (λ ) 0.71069
Å), R ) ∑||Fo| - |Fc||/∑|Fo|, Rw ) [∑[ω(Fo - Fc2)2]/∑[ωFo )2]]0.5
.
2
2
and stirred for 30 min. The solution was then cooled to room
temperature. The white solid formed was washed with hexane and dried
under a vacuum; 540 mg (73%) of 2 was obtained. 1. reaction at 25
°C, yield: 98%. 1H NMR (C6D6): δ 7.80-6.93 (5H, Ph), 5.25(d, 2H,
CH2, J2 ) 25 Hz), 1.88 (sept. J3 ) 7.5 Hz, J2 ) 3.0 Hz, 3H)
PH
HH
PH
0.98 (dd, 18H). 13C{1H} NMR (C6D6): δ 141.4, 130.0, 129.0, 125.1
(Ph), 56.9 (d, CH2, J2PC ) 8.6 Hz), 24.9 (d, CH, J1PC ) 57.6 Hz), 17.7
(d, CH3, J2PC ) 2.2 Hz). 31P{1H} NMR (C6D6): δ 32.8. 2. No evolution
1
of N2 was observed in this case. H NMR (C6D6): δ 7.59-6.88 (5H,
Ph), 5.28 (s 2H, CH2), 1.17 (d, 27H, CH3, J2 ) 10 Hz). 13C{1H}
PH
NMR (C6D6): δ 139.0, 130.2, 129.2, 125.9 (Ph), 64.4 (CH2), 40.4 (d,
P-C, J1 ) 40 Hz), 30.2 (CH3). 31P{1H} NMR (C6D6): δ 50.9. EA
PC
Calcd. for C19H34N3PS: C, 62.09; H, 9.32; N, 11.43. Found: C, 61.96;
H, 9.70; N, 11.47.
Experimental Section
General Data. All preparations were done under an atmosphere of
dry, O2-free N2 employing both Schlenk line techniques and Innovative
Technologies or Vacuum Atmospheres inert atmosphere gloveboxes.
Solvents were purified employing Grubbs type column systems
manufactured by Innovative Technology. All organic reagents were
purified by conventional methods. Guelph Chemical Laboratories Inc.
of Guelph, Ontario performed combustion analyses. N3CH2SPh, t-Bu3P,
i-Pr3P, and anhydrous FeCl2 were purchased from the Aldrich Chemical
Co. NiCl2(DME) was purchased from the Strem Chemical Co.
Synthesis of i-Pr3PNCH2SPh (1) and t-Bu3PN3CH2SPh (2). These
compounds were prepared in a similar fashion using the appropriate
phosphine, and thus only one preparation is detailed. N3CH2SPh (347
mg, 2.1 mmol) was added to t-Bu3P (405 mg, 2 mmol), and white
solid appeared immediately. The reaction mixture was heated at 90 °C
Synthesis of [(i-Pr3PNCH2SPh)FeCl(µ-Cl)]2 (3). A solution of 1
(100 mg, 0.34 mmol) in 5 mL of toluene was added to a suspension of
FeCl2 (43 mg, 0.34 mmol) in 5 mL of toluene. The reaction mixture
was stirred for 20 h. The toluene was removed in vacuo to give a pale
yellow precipitate. A pale yellow solution was extracted with di-
chloromethane. Upon standing, 30 mg of crystals of 3 were obtained
(21%). EA Calcd. for C32H56Cl4Fe2N2P2S2: C, 45.30; H, 6.65; N, 3.30.
Found: C, 45.41; H, 6.93; N, 3.25.
Synthesis of (t-Bu3PN3CH2SPh)2NiCl2 (4) and (t-Bu3PN3CH2-
SPh)2FeCl2 (5). These compounds were prepared in a similar fashion
using FeCl2 or NiCl2(DME), and thus only one preparation is detailed.
FeCl2 (32 mg, 0.25 mmol) was added to a solution of 2 (184 mg, 0.50
mmol) in 2.5 mL of toluene. The reaction mixture was stirred for 20
h at room temperature. An orange complex formed. The solvent was
removed, and the orange powder was washed with hexane. After
recrystallization in a mixture of toluene/hexane, 190 mg of orange
crystals of 5 (88%) were isolated. EA Calcd. for C38H68N6P2S2FeCl2:
C, 52.96; H, 7.95; N, 9.75. Found: C, 52.78; H, 8.36; N, 9.61. 4. Pink
solid, yield: 85%. EA. Calcd. for C38H68N6P2S2NiCl2+0.3C6H6: C,
53.83; H, 7.92; N, 9.46. Found: C, 53.82; H, 8.26; N, 9.44.
X-ray Data Collection and Reduction. The crystals were manipu-
lated and mounted in capillaries in a glovebox. Diffraction experiments
were performed on a Siemens SMART System CCD diffractometer,
collecting a hemisphere of data in 1329 frames with 10 s exposure
times. Crystal data are summarized in Table 1. A measure of decay
was obtained by re-collecting the first 50 frames of each data set. The
intensities of reflections within these frames showed no statistically
significant change over the duration of the data collections. The data
were processed using the SAINT and XPREP processing packages.
An empirical absorption correction based on redundant data was applied
to each data set. Subsequent solution and refinement was performed
(1) Dehnicke, K.; Weller, F. Coord. Chem. ReV. 1997, 158, 103-169.
(2) Dehnicke, K.; Krieger, M.; Massa, W. Coord. Chem. ReV. 1999, 182,
19-65.
(3) Stephan, D. W.; Guerin, F.; Spence, R. E. v. H.; Koch, L.; Gao, X.;
Brown, S. J.; Swabey, J. W.; Wang, Q.; Xu, W.; Zoricak, P.; Harrison,
D. G. Organometallics 1999, 18, 2046-2048.
(4) Stephan, D. W.; Stewart, J. C.; Guerin, F.; Spence, R. E. v. H.; Xu,
W.; Harrison, D. G. Organometallics 1999, 18, 1116-1118.
(5) Reetz, M. T.; Bohres, E.; Goddard, R. Chem. Commun. 1998, 008,
935-936.
(6) Hillhouse, G. L.; Haymore, B. L. J. Organomet. Chem. 1978, 162,
C23-C26.
(7) Hillhouse, G.; Goeden, G. V.; Haymore, B. L. Inorg. Chem. 1982,
21, 2064-2071.
(8) Bieger, K.; Bouhadir, G.; Reau, R.; Dahan, F.; Bertrand, G. J. Am.
Chem. Soc. 1996, 118, 1038-1044.
(9) Katti, K. V.; Cavell, R. G. Organometallics 1991, 10, 539-541.
(10) Katti, K. V.; Santarsiero, B. D.; Pinkerton, A. A.; Cavell, R. G. Inorg.
Chem. 1993, 32, 5919-5925.
10.1021/ic001303x CCC: $20.00 © 2001 American Chemical Society
Published on Web 06/14/2001

3828 Inorganic Chemistry, Vol. 40, No. 15, 2001
Notes
Scheme 1
Figure 1. ORTEP drawing of 3, 20% thermal ellipsoids are shown.
Hydrogen atoms are omitted for clarity. Fe(1)-N(1) 2.033(2); Fe(1)-
Cl(1) 2.2556(12), Fe(1)-Cl(2) 2.3918(10); Fe(1) Cl(2)′ 2.4586(11);
P(1)-N(1) 1.619(3); Fe(1)-S(1) 3.068(2); N(1)-Fe(1)-Cl(1) 125.74(8);
N(1)-Fe(1)-Cl(2) 106.19(7); Cl(1)-Fe(1)-Cl(2) 118.25(5); N(1)-
Fe(1)-Cl(2)′ 112.51(8); Cl(1)-Fe(1)-Cl(2) 99.54(5); Cl(2)′-Fe(1)-
Cl(2) 87.42(3); Fe(1)-Cl(2)-Fe(1) 92.58(3); P(1) N(1) Fe(1) 130.36(14).
2
using the SHELXTL solution package. The reflections with Fo2 > 3σFo
were used in the refinements.
Structure Solution and Refinement. Non-hydrogen atomic scat-
tering factors were taken from the literature tabulations.11 The heavy
atom positions were determined using direct methods, employing the
SHELXTL direct methods routines. The remaining non-hydrogen atoms
were located from successive difference Fourier map calculations. The
refinements were carried out by using full-matrix least squares
techniques on F, minimizing the function ω(|Fo| - |Fc|)2, where the
2
2
weight ω is defined as 4Fo /2σ(Fo ), and Fo and Fc are the observed
and calculated structure factor amplitudes. In the final cycles of each
refinement, all non-hydrogen atoms were assigned anisotropic temper-
ature factors. Carbon bound hydrogen atom positions were calculated
and allowed to ride on the carbon to which they are bonded, assuming
a C-H bond length of 0.95 Å. Hydrogen atom temperature factors
were fixed at 1.10 times the isotropic temperature factor of the carbon
atom to which they are bonded. In the case of 4, a two-site disorder of
one of the thiophenol groups was modeled, although these hydrogen
atom contributions were not included. All other hydrogen atom
contributions were calculated, but not refined. The final values of
refinement parameters are given in Table 1. The locations of the largest
peaks in the final difference Fourier map calculation and the magnitude
of the residual electron densities in each case were of no chemical
significance. Crystallographic data have been deposited as Supporting
Information.
Figure 2. ORTEP drawing of 4, 20% thermal ellipsoids are shown.
Hydrogen atoms are omitted for clarity. Ni(1)-N(3) 2.017(6); Ni(1)-
N(6) 2.031(6); Ni(1)-Cl(2) 2.275(2); Ni(1)-Cl(1) 2.309(2); Ni(1)-
N(1) 2.635(6); Ni(1)-N(4) 2.789(6); P(1)-N(1) 1.675(6); P(2)-N(4)
1.652(6); N(1)-N(2) 1.354(8); N(2)-N(3) 1.287(8); N(4)-N(5)
1.353(8); N(5)-N(6) 1.272(7); N(3)-Ni(1)-N(6) 99.3(2); N(3)-
Ni(1)-Cl(2) 152.83(18); N(6)-Ni(1)-Cl(2) 96.84(17); N(3)-Ni(1)-
Cl(1) 93.22(18); N(6)-Ni(1)-Cl(1) 99.84(17); Cl(2)-Ni(1)-Cl(1)
105.40(9); N(2)-N(1)-P(1) 112.8(5); N(3)-N(2)-N(1) 109.0(5);
N(2)-N(3)-Ni(1) 114.6(5); N(5)-N(4)-P(2) 115.2(5); N(6)-N(5)-
N(4) 110.7(6); N(5)-N(6)-Ni(1) 118.8(4).
Results and Discussion
While it is generally observed that phosphazides are thermally
sensitive intermediates on route to phosphinimines,12,13 the
interceptions of phosphazides have been observed for cases in
which the reagents are sterically demanding.14,15 Similar conclu-
sions are illustrated by the oxidation of the phosphines R3P, R
) i-Pr, t-Bu, with the azide N3CH2SPh. In the case of i-Pr3P,
reaction with N3CH2SPh proceeds with the vigorous evolution
of gas to afford the liquid 1 in 98% yield. This material exhibits
a single 31P NMR resonance at 32.8 ppm. The 1H NMR
resonances at 7.80-6.93, 5.25, 1.88, and 0.98 ppm were
consistent with the presence of phenyl, methylene, methine, and
methyl protons, respectively. The corresponding 13C{1H} NMR
resonances were observed, consistent with the formulation of
the expected product i-Pr3PNCH2SPh 1. In an analogous manner,
the reaction of t-Bu3P under similar conditions proceeds to give
a white product 2 in 73% yield (Scheme 1). This material
exhibits a signal in the 31P{1H} NMR spectrum at 50.9 ppm.
Similar to 1, 1H NMR resonances at 7.59-6.88, 5.28, and 1.17
ppm were consistent with the presence of phenyl, methylene,
and tert-butyl groups. However, the absence of P-coupling to
the methylene protons and the lack of gas evolution in this latter
reaction suggest the formulation of 2 as t-Bu3PN3CH2SPh, a
view that was also supported by elemental analysis. Heating
solutions of 2 to 140° results only in decomposition. Presumably,
the greater steric demands of the t-Bu3P fragment precludes
effective N2 elimination and formation of the corresponding
phosphinimine.
The synthesis of 1 and 2 permits a comparison of these related
ligands. Reaction of compound 1 with FeCl2 in toluene proceeds
with the formation of a pale yellow precipitate, which could be
extracted in dichloromethane. Upon standing, pure yellow
crystalline product 3 was obtained in 21% yield. The relatively
low yield was attributed to the high solubility of the complex.
An X-ray structural determination confirmed the formulation
of 3 as the symmetric dimeric species [(i-Pr3PNCH2SPh)FeCl-
(µ-Cl)]2 3 (Figure 1). In this species, two Fe centers are
coordinated to the phosphinimine-nitrogen atom and a terminal
chloride. Two other chlorine atoms bridge the two Fe centers.
The terminal and average bridging Fe-Cl distances of 2.2556(12)
and 2.4252(11) Å are clearly significantly different, reflecting
the different environments. The Fe-N distance of 2.033(2) Å
is typical of coordinated Fe-phosphinimines.16 Comparison of
the long Fe-thioether S distance of 3.068(2) Å in 3 to those in
(11) Cromer, D. T.; Mann, J. B. Acta Crystallogr., Sect. A 1968, A24, 321-
324.
(12) Staudinger, H. HelV. Chim. Acta 1919, 2, 635.
(13) Pilgram, K.; Goergen, F.; Pollard, G. J. Heterocycl. Chem. 1971, 8,
951-959.
(14) Velasco, M. D.; Molina, P.; Fresneda, P. M.; Sanz, M. A. Tetrahedron
2000, 56, 4079-4084.
(15) Kukhar, V. P.; Kasukhin, L. F.; Ponomarchuk, M. P.; Chernega, A.
N.; Antipin, M. Y.; Struchkov, Y. T. Phosphorus, Sulfur Silicon Relat.
Elem. 1989, 44, 149-153.

Notes
Inorganic Chemistry, Vol. 40, No. 15, 2001 3829
the P-bound N atom, and is electronically favored.8,18 In 4,
(Figure 2), the Ni-N and Ni-Cl distances average 2.024(6) Å
and 2.292(2) Å, respectively. The two phosphorus-bound
nitrogen atoms also approach the Ni center at distances of
2.635(6) and 2.789(6) Å, presumably as a result of a weak
chelate effect. In a similar manner, the two Fe-N and Fe-Cl
distances of complex 5, (Figure 3), average 2.101(4) and
2.2840(14) Å, respectively. The closest approaches of the
P-bound N atoms to Fe are 3.024(6) and 2.971(6) Å. The longer
range metal-nitrogen in 4 and 5 may reflect weak interactions
that contribute to distortions about the metal coordination
spheres. For example, the angles about Fe in 5 range from
96.54(10)° to 125.80(8)°. The nature of binding of the phos-
phazide ligands in 4 and 5 stands in contrast to that seen in
WBr2(CO)3(ArN3PPh3), in which the phosphazide acts as a
bidentate ligand with similar W-N bond lengths (2.163(4),
2.220(5) Å).7
The synthesis of the phosphinimine 1 and phosphazide 2
confirms that steric issues must be considered in developing a
synthetic strategy to bidentate phosphinimine ligands. Moreover,
the crystallographic data for the Fe and Ni complexes 3-5
suggest that chelation via the thioether S or a phosphazide N
atom exhibits only a weak stereochemical effect on a metal
center. These data suggest that synthetic routes to bidentate
ligands derived from phosphinimines may require the incorpora-
tion of both stronger secondary donors and longer donor atom
separations to permit effective chelation. Strategies to such
ligands are under development.
Figure 3. ORTEP drawing of 5, 20% thermal ellipsoids are shown.
Hydrogen atoms are omitted for clarity. Fe(1)-N(4) 2.076(3); Fe(1)-
N(1) 2.122(3); Fe(1)-Cl(2) 2.2587(12); Fe(1)-Cl(1) 2.3093(13); P(1)-
N(3) 1.650(3); P(2)-N(6) 1.660(3); N(1)-N(2) 1.273(3); N(2)-N(3)
1.330(3); N(4)-N(5) 1.288(3); N(5)-N(6) 1.309(4); N(4)-Fe(1)-N(1)
96.54(10); N(4)-Fe(1)-Cl(2) 125.80(8); N(1)-Fe(1)-Cl(2) 108.84(8);
N(4)-Fe(1)-Cl(1) 108.35(8); N(1)-Fe(1)-Cl(1) 101.60(7); Cl(2)-
Fe(1)-Cl(1) 111.99(5); N(2)-N(1)-Fe(1) 124.5(2); N(1)-N(2)-N(3)
112.1(2); N(2)-N(3)-P(1) 115.4(2); N(5)-N(4)-Fe(1) 123.5(2);
N(4)-N(5)-N(6) 112.1(2); N(5)-N(6)-P(2) 116.6(2).
the Fe complexes [Fe(N2H2S2)]2 (2.572(1) Å, N2H2S2 ) 2,2′-
bis(2-mercaptophenylamino)diethyl sulfide)17 and Fe(N2H2S3)-
(CO) (2.279(2) Å, N2H2S3 ) 2′-bis(2-mercaptophenylamino)-
diethyl sulfide)17 suggests a very weak Fe-S interaction in 3.
Reaction of compound 2 with either NiCl2(DME) or FeCl2
affords the paramagnetic complexes (t-Bu3PN3CH2SPh)2NiCl2
4 and (t-Bu3PN3CH2SPh)2FeCl2 5, respectively (Scheme 1).
These compounds were obtained in good yield, and each was
structurally characterized by X-ray crystallography. In both
complexes, the metal centers are bonded to two chlorine atoms
and two nitrogen atoms of the phosphazide ligands adjacent
the methylene group. The coordination of these N atoms is
presumably a result of the lesser steric crowding, compared to
Acknowledgment. Financial support from the NOVA
Chemicals Corporation and NSERC of Canada is gratefully
acknowledged.
Supporting Information Available: Crystallographic X-ray crys-
tallographic files in CIF format for the structure determinations of
compounds 3-5. This material is available free of charge via the
Internet at http://pubs.acs.org.
(16) Mai, H. J.; Wocadlo, S.; Kang, H. C.; Massa, W.; Dehnicke, K.;
Maichle-Moessmer, C.; Straehle, J.; Fenske, D. Z. Anorg. Allg. Chem.
1995, 621, 705-712.
(17) Sellmann, D.; Utz, J.; Heinemann, F. W. Inorg. Chem. 1999, 38, 459-
466.
IC001303X
(18) Widauer, C.; Grutzmacher, H.; Shevchenko, I.; Gramlich, V. Eur. J.
Inorg. Chem. 1999, 1659-1664.