Synthesis and Structures of Nickel and Palladium Salicylaldiminato 1,3,5-Triaza-7-phosphaadamantane (PTA) Complexes

Article abstract of DOI:10.1021/ic0341284

The synthesis of nickel(II) and palladium(II) salicylaldiminato complexes incorporating the water-soluble phosphine 1,3,5-triaza-7-phosphaadamantane(PTA) has been achieved employing two preparative routes. Reaction of the original ethylene polymerization catalyst developed by Grubbs and co-workers (Organometallics 1998, 17, 3149), (salicylaldiminato)Ni(Ph)PPh₃, with PTA using a homogeneous methanol/toluene solvent system resulted in the formation of the PTA analogues in good yields. Alternatively, complexes of this type may be synthesized via a direct approach utilizing (tmeda)M(CH ₃)₂ (M = Ni, Pd), the corresponding salicylaldimine, and PTA. Yields by this method were generally near quantitative. The complexes were characterized in solution by ¹H/¹³C/³¹P NMR spectroscopy and in the solid-state by X-ray crystallography. All derivatives exhibited square-planar geometry with the bulky isopropyl groups on the aniline being perpendicular to the plane formed by the metal center and its four ligands. Such orientation of these sterically encumbering groups is responsible for polymer chain growth during olefin polymerization in favor of chain termination via β-hydride elimination. Polymerization reactions were attempted using the nickel-PTA complexes in a biphasic toluene/water mixture in an effort to initiate ethylene polymerization by trapping the dissociated phosphine ligand in the water layer, thereby eliminating the need for a phosphine scavenger. Unfortunately, because of the strong binding ability of the small, donating phosphine(PTA) as compared to PPh₃, phosphine dissociation did not occur at a temperature where the complexes are thermally stable.

Full text of DOI:10.1021/ic0341284

Inorg. Chem. 2003, 42, 6915−6922
Synthesis and Structures of Nickel and Palladium Salicylaldiminato
1,3,5-Triaza-7-phosphaadamantane (PTA) Complexes
Donald J. Darensbourg,* Cesar G. Ortiz, and Jason C. Yarbrough
Department of Chemistry, Texas A&M UniVersity, P.O. Box 30012, College Station, Texas 77843
Received February 5, 2003
The synthesis of nickel(II) and palladium(II) salicylaldiminato complexes incorporating the water-soluble phosphine
1,3,5-triaza-7-phosphaadamantane(PTA) has been achieved employing two preparative routes. Reaction of the
original ethylene polymerization catalyst developed by Grubbs and co-workers (Organometallics 1998, 17, 3149),
(salicylaldiminato)Ni(Ph)PPh3, with PTA using a homogeneous methanol/toluene solvent system resulted in the
formation of the PTA analogues in good yields. Alternatively, complexes of this type may be synthesized via a
direct approach utilizing (tmeda)M(CH ) (M ) Ni, Pd), the corresponding salicylaldimine, and PTA. Yields by this
3 2
1
method were generally near quantitative. The complexes were characterized in solution by H/13C/31P NMR
spectroscopy and in the solid-state by X-ray crystallography. All derivatives exhibited square-planar geometry with
the bulky isopropyl groups on the aniline being perpendicular to the plane formed by the metal center and its four
ligands. Such orientation of these sterically encumbering groups is responsible for polymer chain growth during
olefin polymerization in favor of chain termination via â-hydride elimination. Polymerization reactions were attempted
using the nickel−PTA complexes in a biphasic toluene/water mixture in an effort to initiate ethylene polymerization
by trapping the dissociated phosphine ligand in the water layer, thereby eliminating the need for a phosphine
scavenger. Unfortunately, because of the strong binding ability of the small, donating phosphine(PTA) as compared
to PPh3, phosphine dissociation did not occur at a temperature where the complexes are thermally stable.
Introduction
adhesive properties. In addition, catalytic activities and
polymer molecular weights obtained by employing these
catalysts rival those achieved by Ziegler-Natta and metal-
locene catalytic systems. A drawback of these systems
involves competing â-elimination which leads to low mo-
lecular weight polymers. This is the premise of the SHOP
(Shell higher olefin polymerization) process using the
[(OCH(R1)CH(R2)PPh2)Ni(L)(R)] catalyst developed in the
late 1960s by Keim and co-workers.3 This latter process is
currently used to produce higher molecular weight R-olefins
which may be later converted to other useful products (e.g.
detergents).3
Presently, most R-olefins are polymerized through the use
of heterogeneous Ziegler-Natta catalysts or metallocene
catalysts to achieve high molecular weight polymers.1 Both
of these systems are based on early transition metals, and
although these catalysts are very active, they are highly
oxophilic and vulnerable to decomposition. Therefore, extra
purification costs must be added to the industrial polymer-
ization process to ensure high conversions and high molec-
ular weight polymers. Consequently, attention has been
turned to late transition metals which are less oxophilic and
therefore able to tolerate functional-containing monomers.2
The copolymerization of functionalized monomers with
ethylene has led to new polymeric materials with enhanced
One of the most notable and well-understood ethylene
polymerization systems utilizes Brookhart’s cationic â-di-
imine catalysts [(ArNdC(R)C(R′)dNAr)M(L)(CH3)]+ (M )
* To whom correspondence should be addressed. E-mail: djdarens@
mail.chem.tamu.edu.
(2) For a comprehensive review, see: (a) Brookhart, M.; Ittel, S. D.;
Johnson, L. K. Chem. ReV. 2000, 100, 1169. (b) Mecking, S. Coord.
Chem. ReV. 2000, 203, 325. (c) Gibson, V. C.; Britovsek, G. J.; Wass,
D. F. Angew. Chem., Int. Ed. 1999, 38, 428.
(3) (a) Keim, W.; et al. (Shell Dev.). U.S. Patents 3635937, 3647914,
3686159, 3644563, 3647914, 1974. (b) Keim, W. Angew. Chem., Int.
Ed. Engl. 1990, 29, 235. (c) Keim, W.; Peuckert, M. Organometallics.
1983, 2, 594. (d) Skupinska, J. Chem. ReV. 1991, 91, 613.
(1) For a comprehensive review, see: (a) Kuran, W. Principles of
Coordination Polymerization; Wiley-VCH: New York, 2001; pp 53-
215. (b) Coville, N. J.; Mo¨hring, P. C. J. Organomet. Chem. 1994,
479, 1. (c) Hlatky, G. Coord. Chem. ReV. 2000, 199, 235. (c) Alt, H.
G.; Ko¨ppl, A. Chem. ReV. 2000, 100, 1205. (d) Bendikt, G. M., Ed.
Metallocene Technology in Commercial Application; Plastics Design
Library: Norwich, NY, 1999.
10.1021/ic0341284 CCC: $25.00 © 2003 American Chemical Society
Published on Web 09/17/2003
Inorganic Chemistry, Vol. 42, No. 21, 2003 6915

Darensbourg et al.
Figure 2. PTA, meta-TPPTS, and para-TPPTS water-soluble phosphines.
process is that this cocatalyst is highly air-sensitive and prone
to autocatalytic decomposition.
The purpose of this study is to explore the effect of
incorporating a water-soluble phosphine (WSP) into 1 and
using a biphasic toluene/water solvent mixture to allow
irreversible phosphine dissociation, thereby eliminating the
need for a phosphine scavenger. Although it has been shown
in Grubb’s system that initiation can take place without the
need of a cocatalyst by using a more sterically demanding
group in the R1 position of the ligand, many of the active
derivatives require cocatalysts.6a The polymerization of
various olefins has been achieved in water; however, to our
knowledge, this is the first attempt to use a biphasic solvent
system to generate the active catalyst in the organic phase
while allowing the WSP to enter and remain in the aqueous
phase.7
The use of WSP in catalysis was first commercialized by
Ruhrchemie in 1984 for the hydroformylation of higher
molecular weight R-olefins to predominantly form terminal
aldehydes.8 The system uses the trisulfonated triphenylphos-
phine ligand meta-TPPTS (Figure 2),9 coordinated to rhod-
ium to form the active RhH(CO)(TPPTS)3 catalyst. Another
WSP, which is of main interest to this study, is the hetero-
cyclic, aliphatic 1,3,5-triaza-7-phosphaadamantane (PTA).10
Many complexes incorporating this ligand have been pre-
pared, and the ruthenium and rhodium derivatives have been
found to be effective catalysts in the hydrogenation of various
unsaturated hydrocarbons.11 Herein, we wish to report the
synthesis of various PTA derivatives of 1 by two routes, (a)
ligand exchange reaction and (b) direct synthesis, as well as
the solid-state characterization of many of these derivatives.
Figure 1. Nickel(II) salicylaldimine catalyst.
Ni, Pd).2a,4 Optimization of the catalyst’s activity over the
years by ligand modification has led to the industrial
implementation of the catalyst for the polymerization of
ethylene.5 Of primary interest to this study is the catalyst
developed by Grubbs and co-workers which is based on the
salicylaldimine ligand framework (Figure 1).6 Similar to the
SHOP catalyst, the ligand is monoanionic, rendering a neutral
catalyst which is void of bulky counterions. However, unlike
the SHOP catalyst, the incorporation of sterically demanding
groups on the ligand effectively shields the axial sites of the
metal center thereby allowing enchainment to predominate
over â-hydride elimination processes. As with most olefin
polymerization catalysts, a cocatalyst is usually needed to
initiate polymerization. In the case of metallocene1d or
Brookhart’s4a catalysts, an excess of an aluminum cocatalyst
such as MAO must be used. For 1, a phosphine scavenger
such as Ni(COD)2 (COD ) 1,5-cyclooctadiene) is needed
to remove the phosphine from the metal center and allow
alkene coordination.6 An obvious drawback to this latter
(4) (a) Brookhart, M.; Johnson, L. K.; Killian, C. M. J. Am. Chem. Soc.
1995, 117, 6414. (b) Brookhart, M.; Johnson, L. K.; Mecking, S. J.
Am. Chem. Soc. 1996, 118, 267. (c) Brookhart, M.; Killian, C. M.;
Tempel, D. J.; Johnson, L. K. J. Am. Chem. Soc. 1996, 118, 11664.
(d) Brookhart, M.; Killian, C. M.; Johnson, L. K. Organometallics.
1997, 16, 2005. (e) Brookhart, M.; Svejda, S. A.; Johnson, L. K. J.
Am. Chem. Soc. 1999, 121, 10634. (f) Brookhart, M.; Tempel, D. J.;
Johnson, L. K.; Huff, R. L.; White, P. S. J. Am. Chem. Soc. 2000,
122, 6686.
(5) (a) Arthur, S. D.; Bennett, A. M. A.; Brookhart, M. S.; Coughlin, E.
B.; Feldman, J.; Ittel, S. D.; Johnson, L. K.; Killian, C. M.; Kreutzer,
K. A. [DuPont (Polymerizations)]. U.S. Patent 5866663, Feb 2, 1999.
(b) Brookhart, M. S.; Ittel, S. D.; Johnson, L. K.; Killian, C. M.;
Kreutzer, K. A.; McCord, E. F.; McLain, S. J.; Tempel, D. J. [DuPont
(Polymer Compositions)]. U.S. Patent 5880241, May 3, 1999. (c)
Brookhart, M. S.; Johnson, L. K.; Killian, C. M.; Wang, L.; Yang, Z.
Y. [DuPont (R-Olefins)]. U.S. Patent 5880323, March 9, 1999. (d)
Arthur, S. D.; Bennett, A. M. A.; Brookhart, M. S.; Coughlin, E. B.;
Feldman, J.; Ittel, S. D.; Johnson, L. K.; Killian, C. M.; Kreutzer, K.
A.; Parthasarathy, A.; Temple, D. J. [DuPont (Ligand Compositions)].
U.S. Patent 5886224, March 23, 1999. (e) Arthur, S. D.; Brookhart,
M. S.; Hohnson, L. K.; Killian, C. M.; McCord, El F.; McLain, S. J.
[DuPont (Copolymers)]. U.S. Patent 5891963, April 6, 1999.
(6) (a) Grubbs, R. H.; Wang, C.; Friedrich, S.; Younkin, T. R.; Li, R. T.;
Bansleben, D. A.; Day, M. W. Organometallics 1998, 17, 3149. (b)
Grubbs, R. H.; Younkin, T. R.; Connor, E. F.; Henderson, J. I.;
Friedrich, S.; Bansleben, D. A. Science 2000, 287, 460.
(7) Mecking, S.; Held, A.; Bauers, F. Angew. Chem., Int. Ed. 2002, 41,
545.
(8) (a) Cornils, B.; Wiebus, E. CHEMTEC 1995, 25, 33. (b) Cornils, B.
Angew. Chem., Int. Ed. Engl. 1995, 34, 1575. (c) Cornils, B.; Kuntz,
E. G. J. Organomet. Chem. 1995, 502, 177. For a comprehensive
review, see: (d) Joo´, F.; Katho´, AÄ . J. Mol. Catal., A: Chem. 1997,
116, 3. (e) Cornils, B.; Herrmann, W. A.; Eckl, R. W. J. Mol. Catal.,
A: Chem. 1997, 116, 27.
(9) (a) Herrmann, W. A.; Kohlpaintner, C. W. Inorg. Synth. 1998, 32, 8.
For a comprehensive review on the use of TPPTS, see: (b) Joo´, F.
Aqueous Organometallic Catalysis; Kluwer Academic: Boston, MA,
2001. (c) Cornils, B., Herrmann, W. A., Eds. Aqueous-Phase Orga-
nometallic Catalysis; Wiley-VCH: Weinheim, Germany, 1998.
(10) (a) Daigle, D. J.; Peppermann, A. B.; Vail, S. L. J. Heterocycl. Chem.
1974, 11, 407. (b) Daigle, D. J. Inorg. Synth. 1998, 32, 40.
(11) (a) Darensbourg, D. J.; Joo´, F.; Daigle, D. J.; Kannisto, M.; Katho´,
A.; Reibenspies, J. H. Inorg. Chem. 1994, 33, 200. (b) Darensbourg,
D. J.; Joo´, F.; Kannisto, M.; Katho´, AÄ .; Riebenspies, J. H. Organo-
metallics. 1992, 11, 1990. (c) Darensbourg, D. J.; Stafford, N. W.;
Joo´, F.; Reibenspies, J. H. J. Organomet. Chem. 1995, 488, 99. (d)
Joo´, F.; Na´dasdi, L.; Be´nyei, A. C.; Darensbourg, D. J. J. Organomet.
Chem. 1996, 512, 45. (e) Akbayeva, D. N.; Gonsalvi, L.; Oberhauser,
W.; Peruzzini, M.; Vizza, F.; Bru¨ggeller, P.; Romerosa, A.; Sava, G.;
Bergamo, A. Chem. Commun. 2003, 264.
6916 Inorganic Chemistry, Vol. 42, No. 21, 2003

Nickel and Palladium Salicylaldiminato Complexes
PTA complexes were prepared in an analogous fashion. Complexes
Experimental Section
3a-e and 4a-e were all obtained as yellow solids in good yields.
Materials and Methods. Unless otherwise indicated, all reac-
tions were carried out under an inert argon atmosphere using
standard Schlenk and drybox techniques. Prior to their use, all
solvents were distilled using standard techniques. 2-Hydroxy-3-
phenylbenzaldehyde was prepared from the corresponding phenol,12
and all other benzaldehydes were purchased from Aldrich Chemi-
cals. Ligands a-g were prepared by the condensation reaction of
the corresponding aldehyde with commercially available 2,6-
diisopropylaniline. PTA,10 (TMEDA)Ni(CH3)2,13a (TMEDA)Pd-
(CH3)2,13b and 1a6a were prepared according to literature procedures.
1H, 13C, and 31P NMR data were obtained using a Varian Unity+
300 MHz NMR instrument. 1H and 13C chemical shifts were
referenced according to the deuterated solvent used. The 31P
chemical shifts were referenced using an external 85% H3PO4
sample. Elemental analysis was conducted by Canadian Microana-
lytical Inc.
3a (R3 ) NO2, R1 ) R2 ) R4 ) H): 1H NMR (300 MHz, C6D6,
3
δ) -1.33 (s, 3H, Ni-CH3), 0.98 (d, JHH ) 6.60 Hz, 6H, CH-
(CH3)2), 1.35 (d, 3JHH ) 6.90 Hz, 6H, CH(CH3)2), 3.58 (sept, 3JHH
2
) 6.90 Hz, 2H, CH(CH3)2), 3.73 (s, 6H, NCH2N), 4.09 (dd, JHP
2
3
) 32.69 Hz, JHH ) 12.90 Hz, 6H, PCH2N), 6.41 (t, JHH ) 9.60
Hz, 1H, Ar), 7.05-7.13 (m, 3H, Ar), 7.46 (s, 1H, HNdC), 7.98
3
3
3
(d, JHH ) 2.94, 1H, Ar), 8.11 (dd, JHH ) 2.94 Hz, JHH ) 9.60
Hz, 1H, Ar); 13C NMR (75 MHz, C6D6, δ) -15.84 (d, 2JCP ) 4.25
Hz, Ni-CH3), 23.47, 25.09, 29.00, 50.85 (d, 1JCP ) 5.93 Hz, P-C-
3
N), 73.93 (d, JCP ) 4.85 Hz, N-C-N), 122.93, 124.14, 127.36,
129.18, 132.84, 141.01, 165.92; 31P NMR (121 MHz, C6D6, δ)
-47.10.
3b (R1 ) OMe, R2 ) R3 ) R4 ) H): 1H NMR (300 MHz,
3
C6D6, δ) -1.26 (s, 3H, Ni-CH3), 1.04 (d, JHH ) 6.30 Hz, 6H,
CH(CH3)2), 1.40 (d, 3JHH ) 6.60 Hz, 6H, CH(CH3)2), 3.42 (s, 3H,
3
OCH3), 3.82 (sept, JHH ) 6.60 Hz, 2H, CH(CH3)2), 4.00 (s, 6H,
Preparation of Nickel Salicylaldiminato PTA Complexes by
Ligand Exchange with 1 (2a). To a 50 mL Schlenk flask
containing 1a (100 mg, 0.138 mmol) in 3 mL of toluene was added
a concentrated methanol solution of PTA (23.7 mg, 0.152 mmol,
in 5 mL of MeOH). A yellow precipitate immediately formed, and
the reaction was stirred overnight. After filtration and washing with
pentane, the solid was redissolved in CH2Cl2 and the solution was
filtered. After removal of the solvent under vacuum, 2a was
obtained.
NCH2N), 4.08 (dd, 2JHP ) 22.79 Hz, 2JHH ) 13.20 Hz, 6H, PCH2N),
3
3
6.46 (t, JHH ) 7.80 Hz, 1H, Ar), 6.59 (d, JHH ) 7.50 Hz, 1H,
3
Ar), 6.67 (d, JHH ) 7.20 Hz, 1H, Ar), 6.94-7.48 (m, 3H, Ar),
7.90 (s, 1H, HCdN); 13C NMR (75 MHz, C6D6, δ) -19.71 (d,
2JCP ) 4.36 Hz, Ni-CH3), 21.10, 22.82, 26.50, 48.38 (d, JCP
)
1
3
6.03 Hz, P-C-N), 53.46 (d, JCP ) 4.90 Hz, N-C-N), 71.29
(OCH3), 111.10, 116.88, 121.58, 123.80, 124.48, 128.34, 132.51,
139.29, 147.16, 151.42, 156.86, 163.32; 31P NMR (121 MHz, C6D6,
δ) -54.56; yield 51.1%. Anal. Calcd for C27H39N4OPNi: C, 59.91;
H, 7.26; N, 10.35. Found: C, 59.86; H, 7.10; N, 10.55.
2a (R3 ) NO2, R1 ) R2 ) R4 ) H): 1H NMR (300 MHz, CD2-
3
3
Cl2, δ) 1.00 (d, JHH ) 6.60 Hz, 6H, CH(CH3)2), 1.29 (d, JHH
)
3c (R3 ) CH(CH)2CH ) R4, R1 ) R2 ) H): 1H NMR (300
MHz, C6D6, δ) -1.26 (s, 3H, Ni-CH3), 1.06 (d, 3JHH ) 6.60 Hz,
6H, CH(CH3)2), 1.39 (d, 3JHH ) 6.60 Hz, 6H, CH(CH3)2), 3.83 (s,
3
6.90 Hz, 6H, CH(CH3)2), 3.52 (sept, JHH ) 6.90 Hz, 2H,
2
CH(CH3)2), 3.91 (s, 6 H, NCH2N), 4.313 (dd, JHP ) 19.79 Hz,
2JHH ) 12.90 Hz, 6H, PCH2N), 6.50-6.59 (m, 3H, Ar), 6.79-
6.85 (m, 3H, Ar), 6.89-6.94 (m, 3H, Ar), 7.17-7.25 (m, 1H, Ar),
3
6H, NCH2N), 3.90 (sept, JHH ) 6.90 Hz, 2H, CH(CH3)2), 4.02
2
2
(dd, JHP ) 34.19 Hz, JHH ) 13.20 Hz, 6H, PCH2N), 6.96-7.14
8.00 (d, 4JHP ) 8.10 Hz, 1H, HCdN), 8.11-8.17 (m, 2H, Ar); 13
C
)
(m, 6H, Ar), 7.46-7.52 (m, 3H, Ar), 8.87 (s, 1H, HCdN); 13C
NMR (75 MHz, C6D6, δ) -18.68 (d, JCP ) 4.30 Hz, Ni-CH3),
1
NMR (75 MHz, CD2Cl2, δ) 22.23, 25.61, 28.74, 50.46 (d, JCP
2
14.86 Hz, P-C-N), 73.20 (d, 3JCP ) 6.26 Hz, N-C-N), 118.03,
122.68, 122.74, 126.10, 126.19, 128.67, 132.33, 136.02, 137.18,
137.23, 139.97, 142.49, 143.23, 148.73, 165.82, 171.03; 31P NMR
(121 MHz, CD2Cl2, δ) -57.76; yield 58.3%. Anal. Calcd for
C31H38N5O3PNi: C, 60.22; H, 6.19; N, 11.33. Found: C, 61.32;
H, 6.07; N, 10.70.
21.14, 23.03, 26.51, 48.33 (P-C-N), 71.39 (N-C-N), 107.33,
116.44, 120.09, 121.67, 123.70, 124.52, 125.05, 127.41, 132.99,
133.52, 139.67, 148.00, 157.31, 165.72; 31P NMR (121 MHz, C6D6,
δ) -60.88; yield 55.0%. Anal. Calcd for C30H39N4OPNi: C, 62.41;
H, 6.81; N, 9.70. Found: C, 62.83; H, 6.74; N, 10.48.
3d (R1 ) R3 ) Cl, R2 ) R4 ) H): 1H NMR (300 MHz, C6D6,
Direct Synthetic Approach for the Preparation of Nickel and
Palladium Salicylaldiminato PTA Complexes (3a-f and 4a-
f). To a 50 mL Schlenk flask containing (TMEDA)Ni(CH3)2 (200
mg, 0.976 mmol) in 10 mL of toluene at -30 °C was introduced
PTA (170 mg, 1.07 mmol) in 5 mL of methanol via cannula. To
this mixture, Ha (318 mg, 0.976 mmol) in 10 mL of toluene at
-30 °C was slowly cannulated into the flask, and the solution was
stirred for 30 min. Subsequently, the temperature was raised to room
temperature, and the light red solution was further stirred overnight.
After the solution was stirred overnight, the solvent was removed
in vacuo until approximately 5 mL remained, and 20 mL of cold
(-78 °C) pentane was added, resulting in the formation of a yellow
precipitate. The solid was collected by cold cannula filtration and
washed (3 × 5 mL) with cold (-78 °C) pentane, affording 3a in
60% yield (350 mg). The other (salicylaldiminato)nickel and
-palladium ((TMEDA)Pd(CH3)2 was used as the palladium source)
3
δ) -1.29 (s, 3H, Ni-CH3), 0.99 (d, JHH ) 6.60 Hz, 6H, CH-
(CH3)2), 1.38 (d, 3JHH ) 6.60 Hz, 6H, CH(CH3)2), 3.66 (sept, 3JHH
2
) 6.60 Hz, 2H, CH(CH3)2), 3.93 (s, 6H, NCH2N), 4.04 (dd, JHP
) 22.79 Hz, 2JHH ) 13.20 Hz, 6H, PCH2N), 6.71 (d, 3JHH ) 2.70
Hz, 1H, Ar), 7.07-7.13 (m, 3H, Ar), 7.38 (d, 3JHH ) 2.70 Hz, 1H,
Ar), 7.54 (s, 1H, HCdN); 13C NMR (75 MHz, C6D6, δ) -18.65
2
1
(d, JCP ) 3.62 Hz, Ni-CH3), 21.06, 22.77, 26.56, 48.61 (d, JCP
) 8.90 Hz, P-C-N), 71.40 (d, 3JCP ) 5.50 Hz, N-C-N), 115.47,
117.90, 121.69, 124.87, 126.47, 129.58, 131.53, 138.81, 146.38,
158.35, 162.81; 31P NMR (121 MHz, C6D6, δ) -59.36; yield 72.4%.
3e (R1 ) C6H5, R2 ) R3 ) R4 ) H): 1H NMR (300 MHz, C6D6,
δ) -1.27 (s, 3H, Ni-CH3), 1.04 (d, JHH ) 6.60 Hz, 6H, CH-
(CH3)2), 1.37 (d, JHH ) 6.60 Hz, 6H, CH(CH3)2), 3.66 (s, 6H,
NCH2N), 3.82 (t, JHH ) 6.90 Hz, 2H, CH(CH3)2), 3.91 (dd, JHP
) 42.29 Hz, 2JHH ) 13.20 Hz, 6 H, PCH2N), 6.54 (t, 3JHH ) 7.50
Hz, 1H, Ar), 6.94 (d, 3JHH ) 7.80 Hz, 1H, Ar), 7.07-7.22 (m, 6H,
3
3
3
2
3
3
Ar), 7.29 (d, JHH ) 6.90 Hz, 1H, Ar), 7.48 (d, JHH ) 6.90 Hz,
(12) Casiraghi, G.; Casnati, G.; Puglia, G.; Sartori, G.; Terenghi, G. J.
Chem. Soc., Perkin Trans. 1 1980, 1862.
(13) (a) Kaschube, W.; Po¨rschke, K. R.; Wilke, G. J. Organomet. Chem.
1988, 355, 525. (b) Byers, P. K.; Canty, A. J.; Jin, H.; Kruis, D.;
Markies, B. A.; Boersma, J.; Koten, G. V.; Hill, G. S.; Irwin, M. J.;
Rendina, L. M.; Puddephatt, R. J. Inorg. Synth. 1998, 32, 167.
2H, Ar), 7.90 (s, 1H, HCdN); 13C NMR (75 MHz, C6D6, δ) -19.43
2
1
(d, JCP ) 4.22 Hz, Ni-CH3), 21.08, 22.87, 26.51, 48.08 (d, JCP
) 12.22 Hz, P-C-N), 71.12 (d, 3JCP ) 4.60 Hz, N-C-N), 112.14,
117.81, 121.63, 124.64, 126.21, 128.35, 132.51, 133.10, 133.67,
Inorganic Chemistry, Vol. 42, No. 21, 2003 6917

Darensbourg et al.
139.31, 139.86, 147.01, 162.98, 163.90; 31P NMR (121 MHz, C6D6,
δ) -55.34; yield 69.7%. Anal. Calcd for C32H41N4OPNi: C, 65.47;
H, 6.98; N, 9.55. Found: C, 65.49; H, 6.95; N, 9.42.
Scheme 1
4a (R3 ) NO2, R1 ) R2 ) R4 ) H): 1H NMR (300 MHz, C6D6,
δ) -0.19 (d, 3JHP ) 3.28 Hz, 3H, Pd-CH3), 1.00 (d, 3JHH ) 6.90
Hz, 6H, CH(CH3)2), 1.32 (d, 3JHH ) 6.90 Hz, 6H, CH(CH3)2), 3.29
(sept, 3JHH ) 6.90 Hz, 2H, CH(CH3)2), 3.83 (s, 6H, NCH2N), 3.99
2
2
(dd, JHP ) 30.59 Hz, JHH ) 13.50 Hz, 6H, PCH2N), 6.65 (d,
3JHH ) 9.60 Hz, 1H, Ar), 7.14 (d, 3JHH ) 3.00 Hz, 1H, Ar), 7.21-
7.17 (m, 2H, Ar), 7.53 (d, 4JHP ) 10.20 Hz, 1H, HCdN), 8.11 (d,
(dd, 2JHP ) 20.09 Hz, 2JHH ) 13.50 Hz, 6H, PCH2N), 6.54 (t, 3JHH
) 6.90 Hz, 1H, Ar), 6.93 (m, 1H, Ar), 7.12 (m, 1H, Ar), 7.24 (t,
3JHH ) 7.20 Hz, 2H, Ar), 7.40 (m, 1H, Ar), 7.59-7.62 (m, 2H,
3
3
3JHH ) 2.70 Hz, 1H, Ar), 8.22 (dd, JHH ) 3.00 Hz, JHH ) 9.60
Hz, 1H, Ar); 13C NMR (75 MHz, C6D6, δ) -7.88 (d, 2JCP ) 17.73
1
Hz, Pd-CH3), 23.32, 25.00, 28.73, 50.95 (d, JCP ) 16.74 Hz,
4
Ar), 7.93 (d, JHP ) 11.70 Hz, 1H, HCdN); 13C NMR (75 MHz,
3
P-C-N), 73.60 (d, JCP ) 6.11 Hz, N-C-N), 118.54, 123.90,
C6D6, δ) -7.88 (d, 2JCP ) 54.59 Hz, Pd-CH3), 23.98, 25.82, 29.22,
127.44, 130.15, 135.21, 136.30, 141.02, 147.35, 166.63, 173.39;
31P NMR (121 MHz, C6D6, δ) -46.18; yield 95.7%. Anal. Calcd
for C26H36N5O3PPd: C, 51.71; H, 5.96; N, 11.60. Found: C, 52.16;
H, 6.30; N, 11.63.
1
3
51.50 (d, JCP ) 15.23 Hz, P-C-N), 73.70(d, JCP ) 7.62 Hz,
N-C-N), 113.53, 120.07, 123.61, 126.42, 126.64, 130.45, 135.83,
136.56, 141.17, 142.02, 147.67, 166.53, 166.34; 31P NMR (121
MHz, C6D6, δ) -46.16; yield 67.6%.
4b (R1 ) OCH3, R2 ) R3 ) R4 ) H): 1H NMR (300 MHz,
Polymerization of Ethylene Using 2a. To a 100 mL glass
miniclave reactor was added approximately 10 mL of degassed
water, followed by the addition of 2a (100 mg, 0.162 mmol) in 10
mL of toluene. Ethylene was added until the total pressure was 8
atm. The mixture was stirred for approximately 5 h at ambient
temperature. After the system was vented, the toluene layer was
separated and toluene was removed by rotoevaporation leaving
behind no polyethylene.
C6D6, δ) -0.18 (d, 3JHP ) 3.60 Hz, 3H, Pd-CH3), 1.01 (d, 3JHH
)
6.90 Hz, 6H, CH(CH3)2), 1.32 (d, 3JHH ) 6.60 Hz, 6H, CH(CH3)2),
3
3.47 (sept, JHH ) 6.90 Hz, 2H, CH(CH3)2), 3.54 (s, 3H, OCH3),
3
3.98-4.08 (m, 12H, PCH2N, NCH2N), 6.43 (t, JHH ) 7.21 Hz,
1H, Ar), 6.71 (m, 2H, Ar), 7.92 (d, 4JHH ) 11.43 Hz, 1H, HCdN);
2
13C NMR (75 MHz, C6D6, δ) -9.33 (d, JCP ) 13.80 Hz, Pd-
1
CH3), 23.36, 25.14, 28.62 51.28 (d, JCP ) 15.24 Hz, P-C-N),
56.46 (OCH3), 73.83 (d, 3JCP ) 7.62 Hz, N-C-N), 112.57, 115.46,
119.49, 123.94, 126.93, 128.61, 141.70, 148.46, 154.17, 161.82,
166.45; 31P NMR (121 MHz, C6D6, δ) -44.74; yield 91.2%. Anal.
Calcd for C27H39N4O2PPd: C, 54.97; H, 6.61; N, 9.50. Found: C,
55.21; H, 6.50; N, 9.41.
Results and Discussion
Initially, we attempted the synthesis of the PTA derivatives
of nickel salicylaldiminato complexes using the commonly
employed protocol of rapidly stirring a biphasic mixture
consisting of the PPh3 analogue complex (e.g., 1a) in toluene
with excess PTA in water at ambient temperature. However,
under these reaction conditions no ligand substitution oc-
curred. This procedure is generally successful since PTA has
smaller steric requirements and is a more donating ligand
than PPh3.14,15 Evidently, in this instance there is little PTA
in the organic phase and vice versa. This conclusion is
supported upon carrying out the reaction in a homogeneous
toluene/methanol mixture in which 1a and PTA were first
dissolved in toluene and methanol, respectively (Scheme 1).
That is, upon addition of the concentrated PTA solution to
a solution of 1a in toluene, a yellow precipitate formed
immediately. The identity of this yellow derivative (2a) was
confirmed to be the PTA analogue of 1a by NMR spectros-
copy, elemental analysis, and X-ray crystallography.
4c (R3 ) CH(CH)2CH ) R4, R1 ) R2 ) H): 1H NMR (300
MHz, C6D6, δ) -0.20 (d, 3JHP ) 3.90 Hz, 3H, Pd-CH3), 1.02 (d,
3
3JHH ) 6.90 Hz, 6H, CH(CH3)2), 1.30 (d, JHH ) 6.90 Hz, 6 Hz,
3
CH(CH3)2), 3.54 (sept, JHH ) 7.20 Hz, 2H, CH(CH3)2), 3.85 (s,
2
2
6H, N-C-N), 3.93 (dd, JHP ) 29.39 Hz, JHH ) 13.20 Hz, 6H,
PCH2N), 6.95-7.13 (m, 5H, Ar), 7.45(d, 3JHH ) 8.10 Hz, 1H, Ar),
3
3
7.52 (d, JHH ) 9.30 Hz, 1H, Ar), 7.57 (d, JHH ) 8.70 Hz, 1H,
Ar), 8.95 (d, JHP ) 11.95 Hz, 1H, HCdN); 13C NMR (75 MHz,
4
C6D6, δ) -8.88 (d, 2JCP ) 15.31 Hz, Pd-CH3), 23.45, 25.37, 28.63,
1
3
51.05 (d, JCP ) 15.31 Hz, P-C-N), 73.74 (d, JCP ) 7.62 Hz,
N-C-N), 119.23, 122.10, 124.04, 126.95, 127.38, 127.46, 128.92,
129.69, 129.89, 136.38, 137.29, 142.13, 149.39, 160.09; 31P NMR
(121 MHz, C6D6, δ) -47.18; yield 98.5%. Anal. Calcd for
C30H39N4OPPd: C, 59.20; H, 6.41; N, 9.21. Found: C, 60.01; H,
6.48; N, 8.81.
4d (R1 ) R3 ) Cl, R2 ) R4 ) H): 1H NMR (300 MHz, C6D6,
δ) -0.19 (d, 3JHP ) 3.30 Hz, 3H, Pd-CH3), 1.00 (d, 3JHH ) 6.90
Hz, 6H, CH(CH3)2), 1.34 (d, 3JHH ) 6.90 Hz, 6H, CH(CH3)2), 3.34
(sept, 3JHH ) 6.60 Hz, 2H, CH(CH3)2), 3.98 (s, 6H, NCH2N), 3.99
(d, 2JHP ) 20.39 Hz, 2JHH ) 12.90 Hz, 6H, PCH2N), 6.77 (d, 3JHH
) 2.70 Hz, 1H, Ar), 7.13-7.18 (m, 3H, Ar), 7.51 (d, 3JHH ) 3.00
Hz, 1H, Ar), 7.57 (d, 4JHP ) 10.50 Hz, 1H, HCdN); 13C NMR (75
MHz, C6D6, δ) -8.40 (d, 2JCP ) 13.73 Hz, Pd-CH3), 23.33, 25.07,
28.67, 51.24 (d, 1JCP ) 16.82 Hz, P-C-N), 73.66 (d, 3JCP ) 7.62
Hz, N-C-N), 116.66, 120.34, 124.04, 127.32, 129.40, 133.93,
134.72, 141.16, 147.69, 162.56, 166.02; 31P NMR (121 MHz, C6D6,
δ) -44.54; yield 82.8%. Anal. Calcd for C26H35N4OPCl2Pd: C,
49.69; H, 5.57; N, 8.92. Found: C, 50.49; H, 5.59; N, 8.92.
4e (R1 ) C6H5, R2 ) R3 ) R4 ) H): 1H NMR (300 MHz, C6D6,
δ) -0.17 (d, 3JHP ) 3.60 Hz, 3H, Pd-CH3), 1.02 (d, 3JHH ) 6.90
Hz, 6H, CH(CH3)2), 1.31 (d, 3JHH ) 6.90 Hz, 6H, CH(CH3)2), 3.47
(sept, 3JHH ) 7.20 Hz, 2H, CH(CH3)2), 3.70 (s, 6H, NCH2N), 3.91
1
The H NMR spectrum of 2a in CD2Cl2 displayed many
characteristic resonances. Importantly, the signals corre-
sponding to the terminal isopropyl groups (CH(CH3)2) were
split into two sets of doublets indicating a rotation barrier
of the aniline moiety upon ligand coordination, consistent
with what is observed for other salicylaldiminato com-
plexes.16,17 Hydrogen resonances due to the PTA ligand are
located between 3.7 and 4.2 ppm, with the NCH2N hydrogens
(14) DeLerno, J. R.; Trefonas, L. M.; Darensbourg, M. Y.; Majeste, R. J.
Inorg. Chem. 1976, 15, 816.
(15) Tolman, C. A. Chem. ReV. 1977, 77, 313.
(16) Darensbourg, D. J.; Rainey, P.; Yarbrough, J. Inorg. Chem. 2001, 40,
986.
(17) Klein, H.; Bickelhaupt, A. Inorg. Chim. Acta 1996, 248, 111.
6918 Inorganic Chemistry, Vol. 42, No. 21, 2003

Nickel and Palladium Salicylaldiminato Complexes
Table 1. Crystallographic data for complexes 1a, 2a, 4a,b, and 3b′
1a
2a
4a
4b
3b′
empirical formula
fw
C43H41O3N2PNi
723.46
C32H40O3N5Cl2PNi
703.27
C26H36O3N5PPd
603.99
C27H39O2N4PPd
588.99
C40H48O4N2Ni
679.08
cryst syst
space group
V, Å3
triclinic
P1h
1780.6(4)
2
monoclinic
P21/n
3169.3(5)
4
triclinic
P1h
1336.4(18)
2
orthorhombic
Pbca
5384(4)
monoclinic
P21/n
3947.3(8)
4
Z
8
a, Å
b, Å
c, Å
9.5718(12)
12.0581(16)
15.784(2)
90.158(2)
99.214(3)
97.906(3)
110
1.349
0.633
4.79
8.79
10.1182(9)
20.5266(18)
15.5267(13)
9.127(7)
10.888(8)
15.293(14)
69.195(14)
76.151(19)
71.950(12)
110
1.501
0.791
8.00
9.41
12.249(6)
18.457(8)
23.812(11)
8.9737(11)
20.255(3)
21.720(3)
R, deg
â, deg
100.643(2)
90.919(3)
γ, deg
T, K
110
110
110
d(calc), g/cm3
abs coeff, mm-1
R,a % [I > 2σ(I)]
Rw,a %
1.474
0.874
8.10
1.453
0.780
3.43
5.23
1.286
0.669
6.11
19.85
14.62
a R ) ∑||Fo| - |Fc||/∑Fo and RwF ) {[∑w(Fo - Fc)2]/∑wFo }1/2
.
2
the M-CH3 protons are consistent with other group 10
complexes, occurring in the 0 to -1 ppm range.17,18 Interest-
ingly, the nickel derivatives, 3, do not exhibit 31P coupling
of the methyl and ketimine hydrogen atoms to PTA. This is
in contrast to the SHOP catalyst developed by Keim and
co-workers, where 31P coupling to the methyl hydrogens was
observed (JHP ∼7.4 Hz).18a Complex 4, however, does exhibit
31P coupling (JHP ∼ 5 Hz) of the Pd-CH3 hydrogen atoms
to PTA. A representative 1H NMR spectrum of 4d is
presented in Figure 3.
The 13C NMR spectra of complexes 3 and 4 also display
unique resonances. The Ni-CH3 carbon resonances are
observed in the -15 to -20 ppm range with 31P coupling
on the order of 4 Hz. In contrast, the Pd-CH3 carbon
resonances are displayed further downfield (-7 to -9 ppm)
with a larger 31P coupling (JCP ∼ 15 Hz). The NCH2N and
PCH2N carbons of the PTA ligand are observed in the 50
and 70 ppm region, respectively. 31P coupling is also
observed for the PCH2N carbons with approximate values
of 8 and 15 Hz for 3 and 4, respectively. As expected, a
smaller 31P coupling constant is associated with the NCH2N
carbon, with JCP values on the order of 5-7 Hz for 3 and 4,
respectively. The 31P NMR resonances for these complexes
are observed at approximately -60 and -45 ppm for 3 and
4, respectively.
The solid-state structure of the nickel(II) salicylaldimato
derivative containing the PPh3 ligand (complex 1e in Figure
1) has been reported by Grubbs and co-workers.6a Herein,
we describe the solid-state structure of an analogous complex,
1a, for comparative purposes. Crystals of 1a suitable for
X-ray analysis were obtained from a solution of 1a in toluene
maintained at -20 °C for approximately 2 weeks. Tables 1
and 2 contain the crystallographic data and selected bond
distances and angles, whereas a thermal ellipsoid representa-
tion of complex 1a may be found in Figure 4. As expected
for four-coordinate d8 metal complexes, the structure of 1a
adopts a nearly ideal square planar geometry with N-Ni-P
and C-Ni-O bond angles of 176.61(7) and 171.64(10)°,
1
Figure 3. Representative H NMR spectrum of 4d.
Scheme 2
appearing as singlets. The resonance due to the PCH2N
hydrogen is observed as a doublet of doublets as a result of
coupling to phosphorus and the geminal proton (2JHP ∼ 25
2
Hz and JHH ∼ 13 Hz). The ketimine (HCdN) hydrogen is
displayed as a doublet near 8.0 ppm with phosphorus
coupling on the order of 8.1 Hz, as previously reported for
the PPh3 derivatives (1).6 The 31P NMR resonance in 2a
(-57.76 ppm) is shifted in CD2Cl2 40.5 ppm downfield from
free PTA at -98.3 ppm in water.
Alternatively, salicylaldiminato PTA complexes may be
prepared via a direct synthetic approach in which (TMEDA)M-
(CH3)2 (M ) Ni, Pd) is used as the metal precursor.13 With
liberation of methane in the process, the metal precursor is
reacted with PTA and the corresponding salicylaldimine at
-30 °C to yield the nickel and palladium complexes (3 and
1
4) in moderate to quantitative yields (Scheme 2). The H
NMR resonance of the ketimine hydrogen in 4 displays
phosphorus coupling (JHP ∼ 11 Hz) with chemical shift
values similar to those observed for 1.6 Resonances due to
(18) (a) Klein, H.; Keim, W.; Heinicke, J.; He, M.; Dal, A.; Hetche, O.;
Flo¨rke, U.; Haupt, H. Eur. J. Inorg. Chem. 2000, 431. (b) Klein, H.;
Wiemer, T. Inorg. Chim. Acta 1988, 154, 21.
Inorganic Chemistry, Vol. 42, No. 21, 2003 6919

Darensbourg et al.
Table 2. Selected Bond Distances (Å) and Angles of Complexes 1a, 2a, 4a,b, and 3b′a
1a
2a
Bond Distances
4a
4b
3b′
M(1)-C(1)
M(1)-P(1)
M(1)-O(1)
M(1)-N(1)
1.893(3)
1.893(6)
2.1345(18)
1.888(4)
1.964(5)
2.024(9)
2.199(3)
2.094(5)
2.097(6)
2.036(4)
2.1995(12)
2.068(2)
2.087(3)
2.1754(8)
1.9141(19)
1.947(2)
1.827(3)
1.907(3)
Bond Angles
169.7(2)
83.85(18)
87.68(13)
92.99(18)
173.33(15)
C(1)-M(1)-O(1)
P(1)-M(1)-C(1)
O(1)-M(1)-N(1)
P(1)-M(1)-O(1)
P(1)-M(1)-N(1)
N(1)-M(1)-N(2)
171.64(10)
84.86(8)
92.86(4)
89.65(6)
176.61(7)
176.1(3)
87.7(3)
88.6(2)
90.44(17)
175.82(19)
177.13(13)
86.26(11)
89.67(11)
90.98(8)
93.37(12)
173.27(8)
178.04(12)
a Estimated standard deviations are given in parentheses.
Figure 5. Thermal ellipsoid representation of 2a showing 50% probability
ellipsoids.
Figure 4. Thermal ellipsoid representation of 1a showing 50% probability
ellipsoids.
respectively. The isopropyl groups on the 2,6-diisopropyl-
benzimine lie perpendicular to the plane created by the N,
P, O, and C atoms. Such positioning allows these groups to
effectively shield the axial faces of the metal center to
enhance the rate of enchainment relative to chain transfer.4,6
The positioning is also in accord with the solution spectro-
scopic data where two sets of doublets are observed for the
terminal isopropyl hydrogens in the 1H NMR spectrum. The
Ni-C bond distance was found to be nearly identical with
that observed in 1e, having a value of 1.893(3) Å. Consistent
with the Ni-P bond distance of 1e (2.172(2) Å), in 1a, the
distance associated with this bond was found to be 2.1754-
(8) Å, indicating little steric interaction of large groups in
the R1 position with PPh3.
Figure 6. Thermal ellipsoid representation of 4a showing 50% probability
ellipsoids.
Using the same crystal growth technique as described for
1a, suitable crystals for X-ray analysis of several of the other
derivatives were obtained. The thermal ellipsoid drawings
of two such complexes (2a and 4a) are shown in Figures 5
and 6, respectively. Selected bond lengths and angles are
listed in Table 2. Similar to the solid-state structures of 1a,e,
these complexes adopt the typical square planar geometry
(e.g., N-M-P angles in 2a and 4a were found to be 173.33-
(15) and 173.27(8)°). The nickel-carbon bond distance to
the phenyl ligand in 2a at 1.893(6) Å is identical to that
seen in complex 1a. Comparing the Ni-P bond distances in
1a of 2.1754(8) Å to that seen in 2a of 2.1345(18) Å, we
notice a shortening of the bond distance in 2a of ap-
proximately 0.05 Å. This decrease in Ni-P bond length upon
going from PPh3 to PTA is consistent with the smaller cone
angle (103°) and more basic nature of PTA as compared to
PPh3. The Pd-P bond distance in 4a was found to be 2.199-
(3) Å, which is slightly shorter than that observed in cis-
PdCl2(PTA)2 of 2.226(5) Å.19 The difference in the Pd-P
bond length in 4a vs the corresponding Ni-P bond distance
in 2a of 0.07 Å is less than the covalent radii difference in
nickel and palladium of 0.11 Å. The Pd-C(methyl) bond
(19) Darensbourg, D. J.; Decuir, T. J.; Stafford, N. W.; Robertson, J. B.;
Draper, J. D.; Reibenspies, J. H.; Katho´, A.; Joo´, F. Inorg. Chem.
1997, 36, 4218.
6920 Inorganic Chemistry, Vol. 42, No. 21, 2003

Nickel and Palladium Salicylaldiminato Complexes
Figure 9. Polymerization of ethylene using 2a in a biphasic toluene/water
biphasic solvent system.
corresponding parameters found in the parent (salicylaldi-
minato)Ni(methyl)(PTA) derivative, 3b, of 1.931(6) and
1.881(5) Å, respectively. These decreases in bond lengths
are anticipated upon loss of the electron-donating phosphine
ligand. The formation of such bis complexes is clearly
undesirable in polymerization processes and has been an
issue for SHOP-type systems which utilize higher temper-
atures and pressure to produce high molecular weight
polyethylene.20
Figure 7. Thermal ellipsoid representation of 4b showing 50% probability
ellipsoids.
Ethylene Polymerization. The polymerization of ethylene
using 1 was first reported in 1998.6 Although derivatives
incorporating large groups in the R1 position of the salicyl-
aldimine (e.g. 1e) did not require a phosphine scavenger,
other active derivatives such as 1a did in fact necessitate
the use of the air-sensitive cocatalyst, Ni(COD)2. Our
approach to bypass the need for a cocatalyst involves the
use of water-soluble phosphines to facilitate the dissociation
process illustrated in Figure 9. The dissociation of PTA into
the aqueous phase would effectively allow the formation of
the active catalyst in the organic phase, initiating the
polymerization. Furthermore, upon PTA dissociation, reen-
tering of this phosphine into the organic phase would not
occur, since PTA is not soluble in toluene. This was
previously quite evident upon failing to synthesize 2 via a
biphasic toluene/water ligand replacement process (vide
supra).
For all attempts to polymerize ethylene at ambient tem-
perature using 2a as catalyst utilizing a biphasic toluene/
water (1/1) solvent system under 8 atm of ethylene pressure,
no polymer formation was observed. Raising the temperature
to 70 °C resulted in Ni(0) formation, and analysis of the
water phase revealed the presence of the phosphine oxide,
PTAdO. The formation of phosphine oxide has also been
observed in other rhodium and ruthenium catalytic systems.11c
Figure 8. Thermal ellipsoid representation of 3b′ showing 50% probability
ellipsoids.
length was determined to be 2.024(9) Å in 4a. The solid-
state structure of the methoxy derivative, complex 4b, was
also determined, and its structure is depicted in the thermal
ellipsoid representation in Figure 7. In this instance, the
electron-donating ability of the OCH3 group is not reflected
in the Pd-P bond distance of 2.1995(12) Å, which is the
same as that observed in 4a (Table 2). However, the Pd-C
bond distance increases by approximately 0.025 Å. Never-
theless, electron-donating effects have been shown to greatly
decrease the activity of the catalyst as evident in a decrease
in the reported turnover number from 253 kg of polyethylene/
mol of Ni for 1a to 13.3 kg of polyethylene/mol of Ni for
1b.6a X-ray structural data for other closely related (salicyl-
aldiminato)M(methyl)(PTA) derivatives (M ) Ni, 3b,d; M
) Pd, 4c,d) are included in the Supporting Information.
In a few cases during crystal growth over extended periods
of time, ligand redistribution occurred with concomitant
formation of the thermodynamically stable bis(salicylaldi-
minato) complexes. These crystals are black in color and
appear to be relatively stable in air. The solid-state structure
of one such isolated species, 3b′, has been determined by
X-ray crystallography. A thermal ellipsoid drawing of 3b′
is shown in Figure 8.
Concluding Remarks
Herein, we have reported the synthesis of methyl and
phenyl derivatives of nickel and palladium salicylaldiminato
complexes containing the water-soluble phosphine (PTA) in
excellent yields. These complexes have all been structurally
characterized in the solid state by X-ray crystallography. Thus
far, we have been unproductive in catalyzing the polymer-
ization of ethylene with (salicylaldiminato)Ni(Ph)(PTA) in
(20) Ittel, S. D.; Calabrese, J.; Klabunde, U.; Mulhaupt, R.; Herskovitz,
T.; Janowicz, A. H. J. Polym. Sci., Part A: Polym. Chem. 1987, 25,
1989.
The Ni-N and Ni-O bond distances observed in complex
3b′ of 1.907(3) and 1.827(3)Å are slightly shorter than the
Inorganic Chemistry, Vol. 42, No. 21, 2003 6921

Darensbourg et al.
a biphasic medium. This is undoubtedly due to the stability
of the Ni-PTA bond, i.e., the high temperature required to
effect phosphine dissociation in this instance.
preparing the meta-TPPTS analogue of complex 2a. This is
most likely due to the significantly larger cone angle in meta-
22
TPPTS of 170° vs that of PPh3 (145°).15 In principle the
approach outlined in Figure 9 appears to be fundamentally
sound if a set of suitable conditions can be found. It is
probable that employing less sterically encumbered water-
soluble triphenylphosphine derivatives as ligands will lead
to metal complexes which have metal-PR3 bond dissociation
energies similar to that for metal-PPh3, thereby making them
effective catalysts for this polymerization process.23
With regard to this latter point, we have attempted to
determine the rate of Ni-PTA bond dissociation in 2a
initially utilizing a large excess (10 equiv) of PPh3 as
incoming ligand. At ambient temperature, as well as at
35 °C, the 31P signal of PTA in 2a at -57.8 ppm was
unaffected over an extended reaction period. On the other
hand, a similar experiment involving the use of the more
basic and less sterically hindered phosphine, PMe3, as enter-
ing ligand resulted in immediate displacement of the PTA
ligand at ambient temperature. This latter process is evidently
taking place via an associative mechanism, a common
occurrence in square-planar nickel(II) complexes. Hence,
qualitatively it is apparent that the dissociation of PTA from
2a is not a facile process at modest reaction temperatures.
The slow initiation step when employing complex 2a as
catalyst precursor for the polymerization of ethylene might
be overcome by preparing nickel(II) derivatives bearing other
water-soluble phosphines. For example, the water-soluble
meta-TPPTS ligand, which is electronically almost identical
with PPh3 and thereby expected to have similar or enhanced
Ni-PR3 dissociation rates, would seem to be quite appropri-
ate.21 Unfortunately, we have thus far been unsuccessful at
Acknowledgment. Financial support from the National
Science Foundation (Grants CHE 99-10342 and CHE 02-
34222; Grant CHE 98-07975 for the purchase of X-ray
equipment) and the Robert A. Welch Foundation is greatly
appreciated.
Supporting Information Available: Tables of anisotropic
thermal parameters, bond distances, and bond angles for complexes
1a, 2a, 3b, 3b′, 3d, and 4a-d, in CIF format, and selected
crystallographic data, bond distances and angles, and thermal
ellipsoid representations for complexes 3b,d and 4c,d, in pdf format.
This material is available free of charge via the Internet at
http://pubs.acs.org.
IC0341284
(22) Darensbourg, D. J.; Bischoff, C. J.; Reibenspies, J. H. Inorg. Chem.
1991, 30, 1144.
(23) Stelzer, O.; Sheldrick, W. S.; Weferling, N.; Herd, O.; Hebler, A.;
Langhans, K. P. J. Organomet. Chem. 1994, 475, 99.
(21) Darensbourg, D. J.; Bischoff, C. J. Inorg. Chem. 1993, 32, 47.
6922 Inorganic Chemistry, Vol. 42, No. 21, 2003