Pentafluoroaryl transfer from tris(pentafluorophenyl)boron hydrate to nickel. Synthesis and X-ray crystal structure of (PPh2CH2C(O)Ph)Ni(C6F5)2

Article abstract of DOI:10.1021/om000606w

The reaction of tris(pentafluorophenyl)boron trihydrate, B(C₆F₅)₃·3H₂O (1), with the nickel dimer [(PPh₂CH double bond C(O)Ph)Ni(Ph)]₂ (3) (synthesized by phosphine abstraction from (PPh

Full text of DOI:10.1021/om000606w

Organometallics 2000, 19, 3983-3990
3983
P en ta flu or oa r yl Tr a n sfer fr om
Tr is(p en ta flu or op h en yl)bor on Hyd r a te to Nick el.
Syn th esis a n d X-r a y Cr ysta l Str u ctu r e of
(P P h 2CH2C(O)P h )Ni(C6F 5)2
Heather A. Kalamarides, Suri Iyer, J ohn Lipian, and Larry F. Rhodes*
The BFGoodrich Company, 9921 Brecksville Road, Brecksville, Ohio 44141
Cynthia Day
Crystalytics Company, P.O. Box 82286, Lincoln, Nebraska 68501
Received J uly 14, 2000
The reaction of tris(pentafluorophenyl)boron trihydrate, B(C6F5)3‚3H2O (1), with the nickel
dimer [(PPh2CHdC(O)Ph)Ni(Ph)]2 (3) (synthesized by phosphine abstraction from (PPh2-
CHdC(O)Ph)Ni(Ph)(PPh3) (2)) yields (PPh2CH2C(O)Ph)Ni(C6F5)2 (4). In this reaction a
number of processes occur: fragmentation of the dimer 3, transfer of two protons from 1,
and transfer of two C6F5 groups from boron to nickel. In solution, 4 loses one C6F5 ligand as
pentafluorobenzene due to proton transfer from the chelating ligand, forming the dimer
[(PPh2CHdC(O)Ph)Ni(C6F5)]2 (5). Compounds 4 and 5 were characterized by NMR and MS
techniques. Compound 4 was independently synthesized by the reaction of PPh2CH2C(O)-
Ph and (η6-toluene)Ni(C6F5)2. The crystal structure of 4 was determined using single-crystal
X-ray diffraction methods. Details concerning the synthesis and characterization of 1-3 are
also given.
In tr od u ction
While B(C6F5)3 can readily be employed in aromatic
hydrocarbon solvents, Siedle and co-workers recognized
the limitations of B(C6F5)3 in aliphatic hydrocarbons and
sought methods of increasing the solubility of this type
of reagent in such media.13 This objective was achieved
by forming adducts of B(C6F5)3 with, for example, long-
chain alcohols and mercaptans. In addition, water
complexes such as B(C6F5)3‚3H2O (1) were also reported.
These adducts of B(C6F5)3 transform this strong Lewis
acid into a Brφnsted acid which can react with dialkyl-
metallocene complexes to form the same catalytically
active, cationic species made with B(C6F5)3. Of course,
Brφnsted acids derived from B(C6F5)3 generate the
cationic metallocenium complexes by protonation of the
metal-alkyl bond with concomitant formation of, in the
case of the alcohol adduct, [ROB(C6F5)3]- as the anion
rather than by the alkyl abstraction reaction preferred
by B(C6F5)3.
Tris(pentafluorophenyl)boron, owing to its pentafluoro-
phenyl substituents, is a very strong Lewis acid.1 As
such, this reagent is effective in a number of chemical
transformations. Among them are aldol and Michael
reactions of silyl enol ethers with carbonyls,2 the Diels-
Alder addition of dienes with R,â-unsaturated alde-
hydes,3 and the stereoselective rearrangement of ep-
oxides.4
Tris(pentafluorophenyl)boron plays an active role in
polymerizations. It can initiate the propagation of
cationically polymerizable monomers such as vinyl
ethers, N-vinylcarbazole,5 and styrene.6
Tris(pentafluorophenyl)boron probably has found the
most utility as an activator in olefin polymerization.
Marks and co-workers have shown that B(C6F5)3 can
abstract the methyl group from (Me5C5)2ZrMe2 to form
a cationic zirconocene complex.7 Others have reported
related reactions with different group 4 alkyl complexes8
which yield catalysts active in styrene,9 propylene,10 and
functional olefin polymerization11 among others.12
(8) (a) Pellecchia, C.; Grassi, A.; Immirzi, A. J . Am. Chem. Soc. 1993,
115, 1160. (b) Pellecchia, C.; Grassi, A.; Zambelli, A. J . Mol. Catal.
1993, 82, 57. (c) Pellecchia, C.; Immirzi, A.; Grassi, A.; Zambelli, A.
Organometallics 1993, 12, 4473. (d) Bochmann, M.; Lancaster, S. J .;
Hursthouse, M. B.; Abdul Malik, K. M. Organometallics 1994, 13, 2235.
(9) (a) Pellecchia, C.; Pappalardo, D. Oliva, L.; Zambelli, A. J . Am.
Chem. Soc. 1995, 117, 6593. (b) Grassi, A.; Zambelli, A.; Laschi, F.
Organometallics 1996, 15, 480. (c) See ref 6.
* To whom correspondence should be addressed. E-mail: rhodes@
brk.bfg.com. Fax: 216-447-5464.
(1) (a) Massey, A. G.; Park, A. J . Organomet. Chem. 1964, 2, 245.
(b) Massey, A. G.; Park, A. J . Organomet. Chem. 1966, 5, 218.
(2) Ishihara, K.; Hanaki, N.; Yamamoto, H. Synlett 1993, 577.
(3) Ishihara, K.; Hanaki, N.; Funahashi, M.; Miyata, M.; Yamamoto,
H. Bull. Chem. Soc. J pn. 1995, 68, 1721.
(4) Ishihara, K.; Hanaki, N.; Yamamoto, H. Synlett 1995, 721.
(5) Wang, Q.; Baird, M. C. Macromolecules 1995, 28, 8021.
(6) Wang, Q.; Quyoum, R.; Gillis, D. J .; Tudoret, M.-J .; J eremic, D.;
Hunter, B. K.; Baird, M. C. Organometallics 1996, 15, 693.
(7) Yang. X.; Stern, C. L.; Marks, T. J . J . Am. Chem. Soc. 1991, 113,
3623.
(10) Bochman, M.; Lancaser, S. J . Makromol. Chem., Rapid Com-
mun. 1993, 14, 807.
(11) Kesti, M. R.; Coates, G. W.; Waymouth, R. M. J . Am. Chem.
Soc. 1992, 114, 9679.
(12) (a) Soga, K.; Deng, H.; Yano, T.; Shiono, T. Macromolecules
1994, 27, 7938. (b) Barsan, F.; Baird, M. C. J . Chem. Soc., Chem.
Commun. 1995, 1065.
(13) (a) Siedle, A. R.; Lamanna, W. M.; Newmark, R. A.; Stevens,
J .; Richardson, D. E.; Ryan, M. Makromol. Chem., Macromol. Symp.
1993, 66, 215. (b) Siedle, A. R.; Lamanna, W. M. U.S. Patent 5296433,
1994.
10.1021/om000606w CCC: $19.00 © 2000 American Chemical Society
Publication on Web 08/23/2000

3984 Organometallics, Vol. 19, No. 19, 2000
Kalamarides et al.
(C2H4)2 were purchased from Strem. PPh3dCHC(O)Ph was
purchased from SPECS. (η6-toluene)Ni(C6F5)2 was synthesized
according to a literature procedure.24 PPh2CH2C(O)Ph was
synthesized by a known procedure25 but was isolated from a
5% ethyl acetate in cyclohexane solution followed by washing
with pentane.
The success of metallocenium catalysts in olefin
polymerization is in part due to the ability of the
gegenion to remain stable and relatively weakly coor-
dinating. Early attempts by J ordan et al. to access
cationic zirconocene complexes by reacting Cp2ZrMe2
with AgPF6 resulted in transfer of the fluoride from PF6
to zirconium.14 To circumvent this problem, BPh4- was
employed as the counterion.15 However, anions of this
type can form π-complexes with coordinately unsatur-
ated metal centers16 and are susceptible to electrophilic
attack.17 Boron-containing anions supported by fluori-
nated aryl groups, on the other hand, are thought to be
relatively inert18 and form the basis for a growing
number of patents and publications on metallocene
olefin polymerization catalysis.19,20
Despite the increasing interest in the reactivity
patterns of B(C6F5)3 and Brφnsted acids derived from
B(C6F5)3 with group 4 metallocenes, relatively few
reports have appeared involving later transition metal
complexes and B(C6F5)3 or its Brφnsted acid deriva-
tives.21 In the late 1980s workers at DuPont described
nickel complexes which, in the presence of phosphine
acceptors, yielded high-molecular-weight polyethylene.22
Given the ability of B(C6F5)3 to form donor/acceptor
complexes with phosphine,23 we investigated the inter-
action of several nickel complexes of the type described
by Klabunde et al. with B(C6F5)3 as well as with the
water adduct of B(C6F5)3. In particular, we found that
B(C6F5)3‚3H2O reacts with the dimer complex [(PPh2-
CHdC(O)Ph)Ni(Ph)]2 by first protonation of the chelat-
ing ligand and the Ni-Ph bond followed by transfer of
an unprecedented two pentafluorophenyl groups from
boron to nickel.
In str u m en ta l An a lysis. The 1H, 19F, and 31P NMR spectra
were recorded on a Bruker AMX-500 NMR spectrometer
operating at 500.14, 470.53, and 202.47 MHz, respectively,
1
unless otherwise specified. Chemical shifts for H NMR spectra
were referenced using internal solvent resonances and are
reported relative to tetramethylsilane. The 19F NMR spectra
were referenced to external CFCl3, while the 31P NMR spectra
were referenced to external H3PO4.
The infrared spectra were recorded on a Perkin-Elmer
Model 1800 spectrometer using polystyrene film as a standard.
Samples were prepared as Nujol mulls on KBr salt plates in
the drybox.
The mass spectra (FD, FI, and CI) were recorded on a
Finnegan MAT 95Q instrument.
B(C6F 5)3. Solid B(C6F5)3 was prepared by the following
method.26 A 3.15 wt % solution of B(C6F5)3 in Isopar E (75.0
mL, 3.33 mmol) was placed in a 200 mL Kjeldahl flask
equipped with a magnetic stirbar. The solution was removed
in vacuo to give a white powder. The powder was sublimed
(95 °C, approximately 0.1 mmHg) to yield white needles. 19F
NMR (C6D6): δ -129.1 (s, 2F), -142.0 (s, 1F), and -160.3 (s,
2F). IR (Nujol): 2953 sh, 2924 vs, 2855 m, 1648 m, 1524 m,
1475 s, 1383 m, 1321 m, 1166 m, 1151 w, 1020 m, 973 s, 784
w, 667 w, 620 w cm-1
.
B(C6F 5)3‚3H2O (1). The synthesis of 1 reported here is
based on a prior publication.27 A 3.15 wt % solution of B(C6F5)3
in Isopar E (50 mL, 2.22 mmol) was placed in a 200 mL
Kjeldahl flask equipped with a magnetic stirbar. Approxi-
mately 50 mL of dry, degassed cyclohexane was added to this
solution followed by 3 equiv of degassed, demineralized water
(0.120 mL, 6.68 mmol). Precipitation of a white, microcrystal-
line solid occurred. The slurry was stirred for 30 min, after
which time the solution was decanted and the solid was dried
in vacuo. Yield: 0.826 g (66%). 19F NMR (C6D6): δ -134.7
(apparent d, 2F), -154.7 (apparent t, 1F), -162.6 (apparent
t, 2F). FI-MS: m/z 512 [M+]; evidently loss of water occurred
in the mass spectrometer. IR (Nujol): 3666 m, 3597 m, 3499
m, 2950 sh, 2920 s, 1647 m, 1602 m, 1520 s, 1468 s, 1379 m,
1288 m, 1111 m, 1098 m, 969 s, 859 w, 797 w, 771 w, 676 w,
Exp er im en ta l Section
Gen er a l Con sid er a tion s. All manipulations of organome-
tallic complexes were carried out under an atmosphere of
prepurified argon or nitrogen using standard Schlenk or
drybox techniques. Toluene was dried by refluxing over sodium
benzophenone ketyl, distilled under nitrogen, and stored over
4A molecular sieves. THF was dried over sodium benzophe-
none ketyl under nitrogen, vacuum-transferred, and stored
over 4A molecular sieves. Cyclohexane was dried by refluxing
over sodium under nitrogen, distilled, and stored over 4A
molecular sieves.
614 w cm-1
(P P h 2CHdC(O)P h )Ni(P h )(P P h 3) (2). The synthesis of 2
reported here is based on two previous publications.28
.
A
toluene slurry (150 mL) of PPh3 (5.00 g, 19.1 mmol) and the
ylide PPh3dCHC(O)Ph (7.30 g, 19.1 mmol) was added to a
chilled (0 °C) toluene slurry (80 mL) of Ni(COD)2 (5.30 g, 19.1
mmol). Afterward, the mixture became a red-brown slurry. The
mixture was warmed to room temperature and stirred for 21
h. This mixture was then heated to 50 °C for 2 h, cooled to
room temperature, and stirred for an additional 16 h. The
mixture was filtered to give a red-brown filtrate which upon
removal of solvent in vacuo gave a brown residue. The residue
was dissolved in toluene (50 mL), and from this solution a tan
precipitate began to form upon addition of 50 mL of hexane.
The mixture was cooled at -20 °C to give a gold-tan solid
which was filtered, washed with hexane, and dried. Yield: 10.5
g (79%). 1H NMR (C6D6): δ 7.57 (m, 12H), 7.00 (m, 20 H), 6.60
Tris(pentafluorophenyl)boron was obtained from Akzo-Nobel
as a 3.15 wt % solution in Isopar E. Ni(COD)2 and Rh(acac)-
(14) J ordan, R. F.; Dasher, W. E.; Echols, S. F. J . Am. Chem. Soc.
1986, 108, 8, 1718.
(15) J ordan, R. F.; Bajgur, C. S.; Willet, R.; Scott, B. J . Am. Chem.
Soc. 1986, 108, 7410.
(16) Bochman, M. Angew. Chem., Int. Ed. Engl. 1992, 31, 1181 and
references therein.
(17) Hlatky, G. G.; Turner, H.; Eckman, R. R. J . Am. Chem. Soc.
1989, 111, 2728.
(18) Brookhart, M.; Grant, B.; Volpe, A. F. Organometallics 1992,
11, 3920.
(19) See, for example: (a) Turner, H. W.; Canich, J . M.; Folie, B. J .
U.S. Patent 5408017, 1995. (b) Lapointe, R. E.; Stevens, J . C.; Nickias,
P. N.; McAdon, M. H. EP Patent 520732, 1992.
(20) Fluorinated borate anions can participate in donor interactions
with coordinatively unsaturated metal cations and undergo C-F
activation in some cases: Kiplinger, J . L.; Richmond, T. G.; Osterberg,
C. E. Chem. Rev. 1994, 94, 373 and references therein.
(21) (a) Taube, R.; Wache, S.; Sieler, J . J . Organomet. Chem. 1993,
459, 335. (b) Cooley, N. A.; Kirk, A. P. EP Patent 619335, 1994.
(22) (a) Klabunde, U.; Mulhaupt, R.; Herskovitz, T.; J anowicz, A.
H.; Calabrese, J .; Ittel, S. D. J . Polym. Sci., Part A: Polym. Chem.
1987, 25, 1989. (b) Klabunde, U. U.S. Patent 4716205, 1987.
(23) Bradley, D. C.; Hawkes, G. E.; Haycock, P. R.; Sales, K. D.;
Zheng, D. H. Philos. Trans. R. Soc. London 1994, 348, 315.
(24) Brezinski, M. M.; Klabunde, K. J . Organometallics 1983, 2,
1116.
(25) Bouaoud, S.-H.; Braunstein, P.; Grandjean, D.; Matt, D.; Nobel,
D. Inorg. Chem. 1986, 25, 3765.
(26) Pohlmann, J . L.; Brinckmann, F. E. Z. Naturforsch. 1965, 20B,
5.
(27) See ref 13b.
(28) a) Keim, W.; Kowaldt, F. H.; Goddard, R.; Kruger, C. Angew.
Chem., Int. Ed. Engl. 1978, 17, 466. (b) Carneiro, T. M. G.; Dupont,
J .; Luke, M.; Matt, D. Quim. Nova 1988, 11, 215.

(PPh2CH2C(O)Ph)Ni(C6F5)2
Organometallics, Vol. 19, No. 19, 2000 3985
(m, 3H), 5.19 (s, 1H). 31P{1H} NMR (C7D8): δ 22.2, 20.5 (second-
The solution was stirred for about 15 min. Removal of the
solvent in vacuo resulted in a yellow powder. H NMR (CD2-
2
1
order AB quartet, J P
) 285 Hz).
P
A
B
Cl2): δ 8.66 (br s, 2 H), 7.81-7.30 (m, 18 H), 4.70 (s, 1 H).
31P{1H} NMR (CD2Cl2): δ 24.8 (s). 19F NMR (CD2Cl2): δ -119.1
(m, 2F), -163.5 (m, 1F), -165.0 (m, 2F). FD-MS: m/z 607 [M+];
exhibited expected isotope pattern for nickel. Anal. Calcd for
[(P P h 2CHdC(O)P h )Ni(P h )]2 (3). The synthesis of 3 is
derived from a previous report.29 In the drybox both compound
2 (1.00 g, 1.43 mmol) and Rh(acac)(C2H4)2 (0.180 g, 0.700
mmol) were weighed into two separate flasks and capped. On
the Schlenk line, toluene was added to each compound: for
the rhodium compound about 5 mL and for the nickel
compound about 20 mL. The solution of Rh(acac)(C2H4)2 was
added to the nickel solution. A precipitate began to form
immediately. The mixture was stirred for 4 h and then filtered
and dried in vacuo to give a brown-yellow solid. Yield: 0.500
g (79%). 1H NMR (C6D6): δ 7.70 (d of d, 4H), 7.45 (d, 2H), 7.30
(d, 2H), 7.05 (m, 6H), 6.90 (m, 3H), 6.55 (t, 1H), 6.46 (t, 2H),
4.87 (s, 1H). 31P{1H} NMR (C6D6): δ 21.3 (s).
C
31H21NPOF5Ni: C, 61.22; H, 3.48. Found: C, 61.28; H, 3.68.
Rea ction of 5 w ith Eth ylen e. Compound 5 (0.0056 g,
0.0053 mmol) was dissolved in CDCl3. Ethylene was bubbled
through this orange solution for about 5 min. The solution was
then allowed to stand at room temperature for about 15 h and
heated to 65 °C overnight to give a cloudy, green-yellow
mixture. 19F NMR analysis of the solution exhibited resonances
consistent with pentafluorostyrene.
X-r a y Cr ysta llogr a p h y. Single crystals of (PPh2CH2C(O)-
Ph)Ni(C6F5)2‚CH2Cl2 (4), grown from methylene chloride at
(P P h 2CH2C(O)P h )Ni(C6F 5)2 (4). Meth od A. Compound 3
(0.150 g, 0.171 mol) was weighed into a 100 mL Kjeldahl flask
in the drybox. Also in the drybox, 0.194 g of B(C6F5)3‚3H2O
(0.342 mol) was weighed into a separate flask. After each solid
was dissolved separately in a minimum of toluene (about 15
mL each), the solution of B(C6F5)3‚3H2O was added to the
solution/slurry of the nickel dimer (it was not completely
soluble in toluene). The mixture changed from cloudy orange
to a translucent red-brown upon addition of the boron reagent.
The solution was stirred for approximately 1 h, after which
time the toluene was removed in vacuo. The yellowish brown
solid was redissolved in a small amount of toluene and cooled
to -20 °C. A yellow solid formed which was filtered and dried
-20 °C, are, at 20 ( 1 °C, monoclinic, space group P21/c-C5
2h
(No. 14) with a ) 14.773 (3) Å, b ) 11.174 (3) Å, c ) 21.175 (6)
Å, â ) 110.84 (2)°, V ) 3267 (2) Å3, and Z ) 4 (dcalcd ) 1.590
g cm -3; µa(Mo KR) ) 0.89 mm-1). A total of 4491 independent
reflections having 2θ(Mo KR) < 45.8° (the equivalent of 0.6
limiting Cu KR spheres) were collected on
a computer-
controlled Nicolet autodiffractometer using full (1.20° wide)
scans and graphite-monochromated Mo KR radiation. The
structure was solved using “Direct Methods” techniques with
the Siemens SHELXTL-PC software package as modified at
Crystalytics Co. The resulting structural parameters have been
refined to convergence (R1(unweighted, based on F) ) 0.041
for 2239 independent reflections having 2θ(Mo KR) < 45.8°
and I > 3σ(I)) using counter-weighted full-matrix least-squares
techniques and a structural model which incorporated aniso-
tropic thermal parameters for all non-hydrogen atoms and
isotropic thermal parameters for all hydrogen atoms. The
hydrogen atoms were fixed at idealized sp3- or sp2-hybridized
positions using a C-H bond length of 0.96 Å. The isotropic
thermal parameter of each hydrogen was fixed at 1.2 times
the equivalent isotropic thermal parameter of the carbon atom
to which it is covalently bonded.
1
in vacuo. Yield: 0.108 g (45%). H NMR (CD2Cl2): δ 7.90 (d,
2H), 7.67 (t, 3H), 7.48 (m, 4 H), 7.20 (m, 6 H), 4.19 (d, J PH
)
5 Hz, 2 H). 31P{1H} NMR (C6D6): δ 26.8 (s). 19F NMR (C6D6):
δ -115.8 (m, 2F), -119.0 (m, 2F), -159.8 (t, 1F), -160.9 (t,
1F), -162.9 (apparent t of d, 2F), -163.7 (apparent t, 2F). FD-
MS: m/z 696 [M+]; exhibited expected isotope pattern for
nickel. CI-MS: m/z 529 [M+ - C6F5]; exhibited expected isotope
pattern for nickel.
Meth od B. (η6-toluene)Ni(C6F5)2 (0.1 g, 0.2 mmol) was
dissolved in 10 mL of toluene under an inert atmosphere. To
this solution was added dropwise PPh2CH2C(O)Ph (0.063 g,
0.2 mmol) in 10 mL of toluene. The solution turned yellow-
brown after 10 min, and a yellow powder precipitated. After
it was stirred at room temperature for 1 h, the solvent was
removed in vacuo, resulting in a yellow solid. This was
dissolved in 10 mL of CH2Cl2, filtered, and kept in the freezer
at -20 °C. Bright yellow crystals of 4 were obtained in
Bond lengths and angles for 4 are found in Tables 1 and 2,
respectively. Atomic coordinates and anisotropic thermal
parameters for non-hydrogen atoms for 4 are found in Tables
3 and 4, while atomic coordinates for hydrogen atoms are found
in Table 5 (see the Supporting Information for Tables 3-5).
Resu lts a n d Discu ssion
quantitative yield after 2 days. Anal. Calcd for C33H18POF10
-
Ni‚1/3CH2Cl2: C, 53.62; H, 2.32. Found: C, 53.84; H, 3.03. The
presence of methylene chloride was confirmed by 1H NMR
spectroscopy.
Syn th esis of Sta r tin g Ma ter ia ls. The previous
synthetic method employed for B(C6F5)3‚3H2O (1), gas-
phase transfer of water to solid B(C6F5)3 under a static
vacuum,30 was somewhat cumbersome for the produc-
tion of larger quantities. Thus, the addition of 3 equiv
of deoxygenated water to a solution of commercially
available B(C6F5)3 (in Isopar E and cyclohexane) was
attempted (eq 1) and found to yield colorless microcrys-
[(P P h 2CHdC(O)P h )Ni(C6F 5)]2 (5). Compound 4 (0.39 g,
0.56 mmol) was dissolved in approximately 50 mL of 1,2-
dichloroethane. The yellow solution was refluxed for 42 h. An
orange color developed during this time. Cooling the solution
to room temperature resulted in an orange powder. The solvent
was removed in vacuo to give an orange solid. Yield: 0.28 g
(93%). Alternatively, large well-shaped crystals of compound
5 can be isolated after allowing 4 to stand in methylene
B(C F ) ‚3H O
B(C6F5)3 + 3 H2O f
(1)
6
5 3
1
2
1
chloride for several days. H NMR (CD2Cl2): δ 7.75 (m, 4H),
7.57 (m, 2H), 7.43 (m, 4H), 7.31 (d, 2H), 7.11 (m, 3H), 4.54 (s,
1H). 31P{1H} NMR (CD2Cl2): δ 19.8 (s). 19F NMR (CD2Cl2): δ
-118.1 (m, 2F), -161.3 (m, 1F), -165.8 (m, 2F). FD-MS: m/z
1058 [M+]; exhibited expected isotope pattern for two nickel
atoms. Anal. Calcd for C52H32P2O2F10Ni2‚11/3CH2Cl2: C, 54.73;
H, 2.99. Found: C, 54.69; H, 3.07. The presence of methylene
chloride in the above ratio was confirmed by 1H NMR
spectroscopy.
tals that were insoluble in aliphatic hydrocarbons.
Despite the fact that the synthesis and the crystal
structure of 1, albeit with sparse details, are available,
no spectroscopic characterization of B(C6F5)3‚3H2O has
been reported until now.
The FI-MS of 1 was rather confusing at first since a
molecular ion of m/z 512, consistent with B(C6F5)3 and
not its trihydrate, was observed. It is clear, however,
from the IR spectrum of 1 that it contained complexed
(P P h 2CHdC(O)P h )Ni(C6F5)(pyr idin e) (6). Approximately
0.20 g of 5 (0.18 mmol) was dissolved in toluene. An excess
(5.0 mL, 5.1 g, 65 mmol) of pyridine was added to the solution.
(29) See ref 22a.
(30) See ref 13a.

3986 Organometallics, Vol. 19, No. 19, 2000
Kalamarides et al.
Ta ble 2. Bon d An gles (d eg) In volvin g
Ta ble 1. Bon d Len gth s (Å) In volvin g
Non -Hyd r ogen Atom s in Cr ysta llin e
(P P h 2CH2C(O)P h )Ni(C6F 5)2‚CH2Cl2 (4)a
Non -Hyd r ogen Atom s in Cr ysta llin e
(P P h 2CH2C(O)P h )Ni(C6F 5)2‚CH2Cl2 (4)a
Complex
Complex
Ni1-P1
Ni1-C21
P1-C1
2.184(3)
1.886(7)
1.821(8)
Ni1-O1
1.939(5)
1.931(8)
P1-Ni1-O1
P1-Ni1-C21
O1-Ni1-C31
82.9(2)
97.4(3)
91.2(3)
C21-Ni1-C31
O1-Ni1-C21
P1-Ni1-C31
89.0(3)
176.7(3)
170.4(2)
Ni1-C31
P1-C3
P1-C9
1.810(7)
1.810(7)
Ni1-P1-C1
Ni1-P1-C3
Ni1-P1-C9
98.8(3)
129.4(3)
110.5(3)
C1-P1-C3
C1-P1-C9
C3-P1-C9
106.8(3)
103.6(3)
104.9(3)
O1-C2
1.252(10)
C1-C2
1.498(10)
Ni1-O1-C2
124.0(4)
O1-C2-C1
O1-C2-C15
C1-C2-C15
117.8(6)
119.7(7)
122.4(7)
C2-C15
C3-C4
C3-C8
C4-C5
C5-C6
C6-C7
C7-C8
1.468(10)
1.390(11)
1.372(12)
1.389(11)
1.373(15)
1.368(15)
1.377(11)
P1-C1-C2
107.0(5)
C9-C10
C9-C14
C10-C11
C11-C12
C12-C13
C13-C14
1.400(11)
1.382(12)
1.384(12)
1.358(16)
1.369(14)
1.374(12)
P1-C3-C4
P1-C3-C8
C4-C3-C8
C3-C4-C5
C4-C5-C6
C5-C6-C7
C6-C7-C8
C3-C8-C7
116.4(6)
124.3(5)
119.3(7)
119.0(8)
120.5(9)
120.6(8)
119.0(9)
121.6(8)
P1-C9-C10
P1-C9-C14
118.8(6)
122.2(6)
119.0(7)
119.1(9)
120.8(8)
120.5(8)
120.0(10)
120.7(8)
C10-C9-C14
C9-C10-C11
C10-C11-C12
C11-C12-C13
C12-C13-C14
C9-C14-C13
C15-C16
C15-C20
C16-C17
1.387(11)
1.386(13)
1.381(11)
C17-C18
C18-C19
C19-C20
1.364(15)
1.373(15)
1.369(11)
C2-C15-C16
C2-C15-C20
C16-C15-C20
C15-C16-C17
118.9(7)
121.6(7)
119.5(7)
118.9(8)
C16-C17-C18
C17-C18-C19
C18-C19-C20
C15-C20-C19
120.4(8)
121.4(8)
118.6(10)
121.1(8)
C21-C22
C21-C26
C22-C23
C23-C24
C24-C25
C25-C26
C22-F22
C23-F23
C24-F24
C25-F25
C26-F26
1.372(12)
1.373(10)
1.381(13)
1.379(12)
1.350(15)
1.372(12)
1.376(9)
1.349(11)
1.406(11)
1.354(9)
C31-C32
C31-C36
C32-C33
C33-C34
C34-C35
C35-C36
C32-F32
C33-F33
C34-F34
C35-F35
C36-F36
1.370(11)
1.363(10)
1.348(12)
1.364(9)
1.367(11)
1.379(12)
1.365(8)
1.360(9)
1.353(10)
1.369(8)
1.375(9)
Ni1-C21-C22
Ni1-C21-C26
C22-C21-C26
C21-C22-C23
C21-C22-F22
C23-C22-F22
C22-C23-C24
C22-C23-F23
C24-C23-F23
C23-C24-C25
C23-C24-F24
C25-C24-F24
C24-C25-C26
C24-C25-F25
C26-C25-F25
C21-C26-C25
C21-C26-F26
C25-C26-F26
124.6(5)
121.7(6)
113.7(7)
124.5(7)
118.6(7)
117.0(8)
118.1(9)
120.4(7)
121.5(9)
120.1(9)
119.2(9)
120.6(7)
119.0(7)
119.0(8)
122.0(8)
124.6(8)
119.3(7)
116.1(7)
Ni1-C31C32
128.0(5)
119.0(6)
113.0(7)
124.5(6)
117.8(7)
117.8(7)
120.3(7)
121.7(6)
118.0(7)
118.9(8)
121.1(7)
120.0(6)
117.8(6)
119.2(7)
123.1(7)
125.6(7)
119.5(8)
114.9(6)
Ni1-C31-C36
C32-C31-C36
C31-C32-C33
C31-C32-F32
C33-C32-F32
C32-C33-C34
C32-C33-F33
C34-C33-F33
C33-C34-C35
C33-C34-F34
C35-C34-F34
C34-C35-C36
C34-C35-F35
C36-C35-F35
C31-C36-C35
C31-C36-F36
C35-C36-F36
1.355(10)
Solvent of Crystallization
1.706(12) Cl2-C1s
Cl1-C1s
1.729(12)
a
The numbers in parentheses are the estimated standard
deviations in the last significant digit. Atoms are labeled in
agreement with Figure 1.
water. Three sharp, medium-intensity bands were
observed at 3666, 3597 (with a shoulder), and 3499 cm-1
and are assigned to OH vibrations. A medium-intensity
band was observed at 1602 cm-1, which was assigned
to a H2O bending mode. The sharpness of the stretching
vibrations is taken as evidence for the highly ordered
environment of the water. In the crystal structure of 1,
one water is coordinated to boron and the other two
waters participate in hydrogen bonding to the bound
water molecule.31 Further evidence that 1 was indeed
distinct from B(C6F5)3 came from its 19F NMR spectrum.
While B(C6F5)3 exhibited resonances at -134.7, -154.7,
and -162.6 ppm, B(C6F5)3‚3H2O resonances appeared
at -128.9, -142,3, and -160.4 ppm. Addition of pure
B(C6F5)3 to a solution of B(C6F5)3‚3H2O results in the
disappearance of the 19F resonances for B(C6F5)3‚3H2O
and the growth of new, broader peaks at -133.3,
-151.7, and -162.8 ppm. Evidently, the water mol-
ecules complexed to B(C6F5)3 undergo a dynamic ex-
change that is fast on the NMR time scale. However,
given the following data which show that 1 is indeed
distinct from B(C6F5)3, the m/z 512 molecular ion can
be explained by complete loss of complexed water from
1 under the high vacuum of the MS experiment.
Furthermore, repeating the reported synthetic method32
for 1 yielded colorless crystals that exhibited the same
19F NMR spectral pattern for 1 synthesized by the
addition of water to a solution of B(C6F5)3. Evidently 1
Solvent of Crystallization
113.0(8)
Cl1-C1s-Cl2
a
The numbers in parentheses are the estimated standard
deviations in the last significant digit. Atoms are labeled in
agreement with Figure 1.
loses water of hydration in the high vacuum of the MS
experiment and therefore only exhibits a molecular ion
for B(C6F5)3.
Although the synthesis of (PPh2CHdC(O)Ph)Ni(Ph)-
(PPh3) (2) has been known for 20 years,33 its spectro-
scopic data were not reported until 1988.34 While the
1H NMR spectrum of 2 is consistent with that published
previously, it was not entirely clear if the 31P NMR data
obtained by us were identical with those reported. For
2, an AB spin system is expected with large phosphorus-
phosphorus coupling since, on the basis of the crystal
structure of 2, the phosphorus atoms are trans. The 31P
NMR pattern for 2 is of higher order; the two inner lines
are much more intense than the outer two lines. This
indicates that the coupling constant between PA and PB
is greater than the difference between their chemical
shifts (i.e., J AB > ∆δAB). This spectrum was successfully
simulated using the calculated chemical shifts of 22.2
(31) See ref 13a.
(32) See ref 13a.
(33) See ref 28a.
(34) See ref 28b.

(PPh2CH2C(O)Ph)Ni(C6F5)2
Organometallics, Vol. 19, No. 19, 2000 3987
and 20.5 ppm and a 2J P
value of 285 Hz (see the
Sch em e 1
APB
Supporting Information).35,36
The complex [(PPh2CHdC(O)Ph)Ni(Ph)]2 (3) can be
synthesized from (PPh2CHdC(O)Ph)Ni(Ph)(PEt3) ac-
cording to the method of Klabunde et al.37 A similar
procedure was followed for 2, the triphenylphosphine
analogue. Thus, addition of the phosphine acceptor Rh-
(acac)(C2H4)2 to 2 gave the dimer complex 3 in good yield
(eq 2). Since no spectroscopic data have been published
(PPh CHdC(O)Ph)Ni(Ph)(PPh )
+
2
3
2
[(PPh CHdC(O)Ph)Ni(Ph)]
Rh(acac)(C2H4)2 f
+
2
2
3
Rh(acac)(PPh3)2 + 2C2H4 (2)
previously for 3, this material was subjected to 31P and
1H NMR analysis, which is consistent with the known
solid-state structure. The results are presented in the
Experimental Section.
-162.0 (2F) ppm, in agreement with data published
previously for C6F5H.38
Rea ction of B(C6F 5)3‚3H2O w ith 3. The reaction of
2 equiv of B(C6F5)3‚3H2O (or 1 equiv of B per Ni) with
[(PPh2CHdC(O)Ph)Ni(Ph)]2 was carried out in toluene
at room temperature and gave yellow microcrystals in
45% isolated yield. The solid was soluble in methylene
chloride and was moderately soluble in toluene. The 31P
NMR spectrum of the product exhibited a single reso-
nance at 26.8 ppm, a shift of only ca. 5 ppm downfield
A reaction sequence for the formation of 4 is presented
in Scheme 1. The proposal takes into account the
production of benzene and pentafluorobenzene. The
Brφnsted acid B(C6F5)3‚3H2O first protonates the me-
thine carbon of the phosphorus/oxygen chelate ligand
in complex 3. A nickel cation and a HOB(C6F5)3- anion
are formed along with loss of 4 equiv of water. The ionic
intermediate is probably a zwitterion or at least a
contact ion pair. Such species arguably would be favored
in toluene.39 The zwitterion could be formed by virtue
of an oxygen donor bond to the nickel between the anion
and cation as shown in Scheme 1. The oxygen donor
bond from the anion replaces the oxygen-nickel donor
bond in the starting material 3, serving to break up the
dimer into its constituent monomeric parts. The close
proximity of the HOB(C6F5)3- anion to the Ni-Ph bond
of the cation increases the probability that a second,
seemingly unfavored, protonation can occur. Loss of
PhH followed by C6F5 transfer from boron to nickel
creates the proposed neutral nickel boronoxy intermedi-
ate in Scheme 1. Reaction of this complex with the
liberated water molecules along with a second C6F5
transfer from boron to nickel generates the observed
final reaction products: 4, C6F5H, and, presumably, H3-
BO3.
Precedent for double protonation of an organometallic
complex by a related boron-containing reagent can be
found for a dimethylzirconocene complex. Siedle et al.40
revealed that [Et3NH][HOB(C6F5)3] reacts with [(Me5C5)2-
ZrMe2] to yield 2 equiv of methane: first, by proton
transfer from the stronger triethylammonium conjugate
acid and, second, by proton transfer from the weaker
boronoxy acid.
Furthermore, Green and co-workers have explored
reactions of group 4 transition-metal complexes41 and
late-transition-metal complexes42 with B(C6F5)3, which
resulted in transfer of one pentafluorophenyl group from
1
from that of the starting material. Its H NMR spectrum
exhibited aryl resonances between 7.90 and 7.20 ppm
and a methylene doublet at 4.19 ppm in a 15:2 ratio,
respectively. The 19F NMR spectrum of the solid is most
revealing. Two distinct pentafluorophenyl groups in a
1:1 ratio are observed and seem to be in similar chemical
environments, given that their chemical shifts are so
similar. A molecular ion of m/z 696 was observed in the
FD-MS of the product, which exhibited the expected
isotopic ratios for nickel. On the basis of these data, the
structure of the reaction product is proposed to be
This proposal was confirmed by independent synthesis
and single-crystal X-ray analysis (see below).
To track down the fate of the missing phenyl group
attached to the starting dimer 3, the volatiles from the
reaction with 1 carried out in toluene-d8 were collected.
NMR analysis of the volatiles showed a significant
amount of benzene, consistent with the loss of the
phenyl ligand in 3 by protonation of the Ni-Ph bond.
The fate of the third pentafluorophenyl group origi-
nating from boron was established by monitoring the
reaction between 1 and 3 using 19F NMR methods.
Besides the resonances for the product, complex 4, three
new resonances appeared at -138.8 (2F), -153.7 (1F),
(38) Bruce, M. I. J . Chem. Soc. A 1968, 1459.
(39) Eisch, J . J .; Pombrik, S. L.; Gurtzgen, S.; Rieger, R. Stud. Surf.
Sci. Catal. 1994, 89, 221.
(40) Siedle, A. R.; Newmark, R. A.; Lamanna, W. M.; Huffman, J .
C. Organometallics 1993, 12, 1491.
(41) Gomez, R.; Green, M. L. H.; Haggit, J . L. J . Chem. Soc., Chem.
Commun. 1994, 2607.
(35) Becker, E. D., High-Resolution NMR, Theory and Chemical
Applications, 2nd ed.; Academic Press: New York, 1980; p 138.
(36) The simulation was carried out using a PC-adapted version of
the program LAOCOON with a 5 Hz line broadening: Castellano, S.
M.; Bothner-By, A. A. J . Chem. Phys. 1964, 41, 3863.
(37) See ref 22a.

3988 Organometallics, Vol. 19, No. 19, 2000
Kalamarides et al.
bond angles involving the nickel and ligand atoms
bonded to nickel total near 360° (observed, 360.5°).
However, there are significant deviations from the
idealized ligand bond angles expected for a square-
planar geometry. For example, the P1-Ni-O1 angle is
more acute than normal (82.9(2)°) and the P1-Ni-C21
angle significantly more obtuse than usual (97.4(3)°).
These deviations are undoubtedly due to the chelate
ligand that ties together the P and O atoms of the
coordination sphere, thus decreasing the angle from the
idealized 90°. Similar, but less dramatic, deviations are
observed for nickel complexes coordinated to P∼O
enolate chelates. For complex 2 and the related complex
(PPri CHdC(O)Ph)Ni(Ph)(PPh3) the corresponding
2
P-Ni-O angle measured 86.5°.
The Ni-P1 distance measures 2.184(3) Å and is
comparable to those found for related compounds. For
example, it is somewhat shorter than the Ni-P dis-
tances in trans-NiBr(C6F5)(PPh2Me)2 and trans-Ni-
(C6F5)2(PPh2Me)2 which average 2.2156(13) and 2.206(1)
Å, respectively.44 The Ni-P1 distance in 4 is somewhat
F igu r e 1. ORTEP drawing of (PPh2CH2C(O)Ph)Ni(C5F6)2‚
CH2Cl2 (4).
longer than in 2 (2.168 Å) and in (PPri CHdC(O)Ph)-
2
Ni(Ph)(PPh3) (2.1756(7) Å). The Ni-O1 distance in 4 is
1.939(5) Å and is significantly longer than that found
for the Ni-O bond (1.895(3) Å) in a related Ni(II) species
with a chelating ketonic ligand, [Ni(HOCMe2CH2C(O)-
the boron to the transition metal. In the former case,
B(C6F5)3 reacts with a zirconium dimethyl species to
yield a Zr-C6F5 bond, while in the latter case CpFe-
(CO)2Me reacts with B(C6F5)3 to yield iron-tetrafluoro-
phenyl derivatives. The synthesis of 4 is the first
example of the transfer of two pentafluorophenyl groups
from B(C6F5)3 to a transition metal. In contrast, Pud-
dephatt and co-workers have shown that B(C6F5)3 reacts
with a dimethylplatinum species by the more traditional
alkyl abstraction route.43
In d ep en d en t Syn th esis of 4. Complex 4 can be
synthesized by an independent route that involved the
addition of the P∼O chelate ligand PPh2CH2C(O)Ph to
a Ni(C6F5)2 precursor. The chelate ligand was synthe-
sized by a known procedure (see Experimental Section)
and was added as a toluene solution to (η6-toluene)Ni-
(C6F5)2. An immediate color change from red-brown to
yellow occurred. Removal of the solvent in vacuo gave
a yellow solid. This solid exhibited the exact NMR
spectral characteristics that were observed for the
reaction product of the trihydrate of B(C6F5)3 and dimer
3 (eq 3). The ability to make 4 by this independent route
Me)2]2+ 45
.
The two Ni-Cipso(C6F5) distances differ significantly
from one another. The Ni-C bond trans to phosphorus
(Ni-C31) is significantly longer (1.931(8) Å) than the
Ni-C bond trans to oxygen (Ni-C21, 1.886(7) Å). These
distances are in agreement with the strong trans
influence of phosphorus.46 There are two crystal struc-
tures of complexes containing cis-oriented bis(pentafluo-
rophenyl) ligands, but they both exhibit Ni-Cipso(C6F5)
distances that are shorter than the Ni-C (Ni-C31)
bond trans to phosphorus in 4: for (norbornadiene)Ni-
(C6F5)2, Ni-C(C6F5) ) 1.914(2) and 1.921(2) Å, and for
(cycloocta-1.5-diene)Ni(C6F5)2, Ni-C(C6F5) ) 1.902(2)
Å.47 The shorter distances are due to the coordination
of olefinic ligands trans to the pentafluorophenyl groups
which are known to exhibit a weaker trans influence
than phosphorus. The least-squares plane containing
the immediate nickel coordination environment (Ni, P1,
O1, C21, and C31) forms dihedral angles of 77.4 and
72.3° with the least-squares planes defined by the
pentafluorophenyl rings, C21-C26 and C31-C36, re-
spectively. This angle, defined as φ by Eyring and
Radonovich, has been correlated with the relative degree
of nickel to C6F5 back-bonding in a series of related cis-
bis(pentafluorophenyl)nickel complexes. As φ approaches
90° for these complexes, the nickel-pentafluorophenyl
bond length decreases. For example, φ ) 72° and Ni-
PPh2CH2C(O)Ph + (η6-toluene)Ni(C6F5)2 f
(PPh2CH2C(O)Ph)Ni(C6F5)2
(3)
4
enabled us to access larger amounts for a thermolysis
study (see below).
Cr ysta l a n d Molecu la r Str u ctu r e of 4. Crystals
of 4 suitable for X-ray analysis were grown from
methylene chloride solution. A perspective plot of the
complex is shown in Figure 1; selected distances and
angles are given in Tables 1 and 2.
The nickel adopts a square-planar geometry involving
the P and O atoms of the chelating phosphino-ketone
ligand and the ipso carbons of the pentafluoroaryl
groups. As expected for a square-planar geometry, the
C
ipso(C6F5) ) 1.917(2) Å in (η4-norbornadiene)Ni(C6F5)2,
φ ) 83° and Ni-Cipso(C6F5) ) 1.902(4) Å in (η4-cycloocta-
1,5,diene)Ni(C6F5)2, and φ ) 89° and Ni-Cipso(C6F5) )
1.891(4) Å in (η6-toluene)Ni(C6F5)2. For complex 4, on
the basis of the φ parameter alone, one would expect
the Ni-Cipso(C6F5) bond distances to hover around 1.92
(44) (a) Churchill, M. R.; Kalra, K. L.; Veidis, M. W. Inorg. Chem.
1973, 12, 1656. (b) Churchill, M. R.; Veidis, M. W. J . Chem. Soc., Dalton
Trans. 1972, 670.
(42) Green, M. L. H.; Haggit, J . L.; Mehnert, C. P. J . Chem. Soc.,
Chem. Commun. 1995, 1853.
(43) Hill, G. S.; Rendina, L. M.; Puddephatt, R. J . J . Chem. Soc.,
Dalton Trans. 1996, 1809.
(45) Foxman, B. M.; Mazurek, H. Inorg. Chem. 1979, 18, 113.
(46) Purcell, K. F.; Kotz, J . C. Inorganic Chemistry, Saunders:
Philadelphia, PA, 1977; p 705.
(47) Eyring, M. W.; Radonovich, L. J . Organometallics 1985, 4, 1841.

(PPh2CH2C(O)Ph)Ni(C6F5)2
Organometallics, Vol. 19, No. 19, 2000 3989
F igu r e 2. Ball and stick drawing of 4.
Å. The fact that the bond lengths are longer (Ni-C31,
1.931(8) Å) and shorter (Ni-C21, 1.886(7) Å) suggests
that the trans influences of phosphorus and oxygen
override the relatively weak nickel to pentafluorophenyl
back-bonding interaction. However given the relatively
high standard deviations of these distances, a definitive
conclusion cannot be made.
F igu r e 3. Ball and stick drawing of [(PPh2CHdC(O)Ph)]-
Ni(C6F5)]2 (5).
pound 5 exhibits a single set of peaks in its 19F NMR
spectrum and a single phosphorus resonance. Integra-
1
tion of the H NMR spectrum of 5 is in agreement with
Figure 2 presents a side view of complex 4 that bisects
the P1-Ni-O1 angle. This perspective accentuates the
puckering of the five-membered P∼O chelate ligand.
This conformation forces the C9-containing phenyl ring
attached to P1 to position ortho hydrogen H10 above
the metal at one of the two axial sites of an octahedral
nickel coordination sphere. While the Ni-H10 separa-
tion, 2.753 Å, is beyond bonding distance, one axial
position of nickel is nevertheless effectively blocked by
the phenyl ring (see the space-filling drawing in the
Supporting Information). Approach to the opposite axial
position is unencumbered. Although the X-ray data
suggest that the methylene protons of the P∼O chelate
are inequivalent, solution NMR data at room temper-
ature exhibit only one resonance for these protons.
Evidently, the P∼O ligand undergoes a dynamic pro-
cess, presumably flipping of the methylene carbon in
the puckered ring from one side of the square plane to
the other, rendering the methylene protons equivalent.
This process would then force the C3-containing phenyl
substituent attached to phosphorus to cover access to
the opposite axial site from that hidden by the C9-
containing phenyl group shown in the static structure
of Figure 2.
loss of one methylene proton from the P∼O chelate in 4
and formation of 1 equiv of pentafluorobenzene (19F
NMR). Formulation of compound 5 as a dimer is based
on (1) the observed molecular ion (m/z 1058) in the MS
experiment, which exhibited isotopic ratios consistent
with two nickel atoms per molecule, and (2) the results
of a preliminary X-ray structure determination for
crystals of 5. Unfortunately, the small size and poor
quality of the crystals for 5 limited the quality and
quantity of the diffraction data and prevented a high-
precision structure determination.48 The data were,
however, of sufficient quality to unambiguously identify
5 as a dimeric species with crystallographic C2 sym-
metry in the solid state. Furthermore, the qualitative
and semiquantitative structural parameters which re-
sulted from the X-ray structure determination are
totally consistent with the proposed structure of 5 and
the other spectroscopic and analytical data for the
compound. As shown in Figure 3, not only does the P∼O
ligand form a chelate to each nickel, it also holds the
dimer together through a dative oxygen to nickel bond
that bridges the two nickel centers. The structure of 5
is analogous to that reported for the starting material,
Th er m olysis of (P P h 2CH2C(O)P h )Ni(C6F 5)2 (4). In
the solid state, 4 undergoes no significant reaction after
1 day of exposure to atmospheric oxygen and moisture
(1H NMR spectroscopy). However, when a 1,2-dichloro-
ethane solution of 4 is maintained at 65 °C for ap-
proximately 40 h, an orange precipitate forms. Spectral
data taken of the orange solid are in agreement with
the synthesis of [(PPh2CHdC(O)Ph)Ni(C6F5)]2 (5). Com-
(48) Crystals of [Ni(C6F5){(C6H5)2PCHCO(C6H5)}]2‚C2H4Cl2 are, at
20 °C, monoclinic, space group P2/c-Ch (No. 13), with a ) 15.512(7) Å,
b ) 8.957(4) Å, c ) 20.828(9) Å, â ) 117.80(4)°, V ) 2560(2) Å3, and Z
) 2 dimeric formula units (dcalcd ) 1.483 g cm-3; µa(Cu KR) ) 3.14
mm-1). The crystals were not only quite small (maximum dimension
0.26 mm) but also invariably twinned. Several crystals were examined
before a relatively single one was found which was used for subsequent
data collection. A total of 2353 independent reflections having 2θ(Cu
KR) < 95.04° (the equivalent of 0.4 limiting Cu KR sphere) were
collected on a computer-controlled Syntex P single-crystal diffracto-
meter using θ/2θ scans with fixed scan rates of 6 or 3°/min and Ni-
filtered Cu KR radiation. “Direct methods” techniques were used to
solve the structure, and the resulting structural parameters have been
refined to convergence (R1(unweighted, based on F) ) 0.105 for 1117
independent reflections having 2θ(Cu KR) < 95.04° and I >2σ(I)) using
counterweighted full-matrix least-squares techniques and a structural
model which incorporated anisotropic thermal parameters for all non-
hydrogen atoms and isotropic thermal parameters for all included
hydrogen atoms. Selected bond distances: Ni1-O1 ) 1.901(10) Å; Ni1-
O1′ ) 1.912(12) Å; Ni1-P1 ) 2.116(6) Å; Ni1-C21 ) 1.91(2) Å; Ni1‚
‚‚Ni1′ ) 2.723(5) Å. Selected bond angles: O1-Ni1-O1′ ) 78.1(5)°;
C21-N1-O1′ ) 99.7(6)°; O1-Ni1-P1 ) 88.4(4)°; C21-Ni1-P1 ) 94.1-
(5)°.

3990 Organometallics, Vol. 19, No. 19, 2000
Kalamarides et al.
Sch em e 2
polymers, ethylene was bubbled through a solution of
dimer 5. After 15 h at room temperature, no reaction
was visible. This was confirmed by NMR analysis of the
solution. However, when the ethylene-saturated solution
was heated to 65 °C overnight, the orange solution
became a cloudy yellow-green mixture. While the 1H
NMR spectrum of this solution was quite complex and
indecipherable, the 19F NMR spectrum showed reso-
nances at -143.4 (2F), -157.9 (1F), and -164.3 (2F)
which are very close to those resonances associated with
pentafluorostyrene.51 Other unidentified resonances are
also observed in this spectrum.
The formation of pentafluorostyrene is consistent with
the migratory insertion of ethylene into the Ni-C6F5
bond, followed by â-hydride elimination. In this process,
a nickel-hydride species should also be generated which
is the purported active catalyst in oligomerization/
polymerization of ethylene. Similar migratory inser-
tion-â-hydride elimination processes have been ob-
served in related Ni-Ph and Ni-C6F5 complexes.52 To
the extent that this nickel-hydride species is formed,
complex 5 should be an active catalyst for the oligomer-
ization and/or polymerization of ethylene under the
right conditions (e.g., increased pressure and temper-
ature).53
dimer 3. Unfortunately, no details regarding the struc-
ture of 3 are available for comparison purposes.
A possible route to 5 involves the protonation of one
of the Ni-C6F5 bonds in 4 to form the observed pen-
tafluorobenzene product by transfer of one of the acidic
methylene protons of the phosphorus/oxygen chelate.
This transformation is detailed in Scheme 2. Therefore,
the chelating ligand is amphoteric in nature; it can
accept protons from strong Brφnsted acids (i.e., B(C6F5)3‚
3H2O) and can donate protons to suitably basic sub-
strates (i.e., Ni-C6F5 bonds). In related palladium
chemistry, Braunstein and co-workers found that the
phosphino-ketone ligand PPh2CH2C(O)Ph can be depro-
tonated with NaH to give the chelating ylide ligand
found in 2 and 3.49 The loss of H+ from PPh2CH2C(O)-
Ph to form the corresponding chelating ylide has been
described for related nickel systems.50
Rea ctivity of [(P P h 2CH2C(O)P h )Ni(C6F 5)]2 (5).
Addition of an excess of pyridine to a toluene solution
of dimer 5 gave a yellow powder after the solvent was
removed under vacuum. The yellow material exhibited
a single resonance at 24.8 ppm in the 31P NMR experi-
ment and a single set of pentafluorophenyl resonances
(ortho, meta, and para) in the 19F NMR spectrum. The
molecular ion observed in the mass spectrum of the
yellow complex is consistent with the formation of the
pyridine adduct (PPh2CH2C(O)Ph)Ni(C6F5)(pyridine)
(6). The two-electron-donor pyridine replaces the oxygen
bridges that hold dimer 5 together, forming mono-
nuclear complex 6 in the process.
Ack n ow led gm en t. We thank Dr. B. Goodall and L.
McIntosh for providing the impetus for carrying out this
work. We are indebted to Dr. R. Cozens and P. Klich
for first exploring the synthesis of 2, to Dr. G. Benedikt
for recording the NMR spectra, to Dr. R. Kinsey for help
with the NMR simulation package, and to Dr. R.
Lattimer for the MS measurements. Dr. A. Siedle
provided much stimulating conversation regarding the
chemistry of B(C6F5)3. Profs. M. Fryzuk and P. Braun-
stein provided comments regarding the purification of
PPh2CH2C(O)Ph. Dr. C. Mehnert is acknowledged for
pointing us in the right direction. The BFGoodrich
Pacing Research Fund and the National Institute of
Standards and Technology Advanced Technology Pro-
gram are acknowledged for financial support.
Su p p or tin g In for m a tion Ava ila ble: Figures giving ex-
perimental and simulated 31P NMR spectra of complex 2 and
space-filling diagrams of complex 4, tables of atomic coordi-
nates and anisotropic thermal parameters for complex 4, and
tables of bond lengths, bond angles, atomic coordinates, and
anisotropic thermal parameters for complex 5. This material
is available free of charge via the Internet at http://pubs.acs.org.
Given the propensity of complexes such as 3 to react
with ethylene to form oligomers and in some cases
OM000606W
(49) See ref 25.
(50) (a) Braunstein, P.; Matt, D.; Nobel, D.; Balegroune, F.; Salah-
Eddine, B.; Grandjean, D.; Fischer, J . J . Chem. Soc., Dalton Trans.
1988, 353. (b) Matt, D.; Huhn, M.; Fischer, J .; De Cian, A.; Klaui, W.
Taktchenko, I.; Bonnet, M. C. J . Chem. Soc., Dalton Trans. 1993, 1173.
(c) Andrieu, J .; Braunstein, P.; Drillon, M.; Dusausoy, Y.; Ingold, F.;
Rabu, P.; Tiripicchio, A.; Ugozzoli, F. Inorg. Chem. 1996, 35, 5986.
(51) See ref 38.
(52) (a) Keim, W.; Behr, A.; Gruber, B. Hoffmann, B.; Kowaldt, F.
H.; Kurschner, U.; Limbacker, B.; Sistig, F. P. Organometallics 1986,
5, 2356. (b) Choe, S.-K.; Kanai, H.; Klabunde, K. J . J . Am. Chem. Soc.
1989, 111, 2875.
(53) Keim, W. Ann. N. Y. Acad. Sci. 1983, 191.