The coordination chemistry of "[BP3]NiX" platforms: Targeting low-valent nickel sources as promising candidates to L3Ni=E and L3Ni≡E linkages

Article abstract of DOI:10.1021/ic049936p

A series of divalent, monovalent, and zerovalent nickel complexes supported by the electron-releasing, monoanionic tris(phosphino)borate ligands [PhBP ₃] and [PhBP^iPr₃] ([PhBP₃] = [PhB(CH₂PPh₂)₃]^-, [PhBP ^iPr₃] = [PhB(CH₂PⁱPr ₂)₃]^-) have been synthesized to explore fundamental aspects of their coordination chemistry. The pseudotetrahedral, divalent halide complexes [PhBP₃]NiCl (1), [PhBP₃]Nil (2), and [PhBP^iPr₃]NiCl (3) were prepared by the metalation of [PhBP₃]Tl or [PhBP^iPr₃]Tl with (Ph ₃P)₂NiCl₂, Nil₂, and (DME)NiCl ₂ (DME = 1,2-dimethoxyethane), respectively. Complex 1 is a versatile precursor to a series of complexes accessible via substitution reactions including [PhBP₃]-Ni(N₃) (4), [PhBP₃] Ni(OSiPh₃) (5), [PhBP₃]Ni(0-p-^tBu-Ph) (6), and [PhBP₃]Ni(S-p-^tBu-Ph) (7). Complexes 2-5 and 7 have been characterized by X-ray diffraction (XRD) and are pseudotetrahedral monomers in the solid state. Complex 1 reacts readily with oxygen to form the four-electron-oxidation product, {[PhB(CH₂P(O)Ph₂) ₂(CH₂PPh₂)]-NiCl} (8A or 8B), which features a solid-state structure that is dependent on its method of crystallization. Chemical reduction of 1 using Na/Hg or other potential 1-electron reductants generates a product that arises from partial ligand degradation, [PhBP ₃]Ni(η²-CH₂PPh₂) (9). The more sterically hindered chloride 3 reacts with Li(dbabh) (Hdbabh = 2,3:5,6-dibenzo-7-azabicyclo[2.2.1]hepta-2,5-diene) to provide the three-coordinate complex [κ²-PhBP^iPr ₃]Ni(dbabh) (11), also characterized by XRD. Chemical reduction of complex 1 in the presence of L-type donors produces the tetrahedral Ni(I) complexes [PhBP₃]Ni(PPh₃) (12) and [PhBP ₃]Ni(CN^tBu) (13). Reduction of 3 following the addition of PMe₃ or tert-butyl isocyanide affords the Ni(I) complexes [PhBP ^iPr₃]Ni(PMe₃) (14) and [PhBP^iPr ₃]Ni(CN^tBu) (15), respectively. The reactivity of these [PhBP₃]Ni^jL and [PhBP^iPr₃]Ni ^jL complexes with respect to oxidative group transfer reactions from organic azides and diazoalkanes is discussed. The zerovalent nitrosyl complex [PhBP₃]Ni(NO) (16) is prepared by the reaction of 1 with excess NO or by treating 12 with stoichiometric NO. The anionic Ni(O) complexes [[κ²-PhBP₃]Ni(CO)₂][ⁿBu ₄N] (17) and [[κ²-PhBP^iPr ₃]Ni(CO)₂][ASN] (18) (ASN = 5-azoniaspiro[4.4]nonane) have been prepared by reacting [PhBP₃]Tl or [PhBP^iPr ₃]Tl with (Ph₃P)₂Ni(CO)₂ in the presence of R₄NBr. The photolysis of 17 appears to generate a new species consistent with a zerovalent monocarbonyl complex which we tentatively assign as {[PhBP₃]Ni(CO)}{ⁿBu₄N}, although complete characterization of this complex has been difficult. Finally, theoretical DFT calculations are presented for the hypothetical low spin complexes [PhBP₃]Ni(N^tBu), [PhBP^iPr ₃]Ni(N^tBu), [PhBP^iPr₃]Ni(NMe), and [PhBP^iPr₃]Ni(N) to consider what role electronic structure factors might play with respect to the relative stability of these species.

Full text of DOI:10.1021/ic049936p

Inorg. Chem. 2004, 43, 4645−4662
The Coordination Chemistry of “[BP ]NiX” Platforms: Targeting
3
Low-Valent Nickel Sources as Promising Candidates to L3NidE and
L3NitE Linkages
Cora E. MacBeth, J. Christopher Thomas, Theodore A. Betley, and Jonas C. Peters*
DiVision of Chemistry and Chemical Engineering, Arnold and Mabel Beckman Laboratories of
Chemical Synthesis, California Institute of Technology, Pasadena, California 91125
Received January 15, 2004
A series of divalent, monovalent, and zerovalent nickel complexes supported by the electron-releasing, monoanionic
iPr
iPr
i
tris(phosphino)borate ligands [PhBP ] and [PhBP ] ([PhBP ] ) [PhB(CH PPh2)3]-, [PhBP ] ) [PhB(CH PPr2)3]-)
3
3
3
2
3
2
have been synthesized to explore fundamental aspects of their coordination chemistry. The pseudotetrahedral,
iPr
divalent halide complexes [PhBP ]NiCl (1), [PhBP ]NiI (2), and [PhBP ]NiCl (3) were prepared by the metalation
3
3
3
iPr
of [PhBP ]Tl or [PhBP ]Tl with (Ph3P)2NiCl2, NiI2, and (DME)NiCl2 (DME ) 1,2-dimethoxyethane), respectively.
3
3
Complex 1 is a versatile precursor to a series of complexes accessible via substitution reactions including [PhBP ]-
3
t
t
Ni(N ) (4), [PhBP ]Ni(OSiPh3) (5), [PhBP ]Ni(O-p-Bu-Ph) (6), and [PhBP ]Ni(S-p-Bu-Ph) (7). Complexes 2−5 and
3
3
3
3
7 have been characterized by X-ray diffraction (XRD) and are pseudotetrahedral monomers in the solid state.
Complex 1 reacts readily with oxygen to form the four-electron-oxidation product, {[PhB(CH P(O)Ph2)2(CH PPh2)]-
2
2
NiCl} (8A or 8B), which features a solid-state structure that is dependent on its method of crystallization. Chemical
reduction of 1 using Na/Hg or other potential 1-electron reductants generates a product that arises from partial
ligand degradation, [PhBP ]Ni(η2-CH PPh ) (9). The more sterically hindered chloride 3 reacts with Li(dbabh) (Hdbabh
3
2
2
iPr
) 2,3:5,6-dibenzo-7-azabicyclo[2.2.1]hepta-2,5-diene) to provide the three-coordinate complex [κ2-PhBP ]Ni(dbabh)
3
(11), also characterized by XRD. Chemical reduction of complex 1 in the presence of L-type donors produces the
t
tetrahedral Ni(I) complexes [PhBP ]Ni(PPh3) (12) and [PhBP ]Ni(CNBu) (13). Reduction of 3 following the addition
3
3
iPr
iPr
t
of PMe3 or tert-butyl isocyanide affords the Ni(I) complexes [PhBP ]Ni(PMe3) (14) and [PhBP ]Ni(CNBu) (15),
3
3
respectively. The reactivity of these [PhBP ]NiIL and [PhBP ]NiIL complexes with respect to oxidative group transfer
iPr
3
3
reactions from organic azides and diazoalkanes is discussed. The zerovalent nitrosyl complex [PhBP ]Ni(NO) (16)
3
is prepared by the reaction of 1 with excess NO or by treating 12 with stoichiometric NO. The anionic Ni(0)
iPr
complexes [[κ2-PhBP ]Ni(CO)2][nBu4N] (17) and [[κ2-PhBP ]Ni(CO)2][ASN] (18) (ASN ) 5-azoniaspiro[4.4]nonane)
3
3
iPr
have been prepared by reacting [PhBP ]Tl or [PhBP ]Tl with (Ph3P)2Ni(CO)2 in the presence of R NBr. The
3
3
4
photolysis of 17 appears to generate a new species consistent with a zerovalent monocarbonyl complex which we
tentatively assign as {[PhBP ]Ni(CO)}{nBu4N}, although complete characterization of this complex has been difficult.
3
t
Finally, theoretical DFT calculations are presented for the hypothetical low spin complexes [PhBP ]Ni(NBu),
3
iPr
t
iPr
iPr
[PhBP ]Ni(NBu), [PhBP ]Ni(NMe), and [PhBP ]Ni(N) to consider what role electronic structure factors might
3
3
3
play with respect to the relative stability of these species.
I. Introduction
the renewed interest has been sparked by the realization that
these types of species can participate in a host of catalytic
group transfer reactions such as aziridinations,1 cyclopropa-
nations,2 and oxidations.3 Moreover, there is hope that new
approaches to catalytic small molecule activation and func-
tionalization reactions that exploit reactive metal fragments
from this sector of the transition block may yet be eluci-
dated.4,5 Small molecule substrates of interest might include
The fundamental coordination chemistry of mid-to-late
first row transition metal ions supported by relatively simple
ligand auxiliaries represents a mature field that has experi-
enced somewhat of a renaissance in recent years. Much of
* Author to whom correspondence should be addressed. E-mail:
jpeters@caltech.edu.
10.1021/ic049936p CCC: $27.50 © 2004 American Chemical Society
Published on Web 06/25/2004
Inorganic Chemistry, Vol. 43, No. 15, 2004 4645

MacBeth et al.
N2, N2O, O2, CO2, CO, and P4.6 One of the most attractive
features of the mid-to-late first row ions concerns their ability
to form metal-to-ligand multiple bond linkages that are
kinetically and thermodynamically reactive. Species of these
types (e.g., MdO, MdCR2, MdNR, MtN) are proposed
as key intermediates in a number of catalytic reactions of
both synthetic and biocatalytic significance.3,7
context, our group has initiated the development of pseudo-
tetrahedral “L3MX” platforms where L3 represents the
anionic tris(phosphine) donor [PhBP3] or [PhBPiPr3] ([PhBP3]
) [PhB(CH2PPh2)3]-, [PhBPiPr3] ) [PhB(CH2PiPr2)3]-) and
M represents a first row ion such as Fe or Co, and in this
paper Ni.14-18 Our interest in the family of nickel complexes
described herein is motivated by recent findings demonstrat-
ing that “L3Co” and “L3Fe” complexes supported by BP3
ligand auxiliaries (i) support later metal-to-ligand triple bond
linkages (e.g., L3CotNR, L3FetNR, L3FetN) and (ii)
participate in multielectron group transfer reactions that
accept and release the multiply bonded functional unit.
Riordan’s group has moreover shown that “L3Ni(I)” species
supported by anionic tris(thioether)borate ligands undergo
oxidative group transfer with dioxygen. For example, the
Ni(I) carbonyl supported by [PhTttBu] reacts with dioxygen
to form a Ni2(µ-O)2 structure in which each nickel center
has been oxidized to Ni(III).19 This structure type was first
elucidated for nickel by Hikichi and co-workers using a tris-
(pyrazolyl)borate auxiliary scaffold.20
Very recently, several groups have reported the isolation
of well-defined later first row complexes that feature metal-
to-ligand multiple bond linkages. For example, using steri-
cally encumbering bis(phosphine) ligands, Hillhouse and co-
workers have shown that L2Ni fragments are compatible with
imide (NR),8 phosphinidene (PR),9 and carbene (CR2)10
functional groups (generally defined as L2NiIIdE species).
These fascinating complexes, which follow on an early
observation from the Jones group that provided strong
evidence for a reactive L2NiIIdS linkage,11 have since been
shown to undergo stoichiometric group transfer processes,
for instance to ethylene.12,13 The L2NiIIdE systems owe their
appreciable stability to a combination of simple electronic
factors and to the considerable steric protection provided by
the bulky bis(phosphine) ligand and the substitution pattern
at the multiply bonded functional group. Four-coordinate
pseudotetrahedral systems that feature MdE and MtE
linkages (i.e., L3MtE systems) are equally intriguing targets,
and a more general understanding of the electronic similari-
ties and differences that dictate the relative stability of later
first row metal MdE and MtE linkages for three-, four-,
and higher coordinate structures is clearly of interest. In this
The present report focuses on the synthesis and charac-
terization of a host of nickel complexes supported by BP3
ligands and on theoretical and reactivity issues pertinent to
the possible construction of L3NidE and L3NitE linkages.
To these ends, we have prepared a series of divalent L3NiX
(X ) I, Cl, N3, OR, or SR) complexes, as well as mono-
and zerovalent nickel precursors. Where applicable, structural
and electronic comparisons are made between these “[PhBP3]-
Ni” and “[PhBPiPr3]Ni” systems and their structurally similar
counterparts supported by other anionic and facially capping
ligand auxiliaries,21,22 or by neutral tris(phosphine) ligand
auxiliaries such as CH3C(CH2PPh2)3 and PhP(CH2PPh2)2.23,24
The results presented demonstrate that tris(phosphino)borate
scaffolds effectively stabilize the lower oxidation states of
nickel (e.g., Ni0, NiI, and NiII) while in most cases maintain-
ing a four-coordinate pseudotetrahedral geometry. Synthetic
(1) (a) Evans, D. A.; Woerpol, K. A.; Hinman, M. M.; Faul, M. M. J.
Am. Chem. Soc. 1991, 113, 726-728. (b) Li, Z.; Quan, R. W.;
Jacobsen, E. N. J. Am. Chem. Soc. 1995, 117, 5889-5890. (c) Muller,
P.; Fruit, C. Chem. ReV. 2003, 103, 2905-2919. (d) Jacobson, E. N.
In ComprehensiVe Asymmetric Catalysis; Jacobsen, E. N., Pfaltz, A.,
Yamamoto, H., Eds.; Springer: Hamburg, 1999.
(2) (a) Doyle, M. P.; McKervey, M. A.; Ye, T. Modern Catalytic Methods
for Organic Synthesis with Diazo Compounds; Wiley: New York,
1998. (b) Brookhart, M.; Studabaker, W. B. Chem. ReV. 1987, 87,
411-432. (c) Doyle, M. P. Chem. ReV. 1986, 86, 919-939.
(3) Collman, J. P.; Hegedus, L. S.; Norton, J. R.; Finke, R. G. Principles
and Applications of Organotransition Metal Chemistry, 2nd ed.;
University Science Books: Mill Valley, CA, 1987.
(14) Jenkins, D. M.; Betley, T. A.; Peters, J. C. J. Am. Chem. Soc. 2002,
124, 11238-11239.
(4) Fisher, B.; Eisenberg, R. J. Am. Chem. Soc. 1980, 102, 7361-7363.
(5) Morgenstern, D. A.; Wittrig, R. E.; Fanwick, P. E.; Kubiak, C. P. J.
Am. Chem. Soc. 1993, 115, 6470-6471.
(15) Brown, S. D.; Betley, T. A.; Peters, J. C. J. Am. Chem. Soc. 2003,
125, 322-323.
(6) For examples, see: (a) Detrich, J. L.; Konecˇny´, R.; Vetter, W. M.;
Doren, D.; Rheingold, A. L.; Theopold, K. H. J. Am. Chem. Soc. 1996,
118, 1703-1712. (b) Cecconi, F.; Ghilardi, C. A.; Midollini, S.;
Moneti, S.; Orlandini, A.; Bacci, M. J. Chem. Soc., Chem. Commun.
1985, 731-733. (c) Ehses, M.; Romerosa, A.; Peruzzi, M. Top. Curr.
Chem. 2002, 220, 107-140. (d) Matsunaga, P. T.; Hillhouse, G. L.;
Rheingold, A. L. J. Am. Chem. Soc. 1993, 115, 2075-2077. (e) Qin,
Z.; Thomas, C. M.; Lee, S.; Coates, G. W. Angew. Chem., Int. Ed.
2003, 42, 5484-5487. (f) Fox, D. J.; Bergman, R. G. J. Am. Chem.
Soc. 2003, 125, 8984-8985. (g) Holland, P. L.; Andersen, R. A.;
Bergman, R. G. J. Am. Chem. Soc. 1996, 118, 1092-1104.
(7) Bioinorganic Catalysis; Reedijk, J., Bouwman, E., Eds.; Marcel
Dekker: New York, 1999.
(16) Betley, T. A.; Peters, J. C. J. Am. Chem. Soc. 2003, 125, 10782-
10783.
(17) Brown, S. D.; Peters, J. C. J. Am. Chem. Soc. 2004, 126, 4538-4539.
(18) Betley, T. A.; Peters, J. C. J. Am. Chem. Soc. 2004, 126, 6252-6254.
(19) Mandimutsira, B. S.; Yamarik, J. L.; Brunold, T. C.; Weiwei, G.;
Cramer, S. P.; Riordan, C. G. J. Am. Chem. Soc. 2001, 123, 9194-
9195.
(20) Hikichi, S.; Yoshizawa, M.; Sasakura, Y.; Akita, M.; Moro-Oka, Y.
J. Am. Chem. Soc. 1998, 120, 10567-10568.
(21) Kubiak and coworkers have examined related [PhBP3]Ni systems:
Kubiak, C. P. et al. Private communication, 2004.
(22) For examples of Ni complexes supported by [Tp] ligands see: (a)
Gutierrez, E.; Hudson, S. A.; Monge, A.; Nicasio, M. C.; Paneque,
M. J. Organomet. Chem. 1998, 551, 215-227. (b) Shirasawa, N.;
Nguyet, T. T.; Hikichi, S.; Moro-oka, Y.; Akita, M. Organometallics
2001, 20, 3582-3598. (c) Uehara, K.; Hikichi, S.; Akita, M.; Muneta,
A. J. Chem. Soc., Dalton Trans. 2002, 3529-3538. (d) Trofimenko,
S. Scorpionates-The Coordination Chemistry of Polypyrazolylborate
Ligands; Imperial College Press: London, 1999.
(8) Mindiola, D. J.; Hillhouse, G. L. J. Am. Chem. Soc. 2001, 123, 4623-
4624.
(9) Melenkivitz, R.; Mindiola, D. J.; Hillhouse, G. L. J. Am. Chem. Soc.
2002, 124, 3846-3847.
(10) Mindiola, D. J.; Hillhouse, G. L. J. Am. Chem. Soc. 2002, 124, 9976-
9977.
(11) Vicic, D. A.; Jones, W. D. J. Am. Chem. Soc. 1999, 121, 4070-4071.
(12) Waterman, R.; Hillhouse, G. L. J. Am. Chem. Soc. 2003, 125, 13350-
13351.
(23) Dapporto, P.; Fallani, G.; Sacconi, L. Inorg. Chem. 1974, 13, 2847-
2850.
(24) (a) Hou, H.; Gantzel, P. K.; Kubiak, C. P. Organometallics 2003, 22,
2817-2819. (b) Hou, H.; Gantzel, P. K.; Kubiak, C. P. J. Am. Chem.
Soc. 2003, 125, 9564-9565.
(13) Waterman, R.; Hillhouse, G. L. Organometallics 2003, 22, 5182-
5184.
4646 Inorganic Chemistry, Vol. 43, No. 15, 2004

The Coordination Chemistry of “[BP3]NiX” Platforms
Scheme 1
access to L3NiIIItE and L3NiIVtE species remains a
synthetic challenge.
II. Results and Discussion
IIa. Synthesis and Characterization of Pseudotetrahe-
II
dral [PhBP3]NiIIX and [PhBPiPr ]Ni X Complexes. Of
Figure 1. Displacement ellipsoid representations (50%) of (A) [PhBP3]-
NiI (2) and (B) [PhBPiPr3]NiCl (3). Hydrogen atoms and solvent molecules
have been omitted for clarity. Selected interatomic distances (Å) and angles
(deg) for 2: Ni-I, 2.498(1); Ni-P1, 2.289(2); Ni-P2, 2.265(2); Ni-P3,
2.267(2); Ni-B, 3.484(6); P1-Ni-P3, 91.52(6); P1-Ni-P2, 93.77(6); P2-
Ni-P3, 92.99(6); I-Ni-P1, 124.25(5); I-Ni-P2, 118.33(6); I-Ni-P3,
127.20(5). For 3: Ni-Cl, 2.1772(9); Ni-P1, 2.2871(8); Ni-P2, 2.2893(9);
Ni-P3, 2.2811(8); Ni-B, 3.474(3); P1-Ni-P2, 94.12(3); P2-Ni-P3,
94.97(3); P1-Ni-P3, 94.16(3); Cl-Ni-P1, 125.30(4); Cl-Ni-P2, 121.92(4);
Cl-Ni-P3, 118.94(3).
3
interest as synthetic precursors to the nickel chemistry
featured herein are the Ni(II) halides [PhBP3]NiX and
[PhBPiPr3]NiX, where X ) Cl or I. Divalent complexes of
these types have already been reported for [PhBP3]- and
[PhBPiPr3]-supported iron and cobalt systems,25 and a syn-
thetic protocol that exploits [PhBP3]Tl or [PhBPiPr3]Tl as a
transmetalation reagent is equally effective for nickel. Thus,
[PhBP3]Tl reacts with the nickel precursors (Ph3P)2NiCl2 or
NiI2 to eliminate TlX (X ) Cl, I) and generate the desired
green [PhBP3]NiCl (1) and red-brown [PhBP3]NiI (2) halide
complexes (Scheme 1) in good yield. NiCl2 could also be
used as a starting material, affording some 1 upon reaction
with [PhBP3]Tl, but undesirable side reactions proved more
problematic. Similarly, the synthesis of the yellow-green S
) 1 complex [PhBPiPr3]NiCl (3) is achieved in high isolated
yield (>90%) by reaction of the starting material (DME)-
NiCl2 (DME ) 1,2-dimethoxyethane) with [PhBPiPr3]Tl.
Evans method measurements of 1, 2, and 3 in benzene
solution are consistent with two unpaired electrons per nickel
center. The attempted synthesis of 3 using other Ni(II)
precursors, such as (Ph3P)2NiCl2 and NiCl2, again provides
some 3; however, the desired product is accompanied by
large amounts of diamagnetic impurities in each case. These
types of electron-transfer reactions would generate a reactive
borane radical that then would be expected to undergo bond
homolysis to liberate a •CH2PR2 radical. The [PhBPiPr3] anion
is inherently more reducing than its [PhBP3] counterpart, and,
as a result, its installation at nickel is somewhat more
sensitive.
The solid-state structures of 2 and 3 were determined by
X-ray diffraction (XRD) studies on single crystals of 2 and
3 (Figure 1). Each of the halide complexes is four-coordinate
and pseudotetrahedral, and the bond lengths and angles are
for the most part unremarkable. The iodide complex 2
features two short (2.265(2) and 2.267(2) Å) and one
modestly longer Ni-P (2.289(2) Å) bond, whereas the
chloride complex 3 exhibits three nearly equidistant bond
lengths (2.2871(8), 2.2893(9), and 2.2811(8) Å). The P-Ni-P
angles are on average slightly larger for 3 (94-95°) than
for 2 (91.5-94°). One comparison worth noting is that the
Ni-I bond length in 2 is appreciably longer (2.498(1) Å)
than the Ni-I bond length in the isostructural but cationic
Scheme 2
[L3NiI]+ species [(CH3C(CH2PPh2)3)NiI][As6I8] (2.414(4)
Å).26 Hence, placing a formal positive charge on the [L3-
NiI]+ fragment attracts the iodide anion closer to the nickel
center. While 2 can be formally described as zwitterionic
with a positive charge on the nickel unit and an anionic
charge on the borate unit, its nickel center nonetheless
appears to be appreciably less electrophilic than in the more
conventional cation [(CH3C(CH2PPh2)3)NiI]+. This conclu-
sion is consistent with other studies we have undertaken to
resolve electronic differences between phosphine-supported
zwitterions and their isostructural cations.27 In general, the
zwitterions appear to be more electron-rich by whichever
measuring stick we use as a qualitative guide.
With complex 1 in hand, we prepared a series of
complexes featuring other X-type ligands by metathetical
routes to determine what geometries would be preferred for
species of the general type [PhBP3]NiX (Scheme 2). For
example, the yellow-green azide complex [PhBP3]Ni(N3) (4)
can be prepared from sodium azide and exhibits a ν(N3)
vibration at 2060 cm-1 (KBr, C6H6), consistent with a
terminal azide functionality that is corroborated by its solid-
(26) Zanello, P.; Cinquantini, A.; Ghilardi, C. A.; Midollini, S.; Moneti,
S.; Orlandini, A.; Bencini, A. J. Chem. Soc., Dalton Trans. 1990,
3761-3766.
(27) (a) Thomas, J. C.; Peters, J. C. J. Am. Chem. Soc. 2003, 125, 8870-
8888. (b) Thomas, J. C.; Peters, J. C. J. Am. Chem. Soc. 2001, 123,
5100-5101.
(25) (a) Jenkins, D. M.; Di Billo, A. J.; Allen, M. J.; Betley, T. A.; Peters,
J. C. J. Am. Chem. Soc. 2002, 124, 15336-15350. (b) Betley, T. A.;
Peters, J. C. Inorg. Chem. 2003, 42, 5074-5084.
Inorganic Chemistry, Vol. 43, No. 15, 2004 4647

MacBeth et al.
Figure 2. Molecular structures of (A) [PhBP3]Ni(N3) (4), (B) [PhBP3]-
Ni(OSiPh3) (5), and (C) [PhBP3]Ni(S-p-tBu-Ph) (7), derived from crystal-
lographic studies. Only the L3NiX subunit of each complex is represented.
Complete structural details can be found in the Supporting Information.
Selected interatomic distances (Å) and angles (deg) for 4: Ni-N1,
1.915(12); Ni-P1, 2.262(2); Ni-P2, 2.261(2); Ni-P3, 2.255(2); P1-Ni-
P2, 95.09(7); P1-Ni-P3, 90.80(8); P2-Ni-P3, 91.67(6); P1-Ni-N1,
126.7(4); P2-Ni-N1, 121.1(3); P3-Ni-N1, 122.4(4). For 5: Ni-O, 1.820-
(2); Ni-P1, 2.275(1); Ni-P2, 2.270(1); Ni-P3, 2.281(1); P1-Ni-P2,
89.77(3); P1-Ni-P3, 94.06(3); P2-Ni-P3, 91.65(3); P1-Ni-O, 123.59(8);
P2-Ni-O, 89.77(3); P3-Ni-O, 130.20(8); Ni-O-Si, 168.08(14). For
7: Ni-S, 2.119(1); Ni-P1, 2.189(1); Ni-P2, 2.188(1); Ni-P3, 2.244(1);
P1-Ni-P2, 89.44(3); P1-Ni-P3, 91.28(3); P2-Ni-P3, 107.87(3); P1-
Ni-S, 151.84(3); P2-Ni-S, 109.79(3); P3-Ni-S, 101.54(3).
Figure 3. Displacement ellipsoid representation (50%) of (A) [PhB(CH2P-
(O)Ph2)2(CH2PPh2)]NiCl (8A) and (B) {[PhB(CH2P(O)Ph2)2(CH2PPh2)]-
NiCl}2 (8B). Hydrogen atoms, phosphine phenyl rings, and solvent
molecules have been omitted for clarity. Selected bond distances (Å) and
angles (deg) for 8A: Ni-Cl, 2.210(1); Ni-O1, 1.926(2); Ni-O2, 1.958(2);
Ni-P3, 2.294(1); O1-Ni-O2, 98.79(7); O1-Ni-P3, 105.28(5); O2-Ni-
P3, 97.64(5); O1-Ni-Cl, 116.74(5); O2-Ni-Cl, 125.31(5); P3-Ni-Cl,
110.07(3). For 8B: Ni1-Cl1, 2.3543(11); Ni1-Cl2, 2.3814(12); Ni2-Cl1,
2.3311(12); Ni2-Cl2, 2.3916(11); Ni1-O1, 2.005(3); Ni1-O2, 1.989(3);
Ni2-O3, 2.006(2); Ni2-O4, 1.980(3); Ni1-P3, 2.3125(13); Ni2-P6,
2.3236(13).
are low spin and exhibit a square planar geometry.33 The
geometry of paramagnetic 7 (Evans method (C6D6, 298 K):
µeff ) 3.02 µB) lies somewhere between a severely distorted
tetrahedral structure type and a cis-divacant octahedron. Most
striking is the dramatic variation in its three P-Ni-S bond
angles, one of which is very large (151.84(3)°) and the other
two (107.87(3)° and 109.79(3)°) of which are much closer
to those of an idealized tetrahedron. Also of note is that the
average Ni-P bond lengths of 7 (av 2.21 Å) are appreciably
shorter than for 2 (av 2.27 Å), 4 (av 2.26 Å), and 5 (av 2.28
Å). Although the solid-state structure of aryloxide 6 has not
state molecular structure.28 The molecular geometry of 4,
depicted in Figure 2, is monomeric and best described as
pseudotetrahedral.29 To our knowledge, 4 represents the only
example of a crystallographically characterized nickel azide
complex in which the azide ligand is coordinated to a single
tetrahedral or pseudotetrahedral nickel center. Other nickel
azides supported by sterically hindered tris(pyrazolyl)borate
ligands have been prepared that present tetrahedral or five-
coordinate structures in which the azide bridges two nickel
centers.30
The green siloxide ([PhBP3]Ni(OSiPh3), 5), the purple
aryloxide ([PhBP3]Ni(O-p-tBu-Ph), 6), and the red arylthi-
olate ([PhBP3]Ni(S-p-tBu-Ph), 7) are all conveniently pre-
pared by transmetalation with thallium reagents. The mo-
lecular structures of 5 and 7 are shown in Figure 2. The
structure of 5 is related to its recently reported cobalt
congener [PhBP3]Co(OSiPh3).31 The Ni-O-Si bond angle
of 168.08(14)° compares well with the Co-O-Si angle of
172.5(1)°. Perhaps structurally more interesting is the red
thiolate complex 7. A large number of nickel thiolates have
been prepared, and their chemistry is of increasing interest
due to their likely role in biocatalytic systems such as NiFe
hydrogenase.32 Four-coordinate nickel complexes that feature
a single Ni-thiolate linkage remain quite rare, and to our
knowledge all such examples that have been characterized
1
been determined, we note that its paramagnetic H NMR
resonances in benzene-d6 are very similar to those observed
for 7, suggesting that they have similar solution (and
presumably solid-state) conformations.
The divalent [PhBP3]NiX complexes 1-7 are all quite
sensitive to dioxygen. While this reactivity has not been
studied explicitly for each case, the product of the reaction
of 1 with dioxygen has been determined. Thus, exposure of
1 to an excess of O2 in benzene results in a rapid color change
from green to purple. This color change is accompanied by
a new and intense ν(PdO) vibration in the IR spectrum (1320
cm-1) confirming a ligand oxidation reaction. We have
observed this type of reactivity for tris(phosphino)borate-
supported Fe(II) and Co(II) systems previously.25 The nickel
oxidation product was isolated as purple crystals at room
temperature by vapor diffusion of petroleum ether into a
benzene solution. XRD analysis of a representative single
crystal established the monomeric and four-coordinate nickel
complex [PhB(CH2P(O)Ph2)2(CH2PPh2)]NiCl (8A) (eq 1 and
Figure 3A), in which two of the phosphine donors have been
oxidized to their corresponding phosphine oxides. Curiously,
yellow rather than purple crystals are obtained if petroleum
(28) Nakamato, K. Infrared and Raman Spectra of Inorganic and Coor-
dination Compounds. Part B: Applications in Coordination, Orga-
nometallic, and Bioinorganic Chemistry, 5th ed.; John Wiley &
Sons: New York, 1997; pp 105-126.
(29) The structure of 4 suffers from modest disorder along its Ni-azide
bond vector, and the thermal ellipsoid of the azide N atom directly
linked to the nickel center is consequently quite large (see Supporting
Information).
(30) Trofimenko, S.; Calabrese, J. C.; Kochi, J. K.; Wolowiec, S.;
Hulsbergen, F. B.; Reedijk, J. Inorg. Chem. 1992, 31, 3943-3950.
(31) Jenkins, D. M.; Peters, J. C. J. Am. Chem. Soc. 2003, 125, 11162-
11163.
(32) (a) Volbeda, A.; Charon, M.-H.; Piras, C.; Hatchikian, E. C.; Frey,
M.; Fontescilla-Camps, J. C. Nature 1995, 373, 580-587. (b)
Sellmann, D.; Geipel, F.; Moll, M. Angew. Chem., Int. Ed. 2000, 39,
561-563.
(33) (a) Sa´nchez, G.; Ruiz, F.; Serrano, J. L.; Ram´ırez de Arellano, M. C.;
Lo´pez, G. Eur. J. Inorg. Chem. 2002, 2185-2191. (b) Clegg, W.;
Henderson, R. A. Inorg. Chem. 2002, 41, 1128-1135. (c) Tucci, G.
C.; Holm, R. H. J. Am. Chem. Soc. 1995, 117, 6489-6496. (d) Kru¨ger,
H.-J.; Holm, R. H. Inorg. Chem. 1989, 28, 1148-1155. (e) Kanatzidis,
M. G. Inorg. Chim. Acta 1990, 168, 101-103.
4648 Inorganic Chemistry, Vol. 43, No. 15, 2004

The Coordination Chemistry of “[BP3]NiX” Platforms
appreciably lower potential (-1.44 V, Figure 4a). An
increase in the solution concentration of 3 (g1 × 10-3 M)
resulted in rapid deposition of material on the glassy carbon
electrode and the disappearance of the reduction wave. The
NiII/I couple for 3 as compared to 1 (∆E1/2 ) 240 mV) reflects
an appreciable difference in electron-releasing character of
the [PhBPiPr3] ligand relative to its [PhBP3] counterpart. This
distinction is consistent with the cyclic voltammograms
obtained previously for the complexes [PhBPiPr3]FeCl and
[PhBP3]FeCl (∆E1/2 ) 322 mV).15,25b Important to note is
that the cyclic voltammograms for complexes 1, 2, and 3 do
not provide any evidence for an oxidation wave within the
detectable range (i.e., below ∼1.0 V).
It is interesting to compare these data to those of several
other L3NiIIX systems that have been characterized electro-
chemically. For example, the cationic tripodal phosphine
system [(CH3C(CH2PPh2)3)NiX][ClO4] has been examined
previously and shows a fully reversible NiII/I redox processes
at -0.53 V (X ) Cl) and -0.43 V (X ) I) in CH2Cl2
solution relative to Fc/Fc+.26,35 The NiII/I couple is thus shifted
approximately 700 mV positive on translating from the
neutral L3NiIIX structure type to the isostructural but cationic
L3NiIIX+ structure type. The magnitude of the difference in
redox couples between the two systems seems large given
the zwitterionic nature of the neutral [PhBP3]NiX system.
An appreciable difference in the degree of solvent reorga-
nization upon oxidation is to be expected, and this difference,
coupled with the increased electron-richness we expect for
zwitterionic systems supported by (phosphino)borates, sug-
gests that a relatively large difference in redox potentials is
in fact quite reasonable. Related infrared carbonyl model
studies from our lab have comparatively examined electronic
factors between zwitterions and their isostructural cations
and revealed unexpectedly large differences in ν(CO) values
between such systems, the zwitterions always giving rise to
much lower CO frequencies.36,37,54 It is also interesting to
compare Riordan’s tris(thioether)borate complex [PhTttBu]-
NiCl ([PhTttBu] ) [PhB(CH2StBu)3]-), which features an
irreVersible NiII/I redox process at -1.3 V relative to Fc/
Fc+ in CH2Cl2.38 The irreversible nature of its one electron
reduction makes a direct comparison to the one electron
reduction potentials of the [PhBP3]NiX systems tenuous. It
is nonetheless qualitatively interesting, and perhaps surpris-
ing, that the reduction potential for Riordan’s sulfur-
supported nickel system is approximately as low as the
phosphine-supported [PhBP3]Ni(II) complexes 1 and 2.
The cyclic voltammetry of the [PhBP3]NiX complexes (X
) -OSiPh3, 5; -O-p-tBu-Ph, 6; -S-p-tBu-Ph, 7) was also
studied between +1.2 and - 3.0 V to examine what effect
X-type ligands of better π-donor strength might have on the
Ni(II/I) redox process and to examine whether such ligands
might better stabilize the Ni(III) oxidation state. As a point
Figure 4. Cyclic voltammograms of (a) 3, (b) 1, and (c) 2 recorded at a
scan rate of 100 mV/s in THF with Bu4NPF6 as the supporting electrolyte.
Potentials are referenced to Fc/Fc+.
ether is allowed to diffuse into a toluene solution of the
oxidation product at low temperature (-40 °C). XRD
analysis of a representative yellow crystal establishes the
dimeric form of the oxidized chloride, {[PhB(CH2P(O)-
Ph2)2(CH2PPh2)]NiCl}2 (8B), in which each of the chloride
ligands now resides in a bridging position (Figure 3B). The
structure of 8B suffers from several disordered solvent
molecules in the lattice; however, the dimeric nickel molecule
is well-defined. The two structures intimate that monomeric
8A is in equilibrium with its dimeric partner 8B and that
the equilibrium likely shifts in favor of the dimer at lower
temperatures. A similar temperature dependent monomer/
dimer equilibrium has been observed for the Ni(II) complex
dichloro[bis(3,5-dimethylpyrazolyl)methane]Ni.34
excess O2
[PhBP3]NiCl
benzene 8
1
{[PhB(CH2P(O)Ph2)2(CH2PPh2)]NiCl}
(1)
8A or 8B
IIb. Electrochemical Data for Divalent Complexes. The
tris(phosphino)borate ligands [PhBP3] and [PhBPiPr3] ef-
fectively stabilize monovalent iron and cobalt complexes that
prove to be effective precursors for the installation of multiple
bond linkages via oxidative group transfer.14-16 To determine
the accessibility of monovalent nickel species we first studied
the cyclic voltammetry of the divalent nickel precursors.
These studies were performed in THF using [nBu4N][PF6]
(0.35 M) as the supporting electrolyte. All data were
referenced to Fc/Fc+ as an internal standard using Ag/AgNO3
as the reference electrode.
Figure 4 compares the first reduction process observed
for the halide complexes 1, 2, and 3. Each process is assigned
as a one-electron reduction of L3NiIIX to [L3NiIX]-. A fully
reversible reduction process is observed for the chloride
complex 1 at -1.20 V (Figure 4b). The NiII/I couple shifts
to -1.12 V (Figure 4c) upon replacement of chloride by the
weaker field iodide ligand in 2. The cyclic voltammetry of
3 shows a quasi-reversible redox process (100 mV/s) at an
(35) The authors report potentials versus SCE and corrected these values
relative to Fc/Fc+.
(36) Betley, T. A.; Peters, J. C. Inorg. Chem. 2003, 42, 5074-5084.
(37) Betley, T. A.; Peters, J. C. Angew. Chem., Int. Ed. 2003, 42, 2385-
2389.
(38) Schebler, P. J.; Riordan, C. G.; Guzei, I. A.; Rheingold, A. L. Inorg.
Chem. 1998, 37, 4754-4755.
(34) Reedijk, J.; Verbiest, J. Transition Met. Chem. 1978, 3, 51-52.
Inorganic Chemistry, Vol. 43, No. 15, 2004 4649

MacBeth et al.
Scheme 3
of rather soft tris(thioether)borate ligands to stabilize Ni(I).
For example, the reduction of [PhTttBu]NiCl by methyllithium
under a blanket of CO produces the Ni(I) species [PhTttBu]-
Ni(CO).39 Ni(I) complexes species supported by betadiketim-
inate ligands have also been reported recently.40
The direct reduction of chloride 2 by Na/Hg in THF led
to the generation of a diamagnetic red organometallic,
[PhBP3]Ni(η2-CH2PPh2) (9), rather than the anticipated
[PhBP3]NiICl- anion suggested by the cyclic voltammetry
of 2. The same product was formed when other strong
reductants were examined, including sodium mirror and
potassium graphite. Complex 9 can be synthesized indepen-
dently in high yield by the direct reaction of 1 with Ph2-
PCH2Li(TMEDA) (Scheme 3), and it appears to be a
thermodynamic sink within the [PhBP3]Ni system. For
instance, when potentially reducing nucleophiles, such as
alkyllithiums or lithium amides, are added to chloride 2 at
low temperature (-78 °C), complex 9 is invariably generated
as the dominant reaction product. Riordan’s [PhTttBu]NiCl
system was reported to behave in a similar fashion in the
presence of alkyllithiums without a trapping ligand such as
CO or PMe3 available.39 An obvious unanswered question
for both systems concerns whether (i) a bona fide NiII-alkyl
(or NiII-amide) is generated but is inherently unstable within
these pseudotetrahedral borate-based systems and somehow
degrades to 9 or (ii) outer-sphere electron transfer occurs as
a first step to generate L3NiI-Cl-, which is unstable to
chloride loss and subsequent degradation to 9. At present
we can only conclude that complexes 1 and 2 are sensitive
to reducing conditions, and that borate ligand degradation
can occur even at low temperatures to produce 9. Degradation
processes are less problematic when a donor ligand trap is
used (vide infra).
Figure 5. Cyclic voltammograms of (a) 5, (b) 6, and (c) 7 recorded at a
scan rate of 100 mV/s. Potentials are referenced to Fc/Fc+.
of reference, we have previously reported that the complex
[PhBP3]Co(OSiPh3) exhibits fully reversible CoIII/II and CoII/I
redox processes, but that its iodide counterpart [PhBP3]CoI
exhibits only a single reversible redox couple (CoII/I) along
with an irreversible CoIII/II oxidation wave.25b,31 Reversible
NiII/I reduction processes are observed for each of the
complexes 5, 6, and 7 (Figure 5). The siloxide complex 5
exhibits the most negative potential at -1.47 V (Figure 5a).
Like its cobalt relative, complex 5 also exhibits an oxidation
wave at higher potential (0.350 V), but in this case the
process is irreversible. Its irreversibility suggests that the
putative [NiIII-OSiPh3]+ species generated decays rapidly
under these conditions. A similar set of observations was
made for the aryloxide complex 6. It exhibits a single and
reversible reduction process (NiII/I) at -1.36 V (Figure 5b),
and an irreversible oxidation event at 0.240 V, suggesting
that complex 6 is both easier to reduce and easier to oxidize
than the siloxide complex 5. The arylthiolate complex 7
displays the most readily accessible NiII/I redox couple (-1.12
V, Figure 5c). Perhaps more interesting is that 7 also shows
a second reduction event at -2.56 V. This latter process is
irreversible, and while we suspect that it corresponds to a
NiI/0 redox change, it may also represent a reduction event
at the arylthiolate ligand. As discussed below, the Ni(0) state
is more readily accessible within these BP3Ni systems when
π-acidic ligands are coordinated to the fourth binding site.
Finally, an irreversible oxidation event is observed at 0.170
V for 7, establishing that it is more easily oxidized than either
5 or 6. Considered collectively, these electrochemical data
establish that monovalent nickel is readily accessible within
these BP3Ni platforms, but that the stabilization of a trivalent
nickel state is likely to require more strongly π-donating
X-type ligands.
A reaction worthy of particular comment is that between
CoCp2 and 1 in the presence of [nBu4N][PF6]. The crude
product of this reaction is a new paramagnetic species (1H
NMR) that is tentatively formulated as the one electron
reduced complex {[PhBP3]NiICl}{nBu4N}. This species is
very unstable in solution and has not been isolated in
analytically pure form. Solution degradation occurs to
produce a mixture of both paramagnetic and diamagnetic
products even at low temperatures (-40 °C). Nonetheless,
its assignment is consistent with the following observation:
rapid reoxidation of the crude product using [Fc][PF6]
regenerates the starting nickel halide 1 in ∼85% yield (1H
NMR).
IIc. Chemical Reduction of [PhBP3]NiX and [PhBPiPr3]-
NiCl Precursors. Given the electrochemical stability of
monovalent nickel within these P3NiX platforms, it was of
interest to investigate the preparation of Ni(I) complexes for
subsequent group transfer studies. Efforts to prepare and
study the reactivity patterns of molecular Ni(I) species have
only recently been pursued in a systematic fashion. For
instance, Hillhouse and co-workers have prepared a series
of L2NiIX complexes that feature halides, pseudohalides, and
alkyls as the X-type ligand using relatively soft phosphine
donor ligands.8-10 Riordan’s group has moreover made use
The related reduction chemistry of [PhBPiPr3]NiCl (3) with
a variety of typical reductants, for example, Na/Hg, sodium
(39) Schebler, P. J.; Mandimutsira, B. S.; Riordan, C. G.; Liable-Sands,
L. M.; Incarvito, C. D.; Rheingold, A. L. J. Am. Chem. Soc. 2001,
123, 331-332.
(40) Holland, P. L.; Cundari, T. R.; Perez, L. L.; Eckert, N. A.; Lachicotte,
R. J. J. Am. Chem. Soc. 2002, 124, 14416-14424.
4650 Inorganic Chemistry, Vol. 43, No. 15, 2004

The Coordination Chemistry of “[BP3]NiX” Platforms
Scheme 4
diisopropylphenyl)}, also suggests the formation of three
coordinate NiII-NR2 structures. These products are likewise
diamagnetic and give highly colored solutions (red and
purple, respectively). It appears that the [PhBPiPr3]NiII
scaffold is more resistant to outer-sphere reduction and
thereby allows cleaner substitution to occur at the Ni(II) state.
However, the pseudotetrahedral geometry is compromised
by dechelation of one phosphine donor.
Figure 6. Displacement ellipsoid representation (50%) of [κ2-PhBPiPr3]-
Ni(dbabh) (11). Hydrogen atoms have been omitted for clarity. Selected
distances (Å) and angles (deg): P1-Ni, 2.1475(7); P2-Ni, 2.1317(7);
N-Ni, 1.748(2); P1-Ni-P2, 93.38(3); P1-Ni-N, 136.86(7); P2-Ni-N,
129.65(7).
naphthalenide, potassium graphite, Mg0, and (9,10-dihydro-
9,10-anthracenediyl)tris(tetrahydrofuran)magnesium, was also
surveyed. These reactions did not result in the formation of
the degradation product [PhBPiPr3]Ni(η2-CH2PiPr2) (10),
which was synthesized independently by reaction of 3 with
LiCH2PiPr2 in THF (Scheme 3). Ill-defined product mixtures
resulted instead.
IId. Synthesis and Characterization of [PhBP3]NiIL and
[PhBPiPr3]NiIL Complexes. The presence of a suitable L-
type trapping ligand can stabilize the Ni(I) state to give syn-
thetically useful precursors (Scheme 4). For example, the
reduction of 1 in the presence of PPh3 by KC8 yields [PhBP3]-
Ni(PPh3) (12) in good yield. PMe3 is also an effective trap.
The core structure of 12 is shown in Figure 7. Similarly, the
reduction of 1 by sodium amalgam in the presence of tBuNC
results in the formation of [PhBP3]Ni(CNtBu) (13). It is ob-
The [PhBPiPr3] anion lowers the reduction potential of the
divalent nickel state considerably as observed in our elec-
trochemical studies, and we therefore suspected that [PhBPiPr3]-
NiCl might give rise to cleaner substitution chemistry with
potentially reducing nucleophiles rather than problematic
redox chemistry. The reactivity of 3 was therefore explored
with a host of lithiated amines. One example concerns the
lithium amide Li(dbabh) (Hdbabh ) 2,3:5,6-dibenzo-7-
azabicyclo[2.2.1]hepta-2,5-diene),41 chosen as a possible
N-atom transfer reagent (vide infra). The reaction between
3 and Li(dbabh) at low temperature results in an inky green-
brown solution from which the three-coordinate nickel(II)
complex [κ2-PhBPiPr3]Ni(dbabh) (11) can be isolated. The
three coordinate nature of 11 in solution is strongly suggested
by its NMR data, which shows a diamagnetic product.
Interestingly, only one phosphine environment is evident by
t
served that the addition of BuNC to 1 prior to addition of
the reductant causes an immediate color change from green
to purple. NMR spectra of the purple reaction mixture (1H
and 31P{1H}) show the formation of a diamagnetic interme-
diate prior to reduction, presumably “[κ2-PhBP3]Ni(Cl)
(CNtBu)”. Complexes 12 and 13 are paramagnetic and
exhibit solution magnetic moments of 1.91 µB and 1.74 µB,
respectively. Complex 13 exhibits an intense ν(CN) band in
its IR spectrum at 2094 cm-1 (KBr, C6H6) suggesting that
the isocyanide ligand is terminal.28 This is corroborated by
its solid-state molecular structure (Figure 7).
The [PhBPiPr3]NiCl (3) precursor can be similarly reduced
to form P3NiIL complexes (Scheme 4). Reduction of 3 using
a sodium/mercury amalgam in the presence of excess PMe3
(5 equiv) results in the formation of analytically pure bright
yellow [PhBPiPr3]Ni(PMe3) (14). Yellow 14 is also formed
in good yield from the reaction of the Ni(0) precursor Ni-
(PMe3)4 with [PhBPiPr3]Tl. This reaction proceeds with
concomitant reduction of Tl(I) to Tl(0), evident by the
precipitation of thallium metal from solution. Crystals of 14
were obtained from this latter reaction by crystallization from
diethyl ether at -30 °C. XRD analysis showed that the
crystals contained a 95:5 mixture of 14 and [PhBPiPr3]Tl.
The presence of a small amount of [PhBPiPr3]Tl in the crystals
was also visible from their 31P{1H} NMR spectrum and by
the presence of a well-refined peak in the difference Fourier
map attributable to the presence of thallium (see Supporting
Information for refinement details). The molecular structure
of 14, depicted in Figure 7, shows a four-coordinate and
pseudotetrahedral Ni(I) center very similar to that for 12.
Measurement of the magnetic moment of analytically pure
14 by the method of Evans in benzene solution provides µeff
) 1.82 µB, consistent with one unpaired electron for the d9
electronic configuration. A [PhBPiPr3]-ligated Ni(I) isocyanide
complex has also been prepared by a similar method.
1
both H NMR and 31P{1H} NMR spectroscopy, indicative
that the phosphine ligands of 11 are in rapid exchange.
Definitive assignment of the structure of 11 is provided by
its solid-state molecular structure (Figure 6). As can be seen,
the amide ligand adopts a conformation such that its available
π-orbital lies in the P2Ni plane. This conformation is similar
t
to a related (dtbpe)Ni-NR2 species (dtbpe ) Bu2PCH2-
CH2PtBu2) prepared by the Hillhouse group.8 When compar-
ing 11 to other Ni-amide complexes, it is interesting to note
that most of their Ni-N distances range from 1.82 to 1.93
Å,42 approximately 0.1 Å longer than that observed for 11.
The Ni-N distance of 11 (1.748(2) Å) is quite similar,
however, to that reported for [(dtbpe)Ni-NH{2,6-
(CHMe2)2C6H3}][PF6] (1.768(14) Å).8 The short Ni-N bond
length suggests a significant π-interaction, consistent with
the orientation of the amide ligand. The dechelation of one
phosphine ligand in 11 is likely favored due to this
π-interaction. In addition to Li(dbabh), the reaction between
3 and lithium anilides, such as Li(NHPh) and Li{NH(2,6-
(41) Mindiola, D. J.; Cummins, C. C. Angew. Chem., Int. Ed. 1998, 37,
945-947.
(42) See ref 8 and references therein.
Inorganic Chemistry, Vol. 43, No. 15, 2004 4651

MacBeth et al.
Figure 7. Molecular structures of (A) [PhBP3]Ni(PPh3) (12), (B) [PhBP3]Ni(CNtBu) (13), (C) [PhBPiPr3]Ni(PMe3) (14), (D) [PhBPiPr3]Ni(CNtBu) (15), and
(E) [PhBP3]NiNO (16), derived from crystallographic studies. Only the L3NiX subunit of each complex is represented. Complete structural details can be
found in the Supporting Information. Selected interatomic distances (Å) and angles (deg) for 12: Ni-P1, 2.28(2); Ni-P2, 2.28(2); Ni-P3, 2.27 (2); Ni-P4,
2.25(2); Ni-B, 3.44(9); P1-Ni-P4, 120.1(8); P2-Ni-P4, 122.1(8); P3-Ni-P4, 123.0(8); P1-Ni-P2, 95.6(8); P2-Ni-P3, 96.8(8); P1-Ni-P3, 92.2(8).
For 13: Ni-C46, 1.861(2); Ni-P1, 2.229(1); Ni-P2, 2.216(1); Ni-P3, 2.252(1); P1-Ni-P2, 91.86(2); P1-Ni-P3, 95.00(2); P2-Ni-P3, 94.58(2); P1-
Ni-C46, 125.59(6); P2-Ni-C46, 107.38(6); P3-Ni-C46, 131.77(6). For 14: Ni-P1, 2.2533(5); Ni-P2, 2.2764(5); Ni-P3, 2.2873(5); Ni-P4, 2.2631(5);
Ni-B, 3.436(2); P1-Ni-P4, 118.67(2); P2-Ni-P4, 121.90(2); P3-Ni-P4, 123.43(2); P1-Ni-P2, 92.76(2); P2-Ni-P3, 98.24(2); P3-Ni-P1, 94.74(2).
For 15: Ni-C28, 1.864(2); Ni-P1, 2.2403(5); Ni-P2, 2.2749(6); Ni-P3, 2.2562(6); Ni-B, 3.431(2); P1-Ni-P2, 96.83(2); P1-Ni-P3, 93.50(2); P2-
Ni-P3, 96.42(2); P1-Ni-C28, 104.56(6); P2-Ni-C28, 124.37(7); P3-Ni-C28, 131.79(7). For 16: Ni-N, 1.624(3); N-O, 1.183(3); Ni-P1, 2.232(1);
Ni-P2, 2.215(1); Ni-P3, 2.234(1); O-N-Ni, 176.0(3); P1-Ni-P2, 92.54(3); P1-Ni-P3, 95.42(3); P2-Ni-P3, 91.74(3); P1-Ni-N, 124.68(9); P2-
Ni-N, 119.84(9); P3-Ni-N, 124.01(9).
Addition of excess (ca. 5 equiv) tBuNC to 3 in THF solution
results in the initial formation of a red, likely square planar
Ni(II) complex, [κ2-PhBPiPr3]Ni(Cl)(CNtBu) (19). Assignment
of this species is based upon its 31P{1H} NMR spectrum (δ
62.7 (br d), 47.7 (br d), 3.2 (br s)). Removal of the reaction
volatiles followed by redissolution in THF and subsequent
reduction using sodium/mercury amalgam provides the
yellow Ni(I) complex [PhBPiPr3]Ni(CNtBu) (15) in good
yield. Attempts to reduce 3 in the presence of excess CNt-
Bu result in the formation of an increased number of side
Figure 8. Cyclic voltammograms of (a) 15 and (b) 14 recorded at a scan
rate of 50 mV/s. Potentials are referenced to Fc/Fc+.
products, presumably due to the competitive reduction of
isocyanide. Complex 15 is more readily prepared in high
yield by the reaction of 14 with excess (ca. 5 equiv) CNtBu,
which substitutes PMe3. The solution magnetic moment of
15 as determined by the Evans method is 1.68 µB. Examina-
tion of the IR spectrum of 15 provides a ν(CN) stretching
frequency of 2056 cm-1 (KBr, THF). The increased electron-
releasing character of [PhBPiPr3] versus [PhBP3] is quite
striking from the ν(CN) vibrations recorded for 15 and 13,
the former being shifted to lower energy by 38 cm-1. The
relatively similar solid-state structures of 15 and 13 are
evident in Figure 7. Hence, this electronic comparison is a
good qualitative benchmark for the [PhBPiPr3] and [PhBP3]
anions. To the best of our knowledge, complexes 13 and 15
are the only structurally characterized, mononuclear Ni(I)
isocyanide complexes.
cyanide ligand. Both complexes show reduction processes,
presumably to Ni(0), at quite low potentials.
On the basis of these electrochemical observations, we
studied the independent chemical reduction of [PhBPiPr3]Ni-
(PMe3) (14). As suggested from the formation of 14 by the
reaction between [PhBPiPr3]Tl and Ni(PMe3)4, the putative
Ni(0) anion {[PhBPiPr3]Ni(PMe3)}- (20) is generated at very
low potential. Anionic {[PhBPiPr3]Ni(PMe3)}- appears to be
accessible by reduction of 14 with strong alkali metal
reductants, such as Na/Hg or KC8, in THF solution.
1
Examination of the H NMR spectra of such reduction
reactions shows complete consumption of paramagnetic 14
and the presence of broadened diamagnetic resonances
consistent with a fluxional Ni(0) anion. In THF solution,
these complexes have limited stability and ultimately reoxi-
dize to Ni(I) 14.
The reduction chemistry of neutral [PhBPiPr3]Ni(CNtBu)
15 was also studied. Chemical reduction of 15 using strong
reductants (e.g., KC8, Na/Hg) results in its complete con-
sumption and the formation of a diamagnetic species (1H
NMR). The solution infrared spectrum of this product shows
a dramatically shifted ν(CN) vibration (1982 cm-1, compared
with 2056 cm-1 for 15), indicating the presence of the Ni-
(0) anion {[PhBPiPr3]Ni(CNtBu)}- (21). This species is
kinetically unstable in benzene or THF solution and gradually
decays to regenerate the Ni(I) species 15 as the predominant
IIe. Access to the Ni(0) State. With Ni(I) complexes in
hand, we studied their proclivity to undergo redox events
by electrochemical methods. The cyclic voltammetry of the
Ni(I) complexes 14 and 15 was studied in THF solution.
The cyclic voltammetry of 14 shows a reversible reduction
to Ni(0) at -1945 mV (Figure 8b). An irreversible oxidation
event occurs at -503 mV. 15 gives rise to a low reduction
potential for the NiI/0 couple (-1848 mV) (Figure 8a), and
an irreversible oxidation event (-396 mV). The shift to more
positive potentials for both the oxidation and reduction
processes for 15 as compared to 14 reflects the exchange of
an electron-donating phosphine ligand for a π-acidic iso-
4652 Inorganic Chemistry, Vol. 43, No. 15, 2004

The Coordination Chemistry of “[BP3]NiX” Platforms
product. Whether some type of disproportionation reaction
or reaction with solvent occurs is presently unclear. The
degradation process to regenerate 15 is quite clean, suggest-
ing the involvement of some external electron acceptor such
as THF solvent.
is achieved by reaction of [PhBPiPr3]Tl with (Ph3P)2Ni(CO)2
in the presence of ASNBr (ASN ) 5-azoniaspiro[4.4]nonane)
in THF solution. The substitution reaction is much slower
in the latter case, but proceeds in reasonable yield after a
period of 3 days at room temperature to form [[κ2-PhBPiPr3]-
Ni(CO)2][ASN] (18). Attempts to accelerate the reaction via
mild heating (50 °C) result in extensive decomposition of
the starting material to unidentified products. Complex 18
displays a characteristic infrared spectrum (2004 (sym) and
1878 cm-1 (antisym); KBr, THF) and 31P{1H} NMR
spectrum (δ 51.1 (s, 2P), 4.8 (s, 1P)). Photolysis of 17 under
a sparge of N2 appears to affect replacement of one of the
CO ligands by the remaining phosphine donor to produce
the κ3 species {[PhBP3]Ni(CO)}{nBu4N} (22). A single sharp
resonance is observed in the 31P{1H} NMR spectrum for 22
at 25.3 ppm, and a single ν(CO) vibration is observed at
1869 cm-1 (KBr, THF). This value is comparable but
appreciably lower than the ν(CO) value of 1890 cm-1 (Nujol)
reported for the neutral Ni(0) complex (triphos)Ni(CO).48
We were surprised to find that attempts to oxidize 17 and
18 by [Fc][PF6] to give well-defined Ni(I) monocarbonyl
products instead led to multiple products as determined by
IR and NMR spectroscopy. Given the stability of Riordan’s
[PhTttBu]NiI(CO) and related Ni(I) carbonyls,39 we suspect
overoxidation to be problematic. Gentler oxidants may yet
prove synthetically effective.
Synthetic access to other zerovalent nickel species sup-
ported by these phosphine anions is possible.43 For example,
treatment of 1 with NO gas generates the diamagnetic nitrosyl
product [PhBP3]Ni(NO) (16) as a crystalline red product (eq
2). This complex is formally an 18-electron species and is
isolobal to its well-known relatives CpNi(NO)44 and Cp*Ni-
(NO).45 The solid-state structure of 16 (Figure 7) shows a
pseudotetrahedral nickel complex with a nearly linear Ni-
N-O bond angle (176.0(3)°) and a very short Ni-N bond
distance of 1.624(3) Å. Complex 16 also exhibits a rather
low ν(NO) vibration at 1737 cm-1 (KBr, C6H6), considerably
lower than observed for the tris(thioether)borate system
[PhTttBu]Ni(NO) (1785 cm-1, CH2Cl2)38 and the Cp* system
Cp*Ni(NO) (1787 cm-1, Nujol).46 Another structurally
related example is Sacconi’s four-coordinate nickel nitrosyl
[(κ3(P)-np3)Ni(NO)][BPh4] (np3 ) N(CH2CH2PPh2)3).47 The
average of the P-Ni bond distances in [(κ3(P)-np3)Ni(NO)]-
[BPh4] is 2.293(6) Å, and its Ni-NO bond distance is 1.59(2)
Å. Thus, all of the Ni-ligand linkages contract for the cation
[(np3)Ni(NO)]+ by comparison to 16. Cationic [(κ3(P)-np3)-
Ni(NO)][BPh4] displays a ν(NO) at 1755 cm-1 (Nujol),
intermediate between those for 16 and [PhTttBu]Ni(NO) or
Cp*Ni(NO). The remarkably low ν(NO) frequency recorded
for 16 indicates a very large degree of π-back-bonding and
is perhaps suggestive that the NidNdO resonance contribu-
tor is meaningful. The N-O bond distance of 1.183(3) Å is
nonetheless similar to those recorded in the other structures.
IIf. Synthetic Attempts To Generate NidE/NitE Link-
ages. One of the primary goals of this study was to assess
the propensity of the low valent nickel precursors described
herein to mediate oxidative group transfer processes that
install NidE/NitE multiple bond linkages at a pseudotet-
rahedral nickel center. We have canvassed numerous reac-
tions that might have led to such species, but to date have
not isolated or spectroscopically verified well-defined ex-
amples of L3NidE/L3NitE. It is nonetheless informative
to briefly summarize these attempts.
NO
8
[PhBP ]Ni(NO)
1 or 12
(2)
3
16
Access to Ni(0) carbonyl complexes was achieved by
reaction of either of the tris(phosphino)borate ligands with
(Ph3P)2Ni(CO)2. Thus, the anionic complex [[κ2-PhBP3]Ni-
(CO)2][nBu4N] (17) is synthesized cleanly by the reaction
between [PhBP3]Tl and (Ph3P)2Ni(CO)2 in the presence of
[nBu4N][Br] in THF over a period of 12 h. Its 31P{1H} NMR
spectrum shows two distinct resonances at 29.2 and -9.46
ppm (2:1 ratio) that correspond to its coordinated and
dissociated phosphine donors, respectively. The solution IR
spectrum of this complex (KBr, THF) shows symmetric and
antisymmetric ν(CO) vibrations at 1994 and 1913 cm-1.
Synthesis of the analogous [PhBPiPr3]-containing complex
One of the more interesting target molecules whose
stability we sought to evaluate concerns the tetravalent L3-
NiIVtN. This hypothetical d6 unit would be both isolobal
and isoelectronic to remarkably stable d6 L3CoIIItNR species
that our lab has prepared previously.14,16 It would moreover
be structurally related to a recently reported d4 L3FeIVtN
system.18 Various routes were canvassed to generate a
putative NiIVtN species, and two are most worthy of
comment. Photolysis of the divalent azide 4 precursor might
effect N2 expulsion and thus nitride transfer to nickel to
produce “[PhBP3]NiIVtN”. We find, however, that the azide
ligand remains intact even under extended photolysis (Hg
lamp) in THF at room temperature. Extended heating of 4
results in a mixture of diamagnetic and paramagnetic
products, but there was no spectroscopic evidence to suggest
that any of these products correspond to a nitride functional
group (bridging or terminal). Another plausible route to the
terminal nitride concerns a two-electron oxidative N-atom
transfer from the dbabh amide to Ni(II). Precedence for this
type of transformation has been provided by Mindiola and
(43) For relevant reviews on the synthesis of transition metal nitrosyl
complexes see: (a) Mingos, M. D.; Sherman, D. J. AdV. Inorg. Chem.
1989, 34, 293-377. (b) Richter-Addo, G. B.; Legzdins, P. Metal
Nitrosyls; Oxford University Press: New York, 1992; pp 34-78.
(44) Crichton, O.; Rest, A. J. J. Chem. Soc., Dalton Trans. 1977, 986-
993.
(45) Schneider, J. J.; Goddard, R.; Krueger, C. Organometallics 1991, 10,
665-670.
(46) (a) Fomitchev, D. V.; Furlani, T. R.; Coppens, P. Inorg. Chem. 1998,
37, 1519-1526. (b) Green, J. C.; Underwood, C. J. Organomet. Chem.
1997, 528, 91-94.
(47) Di Vaira, M.; Ghilardi, C. A.; Sacconi, L. Inorg. Chem. 1976, 15,
1555-1561.
(48) Bianchini, C.; Mealli, C.; Meli, A.; Scapacci, G. Organometallics 1983,
2, 141-143.
Inorganic Chemistry, Vol. 43, No. 15, 2004 4653

MacBeth et al.
Cummins in the synthesis of a CrVItN nitride,41 and more
recently by our lab in the generation of [PhBPiPr3]FeIVtN.18
As described above, the reaction between Li(dbabh) and 3
installs the required NiII-N amide linkage, but the thermo-
dynamically stable product is the trigonal planar three-
coordinate NiII species 11 rather than the four-coordinate
complex. We had hoped that the [PhBPiPr3] ligand might be
sufficiently electron-releasing to induce the loss of anthracene
with concomitant nitride transfer to generate the NiIVtN
subunit. Dechelation of a phosphine arm occurs instead to
give a relatively stable amide. Complex 11 is quite stable at
room temperature in solution, and while its thermolysis at
slightly elevated temperature does produce a new diamag-
of unidentified diamagnetic species. The absence of new
paramagnetic species in the product profile suggests that
terminal NiIIIdCR2 species are not present.
Finally, we note that the reactivity of the Ni(0) species
{[PhBPiPr3]NiL}- 20 and 21 to group transfer was also
canvassed. These species were generated in situ by reduction
of their Ni(I) precursors using sodium/mercury amalgam.
Quantitative consumption of the Ni(I) species is confirmed
by 1H NMR spectroscopy, and subsequent addition of 2 equiv
of an organic azide or diazoalkane leads in each case to rapid
re-formation (<1 h) of the Ni(I) precursor (either 14 or 15)
in high yield. We suspect that the highly reducing Ni(0)-
species are acting as electron-transfer reagents and affecting
radical reduction of the organic substrate. A more systematic
analysis of the organic products of these reactions was not
undertaken.
1
netic species, the H NMR spectrum of the species defini-
tively shows the presence of an intact dbabh ligand, evident
from the signature bridgehead proton resonance (ca. 5.4
ppm). A definitive structural assignment for this thermolysis
product has not yet been made.
IIg. Theoretical Considerations. We undertook the
theoretical analysis of several hypothetical nickel complexes
featuring a multiply bonded linkage to explore the feasibility
of such species from an electronic structure perspective. To
contextualize these results, it is instructive to first consider
the L2NidE fragments of Hillhouse and co-workers which
are already known to give rise to isolable NidE complexes.
Bonding considerations for these complexes have been
addressed previously.49 Figure 9A shows lobal representa-
tions of the frontier orbitals of an L2MdE fragment as
derived from the classic text of Albright, Burdett, and
Whangbo.50 The LUMO orbital is of b2 symmetry (under
C2V) and is π* in character. The much lower lying, filled
MdE π-bonding orbital is not shown. As the transition is
made from the L2MdE fragment to the pseudotetrahedral
L3M-E structure under idealized 3-fold symmetry, the M-E
linkage is more appropriately described as a bona fide triple
bond (MtE) for a d6 configuration (Figure 9B).14,49,51 For
higher d-count configurations (d7 or d8) of similar geometry
the formal bond order is lowered because the e-set is both
σ* (with respect to the phosphines) and π* (with respect to
the multiply bonded ligand) in character. The extent to which
such configurations might be expected to be stable invariably
depends on how energetically destabilized the upper π*
orbitals can lie while still housing an unpaired electron. The
previously reported complexes [PhBP3]CoI14,52 and [PhBP3]-
Co(OSiPh3)31 provide well-defined examples of pseudotet-
rahedral d7 species with 2E ground state configurations that
serve as electronic models for the hypothetical S ) 1/2 [BP3]-
NiIII(NR) systems under present consideration. An obvious
issue concerns the extent to which the imide ligand,
conventionally thought to be quite strongly π-donating, will
By analogy to the previously reported reactions of [BP3]-
MIL precursors (where [BP3] ) [PhBP3] and [PhBPiPr3], M
) Fe and Co, L ) PR3) with aryl and alkyl azide (RN3)
substrates to generate MIIItNR imides,14,15 it was of obvious
interest to canvass similar reactions with the low valent nickel
precursors described herein. To this end, we surveyed a
number of reactions that might have facilitated two-electron
group transfer reactivity to install a NidNR linkage. The
reaction of 12, 13, 14, and 15 with 2 equiv of an organic
azide (typical organic azides used: p-tolyl azide, tert-butyl
azide, 1-adamantyl azide, tosyl azide, and trimethylsilyl
azide) in each case produces a mixture of paramagnetic and
diamagnetic species that is relatively ill-defined. The phos-
phine-capped Ni(I) complexes react relatively quickly at
room temperature (hours). By contrast, the Ni(I) isocyanide
complexes reacted much more slowly. In these latter cases,
starting material is still present even after several days at
room temperature. Attempts to accelerate the rate of these
reactions by heating reaction mixtures of 13 and 15 in the
presence of azide substrates did lead to more rapid consump-
tion of the nickel precursors, but also to an even greater range
of spectroscopically detectable species. A major organic
product of the reaction between 12 and 2 equiv of p-tolyl
1
azide is p-tolyl-NdPPh3 (identified by H NMR and GC/
MS), but it is unclear whether this product is formed by a
nickel-mediated reaction that proceeds through an insipient
“NiIIIdN-p-tolyl” fragment or by the direct reaction between
dissociated PPh3 and p-tolyl azide. The reaction between 13
and p-tolyl azide generates only a small amount of the
carbodiimide p-tolyl-NdCdNtBu (1H NMR, GC/MS).
We also surveyed the reaction between the NiI-L precur-
sors and diazoalkanes as potential two-electron carbene
transfer reagents. For example, we studied the reactivity of
14 and 15 in the presence of 2 equiv of Ph2CN2, Mes2CN2,
ethyl diazoacetate, and trimethylsilyldiazomethane in benzene
solution. Similar to the observations noted above, 14 is
consumed more rapidly (hours) than 15 (days) at ambient
temperature for each substrate except Mes2CN2, which does
not react over a period of several days at room temperature.
The resulting product mixtures invariably contain a variety
(49) (a) Albright, T. A.; Hoffmann, R.; Tse, Y.; D’Ottavio, T. J. Am. Chem.
Soc. 1979, 101, 3812-3821. (b) Albright, T. A.; Hoffmann, R.;
Thibeault, J. C.; Thorn, D. L. J. Am. Chem. Soc. 1979, 101, 3801-
3812. (c) Glueck, D. S.; Wu, J.; Hollander, F. J.; Bergman, R. G. J.
Am. Chem. Soc. 1991, 113, 2041-2054. (d) Cundari, T. R. J. Am.
Chem. Soc. 1992, 114, 7879-7888.
(50) Adapted from: Albright, T. A.; Burdett, J. K.; Whangbo, M.-H. Orbital
Interactions in Chemistry; Wiley & Sons: New York, 1985; pp 365,
383.
(51) Glueck, D. S.; Green, J. C.; Michelman, R. I.; Wright, I. N.
Organometallics 1992, 11, 4221-4225.
4654 Inorganic Chemistry, Vol. 43, No. 15, 2004

The Coordination Chemistry of “[BP3]NiX” Platforms
Figure 9. Qualitative molecular orbital diagrams for trigonal planar and pseudotetrahedral complexes containing NidE linkages. Idealized point groups
and coordinate axes are indicated.
Table 1. Experimental and Calculated Bond Lengths and Angles for 14
and 16
14
exptl
calcd
16
exptl
calcd
Bond Lengths (Å)
Ni-P4
Ni-P1
Ni-P2
Ni-P3
2.263
2.276
2.287
2.253
2.358
2.396
2.379
2.384
Ni-N
1.624
2.232
2.215
2.234
1.665
2.309
2.316
2.315
Ni-P1
Ni-P2
Ni-P3
Angles (deg)
P4-Ni-P1
P4-Ni-P2
P4-Ni-P3
P1-Ni-P2
P1-Ni-P3
P2-Ni-P3
121.90
123.43
118.67
98.24
92.76
94.74
122.45
124.66
117.44
97.50
92.17
95.12
N-Ni-P1
N-Ni-P2
N-Ni-P3
P1-Ni-P2
P1-Ni-P3
P2-Ni-P3
Ni-N-O
124.68
119.84
124.01
92.54
95.42
91.74
176.0
123.43
125.67
122.58
91.54
92.93
91.29
177.9
Figure 10. DFT optimized (Jaguar 5.0: B3LYP/LACVP**) molecular
structures for (A) [PhBPiPr3]Ni(PMe3) (optimized as a doublet ground state)
and (B) [PhBP3]Ni(NO) (optimized as a singlet ground state). Hydrogen
atoms have been omitted for clarity.
destabilize such a system by raising the energy of the high-
lying unpaired electrons for d7 or d8 configurations.
imaginary frequencies through vibrational frequency calcula-
tions. No imaginary frequencies were located.
Figure 9B,C provides qualitative MO splitting diagrams
that we can use to describe the electronic structure of putative
d6 L3NiIVtE and d7 L3NiIIIdE species. To examine the
electronic structure of these structure types more thoroughly,
we have calculated hypothetical structures (DFT) for a series
of [PhBP3] and [PhBPiPr3] nickel complexes. To establish a
reasonable degree of confidence in the theoretical methods
used, we first performed geometry optimizations for the
synthetically isolable and structurally characterized com-
plexes [PhBP3]Ni(NO) (16) and [PhBPiPr3]Ni(PMe3) (14).
Using the Jaguar computational package (B3LYP/LACVP**),
geometry optimizations were performed starting from the
crystallographically determined coordinates of complexes 16
and 14 as the initial starting point in each case. As shown in
Figure 10 and Table 1, the geometry optimized structures
for 16 and 14 compare favorably with the experimentally
determined solid-state structures. The most significant struc-
tural distinction worth noting is that the Ni-P bond distances
are on average ∼0.1 Å longer in the DFT determined
structures by comparison to the observed distances. As an
additional check, the geometry of the optimized structure of
[PhBP3]Ni(NO) was evaluated for the correct number of
We next examined the theoretical structures of several
hypothetical NiIII imides (Jaguar 5.0: B3LYP/LACVP**).
As a starting point geometry for the theoretical optimization
of S ) 1/2 [PhBP3]Ni(NtBu), [PhBPiPr3]Ni(NtBu), and [PhBPiPr3]-
Ni(NMe), we incorporated the x,y,z coordinates for the L3-
Ni-N framework from crystallographically determined
coordinates for S ) 0 [PhBPiPr3]CotNR and [PhBP3]Cot
NR structures.14,16 The DFT-minimized structure for [PhBP3]-
Ni(NtBu) is shown in Figure 11A. The complex features a
rather short Ni-N bond distance (1.690 Å, Table 2) that is
nonetheless appreciably longer (∼0.04 Å) than the Co-N
and Fe-N bond distances observed for [PhBP3]CotNR and
[PhBP3]FetNR imide species, respectively.14-16 This dif-
ference may be a manifestation of the attenuated Ni-N
π-bond order in [PhBP3]Ni(NtBu). Also of note from the
structure is the significant variation in the three Ni-P bond
distances. One Ni-P distance is significantly longer than
the other two (2.429 Å vs 2.277 and 2.343 Å). The Ni-
N-C imide angle is remarkably linear (179.28°). A similar
overall geometry is obtained for the optimized structure of
[PhBPiPr3]Ni(NtBu) (Figure 11B). In this latter structure,
Inorganic Chemistry, Vol. 43, No. 15, 2004 4655

MacBeth et al.
Figure 11. DFT optimized (Jaguar 5.0: B3LYP/LACVP**) molecular
structures for (A) [PhBP3]Ni(NtBu), (B) [PhBPiPr3]Ni(NtBu), and (C)
[PhBPiPr3]Ni(NMe). Each geometry assumes a doublet ground state.
Hydrogen atoms have been omitted for clarity.
Figure 12. Starting point (A) and DFT optimized (Jaguar 5.0: B3LYP/
LACVP**) geometry (B) for S ) 0 [PhBPiPr3]Ni(N). Hydrogen atoms have
been omitted for clarity.
1/2 d7 nickel imide species suggest that L3NidNR may indeed
be synthetically viable targets if (i) the L3 ligand scaffold
employed is sufficiently flexible to accommodate the an-
ticipated distortion and (ii) bimolecular degradation pathways
are not kinetically problematic.
Table 2. Bond Lengths and Angles of DFT Optimized Structures for
[PhBP3]Ni(NtBu), [PhBPiPr3]Ni(NtBu), and ([PhBPiPr3]Ni(NMe)
[PhBP3]Ni(NtBu) [PhBPiPr3]Ni(NtBu) [PhBPiPr3]Ni(NMe)
Bond Lengths (Å)
Ni-N
1.690
2.429
2.343
2.277
1.687
2.519
2.301
2.303
1.692
2.378
2.368
2.320
Ni-P1
Ni-P2
Ni-P3
The other structure type of interest to us concerns the
hypothetical nickel nitride species [PhBPiPr3]NiIVtN. As
mentioned above, this d6 species is formally isolobal and
isoelectronic with the surprisingly stable d6 cobalt imides
that we have described previously.14,16 However, there is an
important distinction to be drawn between the relative
d-orbital splitting energies between L3MtNR and L3MtN
structure types due to the rehybridization at nitrogen that
occurs. As we have recently suggested, the d orbital of a1
symmetry in the d4 complex [PhBPiPr3]FeIVtN is highly
Angles (deg)
Ni-N-C
179.28
89.85
88.81
176.90
90.70
92.25
156.99
94.93
91.90
P1-Ni-P2
P2-Ni-P3
P1-Ni-P3
N-Ni-P1
N-Ni-P2
N-Ni-P3
89.45
90.78
91.26
122.33
126.35
128.38
119.83
124.54
128.40
111.46
114.04
142.75
however, the elongated Ni-P bond distance is extremely
elongated (2.519 Å). The elongation observed for one of the
Ni-P bonds in [PhBPiPr3]Ni(NtBu) and [PhBP3]Ni(NtBu) is
fully consistent with the presence of a single electron in an
orbital that is σ* with respect to a N-P bond vector, as
illustrated in Figure 9C. The distorted geometries we have
previously reported for the structurally characterized low spin
d7 [PhBP3]CoI and [PhBP3]CoOSiPh3 species provide ex-
perimental verification that this type of elongation is to be
reasonably anticipated for such configurations.25,31 That the
DFT-minimized structure of [PhBPiPr3]Ni(NtBu) reveals such
a pronounced distortion relative to [PhBP3]Ni(NtBu) most
likely reflects the respective σ donor character of the two
borate auxiliaries. To probe whether steric factors might also
be significantly contributing to the structure observed, the
methyl imido complex [PhBPiPr3]NiIII(NMe) was examined.
As can be seen in Figure 11C, replacement of the tert-butyl
group in [PhBPiPr3]Ni(NtBu) for a methyl group in [PhBPiPr3]-
Ni(NMe) results in a dramatic distortion of the imido ligand.
The Ni-N-C angle is now quite bent (156.99°). Moreover,
in the [PhBPiPr3]Ni(NMe) structure the three Ni-P bond
distances show comparatively little variation. Thus, stabiliza-
tion of the d7 imide configuration is observed to occur by
one of two processes. For a sterically encumbering imide,
elongation of one Ni-P bond vector appears to be favored.
For an imide that is less bulky, a severe bend in the imide
functionality appears to be more energetically favorable such
that a new geometry, better described as trigonal pyramidal
with the imide ligand occupying an equatorial position, is
obtained. Collectively, the DFT data obtained for these S )
2
destabilized due to a strong σ* interaction between dz and
the nitride σ donor.18 The addition of two electrons to the
frontier orbital diagram of [PhBPiPr3]FeIVtN, which is pro-
vided by replacing the iron center by nickel, provides the
electronic configuration (e)4(a1)2(e)0. Two electrons are there-
fore expected to lie at high energy for [PhBPiPr3]NiIVtN,
perhaps helping to explain why this species is not generated
from [κ2-PhBPiPr3]Ni(dbabh) (11) by loss of anthracene upon
heating, whereas a similar synthetic strategy is quite effective
for the case of iron. Thus, the electronic situation is quite
distinct between a hypothetical [PhBPiPr3]NiIVtN species and
the known d6 BP3CotNR systems. In the latter species, the
2
a1 orbital of dz parentage is nearly degenerate with the
nonbonding e-set.14
The consequences of electronic destabilization in [PhBPiPr3]-
NitN are manifest in its DFT-minimized structure. The
starting and final geometries of this calculation are shown
in Figure 12A,B. As can be seen, the nitride group inserts
in a Ni-P bond in the optimized structure to afford a
complex that is better viewed as a diamagnetic Ni(II)
complex with one phosphinimato and two phosphine link-
ages. Whether this product is kinetically (or even thermo-
dynamically) reasonable is unclear. In solution, bimolecular
pathways to other structures, such as NiIV2N2 and NiI-N2-
NiI, might dominate. Regardless, the overall picture that
begins to emerge is the following: for pseudotetrahedral
structure types L3NitE, imide (NR2-), and perhaps related
carbyne (CR3-) functionalities are more likely to electroni-
cally stabilize tetravalent nickel than terminal oxo or nitrido
functionalities.
4656 Inorganic Chemistry, Vol. 43, No. 15, 2004

The Coordination Chemistry of “[BP3]NiX” Platforms
TlOSiPh3,57 KC8,58 Li-NH-2,6-iPr-Ph,8 and lithium(dbabh)41 were
prepared according to literature methods. Deuterated solvents were
purchased from Cambridge Isotopes Laboratories, Inc., and were
degassed and stored over activated 3-Å molecular sieves prior to
use. All other reagents and solvents were purchased from com-
mercial vendors and used without further purification unless
otherwise noted. CAUTION: Thallium reagents are extremely toxic
and should be handled and disposed of with care.
Conclusions
We have described synthetic entry to the chemistry of
“[PhBP3]Ni” and “[PhBPiPr3]Ni” systems. Most all of the
complexes that have been prepared, for which the formal
oxidation states Ni(0), Ni(I), and Ni(II) have been assigned,
give rise to pseudotetrahedral geometries. For the divalent
nickel complexes [PhBP3]NiIIX in particular, small perturba-
tions of the pseudotetrahedral geometry are observed that
that depend on the nature of the X-type ligand employed.
The unusual Ni(II) thiolate complex [PhBP3]Ni(S-p-tBu-Ph)
(7) is perhaps most distinct in this regard as it adopts a
geometry that lies somewhere between that of a tetrahedron
and a cis-divacant octahedron. The [PhBPiPr3]-containing Ni-
(II) complex [κ2-PhBPiPr3]Ni(dbabh) (11) is in principle a
precursor to a nickel(IV) nitride species but fails to evolve
anthracene. Why this is the case is in part explained by
theoretical considerations that suggest that a d6 L3NitN
would feature a pair of electrons lying in a highly destabilized
σ* orbital. A general characteristic of the Ni(II) systems
described is that access to pseudotetrahedral L3NiIIX com-
plexes where X represents a strong-field ligand, such as an
alkyl or an amide, is problematic. This characteristic is also
true of Riordan’s tris(thioether)borate system [PhTttBu]NiIIX.38
When an amide can be installed, as for 11, the structure type
observed is three-coordinate trigonal planar. Electrochemical
interrogation of the Ni(II) and Ni(I) complexes has revealed
that reversible 1-electron NiII/I and NiI/0 redox couples are
accessible at fairly low potentials. Perhaps most striking is
that the NiIII/II redox process is irreversible in all cases we
have examined, even for ligands that are potentially good
π-donors such as alkoxides and thiolates. This situation
stands in contrast to related BP3Fe and BP3Co systems we
have described, where the trivalent state is quite readily
accessible. Whereas we had hoped that access to trivalent
nickel might be possible by turning to dianionic imide and
carbene functionalities in the fourth site (L3NiIIIdNR or L3-
NiIIIdCR2), the successful synthesis of such species has thus
far eluded us. Simple MO considerations and DFT calcula-
tions nonetheless suggest that these types of complexes may
be electronically accessible. We suspect that our inability to
isolate well-defined examples of these types of species is
due to synthetic nuances that we have yet to fully appreciate.
Physical Methods. Elemental analyses were performed by Desert
Analytics, Tucson, AZ. A Varian Mercury-300 spectrometer was
1
used to record H, 31P{1H}, and 13C{1H} spectra at room temper-
ature unless otherwise noted. 1H and 13C{1H} chemical shifts were
referenced to residual solvent. 31P{1H} NMR are reported relative
to an external standard of 85% H3PO4 (0 ppm). Abbreviations for
reported signal multiplicities are as follows: s, singlet; d, doublet;
t, triplet; q, quartet; m, multiplet; br, broad. IR spectra were recorded
on a Bio-Rad Excalibur FTS 3000 spectrometer controlled by Win-
IR Pro software using a KBr solution cell. Solution magnetic
moments were measured at 298 K in C6D6 following the Evans
method.59 MS data were obtained by injection of a hydrocarbon
solution onto a Hewlett-Packard 1100MSD mass spectrometer
(ES+) or an Agilent 5973 mass selective detector (EI). UV-vis
measurements were recorded on a Varian Cary 50 Bio Spectro-
photometer controlled by Cary WinUV software. All measurements
were recorded using a quartz cell fitted with a Teflon cap. X-ray
diffraction studies (Table 3) were carried out in the Beckman
Institute Crystallographic Facility on a Bruker Smart 1000 CCD
diffractometer. Cyclic voltammetry measurements were recorded
using a BAS CV 100W (Bioanalytical Systems Inc., West Lafayette,
IN) using a glassy carbon working electrode, a platinum wire
auxiliary electrode, and an Ag/AgNO3 nonaqueous reference
electrode filled with THF and Bu4NPF6. All electrochemical
measurements are referenced to Fc/Fc+ as an internal standard and
can be corrected to SCE by adding 0.56 V to reported potential.60
Computational Methods. All calculations were performed using
the hybrid DFT functional B3LYP as implemented in the Jaguar
5.0 program package.61 This DFT functional utilizes the Becke
three-parameter functional62 (B3) combined with the correlation
functional of Lee, Yang, and Parr63 (LYP). The Ni was described
using the LACVP basis set with the valence double-ú contraction
of the basis functions, LACVP**. All electrons were used for the
other elements using Pople’s64 6-31G** basis set. Input coordinates
were derived as described in the text from crystallographically
determined structures. Spin states and molecular charges were
explicitly stated, and no molecular symmetry was imposed. Default
values for geometry and SCF iteration cutoffs were used, and all
(55) Blicke, F. F.; Hotelling, E. B. J. Am. Chem. Soc. 1954, 76, 5099-
5103.
Experimental Section
(56) Jazdzewski, B. A.; Holland, P. L.; Pink, M.; Young, V. G., Jr.; Spencer,
D. J. E.; Tolman, W. B. Inorg. Chem. 2001, 40, 6097-6107.
(57) Harvey, S.; Lappert, M. F.; Raston, C. L.; Skelton, B. W.; Srivastava,
G.; White, A. H. J. Chem. Soc., Chem. Commun. 1988, 17, 1216-
1217.
All manipulations were carried out using standard Schlenk or
glovebox techniques under a nitrogen atmosphere. Petroleum ether,
benzene, toluene, diethyl ether, THF, dichloromethane, and aceto-
nitrile were deoxygenated and dried by sparging with N2 gas and
passage through an activated alumina column. Hydrocarbon and
ethereal solvents were tested with a standard purple solution of
sodium benzophenone ketyl in tetrahydrofuran to ensure dioxygen
and water removal. The compounds [PhBP3]Tl,52 [PhBPiPr3]Tl,25b
Ph2PCH2Li(TMEDA),53 iPr2PCH2Li,54 ASNBr,55 Tl-O-p-tBu-Ph,56
(58) Schwindt, M.; Lejon, T.; Hegedus, L. Organometallics 1990, 9, 2814-
2819.
(59) (a) Evans, D. F. J. Chem. Soc. 1959, 2003-2005. (b) Sur, S. K. J.
Magn. Reson. 1989, 82, 169-173.
(60) For information regarding correcting redox potenials vs Fc+/Fc in
various solvents see: Connelly, N. G.; Geiger, W. E. Chem. ReV. 1996,
96, 877-910.
(61) Jaguar 5.0, Schrodinger, LLC, Portland, Oregon, 2002.
(62) Becke, A. D. J. Chem. Phys. 1993, 98, 5648-5652.
(63) Lee, C.; Yang, W.; Parr, R. G. Phys. ReV. B 1988, 37, 785-789.
(64) (a) Hariharan, P. C.; Pople, J. A. Chem. Phys. Lett. 1972, 16, 217-
219. (b) Francl, M. M.; Pietro, W. J.; Hehre, W. J.; Binkley, J. S.
Gordon, M. S.; DeFrees, D. J.; Pople, J. A. J. Chem. Phys. 1982, 77,
3654-3665.
(52) Shapiro, I. R.; Jenkins, D. M.; Thomas, J. C.; Day, M. W.; Peters, J.
C. Chem. Commun. 2001, 2152-2153.
(53) (a) Peterson, D. J. J. Organomet. Chem. 1967, 8, 199-208. (b) Schore,
N. E.; Benner, L. S.; Labelle, B. E. Inorg. Chem. 1981, 20, 3200-
3208.
(54) Thomas, J. C.; Peters, J. C. Inorg. Chem. 2003, 42, 5055-5073.
Inorganic Chemistry, Vol. 43, No. 15, 2004 4657

MacBeth et al.
Table 3. X-ray Diffraction Experimental Details for 2-5, 7, 8A,B, and 11-16
2
3
4
5
7
chem formula
C45H41BINiP3‚
0.5C6H6
871.14 + 39.05
-177
0.71073
22.502(3)
12.694(2)
29.393(7)
90
90.418(12)
90
8396(3)
P21/c
C27H53BClNiP3
0.84C45H41BN3NiP3‚
0.16C45H41BClNiP3
821.69
C63H56BNiOP3Si‚
1.5C6H6
1136.76
-177
0.71703
C55H54BNiP3S
fw
575.57
-177
0.71073
9.4015(8)
11.5998(10)
29.392(3)
90
90
90
3205.4(5)
P212121
4
909.47
-177
0.71703
38.222(2)
38.2222(2)
12.5712(9)
90
T (°C)
λ (Å)
a (Å)
b (Å)
c (Å)
R (deg)
â (deg)
γ (deg)
V (Å3)
space group
Z
-177
0.71073
40.37(3)
12.887(15)
16.067(12)
90
13.1627(14)
14.3575(15)
17.2349(18)
101.988(2)
112.185(2)
90.580(2)
2936.1(5)
P1h
111.48(6)
90
7779(13)
C2/c
90
90
18366.0(19)
I1/a
16
1.316
6.10
8
8
2
D
calcd (g/cm3)
1.440
13.45
0.0494, 0.0824
1.193
8.52
0.0392, 0.0717
1.403
7.29
0.0489, 0.0708
1.286
4.78
0.0556, 0.0823
µ (cm-1)
R1, wR2a (I > 2σ(I))
0.0523, 0.0649
8A
8B
11
12
chem formula
C45H41BClNiO2P3‚
C6H6
C90H82B2Cl2Ni2O4P6
‚2C7H8‚C6H6‚
CH2Cl1.5
C41H63BNNiP3
C63H56BNiP4‚C9.95
fw
888.76
-177
0.71703
16.2376(12)
16.4690(13)
17.7364(14)
90
113.379(1)
90
4353.6(6)
P21/c
1952.89
-177
0.71703
24.8113(15)
13.9514(8)
30.9581(19)
90
112.5030(10)
90
9900.3
P21/n
4
732.35
-177
0.71703
12.0312(10)
26.359(2)
12.9206(11)
90
99.150(2)
90
4045.4(6)
P21/n
1006.52 + 119.15
-177
0.71073
19.9842(13)
17.8132(11)
33.699(2)
90
T (°C)
λ (Å)
a (Å)
b (Å)
c (Å)
R (deg)
â (deg)
γ (deg)
V (Å3)
space group
Z
90
90
11996.4(13)
Pbca
8
1.247
4.73
4
4
D
calcd (g/cm3)
1.356
6.58
0.0506, 0.0774
1.310
6.25
0.0697, 0.1129
1.202
6.27
0.0513, 0.0943
µ (cm-1)
R1, wR2a (I > 2σ(I))
0.0573, 0.0951
13
14
15
16
chem formula
C50H50BNiP3
0.95C30H62BNiP4‚
0.05C27H53BP3Tl
619.37
C32H62BNNiP3
C45H41BNNiOP3
fw
T (°C)
λ (Å)
a (Å)
b (Å)
c (Å)
R (deg)
â (deg)
γ (deg)
V (Å3)
space group
Z
827.34
-177
623.26
-177
774.22
-177
0.71073
39.395(4)
12.9396(13)
16.1218(16)
90
110.668(2)
90
7682.2(13)
C2/c
-177
0.71703
12.3572(10)
13.6104(11)
15.1576(12)
75.775(1)
70.824(1)
66.409(1)
2187.3(3)
P1h
0.71703
9.6258(5)
21.6538(12)
16.8244(9)
90
0.71703
9.6300(7)
12.2632(9)
15.4425(11)
87.053(1)
80.119(1)
84.968(1)
1788.5(2)
P1h
93.871(2)
90
3498.8(3)
P21/c
2
4
2
8
Dcalcd (g/cm3)
µ (cm-1)
R1, wR2a (I > 2σ(I))
1.256
5.88
0.0411, 0.0669
1.176
9.35
0.0534, 0.0925
1.157
6.97
0.0410, 0.0876
1.339
6.67
0.0492, 0.0717
2
2
2
a R1 ) ∑||Fo| - ∑||Fc||/∑|Fo|, wR2 ) {∑|w(Fo - Fc )2|/∑|w(Fo )2}1/2
.
structures converged under these criteria. The geometry of the
optimized structure of [PhBP3]Ni(NO) was evaluated for the correct
number of imaginary frequencies through vibrational frequency
calculations using the analytic Hessian. Zero imaginary frequencies
correspond to a local minimum, whereas one imaginary frequency
corresponds to a transition structure. No imaginary frequencies were
found for [PhBP3]Ni(NO). Further frequency calculations were
precluded due to the impractical time requirements necessary to
complete frequency calculations on molecules with so many non-
hydrogen atoms.
suspension of (Ph3P)2NiCl2 (0.230 g, 0.352 mmol) in benzene (10.0
mL). The reaction mixture was stirred at room temperature for 24
h and filtered through a Celite pad. The dark green filtrate was
concentrated under reduced pressure to dryness to yield a dark green
solid. The resulting solid was isolated on a medium porosity frit
and washed three times with 4 mL portions of petroleum ether.
The solid was then redissolved in benzene (10 mL) and filtered
again through a Celite pad. Vapor diffusion of petroleum ether into
the resulting green filtrate yielded analytically pure, green crystalline
product (0.225 g, 82.0%). A second crop yielded an additional 0.192
1
Synthesis of [PhBP3]NiCl (1). A benzene solution (30 mL) of
[PhBP3]Tl (0.313 g, 0.352 mmol) was added dropwise to a stirring
g, 12%. H NMR (300 MHz, C6D6): δ 26.1 (s), 8.3 (s), 7.6 (s),
7.5 (s), 1.22 (s), 0.89 (s), -6.8 (br s), -9.9 (s). UV-vis (C6H6)
4658 Inorganic Chemistry, Vol. 43, No. 15, 2004

The Coordination Chemistry of “[BP3]NiX” Platforms
max, nm (ꢀ): 434 (3332). µeff (C6D6, 298 K): 2.91 µB. Anal. Calcd
λ
to dryness under reduced pressure to yield a bright purple solid.
The solid was extracted with benzene (3 mL) and filtered through
a Celite pad. Vapor diffusion of petroleum ether into the benzene
solution afforded the product as purple crystals (0.152 g, 76%). 1H
NMR (300 MHz, C6D6): δ 43.1 (br s), 23.9 (s), 4.3 (s), 3.6 (s),
2.7 (s), 0.3 (s), -4.3 (br s), -8.2 (s). UV-vis (C6H6) λmax, nm (ꢀ):
707 (1116), 524 (2437), 417 (3020). µeff (C6D6, 298 K): 2.87 µB.
Anal. Calcd for C55H54BNiOP3: C, 73.94; H, 6.09. Found: C,
73.67; H, 5.81.
for C45H41BClNiP3: C, 69.32; H, 5.30. Found: C, 69.54; H, 5.69.
Synthesis of [PhBP3]NiI (2). A benzene solution (30 mL) of
[PhBP3]Tl (0.561 g, 0.631 mmol) was added dropwise to a stirring
suspension of NiI2 (0.394 g, 1.26 mmol) in benzene (20 mL). The
reaction mixture was stirred at room temperature for 24 h and
filtered through a Celite pad. The resulting red-brown filtrate was
concentrated under reduced pressure to 50% of original volume
and filtered again through a Celite pad. Vapor diffusion of petroleum
ether into the resulting red-brown filtrate produced brown, crystal-
line needles of the analytically pure product (0.472 g, 86%).1H
NMR (300 MHz, C6D6): δ 26.6 (s), 8.7 (s), 8.3 (s), 7.8 (s), 1.20
(s), 0.84 (s), -6.9 (br s), -11.07 (s). UV-vis (C6H6) λmax, nm (ꢀ):
705 (556), 446 (5612), 329 (7437). µeff (C6D6, 298 K): 2.87 µB.
Anal. Calcd for C45H41BINiP3: C, 62.04; H, 4.74. Found: C, 61.81;
H, 4.93.
Synthesis of [PhBPiPr3]NiCl (3). In a 50 mL round-bottom flask,
(DME)NiCl2 (283.2 mg, 1.289 mmol) was suspended in THF (15
mL). A THF solution (20 mL) of [PhBPiPr3]Tl (882.0 mg, 1.286
mmol) was added dropwise over 10 min with vigorous stirring.
After completion of the addition, the reaction mixture was stirred
for 25 min and then filtered over Celite on a sintered glass frit.
The solids and Celite were washed with benzene (10 mL). The
combined filtrates were concentrated to dryness under reduced
pressure. The resultant green-brown material was dissolved in
benzene (10 mL) and filtered. Lyophilization of the filtrate provided
green solid [PhBPiPr3]NiCl (716.8 mg, 96.8%). Crystals of 3 were
grown from a concentrated solution in diethyl ether at -30 °C. 1H
NMR (300 MHz, C6D6): δ 37.1 (br s), 7.2-7.4 (m), 2.5 (br). µeff
(C6D6, 298 K): 2.97 µB. UV-vis (C6H6) λmax, nm (ꢀ): 720 (330),
425 (3450), 385 (3340). Anal. Calcd for C27H53BClNiP3: C, 56.34;
H, 9.28. Found: C, 56.29; H, 8.92.
Synthesis of [PhBP3]Ni(N3) (4). A solution of 1 (0.171 g, 0.219
mmol) in THF (4 mL) was added to a stirring suspension of NaN3
(1.43 g, 21.9 mmol) in THF (10 mL) at room temperature. After
being stirred for 18 h, the yellow-green reaction mixture was filtered
through a Celite pad and then evaporated to dryness under reduced
pressure. The resulting solid was then extracted into benzene (3
mL) and filtered again through a Celite pad. Vapor diffusion of
petroleum ether into the resulting yellow-green filtrate yielded
yellow-green crystalline product (0.152 g, 88%).1H NMR (300
MHz, C6D6): δ 25.4 (s), 8.02 (s), 7.5 (s), 7.4 (s), 1.25 (s), 0.47 (s),
-6.6 (br s), -9.7 (s). UV-vis (C6H6) λmax, nm (ꢀ): 698 (310),
463 (3143), 328 (6285). µeff (C6D6, 298 K): 2.86 µB. IR: (C6H6)
ν(N3) ) 2060 cm-1. Anal. Calcd for C45H41BN3NiP3: C, 68.74;
H, 5.26; N, 5.34. Found: C, 68.37; H, 4.96; N, 4.93.
Synthesis of [PhBP3]Ni(OSiPh3) (5). A solution of TlOSiPh3
(0.0551 g, 0.115 mmol) in THF (2 mL) was added to a stirring
solution of 1 (0.0897 g, 0.115 mmol) in benzene (4 mL). A white
precipitate formed immediately. The reaction mixture was stirred
for 2 h, filtered through a Celite pad, and evaporated to dryness
under reduced pressure to yield a green solid. The green solid was
extracted with benzene (2 mL) and filtered through a Celite pad.
Vapor diffusion of petroleum ether into the filtrate provided bright
green crystals of the analytically pure product (0.0984 g, 77%). 1H
NMR (300 MHz, C6D6): δ 24.3 (s), 8.7 (s), 7.4 (s), 6.9 (s), 1.39
(s), 0.92 (s), -4.6 (br s), -8.3 (s). UV-vis (C6H6) λmax, nm (ꢀ):
693 (353), 404 (3546). µeff (C6D6, 298 K): 2.95 µB. Anal. Calcd
for C70H64BNiOP3Si: C, 75.62; H, 5.80. Found: C, 76.02; H, 5.60.
Synthesis of [PhBP3]Ni(O-p-tBu-Ph) (6). A solution of Tl(O-
4-tBu-Ph) (0.0794 g, 0.225 mmol) in THF (4 mL) was added
dropwise to a stirring solution of 1 (0.175 g, 0.225 mmol) in THF
(5 mL). The reaction mixture was stirred for 3 h and concentrated
Synthesis of Tl(S-p-tBu-Ph). 4-tert-Butylthiophenol (0.4418 g,
2.66 mmol) was dissolved in petroleum ether (10 mL). The colorless
solution was then treated with Tl(OCH2CH3) (0.663 g, 2.66 mmol)
as a petroleum ether solution (5 mL). The reaction mixture was
stirred for 2 h, and the bright yellow solid that precipitated was
collected on a medium porosity glass frit and washed with petroleum
ether (15 mL) yielding yellow Tl(S-p-tBu-Ph) (0.812 g, 83%). Anal.
Calcd for C10H13STl: C, 32.49; H, 3.54. Found: C, 32.58; H, 3.73.
Synthesis of [PhBP3]Ni(S-p-tBu-Ph) (7). A solution of Tl(S-
4-tBu-Ph) (0.0509 g, 0.138 mmol) in THF (3 mL) was added
dropwise to a stirring solution of 1 (0.107 g, 0.138 mmol) in THF
(4 mL). The reaction mixture was stirred for 3 h and concentrated
to dryness under reduced pressure to yield a dark red solid. The
solid was extracted with benzene (3 mL) and filtered through a
Celite pad. Vapor diffusion of petroleum ether into the benzene
solution yielded the product as red crystals (0.101 g, 81%). 1H NMR
(300 MHz, C6D6): δ 33.4 (br s), 22.8 (s), 8.6 (s), 2.8 (s), 2.7 (s),
0.3 (s), -4.9 (br s), -8.4 (s). UV-vis (C6H6) λmax, nm (ꢀ): 680
(778), 470 (3442). µeff (C6D6, 298 K): 3.02 µB. Anal. Calcd for
C55H54BNiOP3: C, 72.63; H, 5.98. Found: C, 72.54; H, 5.65.
Synthesis of [PhB(CH2P(O)Ph2)2(CH2PPh2)]NiCl (8A). A
solution of 1 (0.0850 g, 0.109 mmol) in benzene (4 mL) was treated
with O2 gas (8.0 mL, 0.327 mmol) with stirring at room temperature.
The color of the reaction mixture changed from bright green to
purple over the course of 6 h. Volatiles were removed under reduced
pressure to yield a dark purple solid. This solid was extracted into
toluene and filtered through a Celite pad. The filtrate was layered
with petroleum ether and stored at -40 °C to afford the bright
yellow crystalline product (0.0712 g, 80%).
{[PhB(CH2P(O)Ph2)2(CH2PPh2)]Ni(µ-Cl)}2 (8B). 8B was pre-
pared as above with the following exception: The dark purple solid
was extracted into benzene and filtered through a Celite pad. Slow
diffusion of petroleum ether into the benzene solution at room
temperature yielded the product as bright purple crystals (0.0782
mg, 94%).1H NMR (300 MHz, C6D6): δ 19.2 (s), 9.6 (s), 8.4 (s),
8.1 (s), 7.8 (s), 5.8 (s), -0.68 (br s), -2.4 (s), -9.9 (s), -15.3 (s).
UV-vis (C6H6) λmax, nm (ꢀ): 901 (323), 505 (634). Anal. Calcd
for C90H82Cl2B2Ni2O4P6‚2C7H8‚C6H6‚CH2Cl1.5: C, 68.67; H, 5.31.
Found: C, 68.58; H, 5.15.
Synthesis of [PhBP3]Ni(η2-CH2-PPh2) (9). Method A. A 0.21%
Na/Hg amalgam was prepared by dissolving metallic sodium (8.1
mg, 0.36 mmol) in mercury (3.70 g). A THF solution (10 mL) of
2 (0.307 g, 0.352 mmol) was added dropwise to a stirring
suspension of the amalgam in THF (5 mL). The reaction mixture
was stirred for 24 h, during which the color changed from red-
brown to bright red. The THF solution was decanted away from
the Hg and removed under reduced pressure to yield a bright red
solid. This solid was extracted into toluene (3 mL) and filtered
through a Celite pad. The resulting deep red filtrate was stored at
-40 °C overnight and afforded red crystalline product (139 mg,
42%). Method B. A solution of 1 (0.114 g, 0.146 mmol) in toluene
(4.0 mL) was cooled to -78 °C. A toluene solution (2 mL) of Ph2-
PCH2Li(TMEDA) (0.047 g, 0.146 mmol) was cooled to -78 °C
and added dropwise to the solution of 1. The reaction mixture was
Inorganic Chemistry, Vol. 43, No. 15, 2004 4659

MacBeth et al.
allowed to warm to room temperature over a 6 h period, during
which the color changed from dark green to blood red. The reaction
mixture was filtered through a Celite pad, and the red filtrate was
layered with petroleum ether and stored at -40 °C overnight to
afford a red microcrystalline product. The product was isolated on
a medium porosity frit and washed twice with 2 mL portions of
a yellow solid. This solid was extracted into benzene and filtered
through a Celite pad. Slow diffusion of petroleum ether into the
yellow filtrate produced yellow crystals of the product (0.711 g,
95%). H NMR (300 MHz, C6D6): δ 42.3 (s), 26.1 (s), 11.7 (s),
9.1 (s), 8.6 (s), 8.0 (s), 7.7 (br s), 6.3 (br), 5.1 (s), 0.60 (br), 0.30
1
(s), -9.9 (s). µeff (C6D6, 298K): 1.91 µB. UV-vis (C6H6) λmax
,
1
diethyl ether (0.109 g, 80%). H NMR (300 MHz, C6D6): δ 8.16
nm (ꢀ): 443 (1927). Anal. Calcd for C64H58NiP4: C, 75.32; H,
(d, 2H, J ) 7.2 Hz), 7.66 (t, 2H, J ) 7.2 Hz), 7.44 (m, 5H), 7.24
(m, 12H), 6.89 (m, 24H), 1.90 (br, 6H), 1.60 (q, 2H, J ) 7.80 Hz).
13C{1H} NMR (75.4 MHz, C6D6): δ 167.1 (m), 141.3 (m), 133.4
(d, JP-C ) 12.1 Hz), 132.5 (m), 132.1(s), 129.8 (d, JP-C ) 3.5
Hz), 128.8 (s), 128.6 (s), 125.1(s), 30.1 (br), -9.5 (m). 31P{1H}
NMR (121.4 MHz, C6D6): δ 22.28 (d, 3P, 2JP-P ) 53 Hz), 29.20
(q, 1P, JP-P ) 53 Hz). UV-vis (C6H6) λmax, nm (ꢀ): 544 (882),
435 (2670). Anal. Calcd for C58H53BNiOP4: C, 73.84; H, 5.66;.
Found: C, 73.49; H, 5.45.
Synthesis of [PhBPiPr3]Ni(η2-CH2PiPr2) (10). A THF solution
(2 mL) of iPr2PCH2Li (23.3 mg, 0.169 mmol) was added dropwise
to a stirring THF solution (2 mL) of 3 (95.9 mg, 0.167 mmol).
The color of the reaction changed from pale green to yellow. After
30 min, volatiles were removed under reduced pressure. The solids
were extracted with benzene (1 mL) and filtered. Volatiles were
removed under reduced pressure, providing yellow 10 (103.1 mg,
92.0%). Attempts to obtain satisfactory elemental analysis for this
compound have not been successful thus far. 1H NMR (300 MHz,
5.73. Found: C, 75.38; H, 5.91.
Synthesis of [PhBP3]Ni(CNtBu) (13). A 0.23 wt % amalgam
was prepared by dissolving sodium (2.7 mg, 0.118 mmol) in
mercury (1.18 g). Liquid tBuNC (9.8 mg, 0.118 mmol) was added
to a THF solution (3 mL) of 1 (0.0916 g, 0.118 mmol) to produce
an immediate color change from dark green to purple. This solution
was then added dropwise to the stirring amalgam, and the reaction
mixture was stirred for 12 h. Over this time period the reaction
mixture changed from purple to burgundy. The burgundy solution
was decanted away from the amalgam and dried under reduced
pressure to yield a solid. The solid was extracted into benzene and
filtered through a pad of Celite. Vapor diffusion of petroleum ether
into the filtrate yielded red block crystals of the product (0.0761 g,
78%). 1H NMR (300 MHz, C6D6): δ 12.52 (s), 8.9 (s), 7.9 (s), 7.7
(s), 0.47 (s), -6.6 (br s), -9.7 (s). UV-vis (C6H6) λmax, nm (ꢀ):
1059 (224), 443 (1409), 342 (8580). µeff (C6D6, 298 K): 1.74 µB.
IR (C6H6, KBr), cm-1, ν(CN): 2094. Anal. Calcd for C50H50-
BNNiP3: C, 72.58; H, 6.09; N, 1.69. Found: C, 72.29; H, 6.07;
N, 1.30.
Synthesis of [PhBPiPr3]Ni(PMe3) (14). In a 20 mL vial, [PhB-
(CH2PiPr2)3]NiCl (227.7 mg, 395.6 µmol) was dissolved in THF
(8 mL). Neat PMe3 (614 µL, 5.93 mmol) was added to the solution,
causing the reaction mixture to become dark red. After 10 min,
the homogeneous reaction mixture was transferred to a vial
containing a stirbar and sodium/mercury amalgam (0.378% w/w,
2.4022 g, 395 µmol of sodium atoms). The reaction mixture was
stirred vigorously for 4 h, during which time the color became
yellow-orange. The reaction mixture was filtered over Celite on
glass fiber filter paper, and the Celite was washed with benzene [3
× 2 mL]. The combined filtrates were concentrated to dryness under
reduced pressure, providing analytically pure yellow-orange solid
14 (239.4 mg, 98.4%). Crystals of 14 were grown from a diethyl
2
3
3
C6D6): δ 8.01 (br d, 2H, JH-H ) 6.6 Hz), 7.49 (t, 2H, JH-H
)
3
7.5 Hz), 7.26 (t, 1H, JH-H ) 7.5 Hz), 1.84 (br m, 6H), 1.13 (m,
36H), 0.75 (m, 12H), -0.28 (s, 2H). 13C{1H} NMR (75.4 MHz,
C6D6): δ 165.6 (q, 1JB-C ) 47 Hz), 132.93, 126.77, 123.39, 26.33
(br), 21.08, 20.85, 20.35 (d, JP-C ) 6.4 Hz), 19.87 (d, JP-C ) 5.1
Hz), 19.71, 19.51 (d, JP-C ) 7.4 Hz), -12.84 (doublet of quartets,
1JP-C ) 15.9 Hz, JP-C ) 5.8 Hz). 31P{1H} NMR (121.4 MHz,
2
C6D6): δ 49 (br, 3P), -18.2 (q, 1P, 1JP-P ) 41 Hz). 11B{1H} NMR
(128.3 MHz, C6D6): δ -15.7.
Synthesis of [K2-PhBPiPr3]Ni(dbabh) (11). A diethyl ether
suspension (2 mL) of Hdbabh41 (11.4 mg, 59.0 µmol) was frozen
in a liquid nitrogen cooled well. To the stirring, thawing suspension
was added a hexanes solution of n-butyllithium (1.6 M, 40 µL, 64
µmol). The reaction mixture was stirred and warmed at room
temperature for 30 min. The resulting solution and a diethyl ether
solution of 3 (33.9 mg, 58.9 µmol) were frozen in a liquid nitrogen
cooled well. The two solutions were thawed, and the thawing
solution of 3 was added to the thawing solution of [Li][dbabh].
The yellow-green reaction mixture was stirred and warmed. As the
mixture warmed, a color change to inky green brown occurred.
After 45 min of warming, the volatiles were removed under reduced
pressure. The dark solids were extracted with petroleum ether (3
mL) and filtered. Crystals were obtained from the petroleum ether
solution by cooling to -30 °C for several days. The crystals were
1
ether/petroleum ether mixture at -30 °C. H NMR (300 MHz,
C6D6): δ 35.3 (br s, 6H), 16.5 (br s, 9H), 11.0 (br s, 6H), 10.1 (br
s, 18H), 9.0 (s, 2H), 8.0 (t, 2H, J ) 7.2 Hz), 7.7 (t, 1H, J ) 7.5
Hz), -0.3 (br s, 18H). µeff (C6D6, 298 K): 1.82 µB. UV-vis (C6H6)
λmax, nm (ꢀ): 430 (1240). Anal. Calcd for C30H62BNiP4: C, 58.47;
H, 10.14. Found: C, 58.50; H, 10.30.
Alternative Synthesis. Solid yellow (COD)2Ni (49.5 mg, 180
µmol) was dissolved in petroleum ether (3 mL). Neat PMe3 (75
µL, 725 µmol) was added with stirring. After 1 h, a benzene solution
(3 mL) of [PhBPiPr3]Tl (123.5 mg, 180.1 µmol) was added to the
pale yellow reaction mixture. Upon addition, the reaction mixture
darkened. Over 48 h, gray solids precipitated from solution.
Volatiles were removed from the reaction mixture under reduced
pressure. The solids were extracted with petroleum ether (10 mL)
and filtered. Crystallization of the petroleum ether solution at -30
°C over 3 days provided yellow crystals of 14. Analysis of the
1
analyzed as 11 (26.4 mg, 61.3%). H NMR (300 MHz, C6D6): δ
3
3
7.75 (d, 2H, JH-H ) 7.5 Hz), 7.41 (t, 2H, JH-H ) 7.5 Hz), 7.21
3
(t, 1H, JH-H ) 7.2 Hz), 6.98 (dd, 4H, JH-H ) 1.8, 2.7 Hz), 6.72
(dd, 4H, JH-H ) 1.8, 3.0 Hz), 5.44 (s), 1.57 (septet, 6H, J ) 7.5
Hz), 1.18 (d, 18H, J ) 7.3 Hz), 1.1 (br m), 0.82 (d, 18H, J ) 7.5
Hz). 31P{1H} NMR (121.4 MHz, C6D6): δ 58 (br). Anal. Calcd
for C41H63BNNiP3: C, 67.24; H, 8.67; N, 1.91. Found: C, 67.18;
H, 8.83; N, 1.48.
1
crystals by H NMR and 31P{1H} NMR spectroscopy showed the
presence of [PhB(CH2PiPr2)3]Ni(PMe3) and [PhB(CH2PiPr2)3]Tl,
1
Synthesis of [PhBP3]Ni(PPh3) (12). Solid PPh3 (0.191 g, 0.727
mmol) was added to a stirring solution of 1 (0.567 g, 0.727 mmol)
in THF (7 mL). A THF solution (3 mL) of KC8 (0.108 g, 0.799
mmol) was added to the reaction mixture and stirred for 12 h. Over
this time period the color changed from bright green to bright
yellow. All volatiles were removed under reduced pressure to yield
in a 95:5 ratio. H NMR (300 MHz, C6D6): δ 35.3 (br s, 6H),
16.5 (br s, 9H), 11.0 (br s, 6H), 10.1 (br s, 18H), 9.0 (s, 2H),
8.0 (t, 2H, J ) 7.2 Hz), 7.7 (t, 1H, J ) 7.5 Hz), 7.1-7.3 (m, [PhB-
(CH2PiPr2)3]Tl), 1.90 (br, [PhB(CH2PiPr2)3]Tl), 1.21 (br, [PhB-
(CH2PiPr2)3]Tl), 1.07 (m, [PhB(CH2PiPr2)3]Tl), -0.3 (br s, 18H).
31P{1H} NMR (121.4 MHz, C6D6): δ 45.35 (d, 1JTl-P ) 5910 Hz).
4660 Inorganic Chemistry, Vol. 43, No. 15, 2004

The Coordination Chemistry of “[BP3]NiX” Platforms
Synthesis of [PhBPiPr3]Ni(CNtBu) (15). To a benzene solution
(3 mL) of 14 (91.3 mg, 148 µmol) was added dropwise a benzene
solution (2 mL) of CNtBu (67.7 mg, 814 µmol). The reaction
mixture darkened to orange-red. The reaction mixture was stirred
for 20 min. The solution was filtered and concentrated to dryness
under reduced pressure. The resulting orange-red material was
dissolved in toluene (1 mL), and concentrated to dryness under
reduced pressure. The toluene evaporation procedure was repeated
twice more, providing yellow-orange solid 15 (78.7 mg, 85.4%).
Crystals of 15 were grown from a diethyl ether/petroleum ether
solution at -30 °C. 1H NMR (300 MHz, C6D6): δ 47.9 (br s, 6H),
12.6 (br s, 6H), 11.3 (br s, 18H), 8.8 (s, 2H), 7.8 (s, 2H), 7.6 (t,
1H, J ) 7.5 Hz), 3.0 (s, 9H), 0.3 (br s, 18H). µeff (C6D6, 298 K):
1.68 µB. UV-vis (C6H6) λmax, nm (ꢀ): 430 (1560). IR (THF, KBr)
cm-1, ν(CN): 2056. Anal. Calcd for C32H62BNNiP3: C, 61.67; H,
10.03; N, 2.25. Found: C, 61.39; H, 10.35; N, 2.00.
Celite pad. All volatiles were removed under reduced pressure to
yield a pale yellow solid. The solid was isolated on a medium
porosity frit and washed twice with 2 mL portions of EtOH.
Dissolving the solid in THF and layering with diethyl ether afforded
crystalline product (0.272 g, 83%). 1H NMR (300 MHz, C6D6): δ
8.15 (t, 2H, J ) 8.0 Hz), 8.0 (m, 4H), 7.69 (d, 2H, J ) 6.3 Hz),
7.55 (m, 6H), 7.27 (t, 2H, J ) 6.0 Hz), 7.06 (m, 17H), 2.79 (br,
8H, [nBu4N]), 2.61 (br, 6H), 1.16 (br, 16H, [nBu4N]), 0.882 (br,
12H, [nBu4N]). 13C{1H} NMR (75.4 MHz, acetone-d6): δ 206.2
(t, JP-C ) 2.1 Hz), 149.0 (m), 138.6 (m), 134.5 (s), 134.3 (s), 134.0
(d, J ) 4.5 Hz), 133.5 (s), 132.8 (s), 132.6 (s), 130.1 (s), 129.8 (s),
129.1 (d, J ) 3.7 Hz), 129.0 (s), 128.2-127.1 (overlapping
resonances), 126.6 (s), 59.5 (t, JC-N ) 2.9 Hz), 24.6 (s), 20.5 (s),
15.7 (br), 13.5 (s). 31P{1H} NMR (121.4 MHz, C6D6): δ 29.2 (s),
-9.46 (br). IR (THF, KBr), cm-1, ν(CO): 1994, 1913. Anal. Calcd
for C63H77BNNiO2P3: C, 72.57; H, 7.44; N, 1.34. Found: C, 72.21;
H, 7.06; N, 1.47.
Alternative Synthesis. Solid green 3 (81.0 mg, 141 µmol) was
dissolved in THF (3 mL). Neat CNtBu (40 µL, 660 µmol) was
added and the reaction mixture was stirred, forming a golden orange
solution. After 20 min, the 31P{1H} NMR spectrum was examined.
31P{1H} NMR (121.4 MHz, THF): δ 62.7 (br d), 47.7 (br d), 3.2
(br s). Volatiles were removed under reduced pressure. The resultant
solids were dissolved in THF (2 mL) and added to a sodium/
mercury amalgam (0.378% w/w, 859.0 mg, 141.2 µmol of sodium
atoms), followed by washing with additional THF (3 mL). The
reaction mixture was stirred vigorously for 3 h. The reaction mixture
was filtered, providing an orange-yellow solution. Volatiles were
removed under reduced pressure. The resultant solids were washed
with petroleum ether [3 × 0.5 mL], removing an orange-red
supernatant and leaving yellow solids. The solids were dried under
reduced pressure, providing 15 (51.3 mg, 58.4%).
Synthesis of [[K2-PhBPiPr3]Ni(CO)2][ASN] (18). Solid [PhBPiPr3]-
Tl (317.1 mg, 0.4625 mmol), (Ph3P)2Ni(CO)2 (296.2 mg, 0.4633
mmol), and ASNBr (96.1 mg, 0.466 mmol) were combined in THF
(7 mL) and stirred in the dark. Monitoring of the reaction mixture
by 31P{1H} NMR showed that all of the [PhBPiPr3]Tl had been
consumed after 4 days. Volatiles were removed under reduced
pressure, and benzene (1 mL) was added to the resulting sticky
solids. A yellow oil separated, and the benzene was decanted. This
procedure was repeated three times, the yellow oil being isolated
each time. The yellow oil was placed under reduced pressure for
several hours, providing sticky yellow solids. The solids were taken
up in a mixture of THF and diethyl ether (2:1), and stored at -30
°C for 24 h. A small amount of white precipitate formed, which
was removed by filtration. This precipitation process was repeated
again. The resulting solution was concentrated to dryness under
reduced pressure, to provide sticky yellow solids whose spectro-
Synthesis of [PhBP3]Ni(NO) (16). Method A. A solution of
12 (0.645 g, 0.641 mmol) in benzene (15 mL) was stirred under 1
atm of NO gas for 2 h. The color of the reaction mixture changed
from bright yellow to dark red over this time period. The reaction
mixture was concentrated to dryness under reduced pressure to yield
a red solid. This solid was extracted into benzene (4 mL) and filtered
through a Celite pad. Vapor diffusion of petroleum ether into the
red filtrate yielded red crystalline product (0.452 g, 91%). Method
B. A solution of 1 (0.215 g, 0.276 mmol) in benzene (4 mL) was
treated with NO gas (13.6 mL, 0.552 mmol) and stirred for 2 h.
Over this time period the reaction mixture changed from dark green
to dark red. The solution was concentrated under reduced pressure
to 50% of its original volume and filtered through a Celite pad.
Vapor diffusion of petroleum ether into the red filtrate yielded
1
scopic properties were consistent with 18 (129.3 mg, 38.7%). H
NMR (300 MHz, THF-d8): δ 7.44 (br s, 2H), 6.87 (t, 2H, J ) 7.5
Hz), 6.67 (t, 1H, J ) 7.5 Hz), 3.49 (m, 8H), 2.19 (m, 8H), 1.91
(septet, 2H, J ) 7.2 Hz), 1.64 (m, 4H), 1.18 (dd, 12H, J ) 6.5, 7.2
Hz), 1.08 (m, 4H), 0.97 (m, 12H), 0.77 (dd, 12H, J ) 6.2, 7.5 Hz),
0.28 (br, 2H). 13C{1H} NMR (75.4 MHz, THF-d8): δ 208.16 (t,
2JP-C ) 1.6 Hz), 169.08 (q, 1JB-C ) 50 Hz), 133.66, 126.11, 122.20,
1
63.86 (t, JN-C ) 3.2 Hz), 28.23, 27.30, 24.92 (d), 23.03, 21.23
(d), 20.71 (d), 20.45 (d), 20.10 (br), 19.81 (br), 19.73 (d). 31P{1H}
NMR (121.4 MHz, THF-d8): δ 51.13 (s, 2P), 4.80 (s, 1P). 11B-
{1H} NMR (128.3 MHz, THF-d8): δ -15.0. IR (THF, KBr) cm-1
ν(CO): 2004, 1878.
,
1
crystalline dark red product (0.193 g, 90%). H NMR (300 MHz,
Photolysis of [K2-PhBP3]Ni(CO)2][nBu4N]. A THF (10 mL)
solution of 17 (0.1274 g, 0.122 mmol) was transferred to a
photolysis cell equipped with an N2 inlet line and a gas outlet line
connected to a mineral oil bubbler. The solution was photolyzed
with a mercury lamp for 8 h. The solution was evaporated to dryness
to yield a pale yellow solid. This solid was extracted into benzene
solution and filtered through a Celite pad. Attempts to crystallize
by slow evaporation of a benzene solution or slow diffusion of
petroleum ether into a THF solution both yielded a yellow oil
containing {[PhBP3]Ni(CO)}{nBu4N} (22) and a paramagnetic
impurity. 1H NMR (300 MHz, C6D6): δ 8.28 (t, J ) 8.15 Hz, 2H),
8.03 (m, 6H), 7.63 (m, 6H), 7.49 (m, 1H), 7.06 (m, 9H), 6.75 (m,
13H) (br, 8H, [nBu4N]), 1.49 (br, 6H), 1.16 (br, 16H, [nBu4N]),
0.882 (b, 12H, [nBu4N]). 31P{1H} NMR (121.4 MHz, C6D6): δ
25.33. IR: (THF, KBr) cm-1, ν(CO): 1869.
C6D6): δ 8.01 (d, 2H, J ) 6.6 Hz), 7.76 (m, 7H), 7.55 (m, 12H),
7.02 (t, 6H, J ) 7.2 Hz), 6.78 (d, 15H, J ) 6.6 Hz), 1.50 (br, 6H).
13C NMR (C6D6, 75.4 MHz, 25 °C): δ 138.3 (m), 132.2-131.0
(overlapping resonances), 131.8 (dd, J ) 4.3, 7.8 Hz) 131.4 (s),
129.5 (s), 129.3 (s), 128.3 (s), 124.9(s), 18.3 (br). 31P{1H} NMR
(121.4 MHz, C6D6): δ 37.3. UV-vis (C6H6) λmax, nm (ꢀ): 913
(159), 433 (2206), 350 (2059). µeff (C6D6, 298K): 2.86 µB. IR
(C6H6, KBr), cm-1, ν(NO): 1737. Anal. Calcd for C45H41-
BNNiOP3: C, 69.81; H, 5.34; N, 1.81. Found: C, 70.21; H, 5.18;
N, 1.53.
Synthesis of [K2-PhBP3]Ni(CO)2][nBu4N] (17). A benzene
solution (3 mL) of [PhBP3]Tl (0.276 g, 0.313 mmol) was added
dropwise to a stirring benzene solution (3 mL) of (PPh3)2Ni(CO)2
(0.200 g, 0.313 mmol). After stirring for 1 h, a THF solution (3
mL) of [nBu4N][Br] (0.101 g, 0.313 mmol) was added to the
reaction mixture and a white precipitate formed immediately. The
reaction mixture was stirred for 12 h and was filtered through a
Reactivity of 1 with Methyllithium. A diethyl ether (6 mL)
solution of 1 (0.0831 g, 0.106 mmol) was cooled to -78 °C. A
solution of methyllithium (1.6 M in hexanes, 1.71 mL, 0.106 mmol)
Inorganic Chemistry, Vol. 43, No. 15, 2004 4661

MacBeth et al.
was added dropwise to this solution. The solution was warmed
slowly to room temperature over the course of 4 h. During this
time the color of the reaction mixture went from dark green to
blood red. The solution was filtered through a pad of Celite to
remove insoluble material. All volatiles were remove under reduced
pressure to yield a dark-red solid. When extracted into benzene
and analyzed by 1H and 31P{1H} NMR, the reaction mixture
exhibited the diamagnetic resonances of complex 9.
was added p-tolyl azide (29.3 mg, 0.22 mmol) drowise as a benzene
solution (2 mL). After 15 min, the color changed from bright yellow
to dark brown. The reaction mixture was stirred for 24 h, and all
solvent was removed under reduced pressure. The only identifiable
product in this reaction was p-tolyl-NdPPh3. 1H NMR (300 MHz,
C6D6): δ 7.81 (m, 4H), 7.04 (m, 15H), 2.20 (s, 3H). 31P{1H} NMR
(121.4 MHz, C6D6): δ -1.17 (s). ESI/MS (m/z): 368.1 (p-tolyl-
NdPPh3 + H+).
Reactivity of 1 with Lithium Amides. A procedure similar to
that reported for the reactions of 1 with methyllithium was followed.
In all procedures, the lithium amides were added cold (-78 °C) to
1. The solutions became dark red upon warming to room temper-
ature. The products exhibited the same spectral signatures as
complex 9.
Reaction of 1 with CoCp2. A THF (5 mL) solution of CoCp2
(5 mL) was added dropwise to a (THF) solution of 1 (0.1889 g,
0.242 mmol). The reaction mixture was stirred for 1.5 h. [nBu4N]-
[Br] (0.0939 g, 0.242 mmol) was then introduced as a THF solution
(3 mL), and the reaction mixture was stirred for an additional hour.
The THF solution was filtered through a Celite pad to remove
insoluble material, and the resulting red-brown solution was
concentrated to dryness. The red-brown solid was collected on a
medium porosity frit and washed with Et2O. Crude 1H NMR (300
MHz, C6D6): δ 21.5 (br), 12.2 (m), 11.8 (m), 8.4 (m), -3.6 (br)
-5.8 (br), -47.6 (br). Attempts to recrystallize this product led to
a mixture of diamagnetic and paramagnetic products.
Acknowledgment. We acknowledge the DOE (PECASE)
and the NSF (CHE-0132216) for financial support of this
work. Larry Henling, Smith Nielsen, and Eric Peters provided
technical assistance. We are grateful to Professors Peter T.
Wolczanski and Daniel Rabinovich for providing insightful
discussions. We also thank Professor Clifford Kubiak for
disclosing results prior to publication. The Beckman Institute
Senior Research Fellows Program (CEM), the Moore Foun-
dation (J.C.T.), the NSF (J.C.T.), and the DOD (T.A.B.) are
each acknowledged for fellowship support.
Supporting Information Available: CIFs and tables of crystal-
lographic data for complexes 2-5, 7, 8A,B, and 11-16, including
fully labeled thermal ellipsoid plots, crystallographic details, atomic
coordinates, equivalent isotropic displacement parameters, aniso-
tropic displacement parameters, and interatomic distances and
angles. This material is available free of charge via the Internet at
http://pubs.acs.org.
Reaction of 12 with p-Tolyl Azide. Complex 12 (0.1103 g,
0.110 mmol) was dissolved in benzene. To this reaction mixture
IC049936P
4662 Inorganic Chemistry, Vol. 43, No. 15, 2004