Coordinating Anions: (Phosphino)tetraphenylborate Ligands as New Reagents for Synthesis

Article abstract of DOI:10.1021/ic0350234

Anionic, electron-releasing phosphines that incorporate a borate counteranion within the ligand framework are promising reagents for organometallic catalysis. This report describes the synthesis of a new class of monodentate tertiary phosphines built upon the commonly employed tetraphenylborate anion. These new phosphines are highly stable and strongly electron-releasing and readily coordinate transition metals. Moreover, they are promising reagents for catalysis, as demonstrated by their ability to promote the Suzuki cross-coupling of aryl chloride substrates.

Full text of DOI:10.1021/ic0350234

Inorg. Chem. 2004, 43, 8−10
Coordinating Anions: (Phosphino)tetraphenylborate Ligands as New
Reagents for Synthesis
Christine M. Thomas 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 August 31, 2003
Scheme 1
Anionic, electron-releasing phosphines that incorporate a borate
counteranion within the ligand framework are promising reagents
for organometallic catalysis. This report describes the synthesis
of a new class of monodentate tertiary phosphines built upon the
commonly employed tetraphenylborate anion. These new phos-
phines are highly stable and strongly electron-releasing and readily
coordinate transition metals. Moreover, they are promising reagents
for catalysis, as demonstrated by their ability to promote the Suzuki
cross-coupling of aryl chloride substrates.
Whereas monodentate neutral phosphine ligands find
utility in nearly all areas of chemical synthesis,1 compara-
tively little attention has been devoted to structurally related
anionic derivatives. The diphenylphosphidoboratabenzene
ligand of Fu provides one of the noteworthy exceptions.2
This system features a triphenylphosphine-type ligand ren-
dered anionic by a boratabenzene subunit, the latter of which
has itself been offered as an intriguing cyclopentadienyl
alternative.3 Given the growing interest in sterically demand-
ing, electron-releasing phosphines (and carbenes) for homo-
geneous catalysis,4 access to electron-rich, anionic phosphines
would provide a timely complement to these increasingly
popular ligands. One conceptual way to generate such spe-
cies, while at the same time preserving desirable properties
inherent to tertiary phosphines, is to embed a borate coun-
teranion within the phosphine donor framework. This ap-
proach has already found utility with respect to developing
catalytically active, zwitterionic organometallic species.5 In
this paper, we introduce a new series of monodentate phos-
phines templated upon the tetraphenylborate anion and briefly
discuss aspects of their stability, their transition-metal binding
affinity, their comparative electron-releasing character, and
their potential as reagents for organic synthesis.
The delivery of lithiated carbanions of methyldiarylphos-
phines (i.e., LiCH2PAr2) and methyldialkylphosphines (i.e.,
LiCH2PR2) to borane electrophiles in the preparation of tri-
and bidentate (phosphino)borates has been explored.6 For
example, addition of 3 equiv of LiCH2PiPr2 to PhBCl2
provides the tridentate anion [PhB(CH2PiPr2)3]-,6a whereas
addition of 2 equiv of LiCH2PiPr2 to Ph2BCl provides the
bidentate anion [Ph2B(CH2PiPr2)2]-.6b A similar protocol ex-
ploiting triphenylborane as the electrophile of choice provides
related monodentate ligands. For example, the addition of
(TMEDA)LiCH2PPh2 to BPh3 generates [(TMEDA)Li]-
[Ph2PCH2BPh3] (1) in good yield.
* To whom correspondence should be addressed. E-mail: jpeters@
caltech.edu.
(1) Dias, P. B.; de Piedade, M. E. M.; Simo˜es, J. A. M. Coord. Chem.
ReV. 1994, 135, 737.
(2) (a) Qiao, S.; Hoic, D. A.; Fu, G. C. J. Am. Chem. Soc. 1996, 118,
6329. (b) Hoic, D. A.; Davis, W. M.; Fu, G. C. J. Am. Chem. Soc.
1996, 118, 8176. (c) Hoic, D. A.; DiMare, M.; Fu, G. C. J. Am. Chem.
Soc. 1997, 119, 7155.
(3) (a) Ashe, A. J., III; Shu, P. J. J. Am. Chem. Soc. 1971, 93, 1804. (b)
Putzer, M. A.; Rogers, J. S.; Bazan, G. C. J. Am. Chem. Soc. 1999,
121, 8112.
(4) (a) Kirchhoff, J. H.; Dai, C.; Fu, G. C. Angew. Chem., Int. Ed. 2002,
41, 1945. (b) Yin, J.; Rainka, M. P.; Zhang, X.; Buchwald, S. L. J.
Am. Chem. Soc. 2002, 124, 1162. (c) Stambuli, J. P.; Kuwano, R.;
Hartwig, J. F. Angew. Chem., Int. Ed. 2002, 41, 4746.
(5) (a) Lu, C. C.; Peters, J. C. J. Am Chem Soc. 2002, 124, 5272. (b)
Betley, T. A.; Peters, J. C. Angew. Chem., Int. Ed. 2003, 42, 2003.
(6) (a) Betley, T. A.; Peters, J. C. Inorg. Chem. 2003, 42, 5074. (b)
Thomas, J. C.; Peters, J. C. Inorg. Chem. 2003, 42, 5055.
8 Inorganic Chemistry, Vol. 43, No. 1, 2004
10.1021/ic0350234 CCC: $27.50 © 2004 American Chemical Society
Published on Web 12/06/2003

COMMUNICATION
Figure 1. Displacement ellipsoid representations (50%) of 3 (left) and 5
(right). Both phosphines were crystallized as ammonium salt derivatives
(cations omitted). See Supporting Information for crystallographic details.
An alternative and potentially more general strategy is to
exchange the methylene linker for an aryl linker. This latter
approach in effect provides a tetraarylborate counteranion
featuring a coordinating phosphine donor. As a number of
known halo-substituted arylphosphines are precursors to
lithio arylcarbanions,7 a diverse family of anionic phosphines
can be envisioned. Several meta- and para-substituted
bromoarylphosphines (2a-d) were prepared to examine this
approach (Scheme 1).
Reaction of the para- and meta-substituted bromoarylphos-
phines with a single equivalent of t-BuLi (-90 °C, THF)
generated the required arylcarbanions. The solutions were
subsequently quenched by low temperature addition of
triphenylborane. This protocol afforded reasonably high
crude yields (>70%) of the desired (phosphino)borate
species, as ascertained by 11B and 31P NMR spectroscopy.
These lithium species were converted in situ to their more
conveniently isolated ammonium salt derivatives, 3-6, by
salt exchange with [NR4][Br] in dichloromethane solution
(NR4 ) NBu4, NEt4, and ASN ) 5-azonia-spiro[4.4]nonane).
The solid-state structures of the ammonium salts of 3 and 5
(Figure 1, also see Supporting Information) establish the
structural integrity of these borates, and compare well with
structural data for related neutral phosphines (e.g., PPh3 and
PhiPr2P).8 For comparative purposes, the isostructural neutral
silane ligands were also prepared (Scheme 1) via low
temperature generation of the arylcarbanion (as above) and
subsequent quenching with triphenylsilyl chloride. This
procedure afforded consistently high crude yields (> 90%
by 31P NMR) of the desired (phosphino)silanes (7-10).
The anionic phosphine salts 3-6 are appreciably soluble
in alcohols (e.g., EtOH), acetonitrile, tetrahydrofuran, chlo-
rinated solvents, and acetone.9 Their stability in CHCl3 and
CH2Cl2 is distinct from those of 1 and related methylene-
bridged (phosphino)borates, which tend to degrade rather
rapidly in such solvents.6b Moreover, ligands 3-6 proved
very stable to both air oxidation and hydrolysis. For example,
3-6 afforded no discernible oxidation products (31P NMR)
over a period of 2 weeks under an atmosphere of air in
acetone solution. This stability again contrasts that of
Figure 2. Displacement ellipsoid representations (50%) of {nBu4N}-
n
m-Ph2
{[Ph3BPP-iPr ]RhX(NBD)} (11, top) and { Bu4N}2{[Ph3BP
]2PtMe2} (12,
2
bottom).10 The two ASN countercations of 12 have been omitted for clarity.
methylene-bridged 1, which was oxidized very rapidly by
air in acetone solution, and also underwent gradual hydrolysis
to release MePPh2 upon exposure to water.
To benchmark the binding affinity of ligands 3-6, their
reactivity with {(NBD)RhCl}2 and (COD)PtMe2 (COD )
cyclooctadiene, NBD ) norbornadiene) was surveyed. Ad-
dition of 6 as a tetra-n-butylammonium salt to {(NBD)RhCl}2
provided the molecular salt {nBu4N}{[Ph3BPp-iPr ]RhCl-
2
(NBD)} (11), whose solid-state structure is shown in Figure
2.10 This anion is a promising precursor to zwitterionic
rhodium species upon formal release of [nBu4N][Cl]. Indeed,
addition of [Tl][PF6] precipitates TlCl instantly to give a
benzene soluble species with a single resonance in the 31P
NMR (1JRh-P ) 166 Hz). The ligands 3, 4, and 5 also reacted
with {(NBD)RhCl}2 to give analogous salt products. With
respect to platinum, addition of 2 equiv of an ammonium
salt of 3-6 to (COD)PtMe2 afforded in each case disubsti-
tution and the cis isomer exclusively: {ASN}2{(Ph3BPm-Ph )2-
2
PtMe2} (12), {NBu4}2{[Ph3BPp-Ph ]2PtMe2} (13), {NBu4}2-
2
p-iPr2
{[Ph3BPm-iPr ]2PtMe2} (14), and {NBu4}2{[Ph3BP
]2-
2
PtMe2} (15).11 The 31P NMR shifts and coupling constants
of complexes 12 and 13 (28.63, ppm, 1JPt-P ) 1947 Hz; and
1
27.47 ppm, JPt-P ) 1935 Hz, respectively) compare well
with literature values for cis-(PPh3)2PtMe2 (27.7 ppm, 1JPt-P
) 1900 Hz).11,12 The isopropyl derivatives 14 and 15 are
(7) See for example: Lustenberger, P.; Diedrich, F. HelV. Chim. Acta
2000, 83, 2865.
(10) The structural data obtained for complex 11 revealed the presence of
a bromide impurity, which was present in the original batch of 6 that
produced the crystal (see Supporting Information). The data are
satisfactorily modeled with the halide occupancy as 60% Br and 40%
Cl.
(11) Nolan, S. P.; Haar, C. M.; Marshall, W. J.; Moloy, K. G.; Prock, A.;
Giering, W. P. Organometallics 1999, 18, 474.
(8) (a) Daly, J. J. J. Chem. Soc. 1964, 3799. (b) Bruckmann, J.; Kruger,
C.; Lutz, F. Z. Naturforsch., B: Chem. Sci. 1995, 50, 351.
(9) Ligand solubility is highly dependent on the ammonium salt used.
For example, the ASN salt of 3 is sparingly soluble in THF, while
the nBu4N salts of 4-6 and the Li(TMEDA)2 salt of 1 are highly
soluble in THF.
Inorganic Chemistry, Vol. 43, No. 1, 2004 9

COMMUNICATION
Table 1. Examination of the Efficiency of Ligands 5, 9, and PiPr2Ph
To Facilitate Suzuki Cross-Coupling between Phenylboronic Acid and
p-Chlorotoluene, p-Chloroacetophenone, or 1,4-Dichlorobenzene
chlorides since such substrates typically require electron-
rich phosphine promoters.13 Under conditions recently re-
ported by Fu,14 we found that [nBu4N][Ph3BPm-iPr ] (5) pro-
2
moted the cross-coupling of the three substrates shown in
Table 1 in modestly good yield. For comparison, the iso-
structural but neutral ligand 9, as well as the more conven-
tional phosphine PiPr2Ph, were screened and also found to
give the cross-coupled products, albeit in yields that were
reproducibly ∼20% lower than the yields obtained using 5.15
The appreciable difference in yields likely reflects the greater
electron-releasing character of 5. Assuming oxidative addi-
tion of the aryl chloride to be rate-limiting, an anionic
[LPd0]- fragment would be expected to undergo oxidative
addition more rapidly than a neutral LPd0 fragment.
yielda (%)
R
L ) 5
L ) 9
L ) iPr2PPh
Me
71%
68%
74%
56%
43%
38%
49%
43%
49%
Cl
COMe
a
Isolated yields reported as the average of two independent runs.
Infrared model studies of ligands 5 and 9 are consistent
with this latter suggestion. For example, refluxing a solu-
tion of either 5 or 9 in a THF solution of Mo(CO)6 led, re-
spectively, to the anionic pentacarbonyl complex {nBu4N}-
shifted further downfield in their 31P NMR spectra (38.11
1
1
ppm, JPt-P ) 1900 Hz and 36.53 ppm, JPt-P ) 1917 Hz,
respectively). An XRD study of crystals of 12 confirmed its
cis coordination (Figure 2, also see Supporting Information).
The dianionic species 12-15 are highly reactive toward
both Bronsted and Lewis acids in THF and acetonitrile
solution. For example, stoichiometric addition of B(C6F5)3
effected the rapid release of 1 equiv of [NR4][Me(B(C6F5)3)]
(1H, 19F NMR) to produce the corresponding trans, mono-
anionic solvento species {NR4}{trans-[Ph3BP′]2Pt(Me)(solv)}.
trans-Disposition of the phosphine ligands was inferred from
{[Ph3BPm-iPr ]Mo(CO)5} (19) and the neutral pentacarbonyl
2
31
(Ph3SiPm-iPr )Mo(CO)5 (20), as confirmed by IR and P NMR
2
spectroscopy, as well as ES/MS. Infrared carbonyl vibrations
for 19 were recorded at 2065 and 1925 cm-1, whereas those
for 20 were recorded at 2070 and 1942 cm-1.16,17 Both the
high and low energy vibrations thus shift to lower energy in
the anionic system 19, likely reflecting an appreciable in-
crease in electron-releasing character of anionic 5 versus
neutral 9.
Future work will examine the utility of these anionic phos-
phines in generating zwitterionic precursors for catalytic
studies.
1
the dramatic increase in the JPt-P coupling constants and
the single resonance observed in the 31P NMR spectra. The
isostructural but neutral dimethyl complexes (Ph3SiPp-iPr )2-
2
PtMe2 (16) and (Ph3SiPp-Ph )2PtMe2 (17) displayed quite
2
distinct reactivity. For example, in the case of 16, methide
Acknowledgment. The NSF (CHE-01232216) and BP
have provided financial support for this work. Larry Henling
and J. Christopher Thomas provided assistance with the
crystallographic studies.
abstraction by B(C6F5)3 required ca. 12 h and led to the
cis-mono(solvento) species [cis-(Ph3SiPp-Ph )2PtMe(solv)]-
2
[Me(B(C6F5)3] (18) exclusively as the kinetic product at RT.
Slow isomerization of 18 to its thermodynamic trans isomer
occurred over a period of days in solution. Given the steric
similarity between ligands 6 and 10, the apparently substan-
tial rate difference displayed with respect to cis f trans
isomerization in these mono-solvento adducts is striking and
most likely electronic in origin. One plausible explanation
is to suggest that the anionic ligand 6 exerts a greater trans-
influence and thus labilizes solvent molecules in the trans
position of the kinetic cis-phosphine product to a larger extent
than for the neutral ligand 10, thereby facilitating rapid
isomerization.
As a final point of interest, we have briefly examined the
ability of 3-6 to promote Suzuki cross-coupling reactions.
Each ligand proved generally effective for the coupling of
PhB(OH)2 with typical aryliodide and arylbromide substrates
(see Supporting Information). More interesting was the ability
of these ligands to facilitate the cross-coupling of aryl
Supporting Information Available: Experimental procedures
(PDF), characterization data, additional Suzuki coupling data, and
crystallographic information (CIF). This material is available free
of charge via the Internet at http://pubs.acs.org.
IC0350234
(13) (a) Miyaura, N.; Suzuki, A. Chem. ReV. 1995, 95, 2457. (b) Suzuki,
A. J. Organomet. Chem. 1999, 576, 147.
(14) (a) Littke, A. F.; Fu, G. C. Angew. Chem., Int. Ed. 1998, 37, 3387.
(b) Littke, A. F.; Dai, C.; Fu, G. C. J. Am. Chem. Soc. 2000, 122,
4020.
(15) The phenyl substituted ligands 1, 3, and 4 were much less effective
in the Suzuki coupling of aryl chlorides. Ligand 6 showed comparable
activity to 5 in several model studies. Also, a point of concern pertains
to whether the tetraarylborate unit of ligand 5 is transferred during
the cross-coupling reactions. While we cannot rule-out this possibility
altogether, we note that tolylboronic acids were also screened briefly
and found to give comparable yields.
(16) Although three IR active vibrations are predicted, only the A12 and E
1
stretches are resolved. The A1 stretch is weak and presumably
coincident with the E vibration.
(12) (a) Konze, W. V.; Scott, B. L.; Kubas, G. J. Chem. Commun. 1999,
1807. (b) Alibrandi, G.; Minniti, D.; Scolaro, M.; Romeo, R. Inorg.
Chem. 1988, 27, 318. (c) Alibrandi, G.; Romeo, R. Inorg. Chem. 1997,
36, 4822.
(17) (a) Cotton, F. A.; Darensbourg, D. J.; Ilsley, W. H. Inorg. Chem. 1981,
20, 578. (b) Cotton, F. A.; Kraihanzel, C. S. J. Am. Chem. Soc. 1962,
84, 4432. (c) Magee, T. A.; Matthews, C. N.; Wang, T. S.; Wotiz, J.
H. J. Am. Chem. Soc. 1961, 83, 3200.
10 Inorganic Chemistry, Vol. 43, No. 1, 2004