Synthesis of hemilabile phosphine-phosphine oxide ligands via the highly selective Pd-catalyzed mono-oxidation of bidentate phosphines: Scope, limitations, and mechanism

Article abstract of DOI:10.1021/om010454k

The first simple and efficient, one-step catalytic method has been developed for the preparation of bis-phosphine monoxides, valuable hemilabile ligands that have proven usefulness in homogeneous catalysis, synthesis, analytical chemistry, cancer and AIDS research, chemistry of materials, etc. Readily available bidentate phosphines (dppm, dppe, dppp, dppb, dppbz, dppfc, and BINAP) are selectively oxidized to the corresponding mono-oxides in 65-90% isolated yield with 1,2-dibromethane in the presence of alkali and catalytic quantities (0.15-2 mol %) of a Pd(II) salt. This anaerobic oxidation smoothly occurs under biphasic conditions at 20-80 °C and atmospheric pressure. In certain cases, the mono-oxidation is promoted by I^-. Stoichiometric studies of the novel process demonstrate that the catalytic loop involves (i) the formation of Pd(0) and a bis-phosphine monoxide (BPMO) upon reaction of a Pd(II) complex of the corresponding bis-phosphine with alkali, (ii) chelation-driven phosphine ligand exchange to displace the coordinated BPMO produced, and (iii) reoxidation of the Pd(0) complex with 1,2-dibromoethane via C-Br oxidative addition followed by β-bromide elimination.

Full text of DOI:10.1021/om010454k

3950
Organometallics 2001, 20, 3950-3961
Syn th esis of Hem ila bile P h osp h in e-P h osp h in e Oxid e
Liga n d s via th e High ly Selective P d -Ca ta lyzed
Mon o-oxid a tion of Bid en ta te P h osp h in es: Scop e,
Lim ita tion s, a n d Mech a n ism
Vladimir V. Grushin†
Department of Chemistry, Wilfrid Laurier University, Waterloo, Ontario, N2L 3C5, Canada
Received May 30, 2001
The first simple and efficient, one-step catalytic method has been developed for the
preparation of bis-phosphine monoxides, valuable hemilabile ligands that have proven
usefulness in homogeneous catalysis, synthesis, analytical chemistry, cancer and AIDS
research, chemistry of materials, etc. Readily available bidentate phosphines (dppm, dppe,
dppp, dppb, dppbz, dppfc, and BINAP) are selectively oxidized to the corresponding mono-
oxides in 65-90% isolated yield with 1,2-dibromoethane in the presence of alkali and catalytic
quantities (0.15-2 mol %) of a Pd(II) salt. This anaerobic oxidation smoothly occurs under
biphasic conditions at 20-80 °C and atmospheric pressure. In certain cases, the mono-
oxidation is promoted by I-. Stoichiometric studies of the novel process demonstrate that
the catalytic loop involves (i) the formation of Pd(0) and a bis-phosphine monoxide (BPMO)
upon reaction of a Pd(II) complex of the corresponding bis-phosphine with alkali, (ii) chelation-
driven phosphine ligand exchange to displace the coordinated BPMO produced, and (iii)
reoxidation of the Pd(0) complex with 1,2-dibromoethane via C-Br oxidative addition followed
by â-bromide elimination.
In tr od u ction
diphosphine and its monoxide and dioxide.10,18-21 Ma¨d-
ing and Scheller19 have reported that basic bidentate
dialkylphosphino substrates (e.g., dmpe) can be selec-
tively mono-oxidized via monoprotonation followed by
oxidation. Due to their poor basicity, the most readily
available and attractive diarylphosphino substrates
cannot be mono-oxidized this way.
Here we report the first general, highly efficient and
convenient method for selective catalytic mono-oxidation
of aromatic diphosphines to the corresponding BPMOs.
Bis-phosphine monoxides (BPMOs) of the general
formula R1R2P-Y-P(O)R3R4, where Y is a divalent
spacer, constitute one of the most important classes of
hemilabile ligands.1 Various BPMOs have already proven
remarkable usefulness in inorganic/organometallic syn-
thesis,2 metal complex catalysis,3-9 cancer10 and AIDS11
research, and analytical chemistry.12 While the syn-
thetic utility and potential of mixed phosphine-phos-
phoryl ligands have long been realized, the chemistry
of BPMOs has been developing slowly due to the lack
of a convenient general method to synthesize these
ligands.
A few methods2t,w,13-15 have been developed to pre-
pare BPMOs from two P1 building blocks. These meth-
ods are specific to the synthesis of BPMOs with only
one or two C atoms separating the P atoms. A two-step
process has been developed16 for the preparation of
BPMOs via monobenzylation of a bidentate phosphine,
followed by alkaline hydrolysis.
(2) (a) Faller, J . W.; Grimmond, B. J .; Curtis, M. Organometallics
2000, 19, 5174. (b) Smith, M. B.; Slawin, A. M. Z. Inorg. Chim. Acta
2000, 299, 172. (c) Faller, J . W.; Patel, B. P.; Albrizzio, M. A.; Curtis,
M. Organometallics 1999, 18, 3096. (d) Slawin, A. M. Z.; Smith, M. B.;
Woollins, J . D. Polyhedron 1999, 18, 1135. (e) Coyle, R. J .; Slovokhotov,
Yu. L.; Antipin, M. Yu.; Grushin, V. V. Polyhedron 1998, 17, 3059. (f)
Brassat, I.; Englert, U.; Keim, W.; Keitel, D. P.; Killat, S.; Suranna,
G.-P.; Wang, R. Inorg. Chim. Acta 1998, 280, 150. (g) Hipler, B.;
Doering, M.; Dubs, C.; Goerls, H.; Huebler, T.; Uhlig, E. Z. Anorg. Allg.
Chem. 1998, 628, 1329. (h) Vicente, J .; Arcas, A.; Bautista, D.; J ones,
P. G. Organometallics 1997, 16, 2127. (i) Mecking, S.; Keim, W.
Organometallics 1996, 15, 2650. (j) Bhattacharyya, P.; Slawin, A. M.
Z.; Smith, M. B.; Woollins, J . D. Inorg. Chem. 1996, 35, 3675. (k)
Slawin, A. M. Z.; Smith, M. B.; Woollins, J . D. J . Chem. Soc., Dalton
Trans. 1996, 4575; 4567; 1283. (l) Smith, M. B.; Slawin, A. M. Z.;
Woollins, J . D. Polyhedron 1996, 15, 1579. (m) Al-J ibouri, S. A.; Al-
Hayaly, L. J .; Al-Allaf, T. A. K. Asian J . Chem. 1995, 7, 471. (n) Colton,
R.; Harrison, K. L.; Mah, Y. A.; Traeger, J . C. Inorg. Chim. Acta 1995,
231, 65. (o) Al-J ibouri, S. A.; Kalay, Z. M.; Al-Allaf, T. A. K. Transition
Met. Chem. (London) 1994, 19, 293. (p) Blagborough, T. C.; Davis, R.;
Ivison, P. J . Organomet. Chem. 1994, 467, 85. (q) Visseaux, M.;
Dormond, A.; Baudry, D. Bull. Soc. Chim. Fr. 1993, 130, 173. (r) Rossi,
R.; Marchi, A.; Marvelli, L.; Magon, L.; Peruzzini, M.; Casellato, U.;
Graziani, R. Inorg. Chim. Acta 1993, 204, 63. (s) Berry, D. E.;
Browning, J .; Dixon, K. R.; Hilts, R. W. Can. J . Chem. 1988, 66, 1272.
(t) Higgins, S. J .; Taylor, R.; Shaw, B. L. J . Organomet. Chem. 1987,
325, 285. (u) Bao, Q. B.; Landon, S. J .; Rheingold, A. L.; Haller, T. M.;
Brill, T. B. Inorg. Chem. 1985, 24, 900. (v) Grim, S. O.; Walton, E. D.
Inorg. Chem. 1980, 19, 1982. (w) Grim, S. O.; Satek, L. C.; Tolman, C.
A.; J esson, J . P. Inorg. Chem. 1975, 14, 656.
The simplest and most attractive way to prepare
BPMOs would be direct selective mono-oxidation of
readily available bidentate phosphines. However, the
direct oxidation of bidentate phosphines with conven-
tional oxidants (e.g., O2, H2O2, Br2/H2O) is usually
nonselective, leading to mixtures of the unreacted
† Present address: Central Research and Development, E. I. DuPont
de Nemours & Co., Inc., Experimental Station E328/306, Wilmington,
DE 19880-0328.
(1) For reviews, see: Bader, A.; Lindner, E. Coord. Chem. Rev. 1991,
108, 27. Slone, C. S.; Weinberger, D. A.; Mirkin, C. A. Prog. Inorg.
Chem. 1999, 48, 233.
10.1021/om010454k CCC: $20.00 © 2001 American Chemical Society
Publication on Web 08/10/2001

Phosphine-Phosphine Oxide Ligands
Organometallics, Vol. 20, No. 18, 2001 3951
A preliminary communication of this novel method has
been published.22
Resu lts
Our research was focused exclusively on aromatic
substrates containing PPh2 moieties because they can-
not be mono-oxidized selectively, using the simple
Ma¨ding-Scheller method.19 The biphasic process de-
veloped (eq 1) readily occurred under mild conditions
to produce the desired products in 50-90% isolated
yield. The mono-oxidation was successfully performed
for bis(diphenylphosphino)methane (dppm), 1,2-bis-
(diphenylphosphino)ethane (dppe), 1,3-bis(diphenylphos-
phino)propane (dppp), 1,4-bis(diphenylphosphino)bu-
tane (dppb), 1,2-bis(diphenylphosphino)benzene (dppbz),
1,1′-bis(diphenylphosphino)ferrocene (dppfc), and 2,2′-
bis(diphenylphosphino)-1,1′-binaphthyl (BINAP) in its
R, S, and racemic forms.
(3) For hydroformylation of olefins3a-f and epoxides,3g,h see: (a)
Huang, I. D.; Westner, A. A.; Oswald, A. A. J ermansen, T. G. PCT
Int. Appl. 8001690, 1980. (b) Abatjoglou, A. G.; Bryant, D. R. Eur. Pat.
Appl. EP 73398, 1983; U.S. Patent 4491675, 1985; U.S. Patent
4593011, 1986. (c) Abatjoglou, A. G.; Billing, E. Eur. Pat. Appl. EP
73961, 1983; U.S. Patent 4400548, 1983; U.S. Patent 4522933, 1985.
(d) Oswald, A. A.; J ermansen, T. G.; Westner, A. A.; Huang, I. D. U.S.
Patent 4687874, 1987. (e) Terekhova, M. I.; Kron, T. E.; Noskov, Yu.
G.; Petrov, E. S. Zh. Obshch. Khim. 1994, 64, 1966. (f) Abu-Gnim, C.;
Amer, I. J . Organomet. Chem. 1996, 516, 235. (g) Weber, R.; Keim,
W.; J aeger, B.; Haas, T.; Vanheertum, R. Eur. Pat. Appl. EP 1000921,
2000. (h) Weber, R.; Keim, W.; Mothrath, M.; Englert, U.; Ganter, B.
Chem. Commun. 2000, 1419.
(4) For carbonylation of alcohols, see: (a) Wegman, R. W.; Schrek,
D. J . Eur. Pat. Appl. EP 173170, 1986. (b) Wegman, R. W. Eur. Pat.
Appl. EP 171804, 1986. (c) Wegman, R. W.; Abatjoglou, A. G. PCT Int.
Appl. 8600888, 1986; U.S. Patent 4670570, 1987. (d) Wegman, R. W.;
Abatjoglou, A. G.; Harrison, A. M. J . Chem. Soc., Chem. Commun.
1987, 1891.
(5) For carbonylation of esters, see: (a) Wegman, R. W. U.S. Patent
4563309, 1984. (b) Wegman, R. W. PCT Int. Appl. 8600889, 1986.
(6) For hydroxycarbonylation of olefins, see: Terekhova, M. I.; Kron,
T. E.; Bondarenko, N. A.; Petrov, E. S.; Tsvetkov, E. N. Izv. Akad. Nauk
SSSR, Ser. Khim. 1992, 2003.
(7) For ethylene oligomerization and copolymerization with CO,
see: Brassat, I.; Keim, W.; Killat, S.; Mothrath, M.; Mastrorilli, P.;
Nobile, C. F.; Suranna, G. P. J . Mol. Catal. A 2000, 157, 41.
(8) Enantioselective hydrosylylation,8a hydroformylation,8b and
cycloaddition8c-e reactions catalyzed by transition metal complexes of
optically active BPMO, e.g., BINAP mono-oxide (BINAP(O))8a-c,e have
been recently reported. An attempt to use BINAP(O) for the hydrovi-
nylation of 2-methoxy-6-vinylnaphthalene has been unsuccessful.8f
Some BINAP(O) complexes of Rh have been reported8g to catalyze
various hydrogenation, hydroboration, and hydroformylation reactions,
albeit with low ee’s. Nonenzymatic kinetic resolution of secondary
alcohols in the presence of BINAP(O) has been recently described.8h
See: (a) Gladiali, S.; Pulacchini, S.; Fabbri, D.; Manassero, M.; Sansoni,
M. Tetrahedron: Asymmetry 1998, 9, 391. (b) Gladiali, S.; Alberico,
E.; Pulacchini, S.; Kolla´r, L. J . Mol. Catal. A 1999, 143, 155. (c) Faller,
J . W.; Parr, J . Organometallics 2000, 19, 1829. (d) Faller, J . W.; Liu,
X.; Parr, J . Chirality 2000, 12, 325. Faller, J . W.; Parr, J . Organome-
tallics 2001, 20, 697. (e) Faller, J . W. Grimmond, B. J .; D’Alliessi, D.
G. J . Am. Chem. Soc. 2001, 123, 2525. (f) Nandi, M.; J in, J .; RajanBabu,
T. V. J . Am. Chem. Soc. 1999, 121, 9899. (g) Gladiali, S.; Medici, S.;
Kegl, T.; Kollar, L. Monatsh. Chem. 2000, 131, 1351. (h) Sekar, G.;
Nishiyama, H. J . Am. Chem. Soc. 2001, 123, 3603.
Ca ta lytic Mon o-oxid a tion of Bid en ta te P h os-
p h in es. As shown below, careful optimization of reac-
tion conditions for each particular bis-phosphine sub-
strate is needed to obtain the desired BPMO product
in good yield. All catalytic reactions (Table 1) were run
under nitrogen to protect Pd(0) catalytic intermediates
from air and avoid noncatalytic, nonselective oxidation
of the substrate.
P h 2P (CH2)2P (O)P h 2 (d p p eO). The mono-oxidation
of dppe to dppeO (eq 1) was successfully performed on
a 0.5-50 g scale in 75-90% isolated yield. Although any
palladium compound capable of forming [(dppe)2Pd]2+
or [(dppe)2Pd] upon mixing with dppe may be used as a
catalyst for this reaction, Pd(OAc)2 was the most
convenient added catalyst to use due to its stability and
solubility in organic solvents. Platinum complexes re-
sulted in slower oxidation. The oxidation was conve-
niently performed in a 1,2-dichloroethane (DCE)-
aqueous NaOH biphasic system under reflux. It took
longer time (days) for the reaction to go to completion
at room temperature. In the absence of a catalyst, a
slow, nonselective oxidation reaction occurred, probably
via the formation of a phosphonium salt from dppe and
1,2-dibromoethane, followed by alkaline hydrolysis.16a
Because of this side-reaction, lower substrate-to-catalyst
ratios resulted not only in longer reaction times but also
in lower yields. For instance, an increase in the dppe to
Pd molar ratio from 225 to 570 lowered the yield of
dppeO from 87 to 82% (Table 1).
(9) (a) Olefin hydroformylation3 and hydroxycarbonylation6 reactions
catalyzed by Rh and Pd BPMO complexes occur with unusually high
selectivities to the desired linear products. In the presence of
Ph2P(O)(CH2)2PPh2, the Rh-catalyzed Monsanto acetic acid process
(MeOH + CO ) AcOH; accounts for 55% of all acetic acid produced
worldwide)9b is as efficient at 80 °C and 50 psi as the BPMO-free
industrial process, which is conventionally run at 200 °C and 500 psi.4a,d
(b) Weissermel, K.; Arpe, H.-J . Industrial Organic Chemistry, 3rd ed.;
Wiley-VCH: Weinheim, 1997.
(10) Berners-Price, S. J .; Norman, R. E.; Sadler, P. J . J . Inorg.
Biochem. 1987, 31, 197.
(11) J iang, S.; Debnath, A. K. PCT Int. Appl. 0055377, 2000.
Debnath, A. K.; Radigan, L.; J iang, S. J . Med. Chem. 1999, 42, 3203.
(12) Lomakina, L. N.; Ignat’eva, T. I.; Pisareva, S. A.; Medved’, T.
Ya.; Kabachnik, M. I. Zh. Anal. Khim. 1980, 35, 86.
(13) Seyferth, D. U.S. Patent 3426021, 1969.
(14) Rudomino, M. V.; Tsvetkov, E. N. Synthesis 1991, 125.
(15) (a) Kabachnik, M. I.; Medved’, T. Ya.; Pisareva, S. A.; Ignat’eva,
T. I.; Lomakina, L. N.; Kozachenko, A. G.; Matrosov, E. I.; Petrovskii,
P. V.; Komarova, M. P. Izv. Akad Nauk SSSR, Ser. Khim. 1980, 673.
(b) Bondarenko, N. A.; Rudomino, M. V.; Tsvetkov, E. N. Izv. Akad.
Nauk SSSR, Ser. Khim. 1990, 2180.
(16) (a) Abatjoglou, A. G.; Kapicak, L. A. Eur. Pat. Appl. EP 72560,
1983; U.S. 4429161, 1984. (b) There have been reports pointing to
difficulties preparing some BPMOs via this route.2t,17
(17) Shcherbakov, B. K. Personal communication.
(18) (a) Ellermann, J .; Schirmacher, D. Chem. Ber. 1967, 100, 2220.
(b) Siegl, W. O.; Lapporte, S. J .; Collman, J . P. Inorg. Chem. 1971, 10,
2158. (c) Lindner, E.; Beer, H. Chem. Ber. 1972, 105, 3261. (d) Brock,
S. L.; Mayer, J . M. Inorg. Chem. 1991, 30, 2138.
(19) Ma¨ding, P.; Scheller, D. Z. Anorg. Allg. Chem. 1988, 567, 179.
(20) (a) The Staudinger-type reaction of 1,2-bis(diphenylphosphino)-
benzene with Me3SiN3 results in the exclusive formation of the
corresponding mono-iminophosphorane,20b which may be hydrolyzed
to the oxide. This case is rather exceptional as diphosphines with a
flexible backbone, such as dppm,20c always react with azides nonse-
lectively. (b) Reed, R. W.; Santarsiero, B.; Cavell, R. G. Inorg. Chem.
1996, 35, 4292. (c) Katti, K. V.; Cavell, R. G. Inorg. Chem. 1989, 28,
413.
(21) It has been reported21a,b that some polydentate phosphines can
be air-oxidized in the presence of cobalt catalysts. This way triphos
has been oxidized to triphosO in 30-45% yield or to triphosO2 in 80-
95% yield.21a The cocatalyzed air-oxidation of dppm and dppe in the
presence of excess sacrificial 3-methylbutanal has been claimed to
produce dppmO and dppeO in ca. 55% yield at 88-100% conversion
(GC; no isolation reported).21b See: (a) Heinze, K.; Huttner, G.; Zsolnai,
L. Chem. Ber./ Recl. 1997, 130, 1393. (b) Mastrorilli, P.; Muscio, F.;
Nobile, C. F.; Suranna, G. P. J . Mol. Catal. A 1999, 148, 17.
(22) Grushin, V. V. J . Am. Chem. Soc. 1999, 121, 5831. Patent
protection: Grushin, V. V. U.S. Patent 5919984, 1999.

3952 Organometallics, Vol. 20, No. 18, 2001
Grushin
Ta ble 1. P a lla d iu m -Ca ta lyzed Mon o-oxid a tion of Bid en ta te P h osp h in es w ith 1,2-Dibr om oeth a n e u n d er
Bip h a sic Con d ition s (Or ga n ic Solven t-Aqu eou s Na OH)
The reaction was monitored by TLC (CH2Cl2-EtOAc,
end of the reaction (ca. 100% conversion of dppe) was
easily determined by a color change of the organic phase
2:1 by volume) and/or by 31P NMR spectroscopy. The

Phosphine-Phosphine Oxide Ligands
Organometallics, Vol. 20, No. 18, 2001 3953
from yellow ([(dppe)2Pd]) to pale yellow ([(dppeO)2Pd-
Br2]).2e At that point the reaction was stopped to avoid
further, much slower oxidation of the dppeO product to
dppeO2. The oxidation of dppeO is considerably slower
than dppe, providing the high selectivity for the catalytic
process. This is due to the fact that in media of low
polarity (1,2-dichloroethane) complexes of the type
[(dppeO)2PdX2] are mostly trans,2e reacting with alkali
more slowly than cis-[(dppe)PdX2] (trans-effect) and
water-soluble cationic [(dppe)2Pd]2+ (see below).
P h 2P (CH2)3P (O)P h 2 (d p p p O). The above descrip-
tion of the dppe mono-oxidation reaction applies to the
preparation of dpppO from dppp; except at room tem-
perature the highest conversion of dppp ever observed
was only ca. 70%. At 80 °C however, good yields of
dpppO were obtained at virtually quantitative conver-
sion (eq 1 and Table 1).
P h 2P (CH 2)4P (O)P h 2 (d p p b O). Due to the long
-(CH2)4- spacer, dppb forms weaker metal chelates
than dppe and dppp. Hence we had expected the mono-
oxidation of dppb to be poorly selective. Indeed, poor
selectivities and yields were observed when the oxida-
tion was performed at 80 °C. At room temperature,
however, good yields (>80%) of dppbO at a high conver-
sion were obtained (eq 1 and Table 1). As the oxidation
occurred, the poorly soluble dppbO formed precipitated
out. Consequently, even at as high conversions of dppb
to dppbO as 80-90% the concentration of dppbO in the
reaction solution remained low, thus keeping the dppbO
product from competing with the as yet unreacted dppb
for coordination to Pd. This, in turn, provided the high
selectivity for the reaction.
P h 2P CH2P (O)P h 2 (d p p m O). When the oxidation of
dppm was carried out under the conditions developed
for dppe (see above), only 60-70% conversions were
obtained, at which point the reaction slowed consider-
ably. Much higher conversions of 90-100% at 80-95%
selectivity were reached when the oxidation was run in
the presence of small quantities of NaI, ca. 5 mol % of
the amount of dppm to be oxidized. The multifunctional
role of the iodide promoter is discussed below. Because
metal complexes of dppm and dppmO exhibit consider-
able C-H acidity of the methylene group (see for
example, refs 2e and 23), highly concentrated alkali
solutions (>20%) should not be used for the dppm
oxidation, to avoid side-reactions due to deprotonation
of the metal intermediates.
1,2-P h 2P C6H4P (O)P P h 2 (d p p bzO). Although 1,2-
(Ph2P)2C6H4 (dppbz) has been mentioned in the litera-
ture twice7,20b its preparation has not been reported. We
found that dppbz is selectively oxidized to dppbzO in
over 80% isolated yield under conditions described above
for the oxidation of dppe to dppeO. The oxidation of
dppbz appeared to be equally selective and efficient in
the absence and in the presence of NaI (Table 1). The
31P NMR spectrum of dppbzO (see the Experimental
Section) was in accord with that reported for a material
tentatively formulated as dppbzO without isolation.20b
P h 2P C5H4F eC5H4P (O)P P h 2 (d p p fcO). Originally
isolated as a minor side-product (11% yield) of the
reaction between dppfc and [CpCo(CO)2],24 dppfcO is an
attractive organometallic BPMO that cannot be pre-
pared by conventional oxidations of dppfc. After careful
optimization of reaction conditions the Pd-catalyzed
oxidation afforded dppfcO in up to 65% isolated yield
(eq 1 and Table 1). Higher selectivities were obtained
when the mono-oxidation was run in dichloromethane
under reflux. Like the analogous reaction of dppm, the
Pd-catalyzed oxidation of dppfc is promoted by I-. While
the use of the Pd(OAc)2/NaI system (see above) led to a
relatively broad variation of yields of dppfcO in the
range of ca. 40-65%, consistently good isolated yields
of dppfcO (>60%) were obtained when PdI2 was em-
ployed as added catalyst (see the Experimental Section).
If necessary, the oxidation of dppfc can be driven to
dppfcO2 if run at a higher temperature (1,2-dichloro-
ethane under reflux) for a longer time.
It has been reported25 that [(dppfc)2Pd] and some
other dppfc Pd species undergo light-induced outer-
sphere oxidation with organic halides. However, no
significant difference in yields of dppfcO was noticed
when the oxidation of dppfc was performed with or
without protection from daylight.
P h 2P C10H6C10H6P (O)P P h 2 (BINAP (O)). The one-
step mono-oxidation of commercially available BINAP
is a good alternative to the recently reported8a multistep
synthesis of racemic BINAP(O). Both (R)- and (S)-
BINAP(O) were obtained by the Pd(OAc)2-catalyzed
oxidation of (R)- and (S)-BINAP, respectively, with 1,2-
dibromoethane/alkali in CH2Cl2 at room temperature.
The reaction was very selective but slow, taking days
to proceed to >80% conversion. Occasionally, early
catalyst deactivation was observed under such condi-
tions, accompanied by color change of the organic phase
from red to yellow. The loss of catalytic activity was
particularly fast when poorly soluble racemic (()-BINAP
was oxidized, with the conversions to (()-BINAP(O)
being as low as 5-15%. All these problems could be
eliminated by using PdI2 (ca. 2 mol %) instead of Pd-
(OAc)2 and running the oxidation under reflux (eq 1 and
Table 1). This way, conversions of 85-95% at 90-100%
selectivity to BINAP(O) were achieved after 8-12 h.
Isolated yields of spectroscopically (NMR) and chro-
matographically (TLC) pure (R)-, (S)-, and (()-BINAP-
(O) were in the range 70-85%.
P h 2P (CH2)5P (O)P h 2 a n d P h 2P (CH2)6P (O)P h 2. Our
numerous attempts to selectively mono-oxidize Ph2P-
(CH2)5PPh2 and Ph2P(CH2)6PPh2 using the above cata-
lytic method failed.26 The oxidations were nonselective,
always producing mixtures of the mono- and dioxides.
Oxid a tion of P h 2P CH2P (P h )CH2P P h 2 (tr ip h os).
This substrate is easily oxidized with 1,2-dibromoethane
and alkali in the presence of Pd(OAc)2. At low conver-
sions of 10-20%, Ph2PCH2P(Ph)CH2P(O)Ph2 and Ph2-
PCH2P(O)(Ph)CH2PPh2 were produced in ca. 2:1 sta-
tistical molar ratio (31P NMR). An immediate interpreta-
tion of this ratio was comparable reactivity of both types
of phosphine groups of the triphos molecule toward the
(24) Kim, T.-J .; Lee, T.-H.; Kwon, S.-C.; Kwon, K.-H.; Uhm, J .-K.;
Lee, H.; Byun, S.-I. Bull. Korean Chem. Soc. 1991, 12, 116.
(25) Kunkely, H.; Vogler, A. J . Organomet. Chem. 1998, 559, 215.
See also: Tanase, T.; Matsuo, J .; Onaka, T.; Begum, R. A.; Hamaguchi,
M.; Yano, S.; Yamamoto, Y. J . Organomet. Chem. 1999, 592, 103.
(26) (a) A different method has been developed for selective mono-
oxidation of Ph2P(CH2)5PPh2 and Ph2P(CH2)6PPh2 to Ph2P(CH2)5P(O)-
Ph2 and Ph2P(CH2)6P(O)Ph2, respectively.26b This method also works
well for other bis-phosphine substrates. (b) Grushin, V. V. Unpublished
results, 1996.
(23) Al-J ibori, S.; Shaw, B. L. Inorg. Chim. Acta 1982, 65, L 123;
1983, 74, 235.

3954 Organometallics, Vol. 20, No. 18, 2001
Grushin
Pd-catalyzed oxidation. As the reaction progressed, both
monoxide products (still containing two phosphine func-
tionalities and hence capable of forming P,P-chelates)
began to compete with the triphos for vacant coordina-
tion sites on Pd. Because of that, the overall reaction
was poorly selective, giving rise to triphosO (two iso-
mers), triphosO2 (two isomers), and triphosO3.
Rea ction Mech a n ism : Step -by-Step Mod elin g of
th e Ca ta lytic Loop . The starting point for the devel-
opment of the catalytic process (eq 1) was a series of
literature reports27 describing the ability of hard ligands,
such as O2-, F-, OH-, and AcO- to promote the
remarkably clean inner-sphere red-ox transformation
of phosphine complexes of Cu, Ni, Pd, and Pt (eq 2).
Pd remains in solution in the form of the corresponding
zerovalent complex, [(L-L)2Pd] (L-L ) diphosphine).
Rendering stoichiometric reactions 3 and 4 catalytic
in Pd would be possible if the Pd(0) formed could be
somehow oxidized back to Pd(II). The catalytic process
would be selective and efficient only if there was an
oxidant capable of rapidly oxidizing the Pd(0) back to
Pd(II), while remaining unreactive toward the bidentate
phosphine substrate. Since organic tertiary phosphines
and their Pd(0) complexes are both strong reductants,
finding such a selective oxidant was obviously the key
problem to be solved. Practical requirements for the
catalysis dictated that the oxidant must oxidize the
Pd(0) rapidly, orders of magnitude faster than the
phosphine substrate under conditions where the
concentration of the palladium catalyst is orders of
magnitude lower than that of the free phosphine
substrate.
Original experiments with some polyhalogenated
organic compounds (e.g., CCl4) and certain metal ions
(e.g., Cu2+ and Fe3+) as potentially suitable oxidants
failed. However, the most promising candidate, 1,2-
dibromoethane, appeared to be an excellent, mild, and
highly selective oxidant for Pd(0) in the presence of a
large excess of bidentate phosphine substrates.
Reaction 2 is most characteristic of Pd(II) com-
plexes,27d-s likely involving coordination of a hard
anionic base X- to the metal center, followed by P-X
reductive elimination. As a result, the Pd(II) is reduced
to Pd(0), while the tertiary phosphine is oxidized to the
corresponding P(V) derivative, such as phosphine oxide
or difluorophosphorane if X ) F.27d Importantly, Pd(II)
complexes of bidentate phosphines undergo the Pd(II)/
P(III) f Pd(0)/P(V) reduction-oxidation to give the
corresponding BPMO or its difluoro derivative (eqs 3
and 4).27d,h In the presence of excess bis-phosphine the
The suitability of 1,2-dibromoethane for reoxidation
of the Pd(0) stemmed from the fact that tertiary phos-
phine complexes of zerovalent palladium are usually
more reactive toward organic halides than free phos-
phines, especially in media of low polarity. Should the
C-Br bond of 1,2-dibromoethane oxidatively add to a
Pd(0) complex, the 2-bromoethylpalladium complex
would undergo â-Br elimination28 to produce Pd(II) and
ethylene. Although the C-Br oxidative addition step
had been expected to occur rather slowly,29 the first tests
of 1,2-dibromoethane for the Pd-catalyzed mono-oxida-
tion of dppe were successful. A series of stoichiometric
modeling studies is presented below in order to provide
an illustration for the grounds on which the catalytic
process was designed and developed.
Step 1. Rea ction of Ad d ed Ca ta lyst w ith d p p e
u n d er Bip h a sic Con d ition s (F igu r e 1). Dissolving
Pd(OAc)2 and dppe (4-fold excess) in DCE resulted in
the fast and clean formation of [(dppe)2Pd]2+ (AcO-)2
(Figure 1A; δ ) 55.7 ppm).31 Soon after the cationic
complex began to precipitate out, the DCE solution was
stirred with water for 5 min to produce two clear phases.
Both layers were analyzed by 31P NMR. Most of the
cationic complex was found in the aqueous layer (Figure
(27) (a) Malatesta, L.; Angoletta, M. J . Chem. Soc. 1957, 1186. (b)
Vinal, R. S.; Reynolds, L. T. Inorg. Chem. 1964, 3, 1062. (c) Berners-
Price, S. J .; J ohnson, R. K.; Mirabelli, C. K.; Faucette, L. F.; McCabe,
F. L.; Sadler, P. J . Inorg. Chem. 1987, 26, 3383. (d) Mason, M. R.;
Verkade, J . G. Organometallics 1990, 9, 864; 1992, 11, 2212. (e) Ioele,
M.; Ortaggi, G.; Scarsella, M.; Sleiter, G. Polyhedron 1991, 10, 2475.
(f) Kozitsyna, N. Yu.; Ellern, A. M.; Antipin, M. Yu.; Struchkov, Yu.
T.; Moiseev, I. I. J . Chem. Soc., Mendeleev Commun. 1991, 92. (g)
Amatore, C.; J utand, A.; M’Barki, M. A. Organometallics 1992, 11,
3009. (h) Ozawa, F.; Kubo, A.; Hayashi, T. Chem. Lett. 1992, 2177. (i)
Grushin, V. V.; Alper, H. Organometallics 1993, 12, 1890; J . Am. Chem.
Soc. 1995, 117, 4305. (j) Mandai, T.; Matsumoto, T.; Tsuji, J .; Saito, S.
Tetrahedron Lett. 1993, 34, 2513. (k) Ozawa, F.; Kubo, A.; Matsumoto,
Y.; Hayashi, T.; Nishioka, E.; Yanagi, K.; Moriguchi, K. Organome-
tallics 1993, 12, 4188. (l) Grushin, V.V. Bensimon, C.; Alper, H. Inorg.
Chem. 1994, 33, 4804. (m) Amatore, C.; Carre´, E.; J utand, A.; M’Barki,
M. A. Organometallics 1995, 14, 1818. (n) Papadogianakis, G.; Peters,
J . A.; Maat, L.; Sheldon, R. A. J . Chem. Soc., Chem. Commun. 1995,
1105. (o) Andrews, M. A.; Gould, G. L.; Voss, E. J . Inorg. Chem. 1996,
35, 5740. (p) Amatore, C.; J utand, A.; Medeiros, M. J . New J . Chem.
1996, 20, 1143. (q) McLaughlin, P. A.; Verkade, J . G. Organometallics
1998, 17, 5937. (r) Csa´kai, Z.; Skoda-Fo¨ldes, R.; Kolla´r, L. Inorg. Chim.
Acta 1999, 286, 93. (s) Amatore, C.; J utand, A.; Thuilliez, A. Organo-
metallics 2001, 20, 3241.
(28) Friedrich, H. B.; Moss, J . R. Adv. Organomet. Chem. 1991, 33,
235.
(29) By the time our project was launched, Fitton and Rick30a had
reported the lack of reaction between [(dppe)2Pd] and PhI. The report
of Amatore et al.30b had not yet appeared in the literature, describing
the lack of reaction between PhI and [(L-L)2Pd] (L-L ) various
bidentate phosphines). Had we been aware of that study30b we would
have been considerably less encouraged to use 1,2-dibromoethane
(which is presumably less reactive than PhI) as an oxidant for the
Pd(0). Very recently, Alcazar-Roman et al.30c reported oxidative addi-
tion of aryl halides to [(L-L)2Pd] (L-L ) dppfc, BINAP).
(30) (a) Fitton, P.; Rick, E. A. J . Organomet. Chem. 1971, 28, 287.
(b) Amatore, C.; Broeker, G.; J utand, A.; Khalil, F. J . Am. Chem. Soc.
1997, 119, 5176. (c) Alcazar-Roman, L. M.; Hartwig, J . F.; Rheingold,
A. L.; Liable-Sands, L. M.; Guzei, I. A. J . Am. Chem. Soc. 2000, 122,
4618.
(31) Lindsay, C. H.; Benner, L. S.; Balch, A. L. Inorg. Chem. 1980,
19, 3503.

Phosphine-Phosphine Oxide Ligands
Organometallics, Vol. 20, No. 18, 2001 3955
F igu r e 2. 31P NMR spectrum of the organic phase after
the DCE-H2O biphasic system containing dppe and
[(dppe)2Pd]2+(OAc-)2 (see Figure 1) was treated with
NaOH. The singlets at -12.8 and 29.9 ppm are from
dppe and [(dppe)2Pd], respectively. The doublets at
-12.4 and 30.8 ppm (J P-P ) 48 Hz) are from the dppeO
produced. The upfield component of the doublet at -12.4
ppm overlaps with the singlet resonance from dppe at
-12.8 ppm. For details, see text and the Experimental
Section.
formed originally,32 followed by chelation-driven ligand
exchange (eq 6).
F igu r e 1. (A) 31P NMR spectrum of the solution of Pd-
(OAc)2 (30 mg) and dppe (213 mg; 4-fold excess) in DCE (5
mL) recorded immediately after the reagents were mixed.
The singlet at -12.8 ppm is from free dppe; the singlet at
55.7 ppm is due to [(dppe)2Pd]2+(OAc-)2. (B) 31P NMR
spectrum of the aqueous phase after the mixture was
stirred with water (5 mL) for 1 min. The singlet resonance
at 57.8 ppm is from [(dppe)2Pd]2+(OAc-)2. For details, see
text and the Experimental Section.
1B; δ ) 57.8 ppm), and only trace amounts of [(dppe)2-
The mechanism of the reaction of tertiary phosphine
complexes of Pd(II) with alkali likely involves nucleo-
philic attack of the OH- on the metal, followed by P-O
reductive elimination.27i
Pd]2+ remained in the organic phase (eq 5).
Step 3. Oxid a tion of [(d p p e)2P d ] w ith 1,2-Dibr o-
m oeth a n e (F igu r e 3). To the NMR sample containing
the organic phase, 1,2-dibromoethane was added. The
tube was placed back in the NMR probe, and 31P NMR
spectra of the sample were recorded every 10-15 min.
The signal at 30.0 ppm from the [(dppe)2Pd] gradually
diminished in intensity while being replaced by a new
31
singlet resonance at 54.3 ppm from [(dppe)2Pd]Br2
(Figure 3). After 2 h the resonance from [(dppe)2Pd]
disappeared altogether. No other changes in the original
31P NMR pattern were observed, pointing to the high
selectivity of the oxidation reaction (eq 7). At certain
Step 2. Rea ction of [(d p p e)2P d ]2+ w ith Alk a li
(F igu r e 2). The organic layer of the biphasic system
turned bright yellow immediately upon addition of
aqueous NaOH. After 1 min of vigorous stirring both
layers were analyzed by 31P NMR spectroscopy. No
signals were found in the spectrum of the colorless
aqueous phase. The spectrum of the yellow organic
phase (Figure 2) contained four resonances, i.e., a singlet
at -12.8 ppm (excess dppe), two doublets at -12.4 and
30.8 ppm with J P-P ) 48.0 Hz (dppe monoxide,
dppeO),22,27c,d,r and a singlet at 29.9 ppm from the
zerovalent complex, [(dppe)2Pd].27d,30a,b Thus, as antici-
pated, the alkali-induced redox reaction readily occurred
to produce the desired product, dppeO, with 100%
selectivity. It is believed that [(dppe)Pd(dppeO)] is
(32) A clear colorless aqueous solution of [(dppe)2Pd]2+ (AcO-)2
turned yellow while growing turbid immediately upon addition of
alkali. After 1.5 h
a microcrystalline, highly air-sensitive orange
precipitate formed, presumably [(dppe)Pd(dppeO)], which was ex-
tracted with O2-free DCE and analyzed by 31P NMR. The spectrum
exhibited two broad resonances at ca. 30 and 22 ppm assigned to Ph2-
PO and Ph2P-Pd, respectively.

3956 Organometallics, Vol. 20, No. 18, 2001
Grushin
Sch em e 1
attention. First, unlike alkyl and benzyl bromides,16 1,2-
dibromoethane is poorly reactive toward dppe and other
tertiary phosphines. Neither phosphonium salts nor any
other observable side-products were formed in step 3,
in which [(dppe)2Pd] was oxidized with 1,2-dibromo-
ethane in the presence of free dppe. Second, given the
diminished reactivity of [(L-L)2Pd] toward PhI,29,30 the
oxidative addition reaction of 1,2-dibromoethane to
[(dppe)2Pd] was surprisingly facile. This unexpectedly
high reactivity may be accounted for by the formation
of a mixed chelate [(dppe)Pd(η2-BrCH2CH2Br)], within
which the oxidation addition occurred. The ability of
organic halides to coordinate to transition metals via
the halogen atom has been established.33 A stable irid-
ium η2-1,2-diiodobenzene chelate has been isolated and
characterized by single-crystal X-ray diffraction.34 After
oxidative addition of one of the two C-Br bonds to Pd,
the bromine of the remaining C-Br bond is probably
coordinated to the metal center,35 facilitating halogen
â-elimination (eq 13).36 As the reaction between
F igu r e 3. 31P NMR spectra recorded 5 (A), 20 (B), 40 (C),
and 120 (D) min after 1,2-dibromoethane was added to the
NMR sample containing dppe, dppeO, and [(dppe)2Pd] (see
Figure 2). The singlets at -12.8, 29.9, and 54.3 ppm are
from dppe, [(dppe)2Pd], and [(dppe)2Pd]Br2, respectively.
The doublets at -12.4 and 30.8 ppm (J P-P ) 48 Hz) are
from dppeO. The upfield component of the doublet at -12.4
ppm overlaps with the singlet resonance from dppe at
-12.8 ppm. For details, see text and the Experimental
Section.
1
point, a weak signal at 5.3 ppm was observed in the H
NMR spectrum of the sample, indicating the formation
of ethylene. When the sample was ejected from the
probe after the reaction had gone to completion, a large
amount of precipitate was found in the NMR tube. The
well-shaped crystals were separated and found to be
identical with an authentic sample of [(dppe)2Pd]Br2.
The catalytic loop thus closed (Scheme 1).
(33) For reviews, see: Kulawiec, R. J .; Crabtree, R. H. Coord. Chem.
Rev. 1990, 99, 89. Plenio, H. Chem. Rev. 1997, 97, 3363. See also:
Butts, M. D.; Scott, B. L.; Kubas, G. J . J . Am. Chem. Soc. 1996, 118,
11831, and references therein.
(34) Crabtree, R. H.; Faller, J . W.; Mellea, M. F.; Quirk, J . M.
Organometallics 1982, 1, 1361.
Concerning the reaction of 1,2-dibromoethane with
[(dppe)2Pd], two important observations merit special

Phosphine-Phosphine Oxide Ligands
Organometallics, Vol. 20, No. 18, 2001 3957
Sch em e 2
Discu ssion
The generality of the catalytic method (eq 1) has been
demonstrated by the high-yield (up to 90%) preparation
of BPMOs, in which the phosphine and phosphoryl
groups are separated by a variety of spacers, i.e., a
saturated hydrocarbon chain of different length (dppm-
O, dppeO, dpppO, and dppbO), a benzene ring (dppbz-
O), the ferrocene framework (dppfcO), and a sequence
of C-C bonds involved in a polyaromatic skeleton
(BINAP(O)). Both high efficiency and selectivity of the
mono-oxidation can be achieved for bis-phosphines with
different bite angles, forming chelates of various strength,
from weak (dppb) to very strong (dppe) and “super-
strong” (dppbz).39 A discussion below deals with differ-
ent factors influencing the catalytic mono-oxidation of
bis-phosphines to BPMOs.
Bis-P h osp h in e Su bstr a tes. The method is suitable
only for bidentate phosphines capable of forming che-
lates with the metal catalyst (Pd). The oxidation may
be selective only if the bis-phosphine chelates of Pd(II)
are more stable than its complexes of the corresponding
BPMOs.
Meta l Com p lex Ca ta lysts. Although Pt compounds
may be used to catalyze the mono-oxidation, less costly
palladium complexes exhibit higher catalytic activity
and selectivity to BPMOs. For instance, under similar
conditions, the oxidation of dppm catalyzed by PtCl2
afforded dppmO in only 45% yield, whereas much higher
yields (up to 84%) were obtained in the presence of Pd
catalysts (see the Experimental Section). Likewise, the
Pt- and Pd-catalyzed oxidations of dppfc afforded dppfcO
in 32% and 65% yield, respectively.
P h a se-Tr a n sfer Ca ta lysts. We have examined the
influence of various phase-transfer catalysts (5-10%)
on the Pd-catalyzed oxidation of bis-phosphines. It has
been found that all phase-transfer agents tested (Bu4N
HSO4, Et3NCH2Ph Br, 18-crown-6) lowered the selectiv-
ity of the process, i.e., the yields of BPMOs decreased
due to the formation of large quantities of bis-phosphine
dioxides. The detrimental effect of the phase-transfer
agents is explained by catalysis of X/OH ligand ex-
change in [(BPMO)2PdX2] (X ) halogen), which results
in oxidation of the BPMO to the dioxide (see above).
Coca ta lysis w ith Iod id e. The iodide cocatalyst is
not required for reactions that proceed via the mecha-
nism outlined in Scheme 1, i.e., for the bis-phosphine
substrates that form cationic complexes of the type [(L-
L)2Pd]2+ (e.g., L-L ) dppe, dppbz, and dppp). These
cationic complexes are water-soluble, readily reacting
with OH- in the aqueous phase of the biphasic system
(see above).
[(dppe)2Pd] and 1,2-dibromoethane was monitored
by 31P NMR spectroscopy, no intermediates were ob-
served.
The mechanism shown in Scheme 1 governs the
mono-oxidation of dppe and other bidentate phosphine
substrates, although some variations may take place.
In certain cases, the structure and composition of the
Pd(II) and Pd(0) bis(phosphine) intermediates may be
different from [(η2-L-L)2Pd]2+ and [(η2-L-L)2Pd]. For
instance, neither Pd(0)30b,c nor Pd(II)27h,37,38 forms stable
bis-chelates with BINAP in solution. Reacting PdBr2
and (S)-BINAP (1:1) in CH2Cl2 gave rise to yellow-
orange [((S)-BINAP)PdBr2] (31P NMR: singlet, δ ) 26.5
ppm). The complex stayed intact and its 31P NMR signal
remained unaffected upon addition of 2 equiv of (S)-
BINAP. Therefore, in sharp contrast with the above
dppe experiments, no formation of cationic bis-chelate
[((S)-BINAP)2Pd]2+ was observed. Nonetheless, when
treated with alkali, the solution containing [((S)-BI-
NAP)PdBr2] and (S)-BINAP turned cherry-red due to
the formation of Pd(0) species (δ ) 27.4 ppm) along with
(S)-BINAP monoxide ((S)-BINAP(O); δ ) -14.6 and 28.4
ppm). Since neutral [((S)-BINAP)PdBr2] (unlike salts of
[(dppe)2Pd]2+) is insoluble in water, the reaction with
alkali likely occurred at the interface rather than in the
bulk of the aqueous phase.38 The red zerovalent Pd
complex readily reacted with 1,2-dibromoethane to
produce [((S)-BINAP)PdBr2] (δ ) 26.5 ppm), the starting
point of the catalytic loop (Scheme 2).
In contrast, the beneficial effect of the iodide is well
pronounced for the oxidation of bis-phosphines such as
dppfc and BINAP, which are not prone to form water-
soluble cationic complexes [(dppfc)2Pd]2+ and [(BINAP)2-
(35) Garcia, M. P.; J imenez, M. V.; Cuesta, A.; Siurana, C.; Oro, L.
A.; Lahoz, F. J .; Lopez, J . A.; Catalan, M. P.; Tiripicchio, A.; Lanfranchi,
M. Organometallics 1997, 16, 1026.
(36) (a) As the catalytic oxidation occurs, ethylene resulting from
the â-halogen elimination may bind to Pd(0) intermediates to form
unstable [(L-L)Pd(C2H4)], from which the olefin ligand is known36b to
be easily displaced by excess diphosphine. (b) Broadwood-Strong, G.
T. L.; Chaloner, P.; Hitchcock, P. B. Polyhedron 1993, 12, 721.
(37) Wolfe, J . P.; Buchwald, S. L. J . Org. Chem. 2000, 65, 1144.
(38) If it exists, the equilibrium between [(BINAP)PdBr2] and [(η2-
BINAP)(η1-BINAP)PdBr]+Br- is shifted toward the former by >99%,
as indicated by the 31P NMR studies described above. It is conceivable
that the reaction between [(BINAP)PdBr2] and NaOH in the presence
of BINAP may be driven by the reversible formation of a cationic
complex, [(η2-BINAP)(η1-BINAP)PdBr]+Br-, which may be somewhat
soluble in water.
Pd]2+ 38
Neutral complexes [(L-L)PdX2] are formed
.
instead, which stay in the organic phase, reacting with
aqueous alkali at the interface (Scheme 2). Under
biphasic conditions, tertiary phosphine palladium(II)
halides undergo facile ligand exchange with halide
(39) (a) Crumpton, D. M.; Goldberg, K. I. J . Am. Chem. Soc. 2000,
122, 962. (b) Mann, G.; Baranano, D.; Hartwig, J . F.; Rheingold, A. L.;
Guzei, I. A. J . Am. Chem. Soc. 1998, 120, 9205.

3958 Organometallics, Vol. 20, No. 18, 2001
Grushin
anions present in the aqueous phase.40 As the iodide-
promoted bis-phosphine oxidation occurs, two different
halide anions are present in the biphasic system, i.e.,
the I- and Br- from 1,2-dibromoethane (eq 1). In accord
with hydration/solvation energy considerations, the
lighter, more strongly hydrated bromide would be
preferentially located in the aqueous phase. Conse-
quently, the heavier iodide would be bound to Pd in the
organic phase of low polarity. In fact, after the PdI2-
catalyzed mono-oxidation of dppfc had gone to ca. 80%
conversion, it is [(dppfc)PdI2], not [(dppfc)PdBr2] that
was isolated from the organic phase by column chro-
matography.41 Only trace amounts of the bromide
[(dppfc)PdBr2] were detected by TLC, indicating that the
Pd(II) binds selectively to I- rather than to Br-, even
at Br-:I- g 45 (see dppfc oxidation, experiment b of the
Experimental Section). Similar observations were made
when the oxidation of BINAP was performed in the
presence of PdI2.
The preferential formation of [(L-L)PdI2] in the
organic phase during the reaction has an important
consequence. Because the Pd-X bond is more labile for
X ) I than for Br,40 the iodo complexes [(L-L)PdI2]
undergo Pd-X ionization and ligand exchange reactions
more readily than their bromide analogues [(L-L)-
PdBr2]. This way the displacement of X for OH in [(L-
L)PdX2] is facilitated for X ) I, resulting in a faster
Pd(II)/P(III) f Pd(0)/P(V) redox process. When dppm
is ozidized, the presence of I- favors the formation of
more reactive Pd(II) complexes containing chelating
rather than more common bridging dppm (A-frame).42
Because a vacant coordination site is required for
â-elimination reactions,43 the â-halogen elimination
from [(L-L)Pd(X)(CH2CH2Br)] (Schemes 1 and 2) should
occur more readily for the more labile iodo complex.
It has been demonstrated44 that in the presence of I-
tertiary phosphine complexes of Pd(0) form electron-rich
anionic complexes [L2PdI]-, which undergo oxidative
addition of C-Hal bonds much faster than analogous
neutral complexes of Pd(0) devoid of an anionic ligand.
Therefore, the iodide may also facilitate oxidative ad-
dition of BrCH2CH2Br to Pd(0) complexes involved in
the catalytic cycle (Schemes 1 and 2).
The oxidation was sluggish when less polar benzene,
toluene, and ether were used as the organic phase
immiscible with water. No reaction occurred in hexane
due to very low solubility of the substrate and catalyst.
Oxid a n t. The best results were obtained with 1,2-
dibromoethane. Other 1,2-dihaloalkanes examined, i.e.,
1-chloro-2-bromoethane, 1-chloro-2-iodoethane, and 1,2-
diiodoethane gave lower yields and selectivities. The
C-Cl bond of 1,2-dichloroethane (DCE) is not suf-
ficiently reactive toward [(L-L)2Pd]. In fact, DCE was
successfully used as the organic phase.
Ba se. Most reactions were run in the presence of
aqueous NaOH. Other bases, such as KOH, Na2CO3,
and NaOAc, were also examined, mostly for dppfc
oxidation. While comparable results were obtained with
KOH, lower yields of dppfcO were obtained when Na2-
CO3 or NaOAc were employed.
At m osp h er e. The reactions were run under N2 to
protect air-sensitive Pd(0) intermediates. An attempt to
perform the otherwise selective Pd-catalyzed mono-
oxidation of (S)-BINAP in air led to a sluggish and
poorly selective reaction. Once formed in the anaerobic
catalytic reaction, the air-stable BPMOs can be suc-
cessfully isolated in air.
Exp er im en ta l Section
A Varian VXR-200 instrument was used for measuring
NMR spectra. Bidentate phosphines and other chemicals were
purchased from Organometallics, Strem, Aldrich, and TCI
chemical companies and used as received. All Pd-catalyzed
oxidation reactions were conducted under nitrogen and moni-
tored by TLC and/or 31P NMR spectroscopy. Isolation and
purification of the BPMOs were performed in air.
P h 2P CH2P (O)P P h 2, d p p m O. (a) A mixture of palladium
acetate (5 mg; 2.2 × 10-2 mmol), dppm (2.00 g; 5.21 mmol),
1,2-dibromoethane (2.0 g; 10.64 mmol), and 1,2-dichloroethane
(10 mL) was stirred for 20 min. To this solution was added
aqueous NaOH (20 wt %; 6 mL) containing NaI (40 mg; 0.27
mmol), and the mixture was vigorously stirred under reflux
for 4 h. The organic phase was filtered through a silica plug,
which was then washed (TLC control) with CH2Cl2/AcOEt (3:1
by volume). After the combined organic solutions were evapo-
rated to dryness, the solid residue was dissolved in the
minimum amount of boiling CH2Cl2 and the warm solution
was treated with ether (100 mL; portionwise) and left at room
temperature for 2 h. Slightly yellowish fluffy needles of
spectroscopically and TLC pure dppmO were separated,
washed with ether, and dried under vacuum. The yield was
Or ga n ic Solven t. Both dichloromethane and 1,2-
dichloroethane are suitable for the biphasic reaction.
1
(40) (a) Grushin, V. V. Angew. Chem., Int. Ed. 1998, 37, 994. (b)
Flemming, J . P.; Pilon M. C.; Borbulevitch O. Ya.; Antipin M. Yu.;
Grushin, V. V. Inorg. Chim. Acta 1998, 280, 87. (c) Grushin, V. V.
Organometallics 2000, 19, 1888.
(41) (a) Due to considerably different Rf characteristics, brown-red
[(dppfc)PdI2] and orange-red [(dppfc)PdBr2] can be easily separated
by TLC or column chromatography. The dppfc Pd complex isolated from
the reaction mixture was identical to an authentic sample of [(dppfc)-
PdI2].41b (b) Brown, J . M.; Cooley, N. A. Organometallics 1990, 9, 353.
Colacot, T. J .; Fair, R. J ., J r.; Boyko, W. J . Phosphorus, Sulfur Silicon
Relat. Elem. 1999, 144-146, 49.
(42) (a) Lindsay, C. H.; Balch, A. L. Inorg. Chem. 1981, 20, 2267.
(b) Hunt, C. T.; Balch, A. L. Inorg. Chem. 1982, 21, 1641. (c) See also:
J anka, M.; Anderson, G. K.; Rath, N. P. Organometallics 2000, 19,
5071.
(43) Collman, J . P.; Hegedus, L. S.; Norton, J . R.; Finke, R. G.
Principles and Applications of Organotransition Metal Chemistry;
University Science Books: Mill Valley, CA, 1987. Crabtree, R. H. The
Organometallic Chemistry of the Transition Metals, 2nd ed.; J ohn Wiley
and Sons: New York, 1994. Spessard, G. O.; Miessler, G. L. Organo-
metallic Chemistry; Prentice Hall: Upper Saddle River, NJ , 1996.
(44) (a) Negishi, E.; Takahashi, T.; Akiyoshi, K. J . Chem. Soc. Chem.
Commun. 1986, 1338. (b) Amatore, C.; Azzabi, M.; J utand, A. J . Am.
Chem. Soc. 1991, 113, 8375. (c) Amatore, C.; J utand, A. J . Organomet.
Chem. 1999, 576, 254, and references therein.
1.53 g (74%). H NMR (CDCl3, 20 °C): δ 3.1 (d, 2H, J ) 12.7
Hz, CH2); 7.1-7.9 (m, 20H, Ph). 31P NMR (CDCl3, 20 °C): δ
-26.5 (d, 1P, J P-P ) 50.5 Hz, PPh2); 29.8 (d, 1P, J P-P ) 50.5
Hz, P(O)Ph2).
(b) A mixture of Pd(OAc)2 (19 mg; 8.5 × 10-2 mmol), dppm
(6.00 g; 15.6 mmol), 1,2-dichloroethane (20 mL), and 1,2-
dibromoethane (4.5 g; 23.9 mmol) was stirred for 20 min. After
a solution of NaOH (2.5 g) and NaI (0.12 g; 0.8 mmol) in water
(20 mL) was added the biphasic mixture was vigorously stirred
under reflux for 3 h 40 min. 31P NMR analysis of the organic
layer indicated 95% conversion of dppm. After the mixture was
kept under N2 at 5 °C overnight the organic layer solidified.
The aqueous layer was separated by a pipet and disposed of.
The solidified organic phase was then stirred with dichlo-
romethane (80 mL) until all solids dissolved. The solution was
put on a silica column, which was washed with CH2Cl2/AcOEt
(3:1 by volume). After the first fraction (Rf ≈ 1) containing
small amounts of unreacted dppm and dibromoethane was
separated, the second orange fraction containing dppmO was
collected and evaporated to dryness. The residue was thor-

Phosphine-Phosphine Oxide Ligands
Organometallics, Vol. 20, No. 18, 2001 3959
oughly washed with ether and then dried to give 4.97 g (80%)
of TLC and spectroscopically pure (NMR), albeit yellowish
dppmO. The dppmO was dissolved in ca. 20 mL of boiling CH2-
Cl2, treated with 200 mL of ether, and kept at -10 °C
overnight. The precipitated dppmO was separated, washed
with ether, and dried under vacuum. The yield after this
recrystallization was 4.68 g (75%).
(c) A mixture of palladium acetate (4 mg; 2 × 10-2 mmol),
dppm (0.65 g; 1.7 mmol), 1,2-dibromoethane (1.0 g; 5.3 mmol),
and 1,2-dichloroethane (5 mL) was stirred for 1 h. To this
solution was added aqueous NaOH (15 wt %; 4 mL) containing
NaI (20 mg; 0.14 mmol), and the mixture was vigorously
stirred under reflux for 4 h. After the standard workup (see
above) the yield of spectroscopically and TLC pure, yellowish
dppmO was 0.55 g (84%).
(d) A mixture of PtCl2 (5 mg; 1.9 × 10-2 mmol), dppm (1.00
g; 2.6 mmol), 1,2-dibromoethane (0.75 g; 4.0 mmol), and 1,2-
dichloroethane (6 mL) was stirred under reflux until the
platinum salt dissolved. To this solution was added aqueous
NaOH (15 wt %; 4 mL), and the mixture was vigorously stirred
under reflux for 12 h. After the standard workup (see above)
the yield of dppmO was 0.47 g (45%).
(b) Palladium acetate (10 mg; 4.45 × 10-2 mmol), dppp (4.00
g; 9.7 mmol), and 1,2-dibromoethane (2.75 g; 14.6 mmol) were
dissolved in 1,2-dichloroethane (15 mL). To this solution was
added aqueous NaOH (25 wt %; 10 mL), and the biphasic
mixture was vigorously stirred under reflux for 6 h until the
originally yellow mixture turned pale yellow. The reaction
mixture was worked up as described in the previous experi-
ment. The yield of dpppO was 3.17 g (76%).
P h 2P CH2CH2CH2CH2P (O)P P h 2, d p p bO. (a) A mixture of
palladium acetate (10 mg; 4.45 × 10-2 mmol), dppb (4.00 g;
9.4 mmol), 1,2-dibromoethane (3.6 g; 19.2 mmol), and dichlo-
romethane (10 mL) was stirred for 30 min. To this suspension
was added aqueous NaOH (20 wt %; 10 mL), and the mixture
was vigorously stirred at room temperature for 3 days.
Dichloromethane (30 mL) was added, and the organic phase
was filtered through a silica plug, which was then washed with
80 mL of CH2Cl2/AcOEt (5:3 by volume). The combined organic
solutions were evaporated to dryness. The solid residue was
dissolved in boiling CH2Cl2 (ca. 30 mL), and the solution was
treated with ether (100 mL; portionwise) and left at room
temperature for 2 h. Well-shaped, colorless crystals were
separated, washed with ether, and dried under vacuum. The
1
yield of dppbO was 3.33 g (80%). H NMR (CDCl3, 20 °C): δ
P h 2P CH2CH2P (O)P P h 2, d p p eO. (a) Palladium acetate (10
mg; 4.45 × 10-2 mmol), dppe (4.00 g; 10.05 mmol), and 1,2-
dibromoethane (2.85 g; 15.2 mmol) were dissolved in 1,2-
dichloroethane (30 mL). To this solution was added aqueous
NaOH (10 wt %; 20 mL), and the biphasic mixture was
vigorously stirred under reflux for 7 h until the originally
yellow mixture turned pale yellow or almost colorless. The
organic phase was filtered through a silica plug, which was
then washed with 60 mL of CH2Cl2/AcOEt (5:3 by volume).
The combined organic solutions were evaporated and treated
with ether, causing precipitation of white crystals of dppeO,
which were washed with ether and dried under vacuum. The
yield was 3.62 g (87%). 1H NMR (CDCl3, 20 °C): δ 2.3 (m, 4H,
CH2); 7.2-7.7 (m, 20H, Ph). 31P NMR (CDCl3, 20 °C): δ -11.5
(d, 1P, J P-P ) 48 Hz, PPh2); 32.3 (d, 1P, J P-P ) 48 Hz, P(O)-
Ph2).
(b) Palladium acetate (5 mg; 2.2 × 10-2 mmol), dppe (5.00
g; 12.6 mmol), and 1,2-dibromoethane (3.6 g; 19.2 mmol) were
dissolved in 1,2-dichloroethane (30 mL). To this solution was
added aqueous NaOH (12 wt %; 20 mL), and the biphasic
mixture was vigorously stirred under reflux for 15 h until the
originally yellow mixture turned pale yellow or almost color-
less. The product (4.29 g; 82%) was isolated as described above.
1.5 (m, 2H, CH2); 1.7 (m, 2H, CH2); 2.0 (m, 2H, CH2); 2.2 (m,
2H, CH2); 7.2-7.8 (m, 20H, Ph). 31P NMR (CDCl3, 20 °C): δ
-16.5 (s, 1P, PPh2); 32.2 (s, 1P, P(O)Ph2).
(b) A mixture of palladium acetate (10 mg; 4.45 × 10-2
mmol), dppb (4.00 g; 9.4 mmol), 1,2-dibromoethane (3.6 g; 19.2
mmol), and dichloromethane (10 mL) was stirred for 30 min.
To this suspension was added aqueous NaOH (20 wt %; 10
mL), and the mixture was vigorously stirred at room temper-
ature for 63 h. Water (5 mL) and dichloromethane (30 mL)
were added, and the mixture was stirred for a few minutes
until all solids dissolved. The organic phase was filtered
through a silica column, which was then washed with 80 mL
of CH2Cl2/AcOEt (5:3 by volume). The first portions of the
eluants containing unreacted dppb (190 mg) were separated
from the main fraction, which was collected and evaporated
to dryness. This residue was dissolved in ca. 40 mL of warm
CH2Cl2. Ether (100 mL) was added to the dichloromethane
solution in four portions of 20-30 mL each. After 3 h the white
crystals of dppbO were separated, wahsed with ether, and
dried under vacuum. The yield was 3.44 g (83%).
1,2-P h 2P C6H4P (O)P h 2, d p p bzO. (a) To a solution of dppbz
(0.501 g; 1.1 mmol), Pd(OAc)2 (2.5 mg; 1.1 × 10-2 mmol), and
1,2-dibromoethane (0.45 g; 2.4 mmol) in 1,2-dichloroethane (4
mL) was added a solution of NaOH (0.58 g; 14.5 mmol) in
water (1.5 mL). The mixture was vigorously stirred under
reflux for 7 h. Silica gel column separation of the organic phase
with CH2Cl2/AcOEt (2:1 by volume) gave 0.43 g (83%) of
spectroscopically pure dppbzO. 1H NMR (CD2Cl2, 20 °C): δ
7.1 (m); 7.4 (m); 7.5 (m); 7.7 (m). 31P NMR (CD2Cl2, 20 °C): δ
-12.8 (d, J P-P ) 14.8 Hz, 1P, PPh2); 30.7 (d, J P-P ) 14.8 Hz,
1P, P(O)Ph2).
(c) Palladium acetate (50 mg; 22.3 × 10-2 mmol), dppe (50.00
g; 125.6 mmol), and 1,2-dibromoethane (36 g; 191.5 mmol) were
dissolved in 1,2-dichloroethane (200 mL). To this solution was
added aqueous NaOH (20 wt %; 125 mL) and the biphasic
mixture was vigorously stirred under reflux for 21.5 h until
the originally yellow mixture turned pale yellow. The product
(40.75 g; 78.3%) was isolated as described above.
P h 2P CH2CH2CH2P (O)P P h 2, d p p p O. (a) Palladium ace-
tate (10 mg; 4.45 × 10-2 mmol), dppp (5.00 g; 12.1 mmol), and
1,2-dibromoethane (3.4 g; 18.9 mmol) were dissolved in 1,2-
dichloroethane (15 mL). To this solution was added aqueous
NaOH (20 wt %; 10 mL) and the biphasic mixture was
vigorously stirred under reflux for 6 h until the originally
yellow mixture turned pale yellow. The organic phase was
filtered through a silica column, which was then washed with
CH2Cl2/AcOEt (5:3 by volume) until the yellow band was about
to come out. The combined organic solutions were evaporated
to give a colorless oily residue. The oil was mixed with CH2-
Cl2 (2 mL) and ether (5 mL) and then treated with pentane
(10 mL) and left for 1.5 h. More pentane (50 mL) was added,
and in 2 h the fluffy white needles of dpppO were separated
and dried under vacuum. The yield was 3.80 g (73%). 1H NMR
(CDCl3, 20 °C): δ 1.8 (m, 2H, CH2); 2.2 (m, 2H, CH2); 2.4 (m,
2H, CH2); 7.2-7.7 (m, 20H, Ph). 31P NMR (CDCl3, 20 °C): δ
-17.2 (s, 1P, PPh2); 32.0 (s, 1P, P(O)Ph2).
(b) Repeating the reaction in the presence of NaI (5 mg)
produced, after 7 h, dppbzO in a similar yield of 81%. No effect
of the iodide on the reaction was observed.
P h 2P C5H4F eC5H4P (O)P P h 2, d p p fcO. (a) A solution of
NaOH (1.5 g; 37.5 mmol) and NaI (50 mg; 0.3 mmol) in water
(6 mL) was added to a solution of dppfc (2.05 g; 3.7 mmol),
Pd(OAc)2 (10 mg; 4.45 × 10-2 mmol), and 1,2-dibromoethane
(2.0 g; 10.6 mmol) in dichloromethane (8 mL). This mixture
was vigorously stirred under reflux (N2) for 23.5 h. At this
point, the organic layer contained dppfcO as the main product
and small amounts of dppfc, [(dppfc)PdI2], C5H5FeC5H4P(O)-
PPh2, and dppfcO2. The organic phase was separated and
evaporated. The dark residue was placed on a silica column,
which was washed first with CH2Cl2 and then with CH2Cl2/
AcOEt (5:3 by volume). Evaporation of the main orange
fraction gave dppfcO, which was washed with cold ether (2 ×
5 mL), pentane (2 × 5 mL), and dried under vacuum. The yield

3960 Organometallics, Vol. 20, No. 18, 2001
Grushin
1
was 1.36 g (65%). H NMR (CDCl3, 20 °C): δ 4.0 (m, 2H, Cp);
128.8 (d, J P-C ) 12.4 Hz, o-C6H5); 131.7 (d, J P-C ) 9.4 Hz,
4.2 (m, 2H, Cp); 4.4 (m, 2H, Cp); 4.6 (m, 2H, Cp); 7.1-7.7 (m,
20H, Ph). 13C NMR (CDCl3, 20 °C): δ 72.9 (s, C5H4); 73.0 (s,
C5H4); 73.1 (s, C5H4); 73.2 (s, C5H4); 73.3 (d, J P-C ) 1.7 Hz,
C5H4); 73.5 (d, J P-C ) 1.7 Hz, C5H4); 128.2 (d, J P-C ) 11.0 Hz,
m-C6H5PCp); 128.4 (d, J P-C ) 16.7 Hz, m-C6H5P(O)Cp); 131.3
(s, p-C6H5PCp); 131.4 (s, p-C6H5P(O)Cp); 131.4 (d, J P-C ) 9.7
Hz, o-C6H5PCp); 133.4 (d, J P-C ) 19.7 Hz, o-C6H5P(O)Cp);
134.5 (d, J P-C ) 105.6 Hz, q-C6H5PCp); 138.8 (d, J P-C ) 10.8
Hz, q-C6H5P(O)Cp). 31P NMR (CDCl3, 20 °C): δ -16.7 (s, 1P,
m-C6H5); 132.1 (d, J P-C ) 2.9 Hz, p-C6H5); 134.7 (d, J P-C
)
106.1 Hz, q-C6H5). 31P NMR (CD2Cl2, 20 °C): δ 28.4 (s).
(()-P h 2P C10H6C10H6P (O)P P h 2, (()-BINAP (O). A mixture
of PdI2 (3 mg; 8 × 10-3 mmol), (()-BINAP (0.10 g; 0.16 mmol),
and dichloromethane (6 mL) was stirred at room temperature
for 0.5 h and then under reflux for 0.5 h until all PdI2 dissolved
to produce a purple-red solution. Additional (()-BINAP (0.50
g; 0.81 mmol) was added, followed by a solution of NaOH (0.31
g; 7.8 mmol) in water (3 mL). After vigorous stirring for 30
min, 1,2-dibromoethane (0.60 g; 3.2 mmol) was added. The
resulting mixture was stirred under reflux for 10 h. At that
point, the mixture consisted of a yellow organic phase, colorless
aqueous phase, and solid unreacted (()-BINAP. 31P NMR
analysis of the organic phase indicated the presence of (()-
BINAP(O) and (()-BINAP in ca. 7:3 ratio, and no (()-BINAP-
dioxide. The content of the NMR tube was shaken with more
PdI2 (finely ground; 3 mg; 0.8 × 10-3 mmol) until the latter
dissolved. The resulting solution was added to the reaction
flask along with 4 mL of dichloromethane. After stirring under
reflux overnight, addition of extra 1,2-dibromoethane (0.60 g;
3.2 mmol) and then stirring under reflux for an additional 24
h, the (()-BINAP had all dissolved. The 31P NMR spectrum of
the organic phase indicated ca. 85% conversion of (()-BINAP
to (()-BINAP(O) at >95% selectivity. Silica column separation
of the organic phase with first pure CH2Cl2 and then CH2Cl2/
AcOEt (3:1 v/v) afforded 0.436 g (71%) of spectroscopically pure
(()-BINAP(O). 1H NMR (CD2Cl2, 20 °C): δ 6.8-7.9 (m). 31P
NMR (CD2Cl2, 20 °C): δ -14.5 (s); 27.9 (s).27h
(S)-P h 2P C10H6C10H6P (O)P P h 2, (S)-BINAP (O), a n d (R)-
P h 2P C10H6C10H6P (O)P P h 2, (R)-BINAP (O). (a) Palladium
acetate (4 mg; 1.8 × 10-2 mmol), (S)- or (R)-BINAP (0.80 g;
1.29 mmol), and 1,2-dibromoethane (1.0 g; 5.3 mmol) were
dissolved in dichloromethane (5 mL). Upon addition of aqueous
NaOH (4.5 wt %; 6 mL) to this solution, the organic phase
rapidly turned cherry-red. After stirring the mixture at room
temperature for 2 days the conversion reached was ca. 60-
70%. More aqueous alkali (0.4 g of NaOH in 1.5 mL of water)
was added, and the mixture was stirred under reflux for 23 h.
At this point, only small amounts of the unreacted BINAP
were detected in the organic phase (TLC). The dichloro-
methane layer was filtered through a short silica plug, which
was then washed with CH2Cl2/AcOEt (10:1 v/v). The combined
organic solutions were evaporated. The remaining solid was
redissolved in dichloromethane and reduced in volume until
a thick oil was obtained. This oily residue was treated with
ether (30 mL) and left at -17 °C overnight. Large, colorless
crystals of BINAP(O) were separated, washed with ether, and
dried under vacuum. The yield was 0.660 g (80%). This
procedure is not applicable to (()-BINAP and may be poorly
efficient for the oxidation of enantiomerically impure samples
of (S)-BINAP and (R)-BINAP. The use of PdI2 instead of Pd-
(OAc)2 (see below) gives consistently high yields of BINAP(O)
independently of enantiomeric purity of BINAP used.45
(b) A mixture of PdI2 (5 mg; 1.4 × 10-2 mmol), (S)-BINAP
(0.50 g; 0.8 mmol), and dichloromethane (6 mL) was stirred
at room temperature for 2 h until all PdI2 dissolved to produce
a purple-red solution. A solution of NaOH (0.47 g; 11.8 mmol)
in water (3 mL) and 1,2-dibromoethane (1.00 g; 5.3 mmol) were
added. The resulting mixture was stirred under reflux for 8
h, until the originally cherry-red solution turned light yellow.
31P NMR analysis of the organic phase indicated 90% conver-
sion of the BINAP to BINAP(O) at >95% selectivity. Silica
column separation of the organic phase with first pure CH2-
Cl2 and then CH2Cl2/AcOEt (10:1 v/v) afforded 0.427 g (83%)
of spectroscopically pure (S)-BINAP(O) as a yellow solid. The
product was dissolved in 2 mL of dichloromethane, and the
resulting solution was reduced in volume to ca. 0.5-1 mL by
blowing nitrogen over its surface. Ether (ca. 3 mL) was added
to the resulting viscous solution, followed by hexane (ca. 5 mL).
After several hours, well-shaped pale yellow crystals of (S)-
CpPPh2); 28.5 (s, 1P, CpP(O)Ph2). Anal. Calcd for C68H58
-
Fe2O3P4 (C34H28FeOP2‚0.5H2O): C, 70.5; H, 5.0. Found: C,
70.7; H, 5.0. This procedure was repeated several times on a
different scale to produce dppfcO in 35-63% yield. The use of
PdI2 instead of Pd(OAc)2/NaI (see below) resulted in consis-
tently better yields (51-65%).45
(b) A mixture of dppfc (1.00 g; 1.8 mmol), CH2Cl2 (6 mL),
and PdI2 (11 mg; 3.1 × 10-2 mmol) was stirred at room
temperature until the Pd salt dissolved (2 h). To the resulting
brown-red solution were added aqueous NaOH (25%; 4 g) and
1,2-dibromoethane (1.00 g; 5.32 mmol), and the mixture was
stirred under reflux for 23 h. 31P NMR analysis of the organic
phase indicated ca. 80% conversion of dppfc to dppfcO (85%)
and dppfcO2 (15%). Silica gel column separation of the organic
phase with first pure CH2Cl2 and then CH2Cl2/AcOEt (3:1 v/v)
gave 0.23 g (23%) of unreacted dppfc and 0.67 g (65%; 85% on
reacted dppfc) of spectroscopically pure dppfcO.
(c) A mixture of dppfc (0.12 g; 0.22 mmol), CH2Cl2 (6 mL),
and PdI2 (6.5 mg; 1.8 × 10-2 mmol) was stirred at room
temperature until the Pd salt dissolved (3 h). To the resulting
brown-red solution were added dppfc (0.88 g; 1.59 mmol) and
aqueous NaOH (25%; 4 g). After 10 min of vigorous stirring,
1,2-dibromoethane (1.00 g; 5.32 mmol) was added, and the
mixture was stirred under reflux for 30 h. 31P NMR analysis
of the organic phase indicated ca. 80% conversion of dppfc.
Silica gel column separation of the organic phase with first
pure CH2Cl2 and then CH2Cl2/AcOEt (3:1 v/v) gave 0.18 g (18%)
of unreacted dppfc and 0.52 g (51%; 62% on reacted dppfc) of
spectroscopically pure dppfcO.
P h 2P (O)C5H4F eC5H4P (O)P P h 2, d p p fcO2. A solution of
NaOH (2.2 g; 55 mmol) in water (8 mL) was added to a solution
of dppfc (2.00 g; 3.6 mmol), Pd(OAc)2 (10 mg; 4.45 × 10-2
mmol), and 1,2-dibromoethane (4.0 g; 21.3 mmol) in 1,2-
dichloroethane (8 mL). After 9.5 h of vigorous stirring under
reflux (N2) 100% conversion of the dppfc was reached (31P
NMR). Dichloromethane (200 mL) was added, at stirring, to
the mixture at room temperature. The organic phase was
filtered through a short silica plug, then through Celite, and
evaporated. The solid residue was recrystallized by addition
of ether (20 mL) to its solution in 20 mL of warm 1,2-
dichloroethane. Well-shaped orange crystals of dppfcO2 were
washed with ether and dried under vacuum. The yield of
spectroscopically pure dppfcO2 was 1.20 g (57%). 1H NMR (CD2-
Cl2, 20 °C): δ 4.3 (dd, 4H, J ) 3.8 and 1.9 Hz, 3,3′,4,4′-Cp);
4.7 (dd, 4H, J ) 3.8 and 1.9 Hz, 2,2′,5,5′-Cp); 7.3-7.8 (m, 20H,
Ph). 13C NMR (CD2Cl2, 20 °C): δ 73.9 (d, J C-P ) 12.4 Hz, C5H4);
74.4 (d, J C-P ) 10.2 Hz, C5H4); 75.0 (d, J C-P ) 114.8 Hz, C5H4);
(45) It is not clear why unsatisfactory reproducibility was occasion-
ally observed when Pd(OAc)2 (or Pd(OAc)2/NaI) rather than PdI2 was
used as catalyst for the oxidation of dppfc and BINAP. The latter two
do not form stable complexes of the type [(L-L)2Pd]2+ and hence
undergo oxidation via the mechanism presented in Scheme 2 (see
above). No reproducibility problem was encountered for the other bis-
phosphines whose oxidation is governed by a different mechanism
(Scheme 1). We cautiously propose that AcO- might have a negative
effect on the catalytic oxidation of dppfc and BINAP. Some impurities
in the materials used for the reaction might somehow scavenge the
acetate anion from Pd(OAc)2 catalyst. The Pd(OAc)2-catalyzed oxidation
reactions of dppfc and BINAP were highly reproducible when run in
glass flasks that had been washed with K2Cr2O7/H2SO4. Although the
glassware was thoroughly rinsed with water after the washings, Cr3+
adsorbed on the glass surface may have trapped the AcO- via
chemisorption.

Phosphine-Phosphine Oxide Ligands
Organometallics, Vol. 20, No. 18, 2001 3961
BINAP(O) were separated, washed with ether, and dried under
vacuum. The yield of the recrystallized product was 0.408 g
(80%).
NMR spectra of the sample were recorded every 10-15 min.
As the reaction occurred, the signal at 30.0 ppm ([(dppe)2
Pd])
gradually diminished in intensity while being replaced by a
singlet at 54.3 ppm ([(dppe)2Pd]2+). After 2 h full conversion
of [(dppe)2Pd] to [(dppe)2Pd]Br2 with >99% selectivity was
observed. The well-shaped crystals formed in the NMR tube
were found to be identical with an authentic sample of
[(dppe)2Pd]Br2.
(c) A mixture of PdI2 (6.5 mg; 1.8 × 10-2 mmol), (R)-BINAP
(0.505 g; 0.8 mmol), and dichloromethane (5 mL) was stirred
at room temperature for 3 h until all PdI2 dissolved to produce
a purple-red solution. A solution of NaOH (0.45 g; 11.3 mmol)
in water (3 mL) and 1,2-dibromoethane (1.00 g; 5.3 mmol) were
added. The resulting mixture was stirred under reflux until
the originally cherry-red solution turned light yellow (10 h).
31P NMR analysis of the organic phase indicated 90% conver-
sion of the BINAP to BINAP(O) at >95% selectivity. The
mixture was treated with ca. 20% H3PO4 until the aqueous
phase turned acidic (pH ) 4). Then, dppe (20 mg) was added
in order to convert the Pd(II) BINAP complexes present in the
(b) A mixture of PdBr2 (21 mg), (S)-BINAP (50 mg; 1 equiv),
and CH2Cl2 (2 mL) was stirred at room temperature for 2.5 h
until the Pd complex dissolved to produce an orange solution.
A 31P NMR spectrum of this solution displayed only one singlet
resonance at 26.5 ppm ([((S)-BINAP)PdBr2]). After an ad-
ditional 2 equiv of (S)-BINAP (100 mg) was added, the solution
was analyzed by 31P NMR again. The spectrum contained only
two sharp singlet resonances, at 26.5 ppm from [((S)-BINAP)-
PdBr2] and -14.8 ppm from free (S)-BINAP, in a 1:2 intensity
ratio. Therefore, no NMR-observable reaction between [((S)-
BINAP)PdBr2] and (S)-BINAP took place. The orange solution
of [((S)-BINAP)PdBr2] (1 equiv) and (S)-BINAP (2 equiv) was
vigorously stirred with 10% aqueous NaOH for 30 min. The
color of the organic phase changed to dark cherry-red due to
the formation of a Pd(0) BINAP complex, while the aqueous
layer remained colorless. The 31P NMR spectrum of the dark
red organic phase contained four singlet resonances at 28.8
(1P), 27.4 (2P), -14.5 (1P), and -14.6 (1P) ppm and an AB
quartet with δ ) 25.4 (0.5 P) and 28.6 (0.5P) ppm (J P-P ) 59
Hz). The singlet at -14.6 ppm was assigned to free (S)-BINAP,
whereas singlets at -14.5 and 28.4 ppm were from the Ph2P
and Ph2P(O) groups of (S)-BINAP(O) formed.27h Taking into
account the solution behavior of [(η2-BINAP)2Pd],30b,c,37 the
singlet at 27.4 ppm and the two doublets are tentatively
assigned to [(η2-BINAP)(η1-BINAP)Pd] and [(η2-BINAP)(η1-
BINAP(O))Pd]. The former would give rise to the signal at 28.4
ppm appearing as a singlet30c due to fast η2-BINAP/η1-BINAP
exchange, whereas the latter might exhibit two doublets from
η1-coordinated BINAP(O) and a singlet from η2-BINAP, which
overlaps with the resonance from [(η2-BINAP)(η1-BINAP)Pd].
The dichloromethane solution in the NMR tube was treated
with excess 1,2-dibromoethane (30 mg) and left at room
temperature for 2 h. This caused a color change from dark
cherry-red to orange. 31P NMR analysis of the sample indicated
the presence of only BINAP(O) (-14.5 and 28.4 ppm), [((S)-
BINAP)PdBr2] (26.5 ppm), and (S)-BINAP (-14.6 ppm) (see
Scheme 2).
organic phase to more easily removable cationic [(dppe)2Pd]2+
.
After 5 min of stirring the organic phase was separated and
placed on a silica column, which was then washed first with
CH2Cl2 to remove the small amounts of unreacted BINAP and
dppe and then with CH2Cl2/AcOEt (10:1 v/v) to isolate the
monoxide. This separation afforded 0.419 g (81%) of spectro-
scopically (NMR) and chromatographically (TLC) pure (R)-
BINAP(O) as a colorless crystalline solid. The product can be
recrystallized from dichloromethane-ether-hexane as de-
scribed above.
Stoich iom etr ic Stu d ies. (a) To a solution of Pd(OAc)2 (30
mg) in 1,2-dichloroethane (5 mL) was added dppe (213 mg;
4-fold excess), and the resulting solution was immediately
analyzed by 31P NMR. The NMR spectrum indicated quantita-
tive formation of [(dppe)2Pd]2+ (AcO-)2 (δ ) 55.7 ppm).31 After
a few minutes the cationic complex [(dppe)2Pd]2+ (AcO-)2 began
to precipitate out. Water (5 mL) was added, and the mixture
was stirred for 5 min to produce two clear phases. 31P NMR
analysis of both layers indicated that most of the cationic
complex was located in the aqueous layer (δ ) 57.8 ppm) with
only trace amounts of [(dppe)2Pd]2+ remaining in the organic
phase. A solution of NaOH (100 mg) in water (2 mL) was added
at stirring. The organic layer of the biphasic system turned
bright yellow within the time of mixing, while the aqueous
phase remained colorless. After 1 min of vigorous stirring both
layers were analyzed by 31P NMR spectroscopy. No signals
were found in the spectrum of the colorless aqueous phase.
The 31P NMR spectrum of the yellow organic phase contained
four resonances: a singlet at -12.8 ppm (excess dppe), two
doublets at -12.4 and 30.8 ppm with J P-P ) 48.0 Hz (dppeO),
and a singlet at 29.9 ppm ([(dppe)2Pd]). No other signals were
observed, indicating >99% selectivity of the reaction. A sample
of the yellow organic phase (0.8 mL) was carefully separated
by a pipet, filtered through cotton, and placed in a standard 5
mm NMR tube. To this sample was added 1,2-dibromoethane
(ca. 50 mg). The tube was placed in the NMR probe, and 31P
Ack n ow led gm en t. Wilfrid Laurier University and
DuPont are acknowledged for support.
OM010454K