reactions, including oxygen atom insertion into the aryl P-C
bond to form phosphinate esters and, in a few cases, arene
epoxidation.5,2b,c Several theoretical studies suggest that
phosphadioxiranes should be electrophilic oxidants.3 How-
ever, the scope of oxygen atom transfer reactions from
phosphadioxirane as well as the regioselectivity and effects
of temperature and solvent are not known. For example, the
photooxidation of arylphosphines with different aryl groups
attached to the phosphorus atom has not been studied; i.e.,
it is not known if the formation of phosphinate esters is
regioselective. Furthermore, there have been no reports of
photooxidation of several other important classes of phos-
phines such as phosphines bearing biaryl ligands. Intramo-
lecular oxygen atom insertion into the P-C bond and arene
epoxidation might ultimately be useful reactions for the
functionalization of such phosphines, if the scope and
regioselectivity of these reactions were known, and if the
competing intermolecular oxygen atom transfer to form
phosphine oxide could be limited. In this paper, we present
a detailed investigation of product distribution, regioselec-
tivity, solvent, and temperature effects in the reactions of
phosphadioxiranes generated from singlet oxygen and a
variety of different phosphines. We show that oxygen atom
insertion in mixed (i.e., bearing different aryl and/or alkyl
groups) phosphines is, in fact, regioselective. We also
investigate how solvent and temperature effects can be used
to maximize the yield of either phosphinate ester or arene
epoxide.
Photooxidation of arylphosphines bearing large substitu-
ents in the ortho position of the aryl ligand predominately
yield phosphinate esters from intramolecular oxygen atom
insertion into the P-C bond.2b Arylphosphines without bulky
substituents give mainly phosphine oxide upon reaction with
singlet oxygen. The intermediate phosphadioxirane is pre-
sumed to be an electrophilic oxidant, but there is little
experimental evidence, as phosphines bearing different aryl
ligands have not been studied. We have therefore begun our
study of the photooxidation of mixed phosphines by inves-
tigating the reaction of bis(o-methoxyphenyl)phenylphos-
phine (1) with singlet oxygen (Scheme 1). Photooxidation
o-methoxyphenylphosphinate (2), and bis(o-methoxyphe-
nyl)phenylphosphine oxide (3).
1
Both products were characterized by H and 31P NMR,
and their identities were confirmed by X-ray molecular
structures (see the Supporting Information). The insertion
reaction leading to formation of 2 is completely regioselec-
tive. We did not observe any oxygen-atom insertion into the
P-C bond bearing the unsubstituted phenyl ring. Further-
more, unlike for binaphthylphosphines,2c we did not observe
any hydroxylation of the aromatic ring.
Dialkylbiarylphosphines have been widely used in Pd-
catalyzed cross-coupling reactions, in part because they are
rather resistant toward oxidation.6,7 Buchwald et al. have
recently suggested that the biaryl ligand as well as substit-
uents on the 2′ and 6′ positions of the biaryl ring play a key
role in the susceptibilitysor lack thereofstoward reaction
with triplet oxygen.6 The reactivity of these phosphines with
singlet oxygen has not been studied to date. A priori, a wide
range of possible products can be envisioned, i.e., oxygen
atom insertion into the P-C bond of either the alkyl or biaryl
ligand, formation of arene epoxides, and simple formation
of phosphine oxide by intermolecular oxygen atom transfer.
Of particular interest is the question of whether alkyl or aryl
group migration is preferred in these compounds; alkyl
migration in a phosphadioxirane has not previously been
reported.
The reaction of unsubstituted 2-di-tert-butylbiphenylphos-
phine (4) with singlet dioxygen in deuterated toluene at room
temperature yields the corresponding phosphine oxide (5)
as the sole product (see Table 1). No epoxidation or insertion
Table 1. Product Distribution and Rates of Singlet Oxygen
Removal (kT) for Photooxidation of Dialkylbiarylphosphines
phosphine
oxideb
(%)
arene
epoxideb
(%)
starting
phosphine
kT × 10-7
phosphinateb
(%)
a
(M-1 s-1
)
4
6
10
1.2 ( 0.1
0.81 ( 0.05
0.63 ( 0.05
2.1 ( 0.1
100 ( 1
13 ( 0.3
65 ( 4
0
0
0
84 ( 0.1
13 ( 5
0
2 ( 0.3
22 ( 4
100 ( 1
14
a kT values are averages of three runs; error is one standard deviation.
b All reactions were performed in an NMR tube using 1 mL solutions of
biphenylphosphine (24 mM) in toluene-d8, concentration of TPP sensitizer
) 0.13 mM. Relative amounts of products were calculated (average of three
runs, error is one standard deviation) using 31P NMR integrations.
Scheme 1
.
Reaction of Bis(o-methoxyphenyl)phenylphosphine
(1) with Singlet Oxygen
products (i.e., via alkyl or aryl migration) were observed.
Even at high temperature where aryl migration may be more
favorable (see below), the reaction of singlet oxygen with
biphenylphosphine 4 in toluene yields phosphine oxide 5 as
the only product.
In marked contrast to the unsubstituted 2-di-tert-butylbi-
phenylphosphine 4, the bulky highly substituted di-tert-
butyltriisopropylbiphenylphosphine 6 reacts with singlet
of 1 at room temperature (solvent ) toluene-d8, sensitizer
) tetraphenylporphyrin (TPP), phosphine concentration )
1-20 mM, tungsten-halogen lamp, cutoff filter at 493 nm)
leads to only two products, namely (o-methoxyphenyl)phenyl
(6) Barder, T. E.; Buchwald, S. L. J. Am. Chem. Soc. 2007, 129, 5096
(7) Martin, R.; Buchwald, S. L. Acc. Chem. Res. 2008, 41, 1461, and
references cited therein
.
(5) Tsuji, S.; Kondo, M.; Ishiguro, K.; Sawaki, Y. J. Org. Chem. 1993,
58, 5055
.
.
Org. Lett., Vol. 12, No. 13, 2010
3101