H. Byrd et al. / Inorganica Chimica Acta 338 (2002) 240Á
/244
241
tribromophenyl)phosphate, (TBPO)3PO, has no active
chlorides. However, we have recently demonstrated that
the only one of these compounds that can be cleanly
prepared is (TBPO)PCl4 [12]. The byproduct of non-
oxidative chlorination reactions with (TBPO)PCl4 is
dichloro(2,4,6-tribromophenyl)phosphate,
(TBPO)P(O)Cl2, and this material has active chlorides
that can react with the alcohols, thiols or amines that are
used in subsequent reactions. This means that large
excesses of these reagents must be used, and this
complicates the workup.
In this paper, we report that the reaction of
(TBPO)PCl4 with o-catechol yields a phosphorane that
can carry out nonoxidative chlorinations of secondary
phosphonates and whose byproduct does not contain
active chlorines. The results of our 31P[1H] NMR studies
of synthesis of this new phosphorane and of its reactions
with secondary phosphonates are reported in this paper.
2.3. Synthesis of dichloro(2,4,6-tribromophenoxy)(1,2-
diphenoxy)phosphorane (TBPO)(DP)PCl2
Typically, 1.14 g (5.48 mmol) of PCl5 and 1.85 g (5.48
mmol) of 2,4,6 tribromophenol was added to a 100 ml
Schlenk flask. To this flask, canula transfer added 20 ml
of dry benzene and N2 was bubbled through the solution
for one hr to remove the HCl byproduct. A solution of
0.610 g (5.48 mmol) of catechol in 30 ml of dry benzene
was then added to the PCl5ÁTBPO mixture by canula
/
transfer. Dry N2 was blown through the solution for 1 h
to remove the HCl byproduct.
2.4. Synthesis of di-n-butyl chlorophosphite
(PCl(OBut)2)
A benzene solution of (TBPO)(DP)PCl2 was prepared
as described above from 9.79 g (47 mmol) of PCl5, 5.54 g
(47 mmol) of 2,4,6 tribromophenol and 5.22 g (47 mmol)
of catechol in 150 ml of dry benzene. To this solution,
8.1 ml (47 mmol) of di-n-butyl chlorophosphite was
added via a 10.0 ml syringe. This reaction gave a
quantitative yield of the di-n-butyl chlorophosphite by
31P[1H] NMR spectroscopy.
2. Experimental
The product was isolated from the reaction mixture as
follows. The reaction was first evaporated to dryness
under aspirator vacuum, and then treated with 100 ml
dry hexane. This mixture was stirred overnight and then
filtered under N2 to remove the insoluble phosphate
byproduct. The hexanes were removed from the filtrate
under aspirator vacuum and the residual liquid was
distilled (73 8C, 0.2 mm Hg) to yield 4.53 g (50%) of the
pure di-n-butyl chlorophosphite.
2.1. Materials and methods
Benzene was dried by refluxing over calcium hydride
for at least 12 h followed by distillation. Chloroform-d
was dried over activated molecular sieves. Very dry
2,4,6-tribromophenol was obtained by drying the mate-
rial under vacuum in a desiccator containing activated
CaSO4 and P2O5 for at least 5 days. Phosphorus
pentachloride was used as received and was handled
exclusively in a Vacuum Atmospheres glove box under a
dry nitrogen atmosphere. Catechol was recrystallized
from dry benzene and was stored and handled exclu-
sively in a Vacuum Atmospheres glove box under a dry
nitrogen atmosphere. The Mo(CO)4(NBD) complex was
prepared by literature methods [13]. All materials were
handled either in the glove box or by standard Schlenk
techniques under a nitrogen atmosphere.
2.5. In situ reaction of (TBPO)(DP)PCl2 with alkyl or
aryl-phosphites and Mo(CO)4(NBD)
A benzene solution of (TBPO)(DP)PCl2 was prepared
as described above in the first section. To this solution,
5.48 mmol of the appropriate phosphite was added via a
1.0 ml syringe. This reaction produces the chloropho-
sphite as observed by 31P[1H] NMR spectroscopy. After
24 h, 0.822 g (2.74 mmol) of Mo(CO)4(NBD) was added
to the Schlenk flask under a vigorous N2 flow. The
31P[1H] NMR spectrum of this mixture indicated the
reaction is quantitative and complete after 1 h.
2.2. Characterization methods
All 31P[1H] NMR spectra were recorded on a Bruker
ARX-300 NMR spectrometer. The 31P[1H] NMR spec-
tra were referenced to external H3PO4 in a coaxial tube
containing chloroform-d1, and chemical shifts downfield
of H3PO4 are reported as positive. A sweep width of
50 000 Hz was used, and 32 768 data points were taken
in this window. From 16 to 32 scans with 308 inverse-
gated decoupling pulse sequences and pulse delays of 30
s were used to obtain quantitative spectra. The resolu-
tion of each spectrum was 1.5 Hz or 0.01 ppm.
3. Results and discussion
3.1. 31P[1H] NMR studies of the reactions of phosphorus
pentachloride, 2,4,6-tribromophenol and catechol
We have demonstrated in a previous paper [12] that
the reactions of phosphorus pentachloride and 2,4,6-
tribromophenol at various stoichiometries produce