Lopin et al.
TABLE 1. Op tim iza tion of th e Rea ction Con d ition s in th e Ca se of P r ecu r sor 19a a n d n -Oct-1-en e (27)
entry procedure n-octene 27 (equiv) 19a (equiv)
hydrogen donora (equiv)
AIBN (equiv) addition time (h) yieldsb (%)
1
2
3
4
A
B
C
D
1
1
1
3
2
3
1
n-Bu3SnH (4) or TTMSSH (4)
TTMSSH (3)
n-Bu3SnH (4)
0.5
0.25
0.5
8
2
0.2
8
60
59
50
20
10
n-Bu3SnH (1)
0.5
a
b
TTMSSH: tris(trimethylsilyl)silane. Isolated yields.
Resu lts a n d Discu ssion
precursor 19a and difluoroalkenes in the presence of a
radical initiator, Motherwell and his collaborators report
a procedure calling for the use of 0.5, 1, 3, and 4 equiv of
azobisisobutyronitrile (AIBN), alkene, 19a , and tri(n-
butyl)tin hydride (n-Bu3SnH), respectively, and an ad-
dition time of 8-10 h (Table 1, entry 1).11c In view of both
the probable higher reactivity of the unfluorinated al-
kenes considered in the present study and the purifica-
tion problems associated with the use of tin compounds,
a somewhat different procedure was worked out with 19a
and n-oct-1-ene (27). Replacement of tris(n-butyl)tin
hydride with tris(trimethylsilyl)silane (TTMSSH),20 and
decreasing the amount of AIBN to 0.25 equiv did not
induce any change in the course of the reaction. The
amount of precursor was also decreased, as was the
addition time of the TTMSSH/AIBN solution. Thus, it
was found that slow addition of 3 equiv of TTMSSH and
0.25 equiv of AIBN over a period of 2 h resulted in the
formation of the desired product in essentially the same
yield as with the original procedure (Table 1, entry 2).
Further diminishing the addition time resulted in a drop
of the yields (Table 1, entry 3). Conditions calling for a
large excess of alkene also produced a deleterious effect
on the course of the reaction (entry 4). Similar results
were obtained with precursors 20a , 21b, and 22b.
The eight precursors were then engaged in reactions
with n-oct-1-ene (27), methylenecyclohexane (28), phenyl
acrylate (29), and n-butyl vinyl ether (30). The results
are compiled in Table 2.
P r ep a r a tion of Ra d ica l P r ecu r sor s 19-22. In
analogy to the reported preparation of 19a , we used an
approach relying on an Arbuzov process on trialkyl
phosphites and O,S,S-trialkyl dithiophosphites. Thus,
triethyl phosphite was reacted with benzeneselanyl
chloride in toluene at 0 °C to afford precursor 19a (77%
yield).14 2-Chloro-1,3,2-phosphinane (24a )15 and 2-chloro-
1,3,2-dithiaphosphinane (24b)16 were quantitatively trans-
formed into the 2-benzyloxy derivatives 25a and 25b,
respectively, by interaction with benzyl alcohols in diethyl
ether, under basic conditions (procedures of Banwarth
and Holmes, respectively).17 To avoid any oxidation of the
phosphorus(III), these compounds were engaged without
further purification in the next step. Not unexpectedly,
benzeneselanyl chloride was found to selectively and
efficiently react on the benzyloxy unit to deliver seleno-
phosphate 19b and selenophosphorodithioate 21b in 76
and 70% isolated yields. Precursor 21a was generated
in a somewhat different manner. Phosphorus trichloride
was sequentially reacted with benzyl alcohol and ethaneth-
iol (2 equiv) under basic conditions. Direct treatment of
the thereby isolated and highly oxygen-sensitive O-
benzyl-S,S-diethyl dithiophophite (26) with benzenese-
lanyl chloride delivered 21a in 54% yield.18 The PdO
bond in all four precursors 19a , 19b, 21a , and 21b were
then converted into a PdS bond by reaction with 1 equiv
of Lawesson’s reagent in refluxing toluene (19a , 19b) or
refluxing CH2Cl2 (21a , 21b); isolated yields were found
to be 72, 58, 56, and 86%, respectively.19
In all cases but one, the selanylated precursors were
found to be totally consumed. Precursor 19b constantly
led to 60-65% consumption, thus demonstrating that the
homolytic cleavage is, in this particular case, a slower
process (see below). The facile homolytic cleavage of the
P-Se bond in the seven other precursors is in line with
the available literature data on the analogous C-Se
bond.21 The expected adducts, when formed, were easily
isolated by chomatography on silica in yields ranging
from 15 to 99%; complete regioselectivity was ob-
served in all cases. Byproducts included n-Bu3SnSePh
or (Me3Si)3SiSePh, and the corresponding phosphites 6
or thiophosphites 7 resulted from partial, competitive
hydrogen quenching of the phosphorus-centered radicals
Ad d ition s Rea ction s of Ra d ica ls 10-13 on to Al-
k en es. In his paper describing the interaction between
(13) For an example of a radical addition on a SdP bond, see:
Romeo, R.; Wozniak, L. A.; Chatgilialoglu, C. Tetrahedron Lett. 2000,
41, 9899-9902.
(14) Reference 11a reports the preparation of compound 19a from
triethyl phosphite and benzeneselanyl bromide. We satisfactorily used
the commercially available analogous chloride to achieve this trans-
formation; see the Experimental Section.
(15) Arbuzov, A. E.; Zoroastrova, V. H. Izv. Akad. Nauk. SSSR, Otd.
Khim. Nauk. 1952, 6, 770-778.
(16) This compound has been described in the past: (a) See refs 8b
and 14. (b) Martin, J .; Robert, J . B.; Taieb, C. J . Phys. Chem. 1976,
80, 2417-2421. However, we found these procedures nonreproducible
and yielding the desired compound in yields <35%. In the end,
dithiaphosphinane 24b was prepared in 45% yield by using the
conditions described for the corresponding phospholane: (c) Martin,
S. F.; Wagman, A. S. J . Org. Chem. 1996, 61, 8016-8023. See the
Experimental Section.
1
(10-100% in the H and 31P NMR spectra).22 The corre-
sponding dithiophosphites and trithiophosphites were not
observed, due to their propensity to disproportionate, to
dimerize and oligomerize, and to the low solubility of
these oligomers.8,10,23 The lower yields observed with
phenyl acrylate may more reflect a competitive polym-
(17) (a) Bannwarth, W.; Trzeciak, A. Helv. Chim. Acta 1987, 70,
175-186. (b) Swamy, K. C. K.; Holmes, J . M.; Day, R. O.; Holmes, R.
R. J . Am. Chem. Soc. 1990, 112, 6092-6094.
(18) Compound 26 was isolated by evaporation of the volatiles and
used without further purification. 1H and 31P NMR data indicated a
purity of 92-95%.
(19) (a) Pedersen, B. S.; Scheibye, S.; Nilsson, N. H.; Lawesson, S.
O. Bull. Soc. Chim. Belg. 1978, 87, 223-228. (b) Scheibye, S.; Pedersen,
B. S.; Lawesson, S. O. Bull. Soc. Chim. Belg. 1978, 87, 229-234. (c)
Horner, L.; Lindel, H. Phosphorus Sulfur 1982, 12, 259-261. (d)
Piettre, S. R.; Raboisson, P. Tetrahedron Lett. 1996, 37, 2229-2232.
(e) Piettre, S. R. Tetrahedron Lett. 1996, 37, 4707-4710.
(20) Ballestri, M.; Chatgilialoglu, C.; Clark, K. B.; Griller, D.; Giese,
B.; Kopping, B. J . Org. Chem. 1991, 56, 678-683.
(21) Although this particular point is not discussed in ref 11b,c, one
can assume that the authors observed a total consumption of precursor
19a .
(22) After evaporation of the volatiles; at least 2 equiv of precursor
was used in each reaction; see the Experimental Section.
9918 J . Org. Chem., Vol. 68, No. 26, 2003