Diethylphosphine oxide (DEPO): High-yielding and facile preparation of indolones in water

Article abstract of DOI:10.1021/ol035173i

(Matrix presented) Indolones are prepared in excellent yield at 80°C in water by radical reaction (aryl radical formation, hydrogen atom abstraction, cyclization, and rearomatization) mediated by the reagent diethylphosphine oxide (DEPO). The reaction fea

Full text of DOI:10.1021/ol035173i

ORGANIC
LETTERS
2003
Vol. 5, No. 16
2971-2974
Diethylphosphine Oxide (DEPO):
High-Yielding and Facile Preparation of
Indolones in Water
Tanweer A. Khan, Re´gis Tripoli, James J. Crawford, Concepcion G. Martin, and
John A. Murphy*
Department of Pure and Applied Chemistry, UniVersity of Strathclyde,
295 Cathedral Street, Glasgow G1 1XL, U.K.
john.murphy@strath.ac.uk
Received June 24, 2003
ABSTRACT
Indolones are prepared in excellent yield at 80 °C in water by radical reaction (aryl radical formation, hydrogen atom abstraction, cyclization,
and rearomatization) mediated by the reagent diethylphosphine oxide (DEPO). The reaction features V-501 as a water-soluble initiator; no
other additives are needed. The process proceeds at a much lower temperature than is required for efficient reaction with toxic tributyltin
hydride in benzene and permits significantly higher isolated yields than the corresponding reaction mediated by ethylpiperidine hypophosphite
(EPHP).
The drive for new reagents to effect clean carbon-carbon
bond formation by radical reactions stems from the well-
known problems with the widely used organotin hydrides,
notably tributyltin hydride (TBTH), which are neurotoxic.
Organotin residues are also noted for difficulties in separation
from desired reaction products. Although numerous methods
continue to be devised for facilitating separation, thereby
diminishing the amounts of tin residues,1 even traces of tin
byproducts pose problems for pharmaceutical manufacture.
Perhaps the most popular alternative reagent is tristrimeth-
ylsilylsilane,2 TTMSS, which has a similar, but not identical,
range of reactivity to the tin reagent; however, it is much
more costly.
tion. These reagents are economical and can be used in water
(hypophosphorous acid, usually with an inorganic base) or
in organic solvents (the lipophilic salt, N-ethylpiperidinium
hypophosphite, EPHP),3,4 and, on workup, separation of
neutral organic reaction products is easily achieved by acid-
and base-wash.
Initial studies in this area used organic solvents such as
benzene and toluene. Jang first showed3b that iodoarenecar-
boxylate salts could be deiodinated by hypophosphorous acid
in alkaline aqueous solution, and we showed3e using aqueous
base that such substrates could be used for C-C bond
formation in water. However, important extensions of this
chemistry to substrates that are not water-soluble (unlike the
carboxylate described above) were made by Oshima and co-
workers3i (EtOH/H2O in the presence of base) and Kita et
al. (H2O + phase-transfer agent).3m The phosphorus reagents
are economical and therefore can be used in excess, but this
brings a problem in terms of waste, unless the unused
phosphorus reagent can be recovered. So far, to our
knowledge, this issue has not been addressed. On reaction
with an iodo-containing substrate, hypophosphorous acid and
Among the newly developed replacement reagents for
TBTH, phosphorus compounds,3,4 particularly hypophos-
phorous acid and its salts, are receiving considerable atten-
(1) See: Clive, D. L.J.; Wang, J. J. Org. Chem. 2002, 67, 1192 and
extensive references therein for amelioration of the tin problem and for a
full listing of alternate reagents.
(2) (a) Chatgilialoglu, C.; Griller, D.; Lesage, M. J. Org. Chem. 1988,
53, 3641. (b) Kulicke, K. J.; Giese, B. Synlett 1990, 91. (c) Chatgilialoglu,
C. Acc. Chem. Res. 1992, 25, 188.
10.1021/ol035173i CCC: $25.00 © 2003 American Chemical Society
Published on Web 07/12/2003

EPHP can form two possible byproducts, IPH(O)(OH) and
I2P(O)(OH); in turn, these can be hydrolyzed to phosphorous
acid and phosphoric acid, respectively. Separation of any of
these four byproducts from each other or from residual
hypophosphorous acid is not trivial, since they are all
relatively strong acids.5
Scheme 1
Recently, our efforts have been directed at finding even
simpler procedures for radical reactions using phosphorus
radicals. The main purpose of the phase-transfer reagent used
in Kita’s work is to lead to a greater interaction between the
water-soluble phosphorus reagent and the water-insoluble
lipophilic substrate. Since the economical phosphorus re-
agents are normally used in 10-20-fold excess, we wondered
whether a phosphorus reagent that is more lipophilic than
hypophosphorous acid but still totally soluble in water might
avoid the need for the phase-transfer agent. Toward this end,
we were attracted by the phosphine oxides, R2P(O)H.
The solubilities of four phosphine oxide examples R2P-
(O)H [R ) Me, Et, n-Bu, and Ph] were screened, and this
showed that diethylphosphine oxide (DEPO) had the best
profile, dissolving fully in water and in many organic solvents
(e.g., DCM, acetone, but being only sparingly soluble in
diethyl ether). DEPO was also chosen in preference to
dimethylphosphine oxide because DEPO has a higher
decomposition temperature.6 DEPO is easily and economi-
cally prepared by addition of ethylmagnesium bromide to
diethyl phosphite. Importantly, DEPO7 has pKa ) 6, and so
qualifies as an almost neutral compound, but yet would be
sufficiently acidic to be extracted into base during workup.
By analogy with other P-H-containing phosphorus com-
pounds, DEPO should react as shown in Scheme 1. We first
examined the reduction of p-iodoanisole under a range of
conditions; in particular, we examined initiation with AIBN,
VA-086, V-50, V-501, di-tert-butylperoxide, and Et3B/O2.
From these preliminary studies, V-501 was singled out as
the initiator of choice and 80 °C as a convenient temperature
for the reduction. This initiator8 has a half-life of 10 h at 69
°C in water and, being a carboxylic acid, dissolves in water
at alkaline pH, and both it and its byproducts can therefore
be easily separated from neutral organic compounds by base-
wash. At the end of the reaction, diethyl ether was added,
and the aqueous phase was made basic (2 M NaOH).
Importantly, DEPO (as its salt) remained in the basic aqueous
layer, while the desired product was extracted into the ether
layer, which was dried and evaporated to afford pure product
directly.
The expected phosphorus byproduct of the reaction,
diethylphosphinyl iodide, must undergo rapid hydrolysis
under the basic workup conditions, as it was not detected in
NMR spectra of the evaporated ethereal layer containing our
reaction product.
To screen this reagent, DEPO, we now chose a reaction
that is important in our work9 on alkaloid synthesis, i.e., the
conversion of iodoarenes 1 to indolones 2. This indolone-
forming sequence of reactions had previously been success-
fully carried out with tributyltin hydride10 but was successful
only at elevated temperatures (160 °C; see below). At 80
°C, the bromide analogue of 1a gave a 20:80 ratio of 2a:3a
as judged by GC10. The higher temperatures were required
because at 80 °C in benzene, abstraction of a hydrogen atom
from TBTH is relatively easy compared to cyclization onto
the aromatic ring.
(3) (a) Barton, D. H. R.; Jang, D. O.; Jaszberenyi, J. Cs. Tetrahedron
Lett. 1992, 33, 5709. (b) Jang, D. O. Tetrahedron Lett. 1996, 37, 5367. (c)
Calderon, J. M. B.; Chicharro, G. J.; Fiandorn, R. J.; Huss, S.; Ward, R. A.
EP2284 970506, 1997. (d) McCague, R.; Pritchard, R. G.; Stoodley, R. J.;
Williamson, D. S. Chem. Commun. 1998, 2691. (e) Graham, S. R.; Murphy,
J. A.; Coates, D. Tetrahedron Lett. 1999, 40, 2415. (f) Graham, S. R.;
Murphy, J. A.; Kennedy, A. R. J. Chem. Soc., Perkin Trans. 1 1999, 3071.
(g) Tokuyama, H.; Yamashita, T.; Reding, M. T.; Kaburagi, Y.; Fukuyama,
T. J. Am. Chem. Soc. 1999, 121, 3791. (h) Martin, C. G.; Murphy, J. A.;
Smith, C. R. Tetrahedron Lett. 2000, 41, 1833. (i) Yorimitsu, H.; Shinokubo,
H.; Oshima, K. Chem. Lett. 2000, 104. (j) Graham, A. E.; Thomas, A. V.;
Yang, R. J. Org. Chem. 2000, 657, 2583. (k) Marotta, E.; Righi, P.; Rosini,
G. Org. Lett. 2000, 2, 4145. (l) Jang, D. O.; Cho, D. H.; Chung, C.-M.
Synlett 2001, 1923. (m) Kita, Y.; Nambu, H.; Ramesh, N. G.; Anilkumar,
G.; Matsugi, M. Org. Lett. 2001, 3, 1157. (n) Yorimitsu, H.; Shinokubo,
H.; Oshima, K. Bull. Chem. Soc. Jpn. 2001, 74, 225. (o) Deprele, S.;
Montchamp, J. L. J. Org. Chem. 2001, 66, 6745. (p) Jang, D. O.; Cho, D.
H. Synlett 2002, 631. (q) Yorimitsu, H.; Shinokubo, H.; Oshima, K. Synlett
2002, 674. (r) Jang, D. O.; Cho, D. H. Synlett 2002, 1523. (s) Jang, D. O.;
Cho, D. H. Tetrahedron Lett. 2002, 43, 5921. (t) Lee, E., Han, H. O.
Tetrahedron Lett. 2002, 43, 7295. (u) Roy, S. C.; Guin, C.; Rana, K. K.;
Maiti, G. Tetrahedron 2002, 58, 2435. (v) Dubert, O.; Gautier, A.;
Condamine, E.; Piettre, S. R. Org. Lett. 2002, 4, 359. (w) Reding, M. T.;
Kaburagi, Y.; Tokuyama, H.; Fukuyama, T. Heterocycles, 2002, 56, 313.
(x) Takamatsu, S. Katayama, S.; Hirose, N.; Naito, M.; Izawa, K.
Tetrahedron Lett. 2001, 42, 7605. (y) Cho, D. H.; Jang, D. O. Bull. Korean
Chem. Soc. 2003, 24, 15. (z) Nambu, H.; Anilkumar, G.; Matsugi, M.; Kita,
Y. Tetrahedron 2003, 59, 77.
Our first efforts used N-ethylpiperidine hypophosphite
(EPHP) and led to the outcomes shown in column 3 of Table
1. The crude reaction mixtures appeared to contain three
Table 1. Yields of Cyclized Oxindoles 2 (and Reduced
Byproducts 3) Prepared from Iodoaryl Precursors 1 at 80 °C in
Water
substrate
DEPO 2 (%)
EPHP 2/3 (%)
1a
1b
1c
1d
97
90
87
82
58/21
49/21
51/23
38/12
(4) For other radical reagents based on phosphorus, see: (a) Barks, J.
M.; Gilbert, B. C.; Parsons, A. F.; Upeandran, B. Tetrahedron Lett. 2001,
42, 3137. (b) Jessop, C. M.; Parsons, A. F.; Routledge, A.; Irvine, D.
Tetrahedron Lett. 2003, 44, 479. (c) Jang, D. O.; Cho, D. O.; Barton, D.
H. R. Synlett 1998, 39.
(5) pKa values: H3PO2, 1.3; H3PO3, 1.3 and 6.78; H3PO4, 2.15, 7.20,
and 12.38. http://www.swbic.org/education/env-engr/chem/acidconstants.h-
tml.
(6) Hays, H. R. J. Org. Chem. 1968, 33, 3690
(7) Determined using a Jenny 3015 pH meter at 25 °C.
aromatic compounds, unreacted 1, product 2, and byproduct
3, as expected. It is seen that considerable amounts of
2972
Org. Lett., Vol. 5, No. 16, 2003

Scheme 2
Scheme 3
reduced compounds 3 are formed, although the ratios are
better than for TBTH at the same temperature. However,
the mass balances on isolated compounds were not always
good, and we established that this was due to the volatility
of the indolones during evaporation of the toluene solvent.
Since the indolones are not expected to have low boiling
points (compound 1a has a boiling point of 180 °C @ 19
mm Hg),11 we can only ascribe this unexpected loss to an
azeotropic effect.
The preparation and isolation in high yield of these
compounds therefore posed a problem, but we reasoned that
they should provide an excellent test for the DEPO protocol,
involving workup with diethyl ether; substrates 1a-d were
therefore tested (see Supporting Information). As seen in
Scheme 3, the reactions proceed in excellent yield, and the
workup directly affords products that are essentially pure.
The most interesting findings, however, were related to the
product profile. When EPHP had been used as a reagent,
significant quantities of reduced compounds 3 were obtained,
while with DEPO, no reduced compound was observed with
1a and very minute quantities with 1b-d. This shows clearly
that DEPO is a superior reagent for this transformation.
However, DEPO has another advantage over EPHP in that
the byproduct, diethylphosphinyl iodide, is easily converted
to diethylphosphinic acid (pKa ) 3.29)12 and separation of
this product from residual DEPO (pKa ) 6.0) is easily
accomplished by adjusting the pH to 5.0.
of cost and toxicity) temperatures of 160 °C were required
with TBTH to obtain good yields of cyclized products.
Having established that DEPO is very useful for indolone
formation, we were keen to investigate whether it would be
equally successful for other radical reactions. The key
reaction for routine radical cyclizations is cleavage of C-I
and C-Br bonds, and as seen above, this is readily achieved
by DEPO. However, phosphorus radicals also readily add
to alkenes4,3n,u; if this occurs more rapidly than C-I cleavage,
then the reagent would have limited value. To test this, the
iodoarenes, 4a-c and 6a-c, were prepared and treated with
DEPO in water at 80 °C. As above, the reactions were
relatively slow (4 days, 20 equiv of DEPO and 0.37 equiv
of V-501 each 12 h). As seen in Scheme 4, this provided
When compared with tributyltin hydride (TBTH), DEPO
also proves to be a better reagent as (quite aside from issues
(8) Yorimitsu, H.; Wakabayashi, K.; Shinkubo, H.; Oshima, K. Tetra-
hedron Lett. 1999, 40, 519.
(9) (a) Lizos, D. E.; Murphy, J. A. Org Biomol. Chem. 2003, 1, 117. (b)
Lizos, D.; Tripoli, R.; Murphy, J. A. Chem. Commun. 2001, 2732.
(10) (a) Beckwith, A. L. J.; Storey, J. M. D. J. Chem. Soc., Chem.
Commun. 1995, 977. (b) Storey, J. M. D. Tetrahedron Lett. 2000, 41, 8173.
(11) Gibson, M. S. J. Chem. Soc. 1961, 2249.
(12) Crofts, P. C.; Kosolapoff, G. M. J. Am. Chem. Soc. 1953, 75, 3379.
Org. Lett., Vol. 5, No. 16, 2003
2973

The similarity of the yields of isolated products from 4a-c
suggests that this is not a problem.
Scheme 4
Finally, we tested the reagent with the acid-sensitive
aliphatic bromide 8 (10 equiv of DEPO, 0.5 equiv of V-501,
12 h). Again, this gave very good yield of the cyclized
product 9 and no evidence of acetal hydrolysis was seen.
In summary, the first examples of indolone formation using
phosphorus reagents are reported. DEPO is introduced as a
reagent that can work in water with no need for additives
(only initiator, substrate, and water are needed). This reagent
affords extremely easy workup and affords high yields of
product, free from phosphorus or initiator-derived byprod-
ucts.
Acknowledgment. We thank EPSRC for funding, CVCP
and University of Strathclyde for ORS and International
studentship awards (to T.K.), Nuffield Foundation for an
undergraduate bursary (to J.J.C.), the Leverhulme Trust for
a Senior Research Fellowship (to J.A.M.), and EPSRC Mass
Spectrometry Service, Swansea, for mass spectra.
Supporting Information Available: Methods of prepara-
tion as well as spectroscopic data for compounds 2a-d, 5a-
c, 7a-c, and 9. This material is available free of charge via
the Internet at http://pubs.acs.org.
the desired products in very good yields. We were particu-
larly keen to compare the terminal alkene case, 4c, with more
highly substituted alkenes 4a,b, since terminal alkenes can
show particular sensitivity to phosphorus radical addition.4
OL035173I
2974
Org. Lett., Vol. 5, No. 16, 2003