Synthesis of biaryls by intramolecular radical transfer: Use of phosphinates

Article abstract of DOI:10.1016/S0040-4039(99)02298-4

Phosphinates 4a-13a give biaryls 4b-13b on heating with stannanes in the presence of a radical initiator. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4039(99)02298-4

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 41 (2000) 1315–1319
Synthesis of biaryls by intramolecular radical transfer:
use of phosphinates
Derrick L. J. Clive ∗ and Shunzhen Kang
Chemistry Department, University of Alberta, Edmonton, Alberta, T6G 2G2 Canada
Received 12 November 1999; accepted 8 December 1999
Abstract
Phosphinates 4a–13a give biaryls 4b–13b on heating with stannanes in the presence of a radical initiator. © 2000
Elsevier Science Ltd. All rights reserved.
Intramolecular transfer of aryl groups by a radical mechanism has been known for a long time,1 but
has recently attracted special attention with a view to evaluating the synthetic possibilities of the process.
Much of the latest work has involved the formation of biaryls,2–4 and has usually been based on the
-SO2- group as a tether between two aromatic rings that are subsequently linked. Reactions such as those
summarized in Scheme 12a,b are typical, and have been studied in detail. The transfer occurs by ipso
substitution, followed by expulsion of the tether.5 In certain cases2a,c,3 (for example, Scheme 22c and
Scheme 33), the initial ipso attack can be of the 6-exo type.
Scheme 1.
Scheme 2.
∗
Corresponding author. E-mail: derrick.clive@ualberta.ca
0040-4039/00/$ - see front matter © 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(99)02298-4
tetl 16226

1316
Scheme 3.
We report the transfer of aromatic units within the phosphinate compound class, according to Eq. (1).
(1)
Our initial experiments involved generation of radicals of type 1 (Scheme 4, R=H, Me, Ph), in order
to establish6 if they would rearrange (1→2), by analogy to the well-known7 β-(phosphatoxy)alkyl and
β-(acyloxy)alkyl radical rearrangements. In the event, the products isolated were of type 3 (R=H, 45%;
R=Me, 50%, R=Ph, ca. 15%). It was quickly established that the presence of the nitrogen subunit is not
essential, and that the reaction is a general one of phosphinates (see Eq. (1) and Table 1).
Scheme 4.
The starting phosphinates are available by acylation of the parent alcohol with the appropriate
diarylphosphinic chloride [Ar2P(O)Cl,9 Et3N, DMAP, CH2Cl2, room temperature, ca. 1 h], and our
general procedure for the radical rearrangement involves slow addition of a solution of stannane (0.03 M,
2 mmol) and AIBN (0.0075 M, 0.5 mmol) to a refluxing solution of the phosphinate (0.015 M, 1 mmol).
We found that reactions done in xylene give better yields than those in lower-boiling hydrocarbons
(toluene or benzene), at least as judged by experiments with 5a, and so we generally used this solvent.
Our preferred workup involves stirring the product mixture with saturated aqueous potassium fluoride,
followed by chromatography.
As shown in Table 1, the presence of electron-withdrawing (CO2Me) or electron-donating (OMe, Me)
substituents is tolerated. The starting alcohol can be primary or secondary. In addition, we have tested
heteroaromatic groups (see Table 1); they undergo the aryl transfer (cf. 10a, 11a) and can also serve as
the location of the initial radical (cf. 12a). The conversion of 13a (which is a mixture of diastereoisomers)
into 13b shows that the phosphinate can carry an alkyl group on phosphorus.
Attempts to extend the process to a primary alkyl radical (derived from 14) were unsatisfactory, as
the major product was that of simple reduction, and the desired phenyl transfer occurred only to a small

1317
Table 1
extent (transfer:reduction 1:210).11 With the naphthalene derivative 15, only the reduction product (H
instead of Br) was isolated (74%). Possibly, peri interactions hinder the desired process in this case.
Mechanistically, it is reasonable to assume that the early stages of the reaction involve the steps

1318
Scheme 5.
shown in Scheme 5, with the initial radical transfer occurring via a six-membered transition state. In this
preliminary work, however, we have not established the mechanism(s) by which 5b is finally liberated.
The suggested precursor 5c,12 made independently, is stable to the workup conditions but, on heating in
xylene (overnight), it affords alcohol 5b (ca. 42%), which can be detected (TLC13) before workup. In
refluxing benzene or toluene (ca. 12 h), 5c is stable but gives 5b (ca. 46%) in benzene (toluene was not
tested) in the presence of tributylstannane and AIBN.
All new compounds were characterized by spectroscopic measurements, including accurate mass
measurement.
Acknowledgements
Acknowledgment is made to the Natural Sciences and Engineering Research Council of Canada and
to Merck Frosst for financial support. We thank Professor M. Klobukowski from this Department for
advice.
References
1. For example, see: Loven, R.; Speckamp, W. N. Tetrahedron Lett. 1972, 1567–1570.
2. (a) Motherwell, W. B.; Pennell, A. M. K. J. Chem. Soc., Chem. Commun. 1991, 877–879. (b) da Mata, M. L. E. N.;
Motherwell, W. B.; Ujjainwalla, F. Tetrahedron Lett. 1997, 38, 137–140. (c) da Mata, M. L. E. N.; Motherwell, W. B.;
Ujjainwalla, F. Tetrahedron Lett. 1997, 38, 141–144.
3. Giraud, L.; Lacôte, E.; Renaud, P. Helv. Chim. Acta 1997, 80, 2148–2156.
4. Leardini, R.; McNab, H.; Nanni, D. Tetrahedron 1995, 51, 12143–12158.
5. For non-sulfone tethers, see, for example: (a) Ref. 3 and (b) Ishibashi, H.; Nakamura, N.; Ito, K.; Kitayama, S.; Ikeda, M.
Heterocycles 1990, 31, 1781–1784. (c) Lee, E.; Lee, C.; Tae, J. S.; Whang, H. S.; Li, K. S. Tetrahedron Lett. 1993, 34,
2343–2346. (d) Lee, E.; Whang, H. S.; Chung, C. K. Tetrahedron Lett. 1995, 36, 913–914.
6. Cf. (a) Evanochko, W. T.; Shevlin, P. B. J. Org. Chem. 1979, 44, 4426–4430. (b) Shahidi, F.; Tidwell, T. T. Can. J. Chem.
1982, 60, 1092–1097.
7. Beckwith, A. L. J.; Crich, D.; Duggan, P. J.; Yao, Q. Chem. Rev. 1997, 97, 3273–3312.
8. Crich, D.; Sun, S. J. Org. Chem. 1996, 61, 7200–7201.
9. Ph2P(O)Cl and (p-MeOC6H4)2P(O)OH were commercial samples. Ph(Me)P(O)Cl was prepared from PhPCl2 (Korpiun, O.;
Lewis, R. A.; Chickos, J.; Mislow, K. J. Am. Chem. Soc. 1968, 90, 4842–4846). The other diarylphosphinic chlorides
were prepared by the action of SOCl2 on the parent acid: (p-MeC6H4)2P(O)OH was made by the general method of
Kosolapoff, G. M.; Struck, R. F. J. Chem. Soc. 1959, 3950–3952; (p-MeO2CC6H4)2P(O)OH was made by oxidation of
(p-MeC6H4)2P(O)OH (Petrov, K. A.; Parshina, V. A.; Daruze, G. L. J. Gen. Chem. USSR 1960, 30, 2972–2975); (2-
furyl)2P(O)OH was made by hydrolysis of (2-furyl)3PO (Allen, D. W.; Hutley, B. G.; Mellor, M. T. J. J. Chem. Soc., Perkin
Trans. 2 1977, 1705–1708).
10. This ratio refers to use of Bu3SnH; with Ph3SnH only the reduction product was isolated.

1319
11. For aryl transfer to alkyl radicals, see: (a) Refs. 1 and 5. (b) Clive, D. J. J.; Boivin, T. L. B. J. Org. Chem. 1989, 54,
1997–2003. (c) Köhler, H. J.; Speckamp, W. N. J. Chem. Soc., Chem. Commun. 1980, 142–143. (d) Studer, A.; Bossart,
M. J. Chem. Soc., Chem. Commun. 1998, 2127–2128. (e) Studer, A.; Bossart, M.; Steen, H. Tetrahedron Lett. 1998, 39,
8829–8832. (f) Köhler, J. J.; Speckamp, W. N. Tetrahedron Lett. 1977, 631–634. (g) Köhler, J. J.; Speckamp, W. N.
Tetrahedron Lett. 1977, 635–638. (h) Speckamp, W. N.; Köhler, J. J. J. Chem. Soc., Chem. Commun. 1978, 166–167. (i)
Wilt, J. W.; Chwang, W. K.; Dockus, C. F.; Tomiuk, N. M. J. Am. Chem. Soc. 1978, 100, 5534–5540. (j) Amii, H.; Kondo,
S.; Uneyama, K. J. Chem. Soc., Chem. Commun. 1998, 1845–1846.
12. Prepared by a general method: Noyce, D. S.; Virgilio, J. A. J. Org. Chem. 1972, 37, 1052–1053.
13. We did not test for the possibility that a hydrolytic step occurs on application of the mixture to silica.