An expedient synthesis of homophthalimides

Article abstract of DOI:10.1039/b207649g

The six-membered heterocyclic subunit of homophthalimides can be obtained by a direct, hitherto unprecedented, radical cyclisation onto an aromatic ring starting from a xanthate precursor.

Full text of DOI:10.1039/b207649g

An expedient synthesis of homophthalimides
Béatrice Quiclet-Sire* and Samir Z. Zard*
Laboratoire de Synthèse Organique associé au CNRS, Ecole Polytechnique, F-91128 Palaiseau Cedex,
France. E-mail: zard@poly.polytechnique.fr; Fax: +33 (0)1 6933 3851; Tel: +33 (0)1 6933 4872
Received (in Cambridge, UK) 5th August 2002, Accepted 9th September 2002
First published as an Advance Article on the web 23rd September 2002
The six-membered heterocyclic subunit of homophthali-
mides can be obtained by a direct, hitherto unprecedented,
radical cyclisation onto an aromatic ring starting from a
xanthate precursor.
Homophthalic acid derivatives are important building blocks
for the synthesis of alkaloids, dyes, and a variety of medicinally
interesting structures.1 They are also convenient precursors of
o-quinonemethide intermediates, which readily undergo Diels–
Alder cycloadditions vith various reactive dienophiles (e.g.
quinones for the synthesis of anthracyclinones as in Scheme 1)2
or of isoquinolines through vigourous heating with zinc powder,
a reaction also pictured in Scheme 1 and first described by Le
Blanc in the 19th century.3
Synthetic routes to these deceptively simple compounds are,
surprisingly, scarce and of very limited scope.1,4 Copper
catalysed aromatic substitution of 2-halobenzoic acids and the
oxidative ring cleavage of indanones are the two main routes to
homophthalic acids, and therefrom to homophthalimides.
Except in the simplest of cases, the substrates needed in both of
these approaches are not generally available. In the present
communication, we describe a practical synthesis based on a
radical cyclisation onto an aromatic ring which leads directly to
homophthalimides.
Scheme 2 Synthesis of homophthalimides.
Our approach is delineated in Scheme 2 and relies on a
versatile radical chemistry of xanthates we have developed over
the past few years.5 The key aspect is the degeneracy of the
reaction of radical 5 with its xanthate precursor 3. This reactive
intermediate is therefore not irreversibly consumed and ac-
quires a comparatively long lifetime, allowing it to overcome
both the rotational barrier of the amide linkage (equilibrium
between 5a and 5b) and the quite slow ring-closure onto the
aromatic ring. This step is then followed by an oxidative
rearomatisation, involving the peroxide or resulting from a
disproportionation process.6 In any case, stoichiometric
amounts of peroxide are required.
the peroxide7 and increases the rate of interchange between the
two rotational isomers of the intermediate radical.8 The
xanthate precursors are readily accessible by reaction of the
After some experimentation, we found that heating xanthates
3 in refluxing o-dichlorobenzene with addition of 1.0 molar
equivalent of di-tert-butyl peroxide brought about the desired
transformation in acceptable yield, as shown by the examples
collected in Fig. 1. The products very often precipitated upon
cooling of the reaction mixture or after evaporation of part of the
solvent under reduced pressure, and could be isolated by mere
filtration.† The use of a high boiling solvent shortens the
reaction time by allowing a convenient decomposition rate of
Scheme 1 Synthesis of isoquinolines from homophthalimides.
Fig. 1
2306
CHEM. COMMUN., 2002, 2306–2307
This journal is © The Royal Society of Chemistry 2002

View Article Online
NH), 8.1 (dd, 1H, J 6 Hz, Ph), 7.3 (m, 2H, Ph), 4.1 (CH2). 13C NMR (75
corresponding benzamides 1 with chloroacetyl chloride, fol-
lowed by displacement of the chlorine in chloracetimide 2 with
commercially available potassium O-ethyl xanthate.
MHz, DMSO-d6) d 170.6, 164.4 (CO), 166.9, 162.9 (CF, JCF 252 Hz),
140.0, 139.8, 130.6, 130.4 (JCF 10 Hz), 121.8 (CPh), 114.9, 114.6, 114.4,
114.1 (J 22 Hz, CHPh), 35.7 (CH2). IR (Nujol, nmax) 1705, 1689 (CO). Mp
244 °C. Anal. Calc. for C9H6FNO2: C 60.34, H 3.38, found: C 60.31, H
3.37%. 4b: 1H NMR (200 MHz, DMSO-d6) d 11.5 (br s, 1H, NH), 8.0 (d,
1H, J 8.9 Hz, Ph), 7.5 (s + d, 2H, Ph), 4.1 (CH2). 13C NMR (62.5 MHz,
DMSO-d6) d 170.4, 164.5 (CO), 138.8, 138.2 (CPh), 129.3, 127.5, 127.4
(CHPh), 123.9 (CPh), 35.7 (CH2). IR (Nujol, nmax) 1702, 1689 (CO). Mp
267–268 °C (decomp.). Anal. Calc. for C9H6ClNO2: C 55.26, H 3.09,
found: C 55.21, H 3.28%. 4c: 1H NMR (250 MHz, DMSO-d6) d 11.5 (br s,
1H, NH), 8.03 (d, 1H, J 8.9 Hz, Ph), 7.7 (s + d, 2H, Ph), 4.15 (CH2). 13C
NMR (62.5 MHz, DMSO-d6) 170.4, 164.6 (CO), 138.9 (CPh), 130.5, 130.3,
129.3 (CHPh), 127.3, 124.2 (CPh), 35.7 (CH2). IR (Nujol, nmax) 1702, 1695
(CO). Mp 278–280 °C (decomp.). Anal. Calc. for C9H6BrNO2: C 45.03, H
2.52, found: C 45.32, H 2.53%. 4d: 1H NMR (250 MHz, DMSO-d6) d 11.5
(br s, 1H, NH), 7.93 (s + d, 2H, Ph), 7.83 (d, 1H, J 8.6 Hz, Ph), 4.1 (CH2).
13C NMR (62.5 MHz, DMSO-d6) d 170.4, 164.9 (CO), 138.5 (CPh), 136.4,
Interestingly, homophthalimides both substituted and un-
substituted at the nitrogen may be obtained. Radical ring closure
of unsubstituted amides is very rare (for an example, see ref. 9a;
it is usually necessary to place a substituent—preferably
bulky—on the nitrogen atom9b) and, as far as we know,
unprecedented for the formation of six-membered rings fused to
an aromatic ring. Such cyclisations would be exeedingly
difficult to perform by more traditional radical methods,
especially those based on stannane chemistry, since premature
capture of intermediate radical 5 would be almost impossible to
avoid. The success of the present procedure hinges to a large
extent on the long lifetime of intermediate radical 5 when
generated using the xanthate technology; moreover, as illus-
trated by examples 4c and 4d, the method tolerates the presence
of bromine and iodine on the aromatic ring, allowing for further
modifications using various powerful transition metal coupling
reactions.
The reaction works best with substrates where the sub-
stituents are symmetrically arranged around the aromatic ring,
to avoid the formation of regioisomers (as for example with
meta-substituted precursors). In the case of ortho-substituted
derivatives we encountered a serious complication from ipso-
substitution (cf. 6) and complex mixtures were observed. This is
presumably due to a peri-type steric repulsion, shown in 7,
which hampers the adoption by the intermediate radical of a
conformation propitious for the desired cyclisation.
136.0, 128.9 (CHPh), 124.4 (CPh), 101.9 (CI), 35.4 (CH2). IR (Nujol, nmax
)
1703, 1698 (CO). Mp 266–267 °C. Anal. Calc. for C9H6INO2: C 37.66, H
2.11, found: C 37.51, H 2.21%. 4e: 1H NMR (300 MHz, DMSO-d6) d 11.6
(br s, 1H, NH), 8.32 (d, 1H, J 9 Hz, Ph), 7.92 (s + d, 2H, Ph), 4.2 (CH2). IR
(Nujol, nmax) 1703, 1698 (CO). Mp 238–239 °C (decomp.). Anal. Calc. for
C
10H6F3NO2: C 52.41, H 2.64, found: C 52.69, H 2.61%. 4f: 1H NMR (300
MHz, DMSO-d6) d 11.37 (s, 1H, NH), 7.47 (s, 1H, Ph), 3.97, 3.96, 3.94 (3s,
9H, OCH3). 3.89 (s, 2H, CH2). 13C NMR (75 MHz, DMSO-d6) 170.7, 164.6
(CO), 152.4, 149.3, 146.1, 123.3, 120.0 (CPh), 105.6 (CHPh), 60.5, 60.4,
55.9 (3 OMe), 31.3 (CH2). MS (IC, NH3, m/z) 252 (MH)+, 269
(MH + NH3)+. IR (nmax, Nujol) 1705, 1686 (CO). Mp 206–207 °C
(decomp.). Anal. Calc. for C12H13NO5: C 57.37, H 5.22, found: C 57.16, H
5.27%.
1 (a) M. Cushman and F. W. Dekow, J. Org. Chem., 1979, 44, 407; (b) D.
Perez, E. Guitian and L. Castedo, J. Org. Chem., 1992, 57, 5911; (c) H.
Heaney and M. O. Taha, Tetrahedron Lett., 2000, 41, 1993; (d) M. S.
Malamas, T. C. Hohman and J. Millen, J. Med. Chem., 1994, 37, 2043;
(e) M. S. Malamas, J. Heterocycl. Chem., 1994, 31, 565; (f) E. S. Lazer,
R. Sorcek, C. J. Cywin, D. Thome, G. J. Possanza, A. G. Graham and L.
Churchill, Bioorg. Med. Chem. Lett., 1998, 8, 1181.
In summary, this approach provides a flexible and rapid route
to otherwise inaccessible homophthalimides and, therefrom, to
homophthalic acids and anhydrides. It allies simplicity with the
use of readily available, cheap substrates and reagents. The
isolation of the crude product by a simple filtration directly from
the reaction mixture is a distinct asset for large scale
preparations.†
2 Y. Tamura, A. Wada, M. Sasho, K. Fukunaga, H. Maeda and Y. Kita, J.
Org. Chem., 1982, 47, 4376.
3 M. Le Blanc, Chem. Ber., 1888, 2299.
4 (a) R. D. Haworth and H. S. Pink, J. Chem. Soc., 1925, 1368; (b) W. J.
Gensler, S. F. Lawless, A. L. Bluhm and H. Dertrouzos, J. Org. Chem.,
1975, 40, 733; (c) S. D. Young, J. M. Wiggins and J. R. Huff, J. Org.
Chem., 1988, 53, 1114; (d) V. H. Belgaonkar and S. N. Usgaonkar, J.
Chem. Soc., Perkin Trans. 1, 1977, 702; (e) I. W. Elliott, J. Heterocycl.
Chem., 1972, 9, 853.
5 For reviews on our work on the radical xanthate transfer reaction, see: S.
Z. Zard, in Radicals in Organic Synthesis, ed. P. Renaud and M. Sibi,
Wiley-VCH, Weinheim, 2001, pp. 90–108; S. Z. Zard, Angew. Chem.,
Int. Ed. Engl., 1997, 36, 672; B. Quiclet-Sire and S. Z. Zard, Phosphorus,
Sulfur Silicon, 1999, 153–154, 137.
6 A. Studer and M. Bossart, in Radicals in Organic Synthesis, ed. P.
Renaud and M. Sibi, Wiley-VCH, Weiheim, vol. 2, ch. 1.4, pp. 63–80.
7 (a) K. Kujimori in Organic peroxides, ed. W. Ando, J. Wiley & Sons,
Chichester, 1992, ch. 7, pp 319–385; (b) C. Walling and E. S. Huyser,
Org. React., 1963, 13, 91.
8 D. P. Curran and J. Tamine, J. Org. Chem., 1991, 56, 2746.
9 (a) T. Sato, S. Ishida, H. Ishibashi and M. Ikeda, J. Chem. Soc., Perkin
Trans. 1, 1991, 353; (b) G. Stork and R. Mah, Heterocycles, 1989, 28,
723.
Notes and references
†
Typical experimental procedure: synthesis of xanthates 3a–f: The
corresponding benzamide 1 (1 mmol) was refluxed in toluene (2 ml) in the
presence of chloroacetyl chloride (1.5 mmol) until complete reaction (TLC
monitoring). The reaction mixture was cooled and the precipitated
chloroacetimide 2 filtered, dried, and used directly in the next step. The
chloroacetimide was treated under argon with potassium O-ethyl xanthate
(1.1 mmol) in acetonitrile (2–2.5 ml). Addition of water gave a precipitate
which was filtered off, washed with water and dried. The xanthate 3 thus
obtained (overall yield 80-95%) was used directly in the next step.
Synthesis of homophthalimides 4: a solution of the xanthate 3a–g (1
mmol) in o-dichlorobenzene (4 ml) was refuxed under argon for 15 min.
Then a solution of di-tert-butyl peroxide (1 mmol.) in o-dichlorobenzene (3
ml) was added over a periode of 60–90 min. After cooling the mixture, the
precipitate was filtered off and dried. Analytical samples were obtained by
sublimation. Chromatography on silica gel were necessary for compounds
4a and 4e.
Spectral and analytical data for new homophthalimides 4a–f (4g has been
described in ref. 1d): 4a: 1H NMR (250 MHz, DMSO-d6) d 11.4 (br s, 1H,
CHEM. COMMUN., 2002, 2306–2307
2307