Beckmann rearrangement of ketoximes on solid metaboric acid: A simple and effective procedure

Article abstract of DOI:10.1016/S0040-4039(02)00282-4

When ketoximes admixed with solid metaboric acid (formed from boric acid at 100°C/0.1 Torr) are heated (～140°C/7-42 h), the corresponding amides or lactams are produced in excellent yields (62-92%) via the Beckmann reaction. Aromatic aldoximes undergo both dehydration to the nitrile as well as (non-stereospecific) rearrangement under the above conditions. The absence of solvent, and the mildness and low toxicity of boric acid, characterise the present procedure.

Full text of DOI:10.1016/S0040-4039(02)00282-4

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 43 (2002) 2455–2457
Beckmann rearrangement of ketoximes on solid metaboric acid:
a simple and effective procedure
Sosale Chandrasekhar* and Kovuru Gopalaiah
Department of Organic Chemistry, Indian Institute of Science, Bangalore 560 012, India
Received 19 December 2001; revised 23 January 2002; accepted 6 February 2002
Abstract—When ketoximes admixed with solid metaboric acid (formed from boric acid at 100°C/0.1 Torr) are heated
(
ꢀ140°C/7–42 h), the corresponding amides or lactams are produced in excellent yields (62–92%) via the Beckmann reaction.
Aromatic aldoximes undergo both dehydration to the nitrile as well as (non-stereospecific) rearrangement under the above
conditions. The absence of solvent, and the mildness and low toxicity of boric acid, characterise the present procedure. © 2002
Elsevier Science Ltd. All rights reserved.
We have recently reported several improved versions of
the Beckmann rearrangement in view of the continuing
general interest in this classical synthetic procedure. The
Initial studies indicated that oximes were unreactive
towards a suspension of boric acid in organic solvents.
However, when the oximes were directly absorbed on
boric acid and the solid mass heated (without a solvent)
above 100°C, the expected rearrangement was
observed, but with competing hydrolysis of the oximes.
Also, droplets of (presumably) water were observed in
the reaction flask, and there was no reaction at lower
temperatures. All this apparently indicated that the
boric acid was being converted under the reaction
conditions to metaboric acid (‘MBA’, Scheme 1), which
was presumably the active species effecting the rear-
rangement, and that the liberated water was responsible
for the competing hydrolysis. (Boric acid is known to
dehydrate sequentially >100°C to yield initially meta-
1
study of the reaction is greatly aided by the simple and
well-established mechanism, but this also implies that
new variants must be unusually novel to be useful. An
enormous variety of reaction conditions have been
reported over the century or so since the reaction was first
discovered, although few of them would meet the sophis-
ticated requirements of the present day, which lay stress
on mildness and environmental concerns.
We were attracted to the possible use of boric acid for
effecting the Beckmann reaction by its mildness (pKa
2
9
.25) and low mammalian toxicity (it has been employed
3,4
2–5
as a bacteriacidal in eyewashes, skin creams, etc.).
boric acid and finally boric anhydride). It was clear
2
Also, borate species are known to add water reversibly,
a characteristic relevant to the Beckmann reaction, which
may be viewed as a dehydration–rehydration sequence
effected on the protonated oxime. (Boric acid has found
that pre-formed metaboric acid would avoid the
hydrolysis and lead exclusively to the Beckmann rear-
rangement and this was indeed confirmed in the event:
several ketoximes upon admixture with metaboric acid
and heating furnished the expected amides or lactams
5
occasional use in organic synthesis. )
Scheme 1.
Keywords: Beckmann; boric acid; formanilide; metaboric acid; rearrangement; solid state.
*
Corresponding author. Fax: (+91 80) 3600 529; e-mail: sosale@orgchem.iisc.ernet.in
0
040-4039/02/$ - see front matter © 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(02)00282-4

2
456
S. Chandrasekhar, K. Gopalaiah / Tetrahedron Letters 43 (2002) 2455–2457
in excellent yields (Scheme 1 and Table 1; note that
rated and the stereochemistry of the rearrangement was
not ascertained, but the presence of multiple carboxam-
2
†
‘
MBA’ is oligomeric ). The stereochemistry of the
ketoximes studied is presumed to be that depicted
Scheme 1 and Table 1) based on their Beckmann
−
1 6
ide bands in the IR (ꢀ1650 cm ) indicated low selec-
tivity. (This was supported by the GC–MS results:
indeed, Beckmann rearrangement of aldoximes is gener-
ally non-stereospecific, essentially because of competing
(
rearrangement in solution, the usual anti migration
pathway thus being observed.
7
stereoisomerisation of the oxime. )
The reaction of aromatic aldoximes with ‘MBA’ appar-
ently gave both the products of dehydration (the
nitriles) and those of the Beckmann rearrangement as
complex mixtures, as indicated by IR, NMR and GC–
MS (Scheme 2, Table 1). The mixtures were not sepa-
A likely mechanism for the above transformation
involves the intermediacy of the metaborate ester 7 of
the oxime, presumably formed via the ate-complex
adduct 6 (Scheme 3, the oligomeric form of ‘MBA’
Table 1. Reaction of the ketoximes 1 (Scheme 1) and aldoximes 3 (Scheme 2) with metaboric acid: reaction conditions,
products and yields
Oxime
Reaction conditions (°C/h)
Product(s)
Yield (%)
1
/3
R1
R2
1a
1b
1c
1d
1e
3a
3b
3c
Ph
Ph
p-Tolyl
Ph
Me
Me
140–145/17
140–145/20
130–135/16
135–140/42
135–140/7
151–155/30
140–145/46
150–155/58
2a
2b
2c
2d
Mixture
4a+5a
4b+5b
4c+5c
92
85
87
62
a
-(CH ) -
2
5
b
b
(+)-Camphor
3,4-(OMe) C H -
4-(OMe)C H -
90
c
H
H
H
84
78
65
2
6
3
c
6
4
c
Naphth-1-yl
a
m-Caprolactam.
b
(neat/cm−1) 2246 and 1663, M+ 149
and 167, for the nitrile and the lactam, respectively): the fragmentation of camphor oxime to nitriles as also the formation of a d-lactam under
This consisted of a nitrile (1.0 part) and possibly a d-lactam (1.4 parts) as seen by IR and GC–MS [w
max
8
9
Beckmann reaction conditions is known, and the mixture was not analysed further.
c
+
4
a:5a=1.0:1.9, 4b:5b=1.0:0.6, 4c:5c=1.0:2.5 (ratios by GC–MS: M was observed in all the cases, and amides 5 were ꢀ1:3 mixtures of the two
−
1
−1
6
possible isomers); wmax (neat/cm ) ꢀ2222 cm (CN), 1648–1668 (amide I).
Scheme 2.
Scheme 3.
†
3
Typical procedures: Boric acid was converted to metaboric acid by heating at 110–115°C/0.1 Torr over P O for 4 h. The oxime (1 mmol) and
2
5
the metaboric acid (3 mmol) were ground together using a pestle and mortar under a blanket of N , the resulting solid mixture was placed in
2
a round-bottomed flask under N and heated under the conditions shown in Table 1. The reaction was followed by TLC and upon completion
2
the solid mixture was taken in CHCl3 and the resulting suspension washed with water. The organic part was dried (MgSO ) and the solvent
4
evaporated in vacuo. The products in the case of the ketoximes 1a–1d were purified by crystallisation (EtOAc–ligroin), and identified by their
1
melting points and spectra (IR, 300 MHz H NMR and MS). In the case of camphor oxime 1e and the aromatic aldoximes 3, the reaction
1
mixtures were analysed by GC–MS to give the product ratios, followed by IR and 300 MHz H NMR which indicated the presence of the
reported products.

S. Chandrasekhar, K. Gopalaiah / Tetrahedron Letters 43 (2002) 2455–2457
2457
2
containing B O3 rings is partially represented). The
References
3
activated ester 7 undergoes Beckmann rearrangement,
the resulting incipient nitrilium ion 8 being recaptured
by the metaborate counterion at the carbon atom to
form the imino-metaborate 9, which is hydrolysed to
the final amide/lactam upon work-up. (Alternatively, 8
may also react with the water produced during the
formation of 7; in either case, boric acid is presumed to
be the final by-product after work up. Also, although
1. Chandrasekhar, S.; Gopalaiah, K. Tetrahedron Lett. 2001,
42, 8123–8125.
2. Greenwood, N. N.; Earnshaw, A. A. Chemistry of the
Elements; Pergamon Press: Oxford, 1984; pp. 230–231.
3. Smith, R. A.; McBroom, R. B. In Kirk-Othmer Encyclope-
dia of Chemical Technology; 4th ed., Kroschwitz, J. I.;
Howe-Grant, M., Eds.; New York: John Wiley, 1992; Vol.
4, p. 377.
‘
3
MBA’ apparently plays a catalytic role, a minimum of
molar equivalents of it vis-a-vis the oxime was found
4. The Merck Index, 12th ed.; Budavari, S., Ed.; Merck:
Whitehouse Station, NJ, 1996; p. 219.
to be required in practice.)
5
. Smith, B. D.; Hughes, M. P. In Encyclopedia of Reagents
for Organic Synthesis; Paquette, L. A., Ed.; John Wiley:
Chichester, 1995; Vol. 1, pp. 644–645.
There are two previous reports of the Beckmann rear-
rangement having been effected by boric acid or its
anhydrides, both involving the high temperature con-
version of cyclohexanone oxime to caprolactam with
6. Kemp, W. Organic Spectroscopy, 3rd ed.; Macmillan:
London, 1991; p. 79.
1
0
either boric acid in a solvent (e.g. tetralin) or B O on
7. Gawley, R. E. Org. React. (N.Y.) 1988, 35, 26–27.
8. Suginome, H.; Furakawa, K.; Orito, K. J. Chem. Soc.,
Perkin Trans. 1 1991, 917–921 and references cited therein.
9. Krow, G. R.; Szczepanski, S. Tetrahedron Lett. 1980, 21,
4593–4596.
10. Yura, S.; Horiguchi, K. Japan Patent (to Honshu Chemical
Industry Co., Ltd.) 75 03,317 (Cl. C 07d), 1975; Chem.
Abstr. 1975, 82, 156979e.
2
3
1
1
‡
a reactor bed (optimally at 600 K); interestingly in
the latter case, a role for metaboric acid was indicated
by the need for a small amount of water to be present.
The present procedure employs relatively low tempera-
tures and no solvent; the simple formation of the
reagent (metaboric acid) from the inexpensive boric
acid, the exceedingly simple work-up and the fact that
the boric acid by-product is relatively harmless are also
noteworthy. Thus, simplicity and low toxicity mark the
present procedure.
11. Immel, O.; Schwartz, H. H.; Starke, K.; Swodenk, W.
Chem.-Ing.-Tech. 1984, 56, 612–613; Chem. Abstr. 1984,
101, 152376u.
‡
We are grateful to a referee for kindly bringing these reports to our
attention.