A one-pot oxidation-imine formation-reduction route from alcohols to amines using manganese dioxide-sodium borohydride: The synthesis of naftifine

Article abstract of DOI:10.1016/S0040-4039(02)01722-7

A new procedure for the one-pot conversion of alcohols into amines is described which utilises manganese dioxide in the presence of sodium borohydride; the scope of this process is outlined, as is its application to the preparation of the topical antifungal agent, naftifine.

Full text of DOI:10.1016/S0040-4039(02)01722-7

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 43 (2002) 7337–7340
A one-pot oxidation–imine formation–reduction route from
alcohols to amines using manganese dioxide–sodium borohydride:
the synthesis of naftifine
Hisashi Kanno and Richard J. K. Taylor*
Department of Chemistry, University of York, Heslington, York YO10 5DD, UK
Received 17 July 2002; accepted 16 August 2002
Abstract—A new procedure for the one-pot conversion of alcohols into amines is described which utilises manganese dioxide in
the presence of sodium borohydride; the scope of this process is outlined, as is its application to the preparation of the topical
antifungal agent, naftifine. © 2002 Elsevier Science Ltd. All rights reserved.
The direct conversion of aldehydes into amines using the
reductive amination procedure is an extremely important
functional group transformation. We recently reported
a related new process which allows the one-pot conver-
sion of alcohols into amines via an in situ oxidation–
imine formation–reduction sequence using manganese
gave encouraging results but was not investigated in
detail in view of the cost factor. Instead, attention was
concentrated on the use of non-supported reductants
that might be compatible with manganese dioxide.
Decaborane, poly(methylhydrosiloxane), sodium acet-
oxyborohydride and sodium cyanoborohydride were
unsuccessful. However, success was achieved with the use
of sodium borohydride in dichloromethane (Eq. (2)).
1
2
dioxide and polymer-supported cyanoborohydride
3
(
PSCBH; Eq. (1)). This method, which utilises the
combination of a heterogeneous oxidant and a heteroge-
neous reductant, has the advantage that the intermediate
aldehydes and imines do not require isolation. The
heterogeneity of the reactants also ensures a straightfor-
ward work-up procedure.
Thus, addition of benzyl alcohol and iso-butylamine to
a mixture of manganese dioxide and sodium borohydride
in dichloromethane at rt followed by heating for 18 h,
cooling and then addition of methanol and work-up gave
N-benzyl(iso-propyl)amine in 89% isolated yield. The
sodium borohydride is essentially insoluble in
dichloromethane, and although some reduction of the
intermediate imine to the corresponding amine is
observed before the addition of methanol, the addition
The drawbacks of this procedure include the cost of the
resin-bound reductant, and the fact that cyanohydrin-
derived by-products arise due to cyanide leaching from
2
,4
the resin. To overcome these problems other reduc5-
6
tants were explored. Polymer-supported borohydride
of the alcoholic solvent speeds up the reduction step.
(
(
1)
2)
*
0
Corresponding author. E-mail: rjkt1@york.ac.uk
040-4039/02/$ - see front matter © 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(02)01722-7

7
338
H. Kanno, R. J. K. Taylor / Tetrahedron Letters 43 (2002) 7337–7340
In view of this success, we went on to study the in situ
oxidation–imination–reduction of a range of substi-
tuted benzylic, allylic and propargylic alcohols with
vii) and a propargylic alcohol example (entry viii). As
before, unactivated alcohols could not be successfully
employed in this sequence.
2
7
primary amines, as shown in Table 1. As can be seen
(
entries
i–v)
the
oxidation–imination–reduction
Although iso-butylamine was used in most examples,
sequence proceeded efficiently with both electron rich
and electron deficient benzyl alcohol derivatives. In
addition, success was achieved using 1-naphthalene
methanol (entry vi), an allylic alcohol example (entry
we also demonstrated that other primary amines could
8
be employed (entries ii and iii; also see later). Finally,
we established that secondary amines do not usually
give good yields in the MnO –NaBH procedure (in
2
4
a
Table 1. One-pot oxidation–imine formation–reduction

H. Kanno, R. J. K. Taylor / Tetrahedron Letters 43 (2002) 7337–7340
7339
2
contrast to the procedure using PSCBH ), although
yields of around 30% were observed in some cases (see
later, Scheme 3).
Further approaches were explored to avoid the methyl-
ation step and generate naftifine directly. Thus, in an
analogous manner to the earlier study, 1-naphthalene-
12
methanol and amine 5 were subjected to the standard
oxidation–imination–reduction sequence (Scheme 2).
Unfortunately, as was anticipated from the preliminary
studies, no naftifine 4 was produced. Use of our poly-
We next went on to utilise this methodology to prepare
9,10
the topical antifungal agent naftifine 4,
as shown in
Scheme 1. Thus, 1-(aminomethyl)naphthalene 1 and
cinnamyl alcohol 2 produced the secondary amine 3 in
This was then methylated using a
published procedure to give naftifine 4 in 91% yield
with fully consistent spectroscopic data [e.g. l
CDCl ) 2.29 (3H, s), 3.30 (2H, d, J 6.5 Hz); lit. lH
CDCl ) 2.32 (3H, s), 3.32 (2H, d, J 6.5 Hz)].
2
mer-supported cyanoborohydride (PSCBH) method,
1
1
8
6% yield.
however, produced naftifine 4 in an acceptable 62%
yield.
1
0
H
9b
(
The alternative coupling mode was also studied
(Scheme 3). The use of the sodium borohydride method
3
(
3
Scheme 1.
Scheme 2.
Scheme 3.

7
340
H. Kanno, R. J. K. Taylor / Tetrahedron Letters 43 (2002) 7337–7340
with the naphthyl amine 6 and cinnamyl alcohol pro-
duced naftifine 4 in 26% yield. This yield was improved
to 72% with the polymer-supported cyanoborohydride
method. These results confirm the supremacy of the
PSCBH method when using secondary amines as
reactants.
Chem. Commun. 1977, 815; using BER and MnO with
2
i
BuNH2 and 4-methoxybenzyl alcohol in CH Cl2 fol-
2
lowed by the addition of methanol gave the expected
amine product in 82% yield.
6. If methanol is present from the start of the reaction no
conversion is observed.
2
7
8
9
. All products gave consistent spectroscopic data which
were comparable to published data where available; com-
pounds 3 and 4 were fully characterised (including high
field NMR spectroscopy and HRMS).
. tert-Butylamine was also employed in the reaction with
benzyl alcohol giving N-benzyl(tert-butyl)amine in 42%
yield, although the reaction was slower then normal and
unreacted starting material was recovered.
In summary, we have shown that activated alcohols
(
benzylic, allylic and propargylic) will undergo the man-
ganese dioxide/sodium borohydride-mediated oxida-
tion–imination reduction sequence with primary amines
to afford secondary amines in good to excellent yields.
This methodology has been applied to a short synthesis
of the topical antifungal agent naftifine 4. We have
established that secondary amines are not normally
good substrates for this one-pot procedure, but that in
certain circumstances tertiary amines can be prepared
in this way. We are currently optimising and expanding
the scope of this sequence and investigating its applica-
. (a) Review: Monk, J. P.; Brogden, R. N. Drugs 1991, 42,
659; (b) for a recent publication, see: Petasis, N. A.;
Akritopoulou, I. Tetrahedron Lett. 1993, 34, 583–586 and
references cited therein.
1
0. Loibner, H.; Pruckner, A.; St u¨ tz, A. Tetrahedron Lett.
984, 25, 2535–2536.
1. Preparation of N-(1-naphthylmethyl)-(E)-cinnamylamine
: Activated manganese dioxide (Aldrich 21,764–6, 0.870
g, 10 mmol) was added to a stirred solution of (E)-cin-
namyl alcohol (0.134 g, mmol), 1-(amino-
9
tions to other antifungal allyl amines and for the
1
synthesis of more complex target molecules.
1
3
Acknowledgements
1
methyl)naphthalene (0.314 g, 2 mmol) and sodium boro-
hydride (0.113 g, 3 mmol) in dichloromethane (20 mL)
,
containing 4 A molecular sieves (ca. 0.2 g). The mixture
We are grateful to the Kureha Chemical Industry Co.
Ltd for studentship support (H.K.).
was stirred at reflux for 17 h and then cooled in an ice
bath. Methanol (5 mL) was added to the reaction mixture
which was stirred for 10 min in an ice-bath and then 30
min at room temperature. The reaction mixture was
References
®
filtered through Celite washing well with EtOAc. The
1
2
3
. Tarasevich, V. A.; Kozlov, N. G. Russ. Chem. Rev. 1999,
8, 55–72 and references cited therein.
. Blackburn, L.; Taylor, R. J. K. Org. Lett. 2001, 3,
637–1639.
combined filtrates were concentrated and then ether (30
6
mL) and a saturated aq. NaHCO solution (30 mL) were
3
added to the residue. A standard ether work-up followed
by silica chromatography (EtOAc–petroleum ether, 2:1 to
EtOAc–MeOH, 10:1) gave the desired compound 3 (0.234
1
. For other routes from alcohols to amines, see: (a) Mura-
hashi, S.; Kondo, K.; Hakata, T. Tetrahedron Lett. 1982,
g, 86%) as a yellow solid, mp 39.5–40.5°C, Rf 0.38
2
3, 229–232; (b) Ohtani, B.; Nakagawa, K.; Nishimoto,
+
(
EtOAc–MeOH, 10:1); [HRMS (CI), MH : 274.159970.
S.; Kagiya, T. Chem. Lett. 1986, 1917–1920; (c) Gasparik,
V.; Dalla, V.; Decroix, B. Synlett 2002, 528–530; (d)
Wood, M. R.; Kim, J. Y.; Books, K. M. Tetrahedron
Lett. 2002, 43, 3887–3890 and references cited therein.
. Foot, J. S.; Kanno, H.; Giblin, G. M. P.; Taylor, R. J. K.
Synlett 2002, 1293–1295.
+
C H N requires MH , 274.159575 (1.4 ppm error)]
20 19
which gave consistent spectroscopic data (for hydrochlo-
ride salt, see: St u¨ tz, A.; Goergopoulos, A.; Granitzer, W.;
Petranyi, G.; Berney, D. J. Med. Chem. 1986, 29, 112–
4
5
125).
. For the preparation of borohydride exchange resin
12. Viallon, L.; Reinaud, O.; Capdevielle, W.; Maumy, M.
(
BER), see: Gibson, H. W.; Bailey, F. C. J. Chem. Soc.,
Tetrahedron 1996, 52, 13605–13614.