The titanocene-catalyzed reduction of acetamides to tertiary amines by PhMeSiH2

Article abstract of DOI:10.1139/v04-063

A variety of acetamide derivatives are reduced in excellent yields to tertiary amines by PhMeSiH₂ in the presence of Cp₂TiX ₂ (X = F or Me) catalysts. The reactions are very clean at 80 °C. At room temperature a secondary reaction, hydrogenolysis of the C(O)-N bond, intervenes and reduces the chemoselectivity. Nevertheless, this chemistry provides a simple methodology for the amide/alkylamine transformation using inexpensive, commercially available reagents.

Full text of DOI:10.1139/v04-063

1244
The titanocene-catalyzed reduction of acetamides
to tertiary amines by PhMeSiH2
Kumaravel Selvakumar, Kesamreddy Rangareddy, and John F. Harrod
Abstract: A variety of acetamide derivatives are reduced in excellent yields to tertiary amines by PhMeSiH2 in the
presence of Cp2TiX2 (X = F or Me) catalysts. The reactions are very clean at 80 °C. At room temperature a secondary
reaction, hydrogenolysis of the C(O)—N bond, intervenes and reduces the chemoselectivity. Nevertheless, this chemis-
try provides a simple methodology for the amide/alkylamine transformation using inexpensive, commercially available
reagents.
Key words: amides, reduction, secondary amides, methylphenylsilane, titanocene, catalysis.
Résumé : Divers dérivés de l’acétamide sont réduits avec d’excellents rendements en amines tertiaires par le PhMe-
SiH2 en présence de catalyseurs du type Cp2TiX2 (X = F ou Me). Les réactions sont très propres à 80 °C. À la tempé-
rature ambiante la chimiosélectivité est réduite par l’intervention d’une réaction secondaire, l’hydrogénolyse de la liai-
son C(O)—N. Néanmoins, cette chimie fournit une méthodologie simple pour la transformation d’amides en
alkylamines faisant appel à des réactifs peu coûteux et commercialement disponibles.
Mots clés : amides, réduction, amides secondaires, méthylphénylsilane, titanocène, catalyse.
[Traduit par la Rédaction] Selvakumar et al. 1248
Introduction
Si—C bonds) have never been observed as products of the
reaction of amides under hydrosilation conditions.
The use of hydrosilation chemistry to achieve the reduc-
tion of carbonyl groups has received a great deal of attention
(1–5). Many studies of ketone hydrosilation have been di-
rected towards enantioselective synthesis of carbinols, and
excellent results have been achieved with both later and
early transition metal homogeneous catalysts (6, 7). Facile
hydrosilations of esters and lactones have also been reported
(8, 9). Among the early transition metal catalysts, titanocene
derivatives are the most effective with respect to both rate
and selectivity for all of these reactions.
In the present paper, we report results for the catalytic re-
duction of acetamides, under hydrosilation conditions, using
titanocene-based catalysts. Although there is still much
scope for improving the specificity and yields of catalytic
amide reduction reactions, the present work shows that ap-
propriate catalytic hydrosilation conditions can be useful for
reduction of tertiary amides to tertiary amines.
Results
Carboxamides are generally more resistant to reduction
than ketones and esters. Nevertheless, the conversion of
amides to the corresponding amines is an important transfor-
mation in organic synthetic methodology. Catalytic hydroge-
nation of amides normally requires high pressures and
elevated temperatures (10). Consequently, stoichiometric re-
duction using expensive reactive main group metal hydride
complexes is the usual method of choice (11–15).
The reaction of acetamide derivatives with PhMeSiH2 at
moderate temperature in the presence of catalytic amounts
of Cp2TiX2 (Cp = η-C5H5; X = F or Me) takes place as
shown in eq. [1].
Literature on the hydrosilation of carboxamides is sparse
(16–19). Kuwano et al. (17) and Igarashi and Fuchikami
(18) reported efficient reduction of aliphatic amides to
amines by catalytic hydrosilation, using late transition metal
catalysts. In a preliminary communication (19), we described
an unusal titanocene-catalyzed deoxygenation/coupling of
aromatic amides to give 1,2-diaminoethanes. Classical hy-
drosilation products (i.e., those containing either Si—OR or
[1]
Received 23 October 2003. Published on the NRC Research Press Web site at http://canjchem.nrc.ca on 19 October 2004.
K. Selvakumar, K. Rangareddy, and J.F. Harrod.1 Chemistry Department, McGill University, 801 Sherbrooke St. W., Montreal,
QC H3A 2K6, Canada.
1Corresponding author. (e-mail: john.harrod@mcgill.ca).
Can. J. Chem. 82: 1244–1248 (2004)
doi: 10.1139/V04-063
© 2004 NRC Canada

Selvakumar et al.
1245
Table 1. Results of Ti-catalyzed reduction of acetamides at
80 °C (reaction [1]).
Entry
1
R1
R2
R3
Method
Yield (%)
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Et
Ph
Ph
Ph
Ph
Ph
Ph
Bn
A
B
A
B
A
B
A
A
93
89
77
83
89
92
91
76a
2
3
[2]
Et
Bn
Bn
Bn
Me
4
5
3,4-(OMe)2Ph
6
7
8
9
Bn
Me
Ph
Ph
Ph
A
A
A
A
40
90
92
cases and complicates purification of the tertiary amine
product. In some cases, this reaction becomes a major con-
sumer of reactant (Table 2, entries 1, 3, and 7), and in one
case (Table 2, entry 10) it is the only reaction observed. In
the latter case, if the reaction is carried out in a closed tube,
ca. 1 equiv. of acetaldehyde is also detected in solution by
1H NMR. In other cases, where secondary amine is pro-
duced in a significant amount, the characteristic quartet of
Me
Me
Me
4-(OMe)Bn
Furfuryl
4-(NO2)Ph
b
Me
—
10
11
12
1-Acetylindoline
1-Phenyl-2-pyrrolidinone
N,N′-Dimethylacrylamide
A
A
88
96
94 (polymer)
Note: Method A, Cp2TiF2 catalyst; method B, Cp2TiMe2 catalyst. Yields
are given for isolated compounds.
1
acetaldehyde is also detected in the H NMR spectrum.
a10% secondary amine formed as deacetylated product.
bNo reaction.
Catalyst concentration also has a significant influence on
the relative contributions of reactions [1] and [2] at room
temperature (Table 2, entries 1–6). A similar effect was de-
tected at 80 °C. At a catalyst loading of 10 mol%, the reduc-
tion of N-benzylacetanilide (Table 1, entry 3) to the tertiary
amine occurs with high chemoselectivity. At 5 mol% cata-
lyst concentration, both the rate of reduction and the
chemoselectivity are similar to those at 10 mol% catalyst.
However, at 2 mol% catalyst concentration, the reaction is
slow (1.5 h for complete conversion) and N-benzylaniline
(24%) is formed along with N-ethyl-N-benzylaniline (76%).
At 80 °C this reaction occurs rapidly and gives excellent
yields of the tertiary amine, for a range of substituents R1,
R2, and R3. This product results from reduction and deoxy-
genation of the carbonyl group and contrasts with the cou-
pling, to give vicinal diamines, observed with aromatic
amides (19). The present results are very similar to those re-
ported for reactions catalyzed by complexes of Rh (17) and
Ru (18). There is, however, a considerable price advantage
in using the relatively inexpensive Cp2TiF2.
The siloxane product was identified by 1H NMR as a typi-
cal mixture of H-terminated oligo(phenylmethylsiloxanes).
The resonances occur at 0.2–0.5 ppm (CH3; broad massif or
many singlet and doublet peaks), 4.6–4.8 ppm (Si–H; over-
lapping quartets), and 6.6–7.5 ppm (Ph; broad massif). The
mass spectrum of a crude extract of the siloxane product re-
corded by chemical ionization mass spectrometry (CI-MS)
showed the presence of oligomers of n up to a value of ca.
10. The siloxane product was not studied further.
The reaction shows some tolerance for functional groups,
as exemplified by entries 5, 7, and 9 in Table 1. Reduction
of secondary amides is very slow compared to that of N,N-
disubstituted acetamides. For example, under similar condi-
tions, acetanilide undergoes reduction to give a small
amount of N-ethylaniline and aniline along with unreacted
acetanilide. Incomplete conversion is observed even after
15 h at 80 °C. N,N-Dimethylacrylamide underwent rapid
polymerization under the conditions used in these experi-
ments, probably due to free radical initiation, and there was
no evidence for reduction of the amide function.
The reaction mechanism
Kuwano et al. (17) suggested a mechanistic cycle for the
Rh(I)-catalyzed reaction [1] involving a series of conven-
tional two-electron oxidative addition/elimination steps, in
conjunction with a σ-bond metathesis step. In principle, the
mechanism of the titanocene-catalyzed reaction could be
analogous to that proposed for Rh. This would require a se-
quence in which the Ti cycles alternately through Ti(II) and
Ti(IV). Although such a two-electron cycle is not unreason-
able for titanocene, a cycle involving TiIIIH and TiIIISi is, in
our opinion, more likely. The facile disproportionation of
titanocene(III) to titanocene(II) and titanocene(IV) and vice
versa makes the distinction between these two possibilities
very difficult to draw, but there is ample evidence that all
three oxidation states are chemically available in
Cp2TiMe2/silane reaction mixtures (21). A tentative set of
pathways that explain the observed chemistry is shown in
Scheme 1. In this scheme, the silane functions as a source of
hydride to generate Ti–H (step a) and as a receptor for O)
but does not give rise to silylated organic products. Addition
of Ti–H across the C=O bond (step b) generates the highly
functionalized alkoxytitanocene intermediate, 1. This inter-
mediate then undergoes reactions to the various observed
products, whose relative importance depends on the substitu-
ents on the substrate, on the catalyst concentration, and on
the temperature. The various titanocene derivatives produced
in the reactions of Scheme 1 will be recycled by appropriate
σ-bond metathesis reactions with PhMeSiH2.
There is little difference in the performance of Cp2TiF2
and Cp2TiMe2, but given the long-term stability of the fluo-
ride, its commercial availability (or ease of preparation from
the commercially available Cp2TiCl2), and its insensitivity to
air and humidity, it is the preferred catalyst (20).
At room temperature, reactivity remains high (Table 2),
but a side reaction involving hydrogenolysis of the C(O)—N
bond (eq. [2]) lowers the efficiency of the reaction in certain
© 2004 NRC Canada

1246
Can. J. Chem. Vol. 82, 2004
Table 2. Results of Ti-catalyzed reduction of acetamides at 25 °C.
Product distribution (%)
Catalyst concn.
(mol%)
Tertiary
amine
Secondary
amine
Entry
R1
R2
R3
1
2
3
4
5
6
7
8
9
2
5
10
2
64
88
93
36
12
7
40
—
—
32
6
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Me
Et
Et
Et
Bn
Bn
Ph
Me
Ph
Me
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Bn
60
5
100
100
68
94
71
10
10
10
10
10
10
10
Furfuryl
2-Py
4-(OMe)Bn
3,4-(OMe)2Ph
29
100
23
15
10
11
12
—
77
85
a
a
13
Me
Me
4-(NO2)Ph
10
—
—
14
15
1-Acetylindoline
1-Phenyl-2-pyrrolidinone
10
2
100
100
—
—
1
Note: Ratios of the products were determined by H NMR Products were not isolated. Amines were characterized
by comparison of the NMR spectra with those of isolated products (Table 1) or commercial samples.
aNo reaction.
Scheme 1. A [Cp2TiH]-mediated reaction sequence for the reduction, deoxygenation, and coupling of amides in the presence of
PhMeSiH2.
A possible mechanism for C—N bond cleavage is shown
in Scheme 1. The favoring of the C—N bond cleavage reac-
tion, relative to that of the C—O bond cleavage, at lower
catalyst concentrations (Table 2, entries 1–6) could be due to
the former involving a rate law that is higher order in cata-
lyst concentration than is the rate law for the latter. Such a
situation could arise if the recycling reactions of the
titanocene products of reactions e and f in Scheme 1 are
slow enough to contribute to the overall rate. Given the
known tendency of titanocene(III) species to form coordina-
tion complexes (21–23), intramolecular coordination of the
amino group to the Ti is a reasonable possibility. The result-
ing complex could then decompose in a unimolecular pro-
cess to give the aldehyde and Cp2TiNR2. The latter could
then undergo a σ-bond metathesis with Si–H to generate Ti–
Si and give the secondary amine product. Finally, Ti–Si can
then be converted by a σ-bond metathesis step to Ti–H and
the silane dimer.
© 2004 NRC Canada

Selvakumar et al.
1247
1
Buchwald and co-workers (24) have described a synthesis
of aldehydes from acetamide derivatives by reaction with
Ti(O-i-Pr)4/Ph2SiH2. In those reactions, the primary product
was found to be the enamine, which was then converted to
aldehyde by mild acid hydrolysis. With our system, the alde-
hyde is the direct product in entry 10 of Table 2, and no in-
termediate vinylamine was detected. In the other reactions
where significant amounts of the secondary amine were pro-
duced, acetaldehyde was also detected, but not vinylamine.
However, the complexity of the reaction products in the lat-
ter cases prevented the definitive exclusion of the possible
presence of vinylamine.
We attribute the reactivity difference between aromatic
amides and acetamide derivatives to the lower stability of
the radical intermediate in step d when R is an alkyl, rather
than an aromatic, substituent. Detailing the mechanisms of
these interesting reactions is the subject of continuing stud-
ies.
N-Methyl-N-(2-phenylethyl)aniline (Table 1, entry 6) H
NMR: 2.82 (s, 3H), 2.78 (t, J = 10.04 Hz, 2H), 3.5 (t, J =
10.04 Hz, 2H), 6.64–6.69 (m, 4H), and 7.13–7.24 (m, 6H);
13C NMR: 33.72 (N–CH3), 38.02 (N–CH2), 54.99 (CH2),
112.33, 116.36, 126.43, and 129.04 (aromatic).
N-Ethyl-N-phenyl-(p-methoxy)benzylamine (Table 1,
entry 7) H NMR: 1.19 (t, J = 6.8 Hz, 3H), 3.44 (q, J =
1
6.8 Hz, 2H) 3.79 (s, 3H), 4.56 (s, 2H), and 6.68–7.19 (m,
9H)
1
N-Ethyl-N-(2-furoylmethyl)aniline (Table 1, entry 8) H
NMR: 1.18 (t, J = 7.1 Hz, 3H), 3.49 (q, J = 7.1 Hz, 2H),
4.58 (s, 2H), 6.69–6.81 (m, 3H), and 7.2–7.4 (m, 5H).
N-Ethylindoline (Table 1, entry 10) 1H NMR: 1.23 (t, J =
10.01 Hz, 3H), 3.02 (t, J = 10.01 Hz, 2H), 3.20 (t, J =
9.2 Hz, 2H), 3.37 (t, J = 9.2 Hz, 2H), 6.55 (d, J = 10.4 Hz,
1H), 6.72 (d, J = 10.4 Hz, 1H), and 7.09 (t, J = 9.2 Hz, 2H);
13C NMR: 28.79 (CH3), 43.50 (CH2), 52.57 (N–CH2), 97.76,
117.89, 124.6, and127.59 (aromatic).
1
N-Phenylpyrrolidine (Table 1, entry 11): H NMR 1.98–
2.07 (m, 4H), 3.41 (t, J = 7.2 Hz, 4H), 6.58 (m, 2H), and
7.21 (m, 3H).
Experimental
N,N′-Dimethylpolyacrylamide (Table 1, entry 12): 13C
NMR (CP-MAS): 37.5 (CH3), 69.02, 77.81, 84.27, 128.2,
(CH(CO)), and 178.03 (CO).
Materials
All of the reactant amides were either obtained commer-
cially from Aldrich Chemical Co. or synthesized by standard
methods. Cp2TiF2 (20) and Cp2TiMe2 (25) were prepared ac-
cording to previously described methods.
Acknowledgements
A typical procedure for the synthesis of tertiary amines is
the following (Table 1, method A). N-Methylacetanilide
(149 mg, 1 mmol), Cp2TiF2 (21 mg, 0.1 mmol), methyl-
phenylsilane (0.28 mL, 2 mmol), and toluene (1 mL) were
heated at 80 °C for 30 min in a Schlenk tube. After the reac-
tion mixture had cooled to room temperature, ether (10 mL)
was added and the solution was extracted with 1 mol/L HCl
solution. Evaporation of the depleted ether solution gave an
oily residue, which was shown by1H NMR to consist of a
mixture of oligo(phenyl-methylsiloxanes) and catalyst resi-
dues. The acid extract was neutralized with 3 mol/L KOH
and extracted with ether. The ether layer was dried with an-
hydrous MgSO4 and evaporated to give analytically pure N-
methyl-N-ethylaniline. In method B of Table 1, Cp2TiMe2
was used instead of Cp2TiF2. The identity of all the products
We are grateful to the Natural Sciences and Engineering
Research Council of Canada (NSERC) and to the Fonds
FCAR (Québec) for their generous financial support.
References
1. I. Ojima. In The chemistry of organic silicon compounds.
Edited by S. Patai and Z. Rappoport. John Wiley & Sons, New
York. 1989. p. 1479, and refs. therein.
2. I. Ojima, Z. Li, and J. Zhu. In The chemistry of organic silicon
compounds. Edited by Z. Rappoport and Y. Apeloig. John
Wiley & Sons, New York. 1998. p. 1687.
3. J.F. Harrod. In Encyclopedia of inorganic chemistry. Vol. 3.
Edited by R.B. King. John Wiley & Sons, London. 1994.
p. 1486.
4. H. Brunner. In Transition metals in organic synthesis. Vol. 2.
Edited by M. Beller and C. Bolm, Wiley-VCH, Weinheim.
1998. p. 131.
5. R.L. Halterman. Chem. Rev. 92, 965 (1992).
6. (a) I. Ojima (Editor). Catalytic asymmetric synthesis. VCH
Publishers, New York. 1993; (b) N. Clos and C. Bastos. Tetra-
hedron, 45, 6901 (1989); (c) I. Ojima and K. Hirai. In Asym-
metric synthesis. Vol. 5. Edited by J.D. Morrison. Academic
Press, Orlando. 1985. p. 103; (d) I. Ojima, K. Yamamoto, and
M. Kumada. In Aspects of homogeneous catalysis. Vol. 3.
Edited by R. Ugo. Reidel, Dordrecht. 1977. p. 185.
7. (a) M.B. Carter, B. Schiøtt, A. Gutiérrez, and S.L. Buchwald.
J. Am. Chem. Soc. 116, 11 667 (1994); (b) S. Xin and J.F.
Harrod. Can. J. Chem. 73, 999 (1995).
1
listed in Table 1 was determined by comparison of H NMR
data to literature values and to those of authentic samples.
N-Ethyl-N-methylaniline (Table 1, entry 1) 1H NMR
(ppm): 1.13 (t, J = 7.2 Hz, 3H), 3.42 (q, J = 7.2 Hz, 2H),
2.91 (s, 3H), 6.70 (m, 2H), and 7.26 (m, 3H).
1
N,N-Diethylaniline (Table 1, entry 2) H NMR (ppm):
1.06 (t, J = 7.2 Hz, 6H), 3.26 (q, J = 7.2 Hz, 4H), 6.61 (m,
2H), and 7.14 (q, J = 4.8 Hz, 3H)
N-Benzyl-N-ethylaniline (Table 1, entry 3) 1H NMR:
1.30 (t, J = 9.6 Hz, 3H), 3.56 (q, J = 9.6 Hz, 2H), 4.6 (s,
2H), 6.76 (m, 4H), and 7.43 (m, 6H); 13C NMR: 45.43
(CH3), 54.19 (N–CH2), 112.4, 116.31, 126.85, 129.5, and
139.58 (aromatic).
N,N-Dibenzylethylamine (Table 1, entry 4) 1H NMR:
1.07 (t, J = 9.2 Hz, 3H), 2.50 (q, J = 9.2 Hz, 2H), 3.57 (s,
4H), and 7.22–7.52 (m, 10H).
8. (a) M.T. Reding and S.L. Buchwald. J. Org. Chem. 60, 7884
(1995); (b) K.J. Barr, S.C. Berk, and S.L. Buchwald. J. Org.
Chem. 59, 4323 (1994).
N-Methyl-N-ethyl-3,4-dimethoxyaniline (Table 1, entry
9. X. Verdaguer, M.C. Hansen, S.C. Berk, and S.L. Buchwald. J.
1
5) H NMR: 1.17 (t, J = 7.4 Hz, 3H), 2.79–3.18 (m, 6H),
Org. Chem. 62, 8522 (1997).
3.78 (q, J = 7.4 Hz, 2H), 3.85 (t, J = 0.9 Hz, 3H), and 6.57–
6.92 (m, 3H).
10. C. Hirosawa, N. Wakasa, and T. Fuchikami, Tetrahedron Lett.
37, 6749 (1996).
© 2004 NRC Canada

1248
Can. J. Chem. Vol. 82, 2004
11. W.G. Brown. In Organic reactions. Vol. VI. Edited by R. Ad-
ams, H. Adkins, F.C. McGrew, A.H. Blatt, C. Niemann, A.C.
Cope, and H.R. Snyder. John Wiley & Sons, New York. 1951.
pp. 469–509.
16. R.A. Benkeser, G.S. Li, and E.C. Mozdzen. J. Organometal.
Chem. 178, 21 (1979).
17. R. Kuwano, M. Takahashi, and Y. Ito. Tetrahedron Lett. 39,
1017 (1998).
12. A.C. Cope and E. Cieganek. In Organic synthesis. Vol. IV.
Edited by N. Rabjohn, R.T. Arnold, N.J. Leonard, T.L. Cairns,
C.C. Price, J. Cason, R.S. Schreiber, A.C. Cope, J.C. Sheehan,
W.S. Johnson, and M. Tischle. John Wiley & Sons, New York.
1963. p. 339.
18. M. Igarashi and T. Fuchikami. Tetrahedron Lett. 42, 1945 (2001).
19. K. Selvakumar and J.F. Harrod. Angew. Chem. Int. Ed. 40(11),
2129 (2001).
20. X. Verdaguer, E.W. Udo, M.T. Reding, and S.L. Buchwald. J.
Am. Chem. Soc. 118, 6784 (1996).
13. A.C. Cope and E. Cieganek. In Organic synthesis. Vol. IV.
Edited by N. Rabjohn, R.T. Arnold, N.J. Leonard, T.L. Cairns,
C.C. Price, J. Cason, R.S. Schreiber, A.C. Cope, J.C. Sheehan,
W.S. Johnson, and M. Tischle. John Wiley & Sons, New York.
1963. p. 354.
14. W.G. Brown and P. Heim. J. Am. Chem. Soc. 86, 3566 (1964).
15. Z.B. Papanastassiou and R.J. Brun. J. Org. Chem. 29, 2870
(1964).
21. J.F. Harrod. Coord. Chem. Rev. 206, 493 (2000).
22. L. Hao, H.G. Woo, A.M. Lebuis, E. Samuel, and J.F. Harrod.
J. Chem. Soc., Chem. Commun. 2013 (1998).
23. L. Hao and J.F. Harrod. Inorg. Chem. Commun. 2, 191 (1999).
24. S. Bower, K.A. Kreutzer, and S.L. Buchwald. Angew. Chem.
108, 1662 (1996); Angew. Chem. Int. Ed. 35, 1515 (1996).
25. K. Claus and K. Bestian. Justus Liebigs Ann. Chem. 654, 8
(1962).
© 2004 NRC Canada