11996 J. Am. Chem. Soc., Vol. 122, No. 48, 2000
Communications to the Editor
Scheme 1
Scheme 3
Scheme 2
path a, the corresponding alcohol would be the expected product,
and if path b were favored, the amine would be the expected
product. The results from this experiment resulted in a 1:1 mixture
of alcohol and amine. This results supports path b in that the
formation of the alcohol can be explained by the presence of trace
water, whereas the amine formation can only be explained via
proposed path b. Furthermore, control experiments attempting the
reduction of the tertiary amide in the presence of tetrabutyl-
ammonium borohydride did not lead to a reduced product as
expected. Further literature precedence for path b comes from a
study carried out by Ganem et al. where secondary amides were
reduced to imines with Cp
2
Zr(H)Cl.11 This observation supports
1
4 and 15) by reduction of the diisopropyl amide and the
piperidinoamide. This was felt to be important, as some previous
methods to reduce sterically demanding tertiary amides to
aldehydes were unable to accomplish this task. The yields, in
these cases, were shown to be only slightly reduced. Of additional
particular interest is the selectiVe chemical reduction of an tertiary
amide in the presence of an ester functionality (entries 11 and
path b, in that the imine being formed from secondary amides
would be neutral and fairly stable and, hence, not as reactive to
water as the iminium ion intermediate, obtained in the reduction
from tertiary amides. An additional interesting point related to
the use of excess reagent is that no further reduction to the amine
was observed.
8
1
2). This selective conversion has no general precedence to the
Another experiment performed to further examine the proposed
9
18
best of our knowledge. As seen in these examples, Cp
2
Zr(H)Cl
mechanism was carried out using a H
2
O
quench as the final
under these conditions can reduce tertiary amides selectively,
giving very good yields of the corresponding aldehydes in short
reaction times.
A general procedure for this reaction is as follows: The
substrate is taken into 5 mL of anhydrous THF under argon. This
step (Scheme 3), the premise being that if path a were followed,
18
18
one would expect no O incorporation, whereas O incorporation
should be expected if path b were followed. The reaction was
1
8
carried out using an H
2
O
quench after reacting the substrate
Zr(H)Cl for 5 min
(p-methoxy-N,N-diethylbenzamide) and Cp
2
1
6
18
solution is then added to 1.5-2.0 equiv of Cp
2
Zr(H)Cl at room
(1:1 mixture of H
2
O
2
and H O ). The resulting product was
analyzed by 13C NMR and MS which showed incorporation of
temperature under argon, which elicits the desired conversion
within 15-30 min. Further workup by short path silica gel
chromatography (hexanes:ethyl acetate) of the concentrated
mixture affords the desired aldehydes in near quantitative yields.
Two major plausible routes through which this reaction may
proceed are shown in Scheme 2. Following path a, I may react
directly with the reagent via a direct hydride transfer proceeding
the label (carbonyl peaks at δ 191.25 and 191.22 in a 1:1 ratio,
+
and a 1:1 ratio for the pairs m/z 165, 163 (M ), and 164, 162,
+
12
(M - 1). Therefore, this experiment suggests that the proposed
reaction path b is more likely than path a.
In summary, the reduction of tertiary amides to aldehydes via
Cp Zr(H)Cl shows several distinct advantages. The reaction
2
through a tetrahedral intermediate, eventually losing the -NR
2
requires very short reaction time ( ∼15 min) and provides very
high yields of the aldehydes with good chemoselectivity. Fur-
thermore, the reaction does not require extensive workup proce-
dures, nor does it require scrupulously dry conditions, if the
aldehyde is the desired product. Additionally, substrate depen-
dence is minimal, and it is therefore expected that this method
will be useful for a wide variety of compounds. Further studies
are currently underway to more fully examine the scope of this
reaction.
group to give the aldehyde II. Alternatively, I can react via path
b to give intermediate III, which can then be reduced (intermedi-
ate IV) to give eventually intermediate V. The iminium ion
intermediate would be expected to be highly reactive, forming
the aldehyde immediately in the presence of water.
The lack of over-reduction of the amide to the alcohol, even
in the presence of excess (>2 equiv) Cp
2
Zr(H)Cl is supportive
of path b, since it is well known that aldehydes will react with
10
Cp
2
Zr(H)Cl to give the corresponding alcohol. To further
Acknowledgment. We thank Dr. Apurba Datta for helpful discussions
as well as the NIH and the NIH Predoctoral Training Grant GM-08545
for financial support.
examine this hypothesis, a tertiary amide was stirred with Cp
2
-
Zr(H)Cl and subsequently treated immediately with tetrabutyl-
ammonium borohydride. If the reaction would proceed through
JA002149G
(
8) Brown, H. C.; Tsukamoto, A. J. Am. Chem. Soc. 1964, 86, 1089-
095.
9) Hudlicky, M. Reductions in Organic Chemistry, 2nd ed.; American
Chemical Society: Washington, DC, 1996.
10) Majoral, J. P.; Zablocka, M.; Igau, A.; Cenac, N. Chem. Ber. 1996,
29, 879-886.
1
(11) Ganem, B.; Godfrey, A. G.; Schedler, D. J. J. Org. Chem. 1996, 61,
4115-4119.
(
(12) Control experiments treating the aldehyde (p-methoxybenzaldehyde)
1
8
(
with H
2
O
showed no incorporation even after 24 h of treatment; thus the
1
background level of incorporation appears minimal.