therefore supposed that NMO could play a dual role in our
envisaged oxidation procedure: (i) stabilization of the
aldehyde hydrate intermediate and (ii) co-oxidant to re-
amounts of water reduce its catalytic nature, thereby
decreasing the reaction rate. Thus, in addition to its
stabilizing effect on aldehyde hydrates, NMO H O also
seems to allow for the presence of water without adverse
effects on the overall catalytic process.
3
2
1
2
cycle the active Ru(VII) catalyst. Scheme 1 shows the
corresponding reaction pathway. The two oxidative steps
are believed to proceed via similar intermediates, Ru(VII)
esters A and B (Scheme 1). The efficiency of the overall
transformation relies on an effective formation and stabi-
2
Table 2. Influence of the Amount of NMO H O and Additional
1
4,15
3
lization of the aldehyde hydrate.
Water on the Oxidation of Octanol and Octanal to Octanoic
Acid
a
NMO
3
b
b
Scheme 1. Proposed Mechanism for the Direct Oxidation of
Primary Alcohols to Carboxylic Acids
2
H O
2
H O
time
(h)
yield
entry substrate
(equiv)
(equiv)
of acid (%)
c
1
2
3
4
5
6
7
octanol
octanol
octanol
octanol
octanol
octanal
octanal
3
5
24
4
44
d
87
d
7.5
10
3
0.8
1
99
c
d
94 (100)
c
7
3
28
c
1
3
41
c
3
3
75
a
Reaction conditions: substrate (1.0 equiv, 0.5 or 1.0 mmol scale),
b
2
TPAP (10 mol %), NMO H O, MeCN (0.25 M). Equivalents added
d
relative to the substrate. Isolated yield of octanoic acid. Conversion
3
c
determined by GCꢀFID.
Even though, in theory (with respect to its role as a co-
oxidant), only2 equivshouldbesufficientfor the oxidation
reaction, we found that 10 equiv of NMO H O was
3
2
We next applied our standard conditions to various
substrates containing different functionalities (Table 3).
Simple aliphatic alcohols are oxidized to the corresponding
carboxylic acids in good to excellent yields in short reaction
times (usually 1 h or less). Alcohols containing double and
triple bonds are alsoviable substrates with, for example, the
cis-double bond of 4-decen-1-ol staying untouched under
the reaction conditions (Table 3, entry 4). β-Branched
alcohols and substrates containing functionalities such as
epoxides, halides, or Boc-protected amines are oxidized to
the corresponding acids in excellent yields. Also, R- and β-
stereocenters remain intact (Table 3, entries 8 and 9).
The method can be employed for benzylic alcohols as
well. However, the efficiency strongly depends on the
nature of the substituents on the aromatic ring. Donor-
substituted benzylicalcoholsprovide onlymoderateyields,
whereas acceptor substituted benzylic alcohols give the
respective benzoic acid derivatives in excellent yields
required (see also Table 1) for the reaction to go to
1
6
completion using octanol as a representative model
substrate (Table 2, entries 1ꢀ4). A 94% yield of octanoic
acid was isolated when the reaction was performed in
1
7
acetonitrile. A slight excess of NMO H O (3 equiv) gave
2
3
only a moderate yield of the acid (44%, Table 2, entry 1).
The same trend was observed when octanal was used as the
substrate(Table2, entries6and 7). Thesedata indicatethat
a significant amount of hydrate-stabilizing agent has to be
added in order to produce a reasonable concentration of
the hydrate intermediate during the lifetime of the Ru(VII)
catalyst. Most importantly, we found that simply adding
water (7 equiv) along with 3 equiv of NMO H O signifi-
3
2
cantly diminished the yield of octanoic acid (28%, Table 2,
entry 5). This result is not surprising since it is known that
the presence of water reduces catalytic turnovers, an effect
that might be explained on the basis of the finding that
(
lytic in colloidal RuO formed during the reaction. Small
stoichiometric) TPAP oxidations are strongly autocata-
8
(
Table 3, entries 13ꢀ18). We have included the weak
results using para-donor-substituted benzylic alcohols
Table 3, entries 14 and 15) as these data not only show
1
2
(
(
14) This has been established for Cr(VI) oxidants: (a) Ro ꢁc ek, J.; Ng,
C.-S. J. Am. Chem. Soc. 1974, 96, 1522. (b) Ro ꢁc ek, J.; Ng, C.-S. J. Org.
Chem. 1973, 38, 3348.
the limitations of our direct oxidation protocol but are also
a negative result in support of the involvement of aldehyde
hydrates. Finally, we applied our method to more complex
substrates (Table 3, entries 19 and 20) including acetonide-
protected galactose which, after esterification of the crude
product using TMS-diazomethane, gave ester 22 in 66%
(
15) A radical mechanism or a BaeyerꢀVilliger-like process are also
possible: B €a ckvall, J.-E., Ed. Modern Oxidation Methods, 1st ed.; Wiley-
VCH: Weinheim, 2004. Another possible pathway may involve the nucleo-
philic addition of NMO to the aldehyde intermediate. The resulting species
ꢀ
þ
0
3
)] would then be oxidized to an active ester followed by
[
RCH(O )(ON R
hydrolysis. Detailed investigations are currently underway.
16) We also found that the amount of water relative to NMO could
be reduced. However, it was found to be more convenient to simply use
NMO H O. Details will be disclosed in a full account.
17) Similarly high yields were obtained when DCM, acetone, or
1
9
yield. The oxidation of an alcohol containing a chiral
(
3
2
(19) This starting material was generously provided by Prof. Dr. H.-U.
Reissig, Freie Universit €a t Berlin, Germany. For details on the synthesis of
this substrate, see: (a) Al-Harrasi, A.; Pfrengle, F.; Prisyazhnyuk, V.;
Yekta, S.; Ko oꢀ ꢁs , P.; Reissig, H.-U. Chem.;Eur. J. 2009, 15, 11632.
(b) Yekta, S.; Prisyazhnyuk, V.; Reissig, H.-U. Synlett 2007, 2069.
(c) Al-Harrasi, A.; Reissig, H.-U. Angew. Chem. 2005, 117, 6383.
Angew. Chem., Int. Ed. 2005, 44, 6227.
(
DMF was used. Low yields were obtained in solvents in which TPAP
and NMO were poorly soluble. See the Supporting Information for a
complete solvent screening.
18) Lee, D. G.; Wang, Z.; Chandler, W. D. J. Org. Chem. 1992,
7, 3276.
(
5
4
166
Org. Lett., Vol. 13, No. 16, 2011