Oxoammonium Salts
J . Org. Chem., Vol. 63, No. 25, 1998 9373
separated on a 30 m Carbowax column. Melting points were
measured on a Kofler hot stage apparatus and are corrected.
Unless noted, all of the substrates were commercial chemicals
and were used without purification. The CH2Cl2 was distilled
and stored over Na2SO4. The silica gel (40 µm for flash
chromatography, Scientific Adsorbants, Inc., Atlanta, GA) was
washed with CH2Cl2 and dried before use. The alumina was
Brockmann 1, basic, 150 mesh, obtained from Aldrich Chemi-
cal Co., Milwaukee, WI. Microanalyses were carried out by
Galbraith Laboratories, Memphis, TN, and NuMega Reso-
nance Labs, San Diego, CA. The 4-amino-2,2,6,6-tetrameth-
ylpiperidine was obtained from Fluka Chemical Co., Milwau-
kee, WI.
Mod ified P r ep a r a tion of 4-Acetyla m in o-2,2,6,6-tetr a -
m eth yl-1-p ip er id in oxy, 3.33 4-Amino-2,2,6,6-tetramethylpi-
peridine (220 g, 1.42 mol) was dissolved in 800 mL of dry ether
in a 3 L flask equipped with a very large magnetic stirring
bar. The mixture was cooled in an ice bath, and with vigorous
stirring, acetic anhydride (445 g, 4.36 mol) in 200 mL of dry
ether was added slowly over about 1 h. Some heat was given
off in the early stages, and there is a tendency for the product
to clump together. These clumps can be broken up with a glass
rod so that the final product is a free-flowing slurry. After the
anhydride addition, the slurry was stirred for 3 h, washed with
dry ether, and filtered to yield, after drying in a hood to
constant weight, 360 g (99%) of 4-acetylamino-2,2,6,6-tetra-
methylpiperidinium acetate.
The acetate (360 g, 1.41 mol) was dissolved in 2.5 L of water,
and the solution was made basic with 317 g of K2CO3 (2.3 mol),
added in small portions. Sodium tungstate (25.0 g, 0.075 mol)
and ethylenediaminetetraacetic acid tetrasodium salt (25 g,
0.060 mol) were added. To the slurry were added five succes-
sive 100-mL portions of 30% H2O2 (about 4.4 mol) at about
3-h intervals. A small amount of heat was given off, and the
mixture tended to foam. For this reason, a beaker or large
mouth container is desirable. After 3 days of stirring, the
mixture was filtered, and the orange precipitate was washed
once with ice water. The washing and filtrate were concen-
trated under vacuum to about half, and a second crop of solid
product, obtained on chilling, was collected by filtration. The
two solids were combined to give 295.0 g of 3 (98%, mp 145-
147 °C, lit. 147-148 °C).9 At other times, yields varied from
90 to 98%.
polymerized using oxoammonium chloride, presumably
through reaction with the enol.21
Easily hydrolyzable acetals are broken down under the
conditions of the oxidations, as are â-lactones (Table 2,
entries 3 and 4). The addition of sodium acetate, as in
PCC oxidations, helps some, but sodium acetate, being
slightly basic, causes problems as discussed above, as well
as adding some acetic acid to the product. An acetonide
does survive the oxidation nicely, and the benzyl alcohol
is very rapidly oxidized (Table 1, entry 12).
As far as other blocking groups for alcohols are
concerned, diphenyl-tert-butylsiloxy (DPTBS) groups are
quite stable to the reagent, although the smaller group,
dimethyl-tert-butylsiloxy group (DMTBS), is not (Tables
1 and 3). Ester groups would appear to be unreactive
since the reactions can be carried out in ethyl acetate,
and of course, the reagent itself contains an amide which
is completely stable.
Benzyl ethers have been reported to react with oxoam-
monium salts.2 However, this has not been a problem in
the oxidation of benzyl alcohols containing benzyloxy
groups. The alcohol oxidation is apparently much faster
than ether oxidation. Ketones have been reported to react
with oxoammonium salts to give diketones,2,3,5 but this
reaction seems to be very slow in CH2Cl2 and was not
encountered in our work.
Solven ts. The solvent of choice is CH2Cl2. The solu-
bilities of the oxidant and its reduced product are ideal;
it is easily evaporated to yield product; and it is inex-
pensive and easily purified. If the substrate is not soluble
in CH2Cl2, ethyl acetate is a poor second choice. The
1
reaction rates are reduced to less than
/
of those in
10
CH2Cl2 (as measured for benzyl alcohol and 2-octanol)
and are only slightly enhanced by silica gel. However, a
mixture of the two solvents (1:1) plus silica gel was used
satisfactorily for the oxidation of piperonyl alcohol (Table
1, entry 7). The isolation procedure is the same as that
for methylene chloride.
P r ep a r a t ion of Oxoa m m on iu m Sa lt P er ch lor a t e, 1.
The nitroxide (50 g, 0.23 mol) was slurried with 100 mL of
H2O, and 33.0 g of 70% HClO4 (0.23 mol) in 25 mL of H2O
was added as slowly as possible (about 1 h). Commercial bleach
(181.5 g, 5.25% NaOCl, 0.125 mol) was slowly added (about 2
h). The reaction mixture was cooled in ice and filtered, and
the yellow, crystalline precipitate was washed with ice-cold
5% NaHCO3 (100 mL) ice water (100 mL), and two 100-mL
portions of CH2Cl2. The solid was dried to constant weight in
air to yield 61.0 g (83%) of product, mp 172-174 °C (dec). The
melting point-decomposition point of this compound is some-
what variable, depending on crystal flaws, and can be from
168 to 175 °C. The assay against 2-octanol was 100%.34 The
analytical sample was recrystallized from H2O and melted at
177-178 °C35 Anal. Calcd for C11H21N2O6Cl: C, 42.24; H, 6.76;
N, 8.95. Found: C, 42.37; H, 6.64; N, 8.83.
The filtrates from the salt were combined, and the CH2Cl2
was separated. The aqueous solution was basified with NaH-
CO3 and stirred overnight with an excess of ethanol (>0.04
mol, ca. 2 g).36 The nitroxide that formed was extracted with
eight 20-mL portions of CH2Cl2.37 The solvent was dried
(MgSO4) and evaporated to give 6.02 g of nitroxide, 3. The
conversion of 3 to oxoammonium salt was 95%. The actual
yield from starting piperidine was about 80%, but the conver-
sion was about 95%.
Tetrahydrofuran is slowly oxidized to butyrolactone (as
shown by GC comparison to a known material) and would
be unsatisfactory except for very fast reactions. Acetone
also reacts slowly with oxoammonium salts.5b Water can
be used, but the solubilities of the oxidant and its reduced
product make product isolation more difficult. At least
in catalytic situations,2 the presence of water leads to the
formation of carboxylic acids rather than aldehydes,
probably through the oxidation of hydrates. Acetonitrile
seems to be quite unreactive to the oxidant, but again,
the solubilities are not very satisfactory for facile product
isolation.
Su m m a r y. We have prepared a stable, nonhygro-
scopic, and readily available oxoammonium salt and
described its reactions with various alcohols. The reac-
tions are colorimetric, proceed in high yields, and give
products of good purity, and the oxidant can be recycled.
However, as cited above and in the reviews,2,3 there are
many other reactions of oxoammonium salts that have
been explored only briefly. Perhaps the availability of 1
will allow a more thorough exploration of some of these,
as well as new reactions.
P r ep a r a tion of Oxoa m m on iu m Sa lt Tetr a flu or obo-
-
r a te, 4 (1, w ith BF 4 Ra th er th a n ClO4-). Using exactly
Exp er im en ta l Section
the same procedure as for 1, 4 was prepared in a yield of 67%.
The analytical sample was recrystallized from water and
melted at 193-194 °C. Anal. Calcd for C11H21N2O2BF4: C,
44.03; H, 7.04; N, 9.33. Found: C, 44.11; H, 6.91; N, 9.37.
Gen er a l P r oced u r es. Gas chromatography was carried out
on a 10 m, 0.53 mm, methyl silicone column using a thermal
conductivity detector. The geraniol-nerol compounds were