Kinetics of oxidation and comproportionation reactions involving an oxammonium ion, benzyl alcohol and methyltrioxorhenium(VII)

Article abstract of DOI:10.1002/(SICI)1097-4601(1999)31:5<381::AID-KIN7>3.0.CO;2-F

The reactions of 4-hydroxy-2,2,6,6-tetramethylpiperidinium N-oxide, an oxammonium ion abbreviated R₂NO⁺, have been studied. The previously unreported triflate salt was used in this study because the anions of the usual chloride and bromide salts can themselves be oxidized. Reactions between R₂NO⁺ and alcohols produce ketones and aldehydes; the rate constant for PhCH₂OH is 4.4×10^-3 L mol^-1 s^-1 in acetonitrile at 298 K. The immediate product is the hydroxylamine, R₂NOH, but its further comproportionation reaction with R₂NO⁺ yields the stable piperidinyl oxyl radical, R₂NO^·. The rate constant of this reaction is 1.78×10³ L mol^-1 s^-1 at 298 K. The possibility of using R₂NO⁺ and MTO as co-catalysts for the oxidation of alcohols was explored, but the competitive rates are such that the resultant is not particularly attractive.

Full text of DOI:10.1002/(SICI)1097-4601(1999)31:5<381::AID-KIN7>3.0.CO;2-F

Kinetics of Oxidation and
Comproportionation
Reactions Involving an
Oxammonium Ion, Benzyl
Alcohol and
Methyltrioxorhenium(VII)
TIMOTHY H. ZAUCHE, JAMES H. ESPENSON
Ames Laboratory and Department of Chemistry, Iowa State University, Ames, Iowa 50011
Received 19 October 1998; accepted 30 December 1998
ABSTRACT: The reactions of 4-hydroxy-2,2,6,6-tetramethylpiperidinium N-oxide, an oxam-
monium ion abbreviated R2NOϩ, have been studied. The previously unreported triflate salt
was used in this study because the anions of the usual chloride and bromide salts can them-
selves be oxidized. Reactions between R2NOϩ and alcohols produce ketones and aldehydes;
the rate constant for PhCH2OH is 4.4 ϫ 10Ϫ3 L molϪ1 sϪ1 in acetonitrile at 298 K. The immediate
product is the hydroxylamine, R2NOH, but its further comproportionation reaction with R2NOϩ
⅐
yields the stable piperidinyl oxyl radical, R2NO. The rate constant of this reaction is 1.78 ϫ
1
1
103 L molϪ sϪ at 298 K. The possibility of using R2NOϩ and MTO as co-catalysts for the
oxidation of alcohols was explored, but the competitive rates are such that the resultant is
not particularly attractive. ᭧ 1999 John Wiley & Sons, Inc. Int J Chem Kinet 31: 381–385, 1999
INTRODUCTION
of these studies. During most oxidations R2NOϩ is
converted to R2NOH without an intermediate piperi-
⅐
Oxammonium ions, such as 4-hydroxy-2,2,6,6-tetra-
methylpiperidinium N-oxide, R2NOϩ in Chart 1, are
strong oxidizing agents. Alcohols are oxidized by
R2NOϩ stoichiometrically and catalytically. Both elec-
trochemical [1,2] and chemical methods have been
used, with a variety of reagents such as HOCl [3–5],
MCPBA [6], and Cu(II) [7]. Primary alcohols react
preferentially over secondaries, possibly for steric rea-
sons [4,8]; for that reason PhCH2OH was used in most
dinyl oxyl radical, R2NO. The comproportionation
⅐
equilibrium, R2NOH ϩ R2NOϩ ϭ 2 R2NO ϩ Hϩ, fa-
vors the radical; thus after the initial stage of the re-
⅐
action R2NO predominates.
Chart 1
OH
OH
OH
Nϩ
N
N
Correspondence to: J. H. Espenson
Contract grant sponsor: National Science Foundation
Contract grant number: CHE-9007283
O
O•
2 ϭ R2NO•
OH
3 ϭ R2NOH
1 ϭ R2NOϩ
᭧ 1999 John Wiley & Sons, Inc.
CCC 0538-8066/99/050381-05

382
ZAUCHE AND ESPENSON
Hydrogen peroxide oxidizes secondary hydroxyl-
amines to nitrones when methyltrioxorhenium
(CH3ReO3, abbreviated as MTO) [9] or Na2WO4 [10]
is used as catalyst. ␣-Tetrasubstituted amines are con-
verted to N-oxyl radicals with MTO–H2O2 [11], per-
haps by the same mechanism (Scheme I). Alcohols are
oxidized with MTO–H2O2 [12,13], albeit very slowly.
dinium triflate, previously unreported, was prepared
by stirring 0.55 mmol silver triflate in 4 mL nitro-
methane with 0.50 mmol 4-HTEMPO. Metallic silver
formed within minutes. It was removed and 25 mL
ether was added to obtain the desired triflate salt of
the oxammonium ion as a yellow solid in 84% yield.
Anal.: Calcd. for C10H18NSO5F3: C, 37.38; N, 4.36;
H, 5.65; Found: C, 37.44, N, 4.33, H, 5.67.
Because H2O2 was used in large excess over the
hydroxylamine, the latter species was the only one that
changed in concentration with time. Thus pseudo-first-
order conditions applied. The rate constant was eval-
uated from the absorbance-time readings by nonlinear
least-squares fitting to the equation
B
A
HO
N9OH
hydroxylamine
k⌿
t
OϪ
Abst ϭ Absϱ ϩ (Abs0 Ϫ Absϱ)eϪ
H2O
Hϩ
HO
ϩN
HO
ϩN"O
The precision of the pseudo-first-order rate constants
so obtained was better than Ϯ2%. In other circum-
stances the method of initial rates was applied. To do
so, the absorbance-time curves were transformed into
concentration-time curves using the known molar ab-
sorptivities. The resulting data were then fit to an nth
order polynomial (n ϭ 5–8): Ct ϭ C0 Ϫ a1t –
a2t2 . . . The numerical value of a1 is equal to the
initial rate of reaction vi. This approach is particularly
advantageous, despite some sacrifice in precision,
when one of the reaction partners, MTO in this case,
slowly decomposes over the reaction time. The values
of the initial rates are precise to Ϯ5–8%.
The presence of acid may interfere with certain re-
action steps. Test reactions were therefore carried out
with an added base. In this case, 2,6-lutidine was used,
since it cannot coordinate to the rhenium(VII) center;
that is, it can function as a Brønsted base, but not a
Lewis base. As it turns out, the addition of 2,6-lutidine
had no effect on the kinetics, and so most kinetics
studies were carried out in acetonitrile without an
added base. This particular system will not tolerate
acid, in contrast to most cases where a high [Hϩ] can
be used to stabilize the rhenium catalyst [19]. Thus,
conditions were selected to minimize the amount of
Hϩ released during any reaction, owing to the involve-
ment of side reactions (to be discussed).
OH
oxammonium ion
Scheme I
This study reports the kinetics for several steps re-
lated to this chemistry: the oxidation of R2NOH by a
peroxorhenium complex derived from MTO, the re-
action between R2NOϩ and benzyl alcohol, and the
comproportionation reaction between R2NHOH and
R2NOϩ. We have also examined the possibility, using
alcohol reactions as the test case, that R2NOϩ might
act in concert with MTO to promote certain oxida-
tions.
EXPERIMENTAL
Some reagents were purchased: 2,2,6,6-tetramethyl-
piperidinium N-oxide (TEMPO), 4-hydroxy-2,2,6,6-
tetramethylpiperidinium N-oxide (HTEMPO), meth-
yltrioxorhenium (CH3ReO3, MTO), 30% H2O2, silver
trifluoromethanesulfonate, (AgCF3SO3), as were
HPLC-grade organic solvents. 2,2,6,6-Tetramethyl-pi-
peridine-1,4-diol and 2,2,6,6-tetramethyl-1-oxo-piper-
idinium chloride were prepared from HTEMPO ac-
cording to the literature procedures [8,14,15]. They
were stored under nitrogen because they are sensitive
to moisture and oxygen [16] .
The customary chloride salt of the oxammonium
ion, 2,2,6,6-tetramethyl-1-oxo-piperidinium chloride,
could not be used in this study. Chloride ions are them-
selves oxidized by hydrogen peroxide in a reaction
that is catalyzed by MTO [17]. The oxidation of chlo-
ride ions produces hypochlorite ions, which must be
avoided because it would interfere [18]. Silver salts
RESULTS AND INTERPRETATION
Peroxides Derived from MTO
Two stepwise equilibrium (but not instantaneous) re-
actions that take place between MTO and hydrogen
peroxide are described later. Because they have re-
ceived attention previously [20], it suffices here to
present reactions (1) and (2) and to show the structural
⅐
are known to oxidize R2NO, but no product has been
isolated. 4-Hydroxy-2,2,6,6-tetramethyl-1-oxo-piperi-

OXIDATION AND COMPROPORTIONATION REACTIONS
383
formulas of the two peroxorhenium products (Scheme
2
2
II), [MeRe(O) (␩ Ϫ O2)] and [MeRe(O)(␩
Ϫ
2
O2) 2(H2O)], that we abbreviate as A and B.
MTO ϩ H2O2 ϭ A ϩ H2O
A ϩ H2O2 ϭ B
(1)
(2)
O
k1 H2O2
Re
kϪ1 H2O
O
O
CH3
MTO
O
O
O
O
O
O
O
O
k2 H2O2
kϪ2
O
CH3
Re
Re
CH3
OH2
Figure 1 Kinetic data for the oxidation of the hydroxyl-
amine derived from 4-hydroxy-2,2,6,6-tetramethylpiperidi-
nium N-oxide by methyl(oxo)(diperoxo)rhenium(VII), B.
This graph shows that the initial reaction rate, vi, is a linear
function of the product of the concentrations of the two re-
actants.
B
A
Scheme II
Kinetics of Oxidation of the Hydroxylamine
The oxidation of R2NHOH by the peroxorhenium
compound B, reaction (3), was monitored at 240 nm
⅐
by recording the buildup of R2NO, ␧ ϭ 1.85 ϫ 103 L
1
molϪ1 cmϪ . This radical resulted from the second step
reaction of B with N-benzyl-N-tert-butylhydroxylam-
1
1
of the sequence, the rapid comproportionation reaction
between R2NOϩ and R2NOH, reaction (4).
ine has k3 ϭ 0.94 L molϪ sϪ in methanol at 298 K
[9]. The similarity of these values substantiates that
the two independent determinations correspond to the
same chemistry.
B ϩ R2NOH !: A ϩ R2NOϩ ϩ OHϪ (3)
R2NOϩ ϩ R2NOH !: 2R2NOD ϩ Hϩ (4)
The reaction sequence found for numerous other
secondary hydroxylamines features two steps. Initially
and rate controlling is the transfer of an oxygen atom
from a peroxorhenium moiety to the electron pair of
the nitrogen. Rapidly following is an elimination re-
action that creates a CϭN double bond in the product
nitrone, Scheme III.
The concentrations employed were 0.6–3.0 mM
MTO, 200 mM H2O2, and 0.7–2.0 mM R2NOH. With
this high concentration of hydrogen peroxide, [B] ϾϾ
[A], only B is present at an appreciable concentration,
whereas A, remains Ͻ2% of the total rhenium. The
kinetic contribution of A can be neglected. (To verify
that point, we showed that the rate constant was in-
dependent of [H2O2] in the range 0.1–0.4 M.) In this
reaction sequence, A is formed as B reacts. Under
these conditions B was rapidly regenerated, however,
since A reacts with hydrogen peroxide to give B; k2
(Scheme II) is 6.2 ϫ 10Ϫ2 L molϪ1 sϪ1 in acetonitrile.
That reaction is sufficiently rapid to maintain [B] at its
equilibrium concentration.
O
RCH2
RCH2
O
O
ϩ
Re
N
O
CH3
O
OH
OH2
O
OϪ
RCH2
RCH2
O
O
Re
ϩ
Nϩ
O
CH3
OϪ
OH
OH2
The second-order rate constant for reaction (3) was
evaluated from a plot of vi against [B], which was
taken as [Re]T for the reasons previously stated. This
plot is displayed in Figure 1. The least-squares slope
RCH2
RCH2
fast
Nϩ
Nϩ9OϪ
affords the value k3 ϭ 3.0 Ϯ 0.1 L molϪ sϪ in ace-
1
1
RCH2
RCH
OH
tonitrile at 298 K. In comparison, the closely-related
Scheme III

384
ZAUCHE AND ESPENSON
The first step of the sequence occurs by nucleo-
philic attack of the nitrogen lone pair on one of the
four peroxo oxygen atoms of B. These peroxides have
been activated by the inductive effect of the highly
electropositive Re(VII) center to which they are co-
ordinated. In effect, the peroxide is activated electro-
philically. The transition state for this process can be
depicted like this:
ϩ
ϩ
O
Re
O
O
O
CH
O
3 H2O
N
Figure 2 The kinetic analysis of the data for reaction (4)
in acetonitrile at 298 K. The values of k’, obtained from eq.
HO
Ϫ
(7), are shown here as a plot of k’ (s 1) vs. ⌬, the difference
OH
in the initial concentrations of the reactants; the slope of this
plot is k4.
Direct Study of the Kinetics of
Comproportionation
The comproportionation reaction between R2NOH and
R2NOϩ, reaction (4), was monitored by stopped-flow
and conventional UV-Vis methods. The initial condi-
tions were 18–32 ␮M R2NOϩ, 46–416 ␮M R2NOH.
A few of these experiments were carried out under
pseudo-first-order conditions, but most were not.
These data were fit to the form for mixed-second-order
kinetics, eq. (5) in terms of concentration and eq. (6)
in terms of absorbance.
Kinetics of Alcohol Oxidation by the
Oxammonium Ion
The reaction between these partners is given by reac-
tion (7).
R2NOϩ ϩ R2CHOH !: R2NOH ϩ R2CO ϩ Hϩ
(7)
We chose benzyl alcohol as the prototypical alco-
hol because, in general, it is often more easily oxi-
dized. The buildup of benzaldehyde was monitored by
recording the absorbance at 244 nm, with 1–5 mM
R2NOϩ and 60–200 mM PhCH2OH. These measure-
R Ϫ 1
[R2NOϩ]t ϭ [R2NOϩ]0 ϫ
(5)
(6)
Јt
R ϫ ek Ϫ 1
(R Ϫ 1) ϫ (Abs0 Ϫ Absϱ)
3
1
ments afforded k8 ϭ (4.4 Ϯ 0.5) ϫ 10Ϫ L molϪ1 sϪ
Abst ϭ Absϱ ϩ
R ϫ ek Ϫ 1
Јt
in acetonitrile at 298 K. The overall reaction stoichi-
ometry is 2R2NOϩ:1PhCH2OH, because rapid com-
proportionation, reaction (4), occurs immediately
thereafter.
The mechanism just described appears similar to
that given for the oxidation of primary and secondary
alcohols by the oxammonium ion prepared from
TEMPO [21]. The literature on the use of stable or-
ganic nitroxyl radicals for the oxidation of primary and
secondary radicals has recently been reviewed [22].
where R ϭ [R2NOH]0/[R2NOϩ]0 and the rate constant
k’ ϭ k4 ϫ ⌬, where ⌬ ϭ [R2NOH]0 Ϫ [R2NOϩ]0. Data
were obtained in nine experiments. The plot of k’
against ⌬ is depicted in Figure 2; its slope gave k4 ϭ
1
1
(1.78 Ϯ 0.05) ϫ 103 L molϪ sϪ in acetonitrile at
298 K.
The position of the equilibrium in eq. (4) can be
shifted by protonation of the hydroxylamine,
R2NHOHϩ ϭ R2NOH ϩ Hϩ, but doing so provides
no advantage here since the protonated hydroxylam-
monium ion, lacking an unshared electron pair, does
not react with B [9]. The low reactivity of R2NHOHϩ
seems general, which is the reason that catalytic
R2NOϩ oxidations are generally performed under neu-
tral or basic conditions.
Catalyzed Oxidation of Alcohols
With the kinetic data for the preceding reactions now
at hand, the construction of the catalytic cycle was
possible. Scheme IV presents an analysis of the trans-
formations occurring. For the sake of simplicity, steps

OXIDATION AND COMPROPORTIONATION REACTIONS
385
interconverting MTO and A are not shown. At the
outset, the oxidation of PhCH2OH by B is negligible
BIBLIOGRAPHY
relative to its oxidation by the oxammonium ion.
1. Semmelhack, M. F.; Chou, C. S.; Cortes, D. A. J Am
Chem Soc 1983, 105, 4492.
2. Inokuchi, T.; Matsumoto, S.; Fukushima, M.; Torii, S.
Bull Chem Soc Jpn 1991, 64, 796.
3. Anelli, P. L.; Banfi, S.; Montanari, F.; Quici, S. J Org
Chem 1989, 54, 2970.
4. Siedlecka, R.; Skarzewski, J.; Mochowski, J. Tetrahe-
dron Lett 1990, 31.
H2O2
PhCHO PhCH2OH
k2
A
B
R2NOH
PhCHO ϩ Hϩ
A
R2NHOHϩ
R2NOϩ
Hϩ
k8
PhCH2OH
R2NOH, k4
ϩ
•
5. Leanna, M. R.; Sowin, T. J.; Morton, H. E. Tetrahedron
Lett 1992, 33, 5029.
2 R2NO ϩ H
Scheme IV
6. Cella, J. A.; McGrath, J. P.; Kelley, J. A.; Soukarry,
E. O.; Hilpert, L. J Org Chem 1977, 42, 2077.
7. Semmelhack, M. F.; Schmid, C. R.; Cortes, D. A.;
Chou, C. S. J Am Chem Soc 1984, 106, 3374.
8. Yamaguchi, M.; Miyazawa, T.; Takata, T.; Endo, T.
Pure Appl Chem 1990, 62, 217.
9. Zauche, T. H.; Espenson, J. H. Inorg Chem 1997, 22,
5257.
10. Sen, V. D.; Golubev, V. A.; Efremova, N. N. Izv Akad
Nauk, Ser Khim 1982, 61.
11. Murray, R. W.; Iyanar, K.; Chen, J.; Wearing, J. T. Tet-
rahedron Lett 1996, 37, 805.
12. Murray, R. W.; Iyanar, K.; Chen, J.; Wearing, J. T. Tet-
rahedron Lett 1995, 36, 6415.
13. Zauche, T. H.; Espenson, J. H. Inorg Chem, 1998, 37,
6827.
14. Paleos, C. M. J Chem Soc, Chem Commun 1977, 345.
15. Yamaguchi, M.; Miyazawa, T.; Takata, T.; Endo, T.
Pure Appl Chem 1990, 62, 217–222.
The catalytic cycle cannot be sustained for long,
however, because the rapid comproportionation reac-
tion drains off the oxammonium ion. These experi-
ments were run without added Hϩ, so that the equilib-
rium in eq. (4) would lie well to the right. To limit the
occurrence of reaction (4), which limits the effective-
ness of a catalytic mechanism, [MTO] was increased
to 40 mM and [R2NOϩ]0 was decreased to 0.1 mM.
Aldehyde formation was detected, but it arises mainly
from the reaction of B, not R2NOϩ, with PhCH2OH.
Given the rate constants and these observations, it
can be recognized that comproportionation limits the
catalytic applicability of this scheme. Hypochlorite, a
two-electron acceptor, leads to successful oxidations.
It appears that under these conditions the MTO-
TEMPO co-catalysts are relatively ineffective. It was
difficult to sustain the oxidation of alcohol by R2NOϩ
since it is being more rapidly consumed by reaction
(4). Note that the equilibrium in reaction (4) could be
reversed at higher [Hϩ]. With that, the oxyl radical
would be an intermediate, not a dead, end. Conditions
of high [Hϩ] are, however, incompatible with the need
for R2NOH, not the nonreactive R2NHOHϩ.
16. Golubev, V. A.; Sen, V. D.; Kulyk, I. V.; Aleksandrov,
A. L. Izv Akad Nauk, Ser Khim 1974, 10, 2235–2243.
17. Hansen, P. J.; Espenson, J. H. Inorg Chem 1995, 34,
5839.
18. Anelli, P. L.; Banfi, S.; Montanari, F.; Quici, S. J Org
Chem 1989, 54, 2970–2972.
19. Abu-Omar, M.; Hansen, P. J.; Espenson, J. H. J Am
Chem Soc 1996, 118, 4966–4974.
20. Espenson, J. H.; Abu-Omar, M. M. Adv Chem Ser
1997, 253, 99–134.
This research was supported by a grant from the National
Science Foundation (CHE–9007283). Some experiments
were conducted with the use of the facilities of the Ames
Laboratory.
21. Golubev, V. A.; Borislavskii, V. N.; Aleksandrov, A. L.
Izv Akad Nauk, Ser Khim 1977, 2025–2043.
22. De Nooy, A. E.; Besemer, A. C.; Van Bakkum, H. Syn-
thesis 1996, 1153–1174.