Kinetics and mechanisms of reactions of alkyl hydroperoxides with methylrhenium oxides

Article abstract of DOI:10.1021/ic9809052

Aqueous methyldioxorhenium (MDO), prepared from methyltrioxorhenium (MTO) and hypophosphorous acid, abstracts an oxygen atom from tertiary alkyl hydroperoxides. This regenerates MTO and forms the tertiary alcohol with rate constants 3.71 × 10⁴ L mol^-1 s^-1 (t-BuOOH) and 3.47 × 10⁴ L mol^-1 s^-1 (t-AmOOH) at 25.0 °C in aqueous 1.0 M HOTf. MDO reacts with hydrogen peroxide first to form MTO, k = 3.36 × 10⁴ L mol^-1 s^-1, which subsequently reacts with more hydrogen peroxide to form peroxorhenium complexes. In a separate study, the concomitant slow decomposition of alkyl hydroperoxides and MTO (to ReO₄^-) was investigated. The rate law is v = k[MTO][RCMe²OOH]/[H⁺], with k = 7.4 × 10^-5 s^-1 (R = Me) and k = 8.4 × 10^-5 s^-1 (R = Et) at 25.0 °C in aqueous solution at μ 1.0 M. ¹H NMR spectroscopy and GC revealed organic products suggestive of radical reactions. The products from t-BuOOH are acetone, methanol, tert-butyl methyl ether, methane, ethane, and rerf-butyl methyl peroxide. With CH₂DReO₃, it could be shown that both t-BuOOH and MTO were sources of the methane. The rate of decomposition of MTO shows an inverse-first-order dependence on [H⁺] throughout the range pH 1-6.42.

Full text of DOI:10.1021/ic9809052

744
Inorg. Chem. 1999, 38, 744-750
Kinetics and Mechanisms of Reactions of Alkyl Hydroperoxides with Methylrhenium
Oxides
Kimberly A. Brittingham and James H. Espenson*
Iowa State University, 25 Spedding Hall, Ames, IA 50011
ReceiVed July 30, 1998
Aqueous methyldioxorhenium (MDO), prepared from methyltrioxorhenium (MTO) and hypophosphorous acid,
abstracts an oxygen atom from tertiary alkyl hydroperoxides. This regenerates MTO and forms the tertiary alcohol
with rate constants 3.71 × 104 L mol-1 s-1 (t-BuOOH) and 3.47 × 104 L mol-1 s-1 (t-AmOOH) at 25.0 °C in
aqueous 1.0 M HOTf. MDO reacts with hydrogen peroxide first to form MTO, k ) 3.36 × 104 L mol-1 s-1
,
which subsequently reacts with more hydrogen peroxide to form peroxorhenium complexes. In a separate study,
the concomitant slow decomposition of alkyl hydroperoxides and MTO (to ReO4-) was investigated. The rate
law is v ) k[MTO][RCMe2OOH]/[H+], with k ) 7.4 × 10-5 s-1 (R ) Me) and k ) 8.4 × 10-5 s-1 (R ) Et)
1
at 25.0 °C in aqueous solution at µ 1.0 M. H NMR spectroscopy and GC revealed organic products suggestive
of radical reactions. The products from t-BuOOH are acetone, methanol, tert-butyl methyl ether, methane, ethane,
and tert-butyl methyl peroxide. With CH2DReO3, it could be shown that both t-BuOOH and MTO were sources
of the methane. The rate of decomposition of MTO shows an inverse-first-order dependence on [H+] throughout
the range pH 1-6.42.
Introduction
the strength of the rhenium-oxygen bond in MTO, 464 kJ
mol-1 9
. Hydrogen peroxide converts MDO into the peroxo
An extensive amount of literature now describes how
hydrogen peroxide and MTO (CH3ReO3) form η2-peroxorhe-
nium species, A ) CH3Re(η2-O2)(O)2 and B ) CH3Re(η2-O2)2-
(O)(H2O), eq 1.1-4 Both are active catalysts for hydrogen
peroxide oxidations.5
complexes A and B. Alkyl hydroperoxides, on the other hand,
yield only MTO, allowing a kinetic analysis of the first stage:
ROOH + CH3ReO2 f ROH + CH3ReO3 (k4)
(4)
H2O2
CH3ReO3 {H O } CH3Re(η2-O2)(O)2 (A) {
}
The slow decomposition of CH3ReO3 in aqueous solutions
containing Me3COOH (t-BuOOH) or EtCMe2OOH (t-AmOOH)
has been studied. The different rhenium and organic products
were used to assign a mechanism that clearly involves free
radicals. Cumyl hydroperoxide, C6H5(CH3)2COOH, is decom-
posed heterolytically by MTO in chlorobenzene; <1% of the
process involves radicals.10
2
2
H2O
CH3Re(η2-O2)2O(H2O) (1)
In solutions containing hydrogen peroxide, MTO is prone to
irreversible decomposition by the nucleophilic hydroperoxide
anion, HO2 (eq 2)6 and OH- (eq 3).6,7
-
Experimental Section
-
HO2- + CH3ReO3 f CH3OH + ReO4
(2)
(3)
Materials. A Millipore-Q water-purification system provided high-
purity water. Deuterium oxide (99%) was used as the NMR solvent.
Solutions were acidified with trifluoromethanesulfonic acid (HOTf,
triflic acid) diluted from the commercially available material (98%).
Perchloric acid or its salts could not be present, owing to the reaction
between MDO and ClO4-.8 Where needed, ionic strength was main-
tained with lithium triflate.
t-Butyl hydroperoxide was obtained commercially as a 70% solution
in water. t-Amyl hydroperoxide was synthesized and analyzed as
previously described.11 CH2DReO3 was synthesized from dirhenium
heptoxide, trifluoroacetic acid, and CH2DSn(n-butyl)3,12 kindly donated
by Dr. Michael T. Ashby. Other reagents were available com-
mercially: MTO, hypophosphorous acid (50% in water), phosphorous
acid, tert-butyl alcohol, hydrogen peroxide (30% in water), t-amyl
alcohol, 2-methyl-1-phenyl-2-propanol, HPLC grade acetonitrile, dirhe-
nium heptoxide, and trifluoroacetic acid. Methyldioxorhenium (MDO)
OH- + CH3ReO3 f CH4 + ReO4
-
The rhenium(V) compound methyldioxorhenium (MDO,
solvated or ligated) abstracts an oxygen atom from many oxygen
donors XO:7-9 XO + CH3ReO2 f X + CH3ReO3, driven by
(1) Herrmann, W. A.; Fischer, R. W.; Scherer, W.; Rauch, M. U. Angew.
Chem., Int. Ed. Engl. 1993, 32, 1157.
(2) Yamazaki, S.; Espenson, J. H.; Huston, P. Inorg. Chem. 1993, 32,
4683.
(3) Espenson, J. H.; Pestovsky, O.; Huston, P.; Staudt, S. J. Am. Chem.
Soc. 1994, 116, 2869.
(4) Hansen, P. J.; Espenson, J. H. Inorg. Chem. 1995, 34, 5839.
(5) Espenson, J. H.; Abu-Omar, M. M. AdV. Chem. Ser. 1997, 253, 99-
134.
(6) Abu-Omar, M. M.; Hansen, P. J.; Espenson, J. H. J. Am. Chem. Soc.
1996, 118, 4966.
(7) Zhu, Z.; Espenson, J. H. J. Mol. Catal. 1995, 103, 87.
(8) Abu-Omar, M. M.; Espenson, J. H. Inorg. Chem. 1995, 34, 6239.
(9) Abu-Omar, M. M.; Appleman, E. H.; Espenson, J. H. Inorg. Chem.
1996, 35, 7751.
(10) Smurova, L. A.; Kholopov, A. B.; Nikitin, A. V.; Brin, E. F.; Rubailo,
V. L. Kinet. Catal. (Transl. of Kinet. Katal.) 1996, 37, 361.
(11) Leslie, J. P., II.; Espenson, J. H. J. Am. Chem. Soc. 1976, 98, 4839.
(12) Herrmann, W. A.; Ku¨hn, F. E.; Fischer, R. W.; Thiel, W. R.; Roma˜o,
C. C. Inorg. Chem. 1992, 31, 4431.
10.1021/ic9809052 CCC: $18.00 © 1999 American Chemical Society
Published on Web 02/04/1999

Reactions of Methylrhenium Oxides
Inorganic Chemistry, Vol. 38, No. 4, 1999 745
was generated in situ from hypophosphorous acid and MTO in strongly
acidic aqueous solution, k5 ) 2.8 × 10-2 L mol-1 s-1 at 25.0 °C:9
CH3ReO3 + H3PO2 f CH3ReO3+ H3PO3 (k5)
(5)
With higher (∼10 mM) concentrations of MDO for NMR studies,
a blue color indicated the onset of MDO oligomerization.8,9 The addition
of alkyl hydroperoxide immediately restored colorless MTO. We
surmise that the disappearance of oligomer in this case is analogous to
perchlorate reduction by dimeric MDO.8 Oligomerization might have
been largely avoided by maintaining pH 0, but this was not feasible
because of NMR shimming difficulties.
Instrumentation. Spectrophotometric data were recorded using
Shimadzu UV-2501PC and UV-3101PC spectrophotometers. Kinetic
data for fast reactions were collected with an Applied Photophysics
Sequential DS-17MV stopped-flow spectrophotometer. 1H NMR spec-
tra, used to monitor MTO (s, 2.46 ppm) and organic products over
time in D2O, were obtained with Varian VXR-300 or Bruker DRX-
400 spectrometers. Acetonitrile was the internal standard (δ 2.09 ppm).
Figure 1. Kinetic data for the reaction of tert-butyl hydroperoxide
with methyldioxorhenium in aqueous solutions of triflic acid (0-1.0
M) at 25.0 °C and ionic strength 1.0 M. Shown is a plot of the pseudo-
first-order rate constant against the average concentration of the
hydroperoxide, the reagent in excess.
1
31P chemical shifts were referenced to 85% H3PO4. H NMR kinetics
were recorded with the Bruker spectrometer at 22.5 ( 0.2 °C. The
chemical shifts of the proton resonances for the organic reactants and
products are as follows:
of H3PO2 and H+ were varied modestly, 0.05-0.1 M and
0.032-1.0 M, respectively, to confirm the absence of kinetic
effects from these variables. Ionic strength was maintained at
1.0 M with lithium triflate. The absorbance-time data in each
experiment were fit to first-order kinetics, eq 6, since the alkyl
hydroperoxide was taken in substantial excess.
(CH3)2CO δ 2.25 (s, 6H)
(CH3)3COH δ 1.271 (s, 9H)
(CH3)3COCH3 δ 3.25 (s, 3H)
CH3OOH δ 3.887 (s, 3H)
(CH3)3COOH δ 1.266 (s, 9H)
CH3OH δ 3.37 (s, 3H)
CH3CH2(CH3)2COOH δ 0.893 (t, 3H)
CH3CH2(CH3)2COOH δ 1.601 (q, 2H)
CH3CH2(CH3)2COOH δ 1.215 (s, 6H)
CH3CH2(CH3)2COH δ 0.86 (t, 3H)
CH3CH2(CH3)2COH δ 1.485 (q, 2H)
CH3CH2(CH3)2COH δ 1.166 (s, 6H)
Abst ) Abs∞ + (Abs0 - Abs∞)e-k t
(6)
ψ
(CH3)3COOCH3 δ 3.86 (s, 3H) CH3CH2(CH3)2COCH3 δ 3.217 (s, 3H)
CH2O δ 5.092 (s, 2H) CH3CH2(CH3)2COOCH2CH3 δ 3.678 (q, 2H)
The variation of kψ with [t-BuOOH]av is presented in Figure
1. A linear least-squares fit gave k4 at 25.0 °C as the slope:
(3.71 ( 0.10) × 104 L mol-1 s-1. For t-AmOOH, k4 ) (3.47 (
The chemical shifts for the commercially available compounds were
assigned on the basis of authentic samples. The CH3OOH was prepared
by literature methods.13 The (CH3)3COOCH3 was prepared from
dimethyl sulfate and tert-butyl hydroperoxide on the basis of the same
procedure.13 A quartet resonance among the products of the reaction
of t-AmOOH and MTO is the result of an unassigned methylene group
of an ethyl group that has been shifted downfield by an oxygen. We
take this resonance to be that of t-amyl ethyl peroxide because the
t-BuOOH reaction forms the analogous dialkyl peroxide.
Methane, ethane, ethylene, and propane were detected by gas
chromatography on a VZ-10 column. Perrhenate ions were detected
with a Perkin-Elmer Sciex API electrospray mass spectrometer with a
syringe injector. GC-MS experiments were performed with a TSQ700
mass spectrometer (Finnigan MAT). The mass spectra were obtained
by scanning the first quadrupole from m/z 10 to m/z 20 for 0.5 s. The
second and third quadrupoles were maintained in the RF only mode.
The system was configured in the electron impact ionization mode. A
DB1 column was used. The injector and transfer-line temperatures were
maintained at 260 and 240 °C, respectively.
0.07) × 104 L mol-1 s-1
.
The reactions in eqs 4 and 5 constitute a catalytic cycle, the
faster step destroying MDO and the other reforming it more
slowly. In effect, then, MTO catalyzes the reaction between
hypophosphorous acid and alkyl hydroperoxide:
ROOH + H3PO2 cat. MTO8 ROH + H3PO3
(7)
This is a spontaneous reaction in the thermodynamic sense,
but it does not occur under these conditions without a catalyst.
In the course of checking the catalytic reaction, other reactions
were found to take place concurrently over longer times (here
limited by the lower rate of reaction 5). The additional reactions
entailed the concurrent decomposition of MTO and ROOH, as
discussed later.
Kinetics of the Reaction of MDO with H2O2. In principle,
these experiments were the same as those just reported for
ROOH, except that reaction 4 is followed by others, leading to
further products, eq 1. When MDO ([MTO]0 ) 0.1 mM, [H3-
PO2]0 ) 50 mM) was mixed with H2O2 (10 mM), the
absorbance at 270 nm for MTO increased with time according
to first-order kinetics. For that stage, the rate constant for the
reaction between MDO and H2O2 is k4 ) (3.36 ( 0.02) × 104
L mol-1 s-1. Over longer times, the formation of B was
monitored at 360 nm. The pseudo-first-order rate constant of
this second component agreed with that determined directly for
H2O2 + MTO. These findings provide spectroscopic and kinetic
evidence that MTO was formed in the first step.
Kinetics of the Concurrent Decomposition of MTO and
RCMe2OOH. In the strongly acidic medium needed to stabilize
MDO, pH 0-1, MTO remains fairly stable without the
hydroperoxide. With hydroperoxide, however, both MTO and
the alkyl hydroperoxide slowly decompose concurrently at a
pH dependent rate.
Results
Kinetics of the Reaction of MDO with ROOH. Kinetic data
were obtained with the stopped-flow technique at 25.0 ( 0.2
°C in aqueous solution. The MDO solution was prepared from
MTO and hypophosphorous acid in one of the SF syringes
containing HOTf and/or LiOTf. After 5 half-lives for MDO
formation had elapsed, k5 ) (2.8 ( 0.2) × 10-2 L mol-1 s-1
,
the MDO solution was mixed in the SF apparatus with the
contents of the other syringe holding the appropriate concentra-
tions of ROOH, HOTf, and LiOTf. The buildup of MTO was
monitored at 270 nm (ꢀ ) 1300 L mol-1 cm-1). The reactant
concentrations were varied over these ranges: 50-200 µM
MDO and 0.5-10 mM t-BuOOH. In addition, the concentrations
(13) Davies, D. M.; Deary, M. E. J. Chem. Soc., Perkin Trans. 2 1992,
559.

746 Inorganic Chemistry, Vol. 38, No. 4, 1999
Brittingham and Espenson
Table 1. Rate Constants for the Decrease of MTO and Increase of
Products During the Decomposition of MTO (17 mM) and
t-BuOOH (200 mM) at 22.5 ( 0.2 °C
species
kdec/10-5 s-1
final concna/mM
MTO
CH3OH
(CH3)2CO
t-BuOCH3
t-BuOOCH3
6.2 ( 0.2
5.55 ( 0.07
6.1 ( 0.1
5.9 ( 0.1
5.2 ( 0.2
0.4 ( 0.1
4.44 ( 0.02 (26%)b
2.78 ( 0.2 (16%)
2.37 ( 0.02 (14%)
1.85 ( 0.04 (11%)
a From least-squares fitting of the experimental data. b Percent yield
relative to [MTO]0.
mM HOTf, and 10 mM CH3CN as the NMR standard. The
reagents were added except t-BuOOH, and the NMR spectrum
recorded. Then t-BuOOH was added and the spectrum recorded
over approximately 4 h. The concentration-time data for the
decrease in MTO and the increase in reaction products were fit
to first-order kinetics. The rate constants are reported in Table
1.
According to Table 1, the pseudo-first-order rate constants
for the decomposition of MTO and the buildup of methanol,
acetone, tert-butyl methyl ether, and tert-butyl methyl peroxide
are the same within error. The weighted mean14 of these first-
order rate constants is kdec ) (5.76 ( 0.05) × 10-5 s-1. Table
1 also reports the final concentration of each product. It is
difficult to calculate the mass balance of products because of
the complicated radical decomposition process, which is dis-
cussed later.
Methane and ethane were monitored by gas chromatography.
MTO (10 mM) and t-BuOOH (200 mM) were mixed together
in 0.1 M HOTf at room temperature. The total volume of
solution was 7 mL. The reaction was carried out in a 20-mL
bottle sealed with a septum. No care was taken to exclude
oxygen. The solution was shaken vigorously, and then the
headspace was injected into the GC. Data seemed to indicate
that the these products were formed much faster than those
detected by NMR spectroscopy. It should be noted that these
experiments were done to detect methane and ethane and are
of qualitative significance only.
Figure 2. Dependence of second-order rate constant, kapp, on 1/[H+]
at 25 °C. Slope is kf ) (7.4 ( 0.1) × 10-5 s-1. Inset shows a plot of
log k vs pH, with a slope 0.93 ( 0.03. Points at pH 1, 2, 3.44, and
4.72 were at µ ) 1.0 M; the others were at µ ) 0.05 M.
The decomposition of MTO (0.5 mM) by RCMe2OOH (10
mM) was monitored spectrophotometrically at 270 nm at pH
1.0-6.42. The data were fit to first-order kinetics at each pH.
Values of an apparent second-order rate constant were calculated
as kapp ) kψ/[RCMe2OOH]. The plot of log kapp versus pH was
linear, Figure 2 inset, with a slope of 0.93 ( 0.03. Thus, the
decomposition is inverse-first-order with respect to [H+].
Reactions of t-AmOOH were monitored at pH 1, 2, and 3.25.
The pattern closely resembles that of t-BuOOH. The rate
equation for both is
[MTO][RCMe2OOH]
V ) kf
(8)
[H+]
The plot of kapp versus 1/[H+] is linear, Figure 2. The data in
this plot were taken from experiments at ionic strengths of 0.05
and 1.0 M. The kf values from these fits are (7.4 ( 0.1) × 10-5
s-1 (t-BuOOH) and (8.4 ( 0.4) × 10-5 s-1 (t-AmOOH) at 25.0
°C.
To determine whether MTO or t-BuOOH or both were a
source of Me•, 98% CH2DReO3 was employed. The GC of the
argon-saturated solution indicated peaks for methane and water.
The MS of the methane peak showed a significant amount of
the CH3D+ ion at m/z 17. It was not possible to determine the
ratio CH3D/CH4 to sufficient accuracy, but it was evident that
both were present. Thus, the methane is composed of methyl
radical that is partly derived from the â-scission of the tert-
butoxyl radical and partly from MTO decomposition.
1H NMR was used to study the dependence of the same
reaction (µ ) 0.1 M, [H+] ) 0.1 M) on [tert-BuOOH] at 22.5
°C. The values of kψ for the disappearance of MTO were plotted
against [t-BuOOH] to give kapp ) (2.41 ( 0.07) × 10-4 L mol-1
s-1, where kapp ) kf/[H+]. Therefore, according to these NMR
experiments, kf ) (2.41 ( 0.07) × 10-5 s-1. These values for
kf, as determined by UV-vis spectrophotometry and NMR,
differ by a factor of 3. This discrepancy arises from the use of
quite different ranges of reactant concentrations, a different
experimental technique, a different reaction medium, and a
modestly different temperature.
Products of the Decomposition of MTO and t-AmOOH.
Acetone, methanol, t-amyl methyl ether, and t-amyl ethyl
1
peroxide were detected by H NMR. The rate constants, kdec
,
Products of Decomposition of MTO and t-BuOOH. The
decomposition of MTO and t-BuOOH might be anticipated to
parallel reaction 2. Perrhenate ions and tert-butyl methyl ether
would be the products from attack of the tert-butyl peroxide
anion on MTO. These products were indeed found, but others
were formed as well. The loss of MTO and t-BuOOH under an
Ar atmosphere was monitored by NMR. GC was also employed
to detect alkanes in the gas phase. The products included (CH3)2-
CO, CH3OH, t-BuOCH3, CH4, C2H6, and t-BuOOCH3. Although
for the decrease in MTO and increase in organic products, Table
2, show a weighted mean14 of (2.36 ( 0.04) × 10-4 s-1
.
Gaseous products, namely methane, ethane, ethylene, and
propane, were detected by GC. Methane was formed in a higher
yield than ethane. Ethylene and propane were minor. Quantita-
tive statements cannot be made.
Experiments in the Presence of Oxygen. The reaction
between MTO (10 mM) and t-BuOOH (200 mM) under O2 gave
two additional organic products, methyl hydroperoxide and
formaldehyde.
1
t-BuOH is seen as a product by H NMR, its signal overlaps
that for t-BuOOH, and therefore the change in its peak intensity
could not be monitored. The perrhenate ion was also qualita-
tively detected using electrospray mass spectroscopy. The initial
concentrations were: ∼10 mM MTO, 200 mM t-BuOOH, 100
(14) Bevington, P. R.; Robinson, D. K. Data Reduction and Error Analysis
for the Physical Sciences. 2nd ed.; McGraw-Hill: St. Louis, 1992.

Reactions of Methylrhenium Oxides
Inorganic Chemistry, Vol. 38, No. 4, 1999 747
Table 2. Rate Constants for the Decrease of MTO and Increase of
Products During the Decomposition of MTO (7.0 mM) and
t-AmOOH (200 mM) at 22.5 ( 0.2 °C
species
kdec/10-4 s-1
final concna/mM
MTO
CH3OH
(CH3)2CO
2.13 ( 0.08
2.53 ( 0.07
1.7 ( 0.2
1.2 ( 0.1
3.89 ( 0.05 (56%)b
5.21( 0.05 (74%)
0.83 ( 0.02 (12%)
4.0 ( 0.7 (57%)
Equations 4 and 9 together account for rapid MTO buildup.
One cannot say directly from the data whether the rate-
controlling step is the coordination of the alkyl hydroperoxide
or the breaking of the O-O bond. Given the high rate of MTO
ligation reactions,16 however, the rate-controlling step in reaction
9 seems most probably to be the breaking of the peroxide bond,
as in the second step. The value of k4 for t-BuOOH is only 7%
larger than that for t-AmOOH, and the k4 for H2O2 is slightly
lower than that for either alkyl hydroperoxide. Two factors might
contribute to these small differences in k4: The more electron-
donating R group provides a more nucleophilic peroxide anion,
and thus a larger k4, if the first step is rate-controlling. Also,
t-AmOOH is slightly more sterically hindered than t-BuOOH,
but this is at best a minimal determinant of the rate of MTO
formation.
Kinetics of Decomposition of MTO and ROOH. The
concurrent decomposition of MTO and H2O2 showed a 1/[H+]
kinetic dependence at µ ) 0.05 M. Methanol and perrhenate
ions may be formed by one of these kinetically indistinguishable
pathways:6 (a) nucleophilic attack of HOO-on MTO with ka )
2.03 × 108 L mol-1 s-1
t-AmOCH3
t-AmOOCH2CH3
2.59 ( 0.07
1.3 ( 0.2
a From least-squares fitting of the experimental data. b Percent yield
relative to [MTO]0.
Table 3. Rate Constants and Final Concentrations for the
Decomposition of MTO and the Buildup of Products in the
Reaction of MTO (10 mM), H3PO2 (100 mM), and t-BuOOH (200
mM) at 22.5 ( 0.2 °Ca
species
kdec/10-5 s-1
final concnb/mM
MTO
CH3OH
(CH3)2CO
t-BuOCH3
t-BuOOCH3
15 ( 2
17.1 ( 0.8
50 ( 2
9.4 ( 0.7
54 ( 1
4.3 ( 0.2
3.84 ( 0.05 (38%)c
73 ( 2 (730%)
0.88 ( 0.02 (9%)
2.46 ( 0.03 (25%)
a [HOTf]0 ) 100 mM. b From least-squares fitting of the experi-
mental data. c Percent yield relative to [MTO]0.
Phosphoric Acid. No conditions have been found under
which H3PO2 or H3PO3 could be oxidized to phosphate by MTO
catalyzed reactions of hydrogen peroxide. To our surprise then,
phosphoric acid was produced when t-BuOOH was added to a
solution of MDO containing excess H3PO2. This experiment
consisted of adding 200 mM t-BuOOH to a solution prepared
from 10 mM MTO and 100 mM H3PO2 that had been allowed
to stand for 88 s, in which time the blue color for MDO
oligomerization became apparent. The 31P NMR spectrum of
this reaction mixture was used to substantiate H3PO3 (d, 5.49
ppm) and H3PO4 (s, 0.63 ppm).
HOO- + MeReO3 f [Me(O)3Re-OOH]- f
-
MeOH + ReO4
(b) nucleophilic attack of OH- on A with kb ) 3.05 × 109 L
mol-1 s-1
-
OH- + MeRe(O)2(η2-O2) f f MeOH + ReO4
Under these conditions, hypophosphorous acid and tert-butyl
hydroperoxide do not react without MTO. Phosphoric acid was
formed at approximately the same rate when H3PO2 was
replaced by H3PO3. When chlorate ion, ClO3-, was used in place
of t-BuOOH, phosphorous acid was the only product. These
findings suggest that the precursor to phosphoric acid is
phosphorous acid. Also, phosphoric acid formation is probably
(c) a reaction between H2O2 and MTO-OH-, the conjugate
base of MTO (pKa ) 7.5),15 with kc ) 3.1 × 105 L mol-1 s-1
H2O2 + MeRe(O)3(OH)- f f MeOH + ReO4- + H2O
These considerations guide us to an interpretation of the
kinetic data for ROOH. The rate constant for the decomposition
of MTO by t-BuOOH is also proportional to 1/[H+], as shown
in Figure 2. This could be interpreted by mechanism (a) or (c),
but not by (b) since that would require the same rate constant
for H2O2 and RCMe2OOH. The experimental pH dependence
can be expressed by the rate law given by eq 10.
due to radical processes occurring in the t-BuOOH system
-
because the effect was absent with ClO3
.
The rate constants for MTO loss and product buildup are
given in Table 3. The weighted mean14 of the rate constants
for the loss of MTO and the buildup of methanol and tert-butyl
methyl ether is (12.9 ( 0.3) × 10-5 s-1, which is about double
that without H3PO2. Higher values of kdec were found for acetone
and tert-butyl methyl peroxide; the weighted mean is (53.2 (
0.8) × 10-5 s-1. One exceptional finding is the following: the
rate of acetone production, normalized by [MTO], is 14 times
higher with H3PO2 and its final concentration 45 times greater.
ksKa[MTO][ROOH]
V )
(10)
[H+]
If mechanism (a) prevails, then pKa ) 12.8 for t-BuOOH
and ks ) 4.6 × 108 L mol-1 s-1. If mechanism (c) provides the
15
preferred pathway, then pKa ) 7.5 for MTO-OH2 and ks )
Discussion
2.3 × 103 L mol-1 s-1. In eq 10, ks represents the corresponding
bimolecular rate constant for the mechanism indicated, ka or
kc. The value of ks is related to kf (eq 8) by the relationship kf
) ksKa. Although credible arguments can be advanced for (a)
over (c), it is useful to consider an alternative that renders the
distinction moot.17 This scheme involves a complex formed from
MTO and ROOH in which the inductive effect of rhenium-
O-Atom Transfer. Reaction 4 represents O-atom transfer
from the alkyl hydroperoxide to the MDO. For the transfer to
occur, the alkyl hydroperoxide coordinates to MDO, forming a
transient intermediate. This species is analogous to those first
formed when MTO combines with various E-OH species (H-
OH, H-OOH, R-OH, etc.).6,7,12,15
(16) Wang, W.-D.; Espenson, J. H. J. Am. Chem. Soc. 1998, 120, 11335.
(17) We are grateful to a reviewer for this suggestion.
(15) Herrmann, W. A.; Fischer, R. W. J. Am. Chem. Soc. 1995, 117, 3223.

748 Inorganic Chemistry, Vol. 38, No. 4, 1999
Brittingham and Espenson
(VII) enhances the acidity of the ROOH proton. This would
lead to the intermediate, I (R ) t-Bu, t-Am). Its different
cleavage reactions would lead to the various products, as detailed
in the next section.
methane, tert-butyl alcohol, ethane, and tert-butyl methyl
peroxide. These products establish the involvement of methyl,
alkoxyl, and alkylperoxyl radicals. The t-BuO• radical continues
the free-radical chemistry by undergoing â-scission form to
acetone and the methyl radical, eq 16, k16 ) 1.4 × 106 s-1 19
.
The methyl radical then abstracts H• from the t-BuOOH to give
methane and tert-butylperoxyl radical, eq 17 (k17 ) 3 × 104 L
mol-1 s-1).20 The tert-butoxyl radical, itself, can also abstract
a hydrogen atom from t-BuOOH to give tert-butyl alcohol and
tert-butylperoxyl radical, eq 18 (k18 ) 1.2 × 106 L mol-1 s-1).21
Another well-characterized reaction is the combination of two
methyl radicals to form ethane, eq 19 (k19 ) 1.6 × 109 L mol-1
s-1).22
Decomposition of MTO and t-BuOOH. The proposed
sequence of reactions for the concomitant decomposition of
MTO and t-BuOOH under argon is presented in eqs 11-20.
This is the simplest scheme we have been able to formulate to
account for the products and the pH dependence.
The mixed dialkyl peroxide formed in the oxidation of
hydrocarbons by ROOH has been taken to indicate free-radical
involvement.18,23 The dialkyl peroxide arises from the reaction
of alkylperoxyl radicals and carbon centered radicals. By
analogy, tert-butyl methyl peroxide is suggested to form from
methyl radical and tert-butyl peroxyl radical, eq 20. This should
be a diffusion-controlled reaction because both species are
unstable radicals.24
Further evidence in support of the involvement of methyl
radicals comes from experiments done in the presence of O2.
Two additional products were found, methyl hydroperoxide and
formaldehyde. Methyl peroxyl radical is formed in the reaction
between methyl radical and oxygen.25 This radical could undergo
a reaction analogous to eqs 17 and 18 to form methyl
hydroperoxide and t-BuOO•: MeOO• + t-BuOOH f MeOOH
+ t-BuOO•. Formaldehyde might then be a product of the
bimolecular self-reaction of MeOO•: 2 MeOO• f MeOH +
CH2O + O2,21 although a reviewer has noted that it might
instead arise from Lewis acid (Re) catalyzed dehydration of
MeOOH.
MeReO3 + t-BuOOH f [Me(O)3Re‚t-BuOOH] (11)
[Me(O)3Re‚t-BuOOH] a
[Me(O)3ResOOBu-t]- ()I) + H+ (12)
-
I f t-BuOMe + ReO4
(13)
(14)
(15)
(16)
(17)
-
I + H2O f MeOH + t-BuOH + ReO4
-
I f t-BuO• + Me• + ReO4
t-BuO• f Me2CO + Me•
Me• + t-BuOOH f CH4 + t-BuOO•
t-BuO• + t-BuOOH f t-BuOH + t-BuOO•
2 Me• f C2H6
(18)
(19)
Consistent with the rate law and its [H+]-1 dependence, the
partitioning reactions of I are, together, rate controlling. That
option, which involves the breaking of covalent bonds, rather
than a rate-controlling reaction between MTO and t-BuOO-,
is proposed for the formation of I, since precedents suggest the
latter would be extremely rapid. According to that scheme, the
general expression for the reaction rate is
Me• + t-BuOO• f t-BuOOMe
(20)
In this scheme, very rapid coordination of t-BuOOH to MTO
to form a neutral complex (eq 11) leads to the active intermedi-
ate I by acid ionization, eq 12. The partitioning of I among
three reactions leads to the products. Dissociation of I forms
t-BuOCH3 and ReO4-, eq 13, cf. eq 2 with H2O2.6 Formation
of t-BuOMe from t-BuO• and Me• must happen in the transition
state or solvent cage, because separated t-BuO• and Me• remain
at concentrations too low for bimolecular recombination.
K11[MTO][ROOH]
1 + K11[ROOH]
k12
k12 + [H+]
d[P]
dt
) (k13 + k14 + k15)
(21)
Cleavage of I to ReO4-, t-BuO•, and CH3 , eq 15, accounts for
•
Because MTO was detected throughout the reaction and [Me-
(O)3Re‚t-BuOOH] was not observed, K11 < 102 L mol-1. Eigen
and co-workers26 report that the rate constant of proton
association in an acid equilibrium reaction (k12) is on the order
of magnitude of 1010 L mol-1 s-1. Therefore K12 is on the order
of magnitude of 10-10 mol L-1. With these assumptions and
•
t-BuO•, and also for the CH3 that is derived from the methyl
group on MTO. Intermediate I undergoes peroxide bond
homolysis; furthermore, the C-Re bond breaks to give the stable
ReO4-. There is precedent for the bond homolysis of a metal
hydroperoxide. The scheme proposed for the hydroxylation of
cycloalkanes by ROOH using [FeIII2O(TPA)2(H2O)2]4+ involves
O-O bond homolysis.18 Reaction 15 was written as a single
step for simplicity of the scheme, but it is more likely that it
happens sequentially. Indeed, eqs 13 and 15 may not be
independent reactions.17 The third mode by which I reacts is
hydrolysis to methanol, tert-butyl alcohol, and perrhenate ions,
eq 14.
(19) Erben-Russ, M.; Michel, C.; Bors, W.; Saran, M. J. Phys. Chem. 1987,
91, 2362.
(20) Bakac, A.; Wang, W.-D. Inorg. React. Mech. 1998, 1, 65.
(21) Bennett, J. E. J. Chem. Soc., Faraday Trans. 1990, 86, 3247.
(22) Hickel, B. J. Phys. Chem. 1975, 79, 1975.
(23) Nguyen, C.; Guajardo, R. J.; Mascharak, P. K. Inorg. Chem. 1996,
35, 6273.
Other steps, all very fast, complete the sequence: the tert-
butoxyl radical and the methyl radical participate in a series of
independently known steps, eqs 16-20, giving rise to acetone,
(24) Kochi, J. K., Ed. Free Radicals; John Wiley and Sons: New York,
1973; Vol. 1.
(25) Suzuki, H.; Fukui, H.; Moro-oka, Y. in The Role of Oxygen in
Chemistry and Biochemistry; Ando, W., Moro-oka, Y., Eds.; Elsevi-
er: New York, 1988; pp 269-272.
(26) Barshop, B. A.; Wrenn, C. F.; Frieden, C. Anal. Biochem. 1983, 130,
134.
(18) MacFaul, P. A.; Ingold, K. U.; Wayner, D. D. M.; Que, L., Jr. J. Am.
Chem. Soc. 1997, 119, 10594.

Reactions of Methylrhenium Oxides
Inorganic Chemistry, Vol. 38, No. 4, 1999 749
•
•
result of recombination of CH3 and CH3CH2 and the self-
reaction of CH3CH2 .
•
The chemical products in the reactions with t-BuOOH and
t-AmOOH are formed in different relative and absolute yields,
as can be seen from a comparison of the data in Tables 1 and
2. A precise mass balance was not feasible since products such
as methane and ethane could not be determined on a comparable
basis. Not only that, but certain products (e.g., MeOH and RMe2-
COMe) form directly, whereas others (e.g., Me2CO and RMe2-
COOR) result from subsequent steps.
Smurova et al. recently reported that the addition of MTO to
Ph(CH3)2COOH, cumyl hydroperoxide, in chlorobenzene re-
sulted in the heterolytic decomposition of the hydroperoxide
and that <1% of the process included radical pathways.10 The
chemistry of ROOH with MTO is much different in aqueous
solution than in chlorobenzene. In fact, the aqueous chemistry
of the decomposition of CH3ReO3 by t-BuOOH and t-AmOOH
more closely resembles the decomposition of cumyl hydroper-
oxide by manganese(II) and -(III) acetylacetonates in chlo-
robenzene.31
Figure 3. KinSim simulations (‚‚‚) of the buildup in total organic
products for the reaction of MTO at three concentrations of t-BuOOH,
0.60 M (×); 0.47 M (∇); and 0.20 M (+). The quantity on the y axis
is [P] ) ([(CH3)2CO] + [CH3OH] + [t-BuOCH3] + [t-BuOOCH3])/
M.
Phosphoric Acid Formation. This product indicates a special
mechanism operates with the ternary system (MTO, t-BuOOH,
and H3PO2). Certain radicals react with H3PO2 and, especially,
the H3PO3 formed from it. This general notion finds support
from the greatly enhanced rate for acetone buildup and the
substantial amount of acetone formed. These findings signal
the generation of additional t-BuO•, the immediate precursor
of acetone in eq 16.19
the experimental value for kf of 7.4 × 10-5 s-1, it follows that
k13 + k14 + k15 < 4 × 103 s-1
.
The reaction scheme consisting of eqs 11-20 was further
explored by simulations using the KinSim program,26-28
a
routine for generating concentration-time curves given the
initial concentrations and rate constants. The mathematical
algorithms are based on Gear and Runge-Kutta calculations.
Certain rate ratios were then optimized using the FitSim
program.29 In this manner, the agreement between the proposed
reaction scheme and the experimental data could be tested. The
experimental concentration-time curves for the decrease in
MTO and increase in CH3OH, (CH3)2CO, t-BuOCH3, and
t-BuOOCH3 for three separate experiments ([MTO]0 ) 10 mM;
[t-BuOOH]0 ) 200, 470, and 600 mM; [HOTf]0 ) 100 mM)
were included as data in KinSim. Literature values for the known
rate constants were used, except for the estimated value of k20
∼2 × 109 mol L-1 s-1, a choice not critical to the analysis.
Rough estimates for the relative ratios of rate constants for steps
13-15 were calculated from the initial rates for product buildup.
The unknown rate constants were fixed and varied, accordingly,
in FitSim. The actual magnitudes of k13, k14, and k15 could not
The methyl radical is postulated to abstract a hydrogen atom
from H3PO3, eq 22, analogous to known reactions of H•, eq
23.32
•
CH3 + H3PO3(or H3PO2) f
CH4 + [P•(O)(OH)2] (or [HP•(O)(OH)]) (22)
H• + H3PO3(or H3PO2) f
H2 + [P•(O)(OH)2] (or [HP•(O)(OH)]) (23)
The rate constants for k23 are 5.0 × 108 L mol-1 s-1 (H3-
PO3) and 4.2 × 109 L mol-1 s-1 (H3PO2).31 Plausibly, the methyl
radical abstracts H• because the bond strengths of H2 (434.1 kJ
mol-1 33 and CH4 (435 kJ mol-1 34
) ) are virtually the same. There
be established, only the ratios could: k14/k13 ) 1.89, k15/k13
)
1.61, and k15/k14 ) 0.85 (reliability, (20%). KinSim was used
to simulate the total product buildup over time. The precision
in measuring small concentrations of each product separately
was rather low. Therefore, the buildup of the sum of product
concentrations is used to represent the reaction course. Figure
3 shows that the simulated values correspond well with the
experimental data, supporting the proposed mechanism.
Decomposition of MTO and t-AmOOH. To help us derive
the scheme for the decomposition of methyltrioxorhenium and
alkyl hydroperoxides, t-amyl hydroperoxide was employed on
the basis of the observed products, the decomposition process
using t-AmOOH goes by a sequence similar to that gone by
the process using t-BuOOH. There will, of course, be some
minor differences in the overall scheme. The t-AmO• is known
to undergo â-scission at a higher rate because it forms the more
stable ethyl radical.24,30 Also, additional gaseous products (e.g.
propane and ethylene) are formed in the t-AmOOH case as a
are, of course, differences in the rates of H-atom abstraction by
•
H• and CH3 . For example, rate constants for R-hydrogen
abstraction from alcohols are approximately 4 orders of
• 35,36
magnitude greater with H• than with CH3 .
The lower limit of k22 can be estimated from these data to be
on the order of >106 mol L-1 s-1, with an error of 1 order of
magnitude, which is reasonable in that reaction 22 should be
slower than reaction 23. With the long-chain approximation,37
justified from the products formed, k24 >106 L mol-1 s-1. The
phosphorus centered radicals generated in eq 22 are likely to
(31) Abdalla, A. A.-S.; Ivanchenko, P. A.; Solyanikov, V. M. Kinet. Catal.
(Transl. of Kinet. Katal.) 1993, 34, 432.
(32) Muratbekov, M. B.; Seriev, A. S. High Energy Chem. (Transl. of Khim.
Vys. Energ.) 1983, 17, 432.
(33) Cotton, F. A.; Wilkinson, G. AdVanced Inorganic Chemistry, 5th ed.;
John Wiley & Sons: New York, 1988.
(34) Lowry, T. H.; Richardson, K. S. Mechanism and Theory in Organic
Chemistry, 3rd ed.; Harper Collins: New York, 1987.
(35) Thomas, J. K. J. Phys. Chem. 1967, 71, 1919.
(36) Smaller, B.; Avery, E. C.; Remko, J. R. J. Chem. Phys. 1971, 55,
2414.
(37) Espenson, J. H. Chemical Kinetics and Reaction Mechanisms, 2nd
ed.; McGraw-Hill: New York, 1995.
(27) Frieden, C. Trends Biochem. Sci. 1993, 18, 58.
(28) Frieden, C. Methods Enzymol. 1994, 240, 311.
(29) Zimmerle, C. T.; Frieden, C. Biochem. J. 1989, 258, 381.
(30) Walling, C. Pure Appl. Chem. 1967, 15, 69.

750 Inorganic Chemistry, Vol. 38, No. 4, 1999
Brittingham and Espenson
attack the abundant hydroperoxide, eq 24, to form additional
tert-butoxyl radicals that lead, in turn, to acetone,19 eq 16, and
to methyl radicals, further propagating the chain.
Acknowledgment. This work was supported by the National
Science Foundation Grant (CHE-9007283). We are grateful to
Dr. Michael T. Ashby for supplying the CH2DSn(n-butyl)3 and
to Dr. Andreja Bakac, Dr. Kamel Harrata, and Sahana Mollah
for discussions and assistance.
[P•(O)(OH)2] (or [HP•(O)(OH)]) + t-BuOOH f
H3PO4 (or H3PO3) + t-BuO• (24)
IC9809052