Decomposition of tert-butyl hydroperoxide into tert-butyl alcohol and O2 catalyzed by birnessite-type manganese oxides: Kinetics and activity

Article abstract of DOI:10.1016/j.catcom.2014.01.028

Birnessite-type manganese oxides (M-OL-1s, M = K, Mg, Fe, Ni and Cu) are first reported to efficiently catalyze the decomposition of tert-butyl hydroperoxide (TBHP) into tert-butyl alcohol (TBA) and O₂ with a 100% selectivity towards TBA under heterogeneous conditions. The same form of overall second-order kinetic equations is fitted out for the M-OL-1s and explained by the proposed mechanism. Life tests and XRD analyses demonstrate no losses in both the activity and the birnessite-type structure after the reaction.

Full text of DOI:10.1016/j.catcom.2014.01.028

Catalysis Communications 49 (2014) 6–9
Contents lists available at ScienceDirect
Catalysis Communications
journal homepage: www.elsevier.com/locate/catcom
Short Communication
Decomposition of tert-butyl hydroperoxide into tert-butyl alcohol and O2
catalyzed by birnessite-type manganese oxides: Kinetics and activity
a
a,
a
b,
a
Lin Qi , Xingyi Qi ⁎, Lili Wang , Lili Feng ⁎, Shupei Lu
a
School of Chemistry and Environment, Beihang University, Beijing 100191, PR China
Key Laboratory of Urban Stormwater System and Water Environment of Ministry of Education, Beijing University of Civil Engineering and Architecture, Beijing 100044, PR China
b
a r t i c l e i n f o
a b s t r a c t
Article history:
Birnessite-type manganese oxides (M-OL-1s, M = K, Mg, Fe, Ni and Cu) are first reported to efficiently catalyze
Received 23 November 2013
Received in revised form 27 January 2014
Accepted 28 January 2014
Available online 5 February 2014
2
the decomposition of tert-butyl hydroperoxide (TBHP) into tert-butyl alcohol (TBA) and O with a 100% selectiv-
ity towards TBA under heterogeneous conditions. The same form of overall second-order kinetic equations is
fitted out for the M-OL-1s and explained by the proposed mechanism. Life tests and XRD analyses demonstrate
no losses in both the activity and the birnessite-type structure after the reaction.
©
2014 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license
http://creativecommons.org/licenses/by-nc-nd/3.0/).
Keywords:
(
tert-Butyl alcohol
tert-Butyl hydroperoxide
Deperoxidation
Birnessite
Kinetics
1
. Introduction
Oxidations of hydrocarbons are of considerable interest to the fine
to be separated from the obtained mixtures by fractional distillation. Such
treatment constitutes another dilemma of degradation and deactivation
of catalysts. Furthermore, the TBHP decomposition in these studies
tends to be accompanied by the formation of some undesirable impuri-
ties, di-tert-butyl peroxide, acetone, methanol, etc. [9,10,12–14]. There-
fore, the development of highly selective and recyclable solid catalysts
is eagerly expected for this TBHP transformation. Birnessite is a natural
manganese oxide phase with a two-dimensional layered structure
chemicals industry and academia. Many hydrocarbon compounds can
undergo oxidation to afford valuable oxygenated derivatives [1,2]. The
oxidation of isobutane should be exemplified as one of the most impor-
tant oxidations in this regard, which usually generates a mixture of TBA
and TBHP under liquid-phase conditions [3–6]. As is well known, TBA, a
tert-butyl functional reagent, has witnessed its multiple applications
in the syntheses of pharmaceuticals, agrochemicals and other fine
chemicals. Moreover, TBA has received additional attention owing to
the increasing significance of methyl tert-butyl ether (MTBE) as a gaso-
line additive since MTBE can be prepared by etherification of TBA with
methanol [7,8]. Nevertheless, TBHP must be eliminated from the mix-
ture because of its deleterious effects on the above TBA-related usages.
Hence, much research work published hitherto focuses on TBA rather
than TBHP in the aerobic oxidation of isobutane, with an emphasis on
exploring new effective catalytic systems for the direct decomposition
6
consisting of edge-shared MnO octahedra with cations and water
molecules situated in the negative interlayer region; birnessite is des-
ignated as M-OL-1 due to the above octahedral layered (OL) structure,
where M denotes the exchangeable interlayer cation [16].
Herein, we present the results of kinetics and activity for the catalytic
2
decomposition of TBHP into TBA and O with the M-OL-1s (M = K, Mg,
Fe, Ni and Cu; for the XRD patterns and chemical formulas, see Fig. S1
and Table S1).
2. Experimental
2
of TBHP into TBA and O . Until now, there has been some literature
concerning the preparation of TBA via catalytic decomposition of
TBHP, employing transition-metal-containing catalysts mostly in the
form of metal complexes [9–15].
From an industrial viewpoint, there is still plenty of room for im-
provement in many reported cases. One reason for this is clearly attribut-
able to the disadvantage associated with homogeneous catalysis involved
in the TBHP decomposition process where for its final utilization, TBA has
2
.1. Synthesis of the M-OL-1s
K-OL-1 was prepared from the procedure [16] and then M-
exchanged to obtain the other M-OL-1s.
2.2. Kinetic tests for TBHP decomposition
The reaction runs were performed in a 25-mL glass reactor (Fig. S2).
For details, see the Supplementary data.
⁎
Corresponding authors.
http://dx.doi.org/10.1016/j.catcom.2014.01.028
566-7367/© 2014 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/).
1

L. Qi et al. / Catalysis Communications 49 (2014) 6–9
7
3
. Results and discussion
TBHP in the reaction system is thoroughly deperoxidated into TBA
and O ; at a later time t, the concentration of TBHP is [TBHP] and
the subtraction of [TBHP] from [TBHP] is proportional to V with
the same coefficient λ, viz. [TBHP] ; hence the
2
3
.1. Kinetics
0
t
0
− [TBHP] = λV
t
For a real-time kinetic analysis of TBHP decomposition, the change in
first-order integral equation of TBHP decomposition can be arrived
TBHP concentration with time must be accurately traced while the reac-
tion is in progress. Only because of the formation of di-tert-butyl perox-
ide and fragmented products (e.g. acetone and methanol), was the
quantitative measurement of the time-dependent TBHP concentration
therein conducted with iodometric titration [10,14] and in situ UV–vis
spectroscopy [13], respectively. Provided that by an appropriate choice
of catalysts and reaction conditions, TBHP is completely decomposed
at as in Eq. (4) by integrating Eq. (3):
ꢀ
ꢁ
1
′
1
V∞
ln
¼ k t þ ln
:
ð4Þ
V∞−Vt
Consequently, the pseudo-first-order kinetics of TBHP in Eq. (3)
leads to a straight line of ln[1 / (V )] versus t defined by Eq. (4),
∞
− V
t
into TBA and O
2
, the rate of TBHP disappearance will be followed by
(V ) evolved from the reaction system
whose slope is the pseudo-first-order rate constant k′ in Eqs. (3)
and (4).
monitoring the volume of O
2
t
with time (t). The case is true here in our studies due to the observation
that TBA was yielded as a single organic product (Table 1). In addition,
neither induction period nor deactivation occurred for the M-OL-1s
under the selected conditions.
The catalytic decomposition of TBHP with the M-OL-1s is supposed
to have a rate law equation involving TBHP and M-OL-1 as in Eq. (1):
In contrast, if the order α in Eq. (2) is assumed to be 2, the resultant
pseudo-second-order integrated rate expression can be written
into Eq. (5), thus a straight line being predicted with a slope of λk′
when 1 / (V
∞
t
− V ) is plotted against t:
1
′
1 :
¼
λk t þ
ð5Þ
V −V
V
∞
∞
t
d½TBHPꢀ
α
β
−
¼ k ½TBHPꢀ ½M‐OL‐1ꢀ
ð1Þ
a
dt
Fig. 1 represents the results of the TBHP decomposition catalyzed by
Cu-OL-1 at four [Cu-OL-1]s, where ln[1 / (V − V )] was plotted against
∞
t
a
where k is the apparent rate constant, α the order in TBHP, β the order
t. As shown in Fig. 1, the four sets of kinetic data can be linearly fitted by
the least squares, suggesting that the order α in Eqs. (1) and (2) has a
value of 1; although the net reaction does not implicate Cu-OL-1, k′,
which was measured from the slope, increases with [Cu-OL-1] for a
in M-OL-1 (the overall order the sum of α and β), [TBHP] the concentra-
tion of TBHP and [M-OL-1] the formal concentration of M-OL-1 (sieved
to −300/+500 mesh); [M-OL-1] was calculated by dividing the desired
loading of M-OL-1 (mg) by the total volume of TBHP and acetonitrile
given [TBHP]
0
and a given temperature.
(
mL). If no deactivation of M-OL-1 exists and [M-OL-1] is unchanged
β
In consideration of k′ = k
a
[Cu-OL-1] , a new form of k′ expression is
β
throughout the course of the reaction, [M-OL-1] in Eq. (1) is so left
deduced as in Eq. (6):
constant and absorbed into k
a
, giving rise to the underlying rate law
equation:
′
lnk ¼ β ln½Cu‐OL‐1ꢀ þ lnk
ð6Þ
a
d½TBHPꢀ
′
α
−
¼ k ½TBHPꢀ
ð2Þ
so that another type of straight line may arise from a plot of lnk′ versus
ln[Cu-OL-1], its slope being the order β. Indeed, the plot of lnk′ versus ln
[Cu-OL-1] (Fig. 2) gives a straight line with a slope of 1.0690 ± 0.0291.
The order β in Eq. (6) is then fitted to be equal to 1. As a result, the TBHP
decomposition has been determined kinetically to be first order in both
TBHP and Cu-OL-1, i.e. second order overall. As usual, the Arrhenius-
dt
β
a
where k′ = k [M-OL-1] is pseudo rate constant. According to Eq. (2),
the simplest rate law concerns a first-order reaction of TBHP decompo-
sition (α = 1) as in Eq. (3):
type behavior was exhibited between k′ and reaction temperature (T/K)
and so an apparent activation energy E (142 ± 9 kJ mol ) was calcu-
a
d½TBHPꢀ
′
−
¼ k ½TBHPꢀ
ð3Þ
−1
dt
lated from the Arrhenius plot of lnk′ versus 1 / T (Figs. S3,S4). The
success in these linear fittings negates the possibility of coexistence of
other side reactions otherwise the deviation from the specific form of
which is then integrated in the time interval of zero to t. Noting that
at t = 0, the initial concentration of TBHP, [TBHP] is proportional to
, viz. [TBHP] = λV , where V is the volume of O evolved when
0
V
∞
0
∞
∞
2
-
0.90
-1.05
1.20
-1.35
Table 1
Kinetics and activity for the M-OL-1s to catalyze the decomposition of TBHP into TBA and
O .
2
(
d)
-
a
a
a
4b
M-OL-1
α
β
E
(
a
k′ × 10
Conversion
(%)
kJ mol−1)
(s
−1
)
-
-
1.50
1.65
K-OL-1
1
1
1
1
1
1
1
1
1
1
83
6.8792
7.1120
7.2875
8.3583
25.9000
78c
(c)
(b)
d
(
85
76)
c
-1.80
-1.95
Mg-OL-1
Fe-OL-1
Ni-OL-1
Cu-OL-1
98
95
59
d
d
d
(
82
83)
c
-
2.10
-2.25
2.40
-2.55
(
87
80)
c
(a)
(
89)
-
c
142
98
96)d
(
a
0
50
100
150
200
250
300
350
400
a
α, β and E were determined from Figs. 1,2 and S3–S20.
b
time (s)
2
k′ was obtained under the conditions where T 338 K, TBHP (65 wt.% in H O) 1.00 mL,
−
1
[
M-OL-1] 8.33 mg mL and acetonitrile 5.00 mL, respectively.
c
t 1 h, the other conditions were the same as in footnote b; the TBHP conversion was
determined by GC with a 100% selectivity towards TBA for all the M-OL-1s.
Fig. 1. Plots of ln[1 / (V∞ − V )] versus t. Reaction conditions: T 338 K, TBHP (65 wt.% in
t
H O) 1.00 mL, [Cu-OL-1] 1.67 mg mL−1, [Cu-OL-1] 3.33 mg mL−1, [Cu-OL-1]c
2
a
b
d
−1
−1
The parenthesized TBHP conversion was determined by the V
t
/ V
∞
ratio (t 1 h).
5.00 mg mL , [Cu-OL-1]d 8.33 mg mL , acetonitrile 5.00 mL.

8
L. Qi et al. / Catalysis Communications 49 (2014) 6–9
-5.8
-6.0
-6.2
-6.4
-6.6
-6.8
-7.0
-7.2
-7.4
-7.6
-7.8
(d)
β = B (slope) = 1.0690 ± 0.0291; R = 0.9993.
(
c)
Scheme 1. The mechanism for the catalytic decomposition of TBHP into TBA and O
2
with
Cu-OL-1. k ≪ k−1 and k ; [{TBHP∙∙∙Cu-OL-1}] ≪ [TBHP] and [Cu-OL-1].
1
2
(b)
In accordance with the above mechanism, the rate law of the catalytic
decomposition of TBHP into TBA and O with Cu-OL-1 can be formulated
by the steady-state approximation as in Eq. (7):
2
(a)
d½TBHPꢀ
′
−
¼ k K½TBHPꢀ½Cu‐OL‐1ꢀ ¼ k ½TBHPꢀ
ð7Þ
1
dt
0
.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2
where K, equilibrium constant is the k
constant (k ) in Eq. (1), and k′ = k
2
/ k−1 ratio, k
1
K the apparent rate
[Cu-OL-1] the
ln[Cu-OL-1]
a
a
[Cu-OL-1] ≈ k
a
0
pseudo-first-order rate constant in Eqs. (3) and (4). As we have seen
now, the expression in Eq. (7) conforms to the first-order behavior
already noticed in both TBHP and Cu-OL-1. This suggests that the overall
second-order kinetics of the Cu-OL-1-catalyzed TBHP decomposition can
be rationalized by the mechanism proposed in Scheme 1.
Fig. 2. Plot of ln k′ versus ln[Cu-OL-1].
kinetics (e.g. α = 1 in Fig. 1) would be brought about to some extent. In
this context, the argument above was further confirmed by the data
collated in Table 1 where TBA was detected to be the exclusive organic
product by GC. Moreover, no linear relationship was established
3.3. Comparison of kinetics and activity
As shown in Table 1, the M-OL-1s have the same α and β, viz. α = 1
and β = 1 (Figs. 1,2 and S3–S20), revealing the generality of the mech-
anism for the TBHP decomposition; although the reason for disparities
among kinetic details and activity (e.g. E
the effective transformation of TBHP into TBA and O
between 1 / (V
of 1 / (V − V ) with t was observed (Fig. 3), thus rejecting a second
order of α in TBHP.
∞
− V
t
) and t but a perfect exponential growth
∞
t
a
and k′) is unknown yet now,
proceeds smoothly
2
3
.2. Mechanism
over the M-OL-1s; the preferred Cu-OL-1 possesses the highest activity
in terms of both k′ and the TBHP conversion while no deperoxidation of
TBHP was observed for cupric nitrate (used with the same Cu loading as
In order to accommodate the obtained kinetic fittings, the mecha-
nism for the Cu-OL-1-catalyzed TBHP decomposition is postulated as
in Scheme 1.
d
that of [Cu-OL-1] in Fig. 1), implying the occurrence of some synergetic
effect between the interlayer Cu and framework Mn cations; the com-
parability between two forms of the TBHP conversion values listed in
Table 1 coincides well with a 100% selectivity towards TBA for all the
reaction runs (as indicated in footnotes c and d), further indicating no
other side reactions (if not so, the GC-determined TBHP conversion
In Scheme 1, {TBHP∙∙∙Cu-OL-1}, an activated intermediate is formed
through the simple bimolecular reaction between TBHP and Cu-OL-1
in the pre-equilibrium step; then {TBHP∙∙∙Cu-OL-1} decays into TBA
and O
step; k
2
and releases its moiety of Cu-OL-1 in the rate-determining
is rather small in value as compared with k−1 and k which
t ∞
would become inconsistent with that from the V / V ratio); with
1
2
are both responsible for the pre-equilibrium; {TBHP∙∙∙Cu-OL-1} is in a
so-called steady state, viz. d[{TBHP∙∙∙Cu-OL-1}] / dt = 0; the contribu-
tion of [{TBHP∙∙∙Cu-OL-1}] to [TBHP] and [Cu-OL-1] may be neglected
in setting up the rate law equation for {TBHP∙∙∙Cu-OL-1}.
the TBHP conversion almost unaltered, there appears no sign of deacti-
vation of K-OL-1 and Cu-OL-1 when examined up to six cycles
(Table S2); the two birnessite-type XRD peaks remain intact for the
two spent M-OL-1s above and so the OL-1 structure bears an intrinsic
0
0
0
0
0
0
0
0
0
0
.30
.28
.26
.24
.22
.20
.18
.16
.14
.12
-
-
-
-
-
-
-
1.20
1.35
1.50
1.65
1.80
1.95
2.10
(a)
(b)
0
50
100
150
200
250
300
350
400
time (s)
Fig. 4. XRD patterns of the two fresh and spent M-OL-1s (M = K and Cu, one cycle). Reac-
−
1
Fig. 3. Two fittings of the data of V
t
and V
∞
. (a): linear fitting of ln[1 / (V
∞
− V
t
)] versus t;
tion conditions: T 338 K, t 3 h, TBHP (65 wt.% in H
acetonitrile 5.00 mL.
2
O) 1.00 mL, [M-OL-1] 8.33 mg mL
,
(
b): exponential fitting of 1 / (V − V
∞
t
) versus t. The line of (a) here is that of (d) in Fig. 1.

L. Qi et al. / Catalysis Communications 49 (2014) 6–9
9
endurance enough under the selected heterogeneous conditions
Fig. 4); ICP-AES analyses also show no leaching of K, Cu and Mn in
the mother liquors removed from the spent K-OL-1 and Cu-OL-1.
Appendix A. Supplementary data
(
Supplementary data to this article can be found online at http://dx.
doi.org/10.1016/j.catcom.2014.01.028.
4
. Conclusions
References
On a commercial scale, TBA is mainly coproduced with propylene
[
[
[
1] A. Gunay, K.H. Theopold, Chem. Rev. 110 (2010) 1060–1081.
2] A.K. Suresh, M.M. Sharma, T. Sridhar, Ind. Eng. Chem. Res. 39 (2000) 3958–3997.
3] U. Shah, S.M. Mahajani, M.M. Sharma, T. Sridhar, Chem. Eng. Sci. 55 (2000) 25–35.
epoxide from the metal-catalyzed epoxidation of propylene using
TBHP as an oxidant. However, a more convenient synthetic route to
TBA may be developed based on the effective one-step transformation
of TBHP into TBA. Therefore, the M-OL-1s with such a capacity as above
highlight a possibility of their practical applications in obtaining TBA di-
rectly from the TBHP decomposition. And the results here may provide
some enlightening information for efficient, catalytic deperoxidations
of other alkyl hydroperoxides into their corresponding alcohols.
[4] K.T. Moore, I.T. Horváth, M.J. Therien, Inorg. Chem. 39 (2000) 3125–3139.
[
[
[
5] K.T. Moore, I.T. Horváth, M.J. Therien, J. Am. Chem. Soc. 119 (1997) 1791–1792.
6] T.M. Nenoff, M.C. Showalter, K.A. Salaz, J. Mol. Catal. A Chem. 121 (1997) 123–129.
7] E. C. Nelson, D. A. Storm, M. S. Patel. US Patent 4 918 244 (1990).
[8] J. F. Knifton, J. R. Sanderson. US Patent 5 220 078 (1993).
[9] J. R. Sanderson, J. F. Knifton, E. T. Marquis. US Patent 5 414 163 (1995).
10] R. Hiatt, K.C. Irwin, C.W. Gould, J. Org. Chem. 33 (1968) 1430–1435.
11] T. Wijesekera, J. E. Lyons, P. E. Ellis, Jr., M. V. Bhinde. US Patent 5 767 272 (1998).
12] J. R. Sanderson, E. T. Marquis, R. A. Meyer, T. L. Renken. US Patent 4 704 482 (1987).
[
[
[
Acknowledgments
[13] N. Turrá, U. Neuenschwander, A. Baiker, J. Peeters, I. Hermans, Chem. Eur. J. 16
2010) 13226–13235.
(
[
[
14] W.H. Richardson, J. Am. Chem. Soc. 88 (1966) 975–979.
15] H. Berger, A.F. Bickel, Trans. Faraday Soc. 57 (1961) 1325–1335.
This work was supported by the National Natural Science Foundation
of China (No. 51206009).
[16] Y. Ma, J. Luo, S.L. Suib, Chem. Mater. 11 (1999) 1972–1979.