On the activation of Mo2B and MoB catalysts in oct-1-ene epoxidation with tert-butyl hydroperoxide

Article abstract of DOI:10.1134/S0023158414020062

The activation of Mo₂B and MoB catalyst in the epoxidation of oct-1-ene with tert-butyl hydroperoxide is reported. The activation process is described by the Avraami-Erofeev topokinetic equation and includes two successive steps, viz.; the nucleation and formation of a new active phase. The epoxide is produced only when the activated form of the catalyst is involved. The effective and topochemical constants of the process have been determined.

Full text of DOI:10.1134/S0023158414020062

ISSN 0023ꢀ1584, Kinetics and Catalysis, 2014, Vol. 55, No. 2, pp. 212–216. © Pleiades Publishing, Ltd., 2014.
Original Russian Text © M.V. Nykypanchuk, Z.M. Komarenskaya, M.O. Chernii, 2014, published in Kinetika i Kataliz, 2014, Vol. 55, No. 2, pp. 221–225.
On the Activation of Mo B and MoB Catalysts in Octꢀ1ꢀene
2
Epoxidation with tertꢀButyl Hydroperoxide
M. V. Nykypanchuk, Z. M. Komarenskaya*, and M. O. Chernii
Lviv Polytechnic National University, Lviv, 79013 Ukraine
*
eꢀmail: zkomaren@mail.ru
Received October 14, 2012; in final form, June 13, 2013
Abstract—The activation of Mo B and MoB catalyst in the epoxidation of octꢀ1ꢀene with tertꢀbutyl hydroꢀ
2
peroxide is reported. The activation process is described by the AvraamiꢀErofeev topokinetic equation and
includes two successive steps, viz., the nucleation and formation of a new active phase. The epoxide is proꢀ
duced only when the activated form of the catalyst is involved. The effective and topochemical constants of
the process have been determined.
DOI: 10.1134/S0023158414020062
Molybdenum boride catalysts are very efficient in
The solvent was toluene (reagent grade), which was
olefin epoxidation with tertꢀbutyl hydroperoxide [1– purified according to a standard procedure.
4
]. With these catalysts, the hydroperoxide consumpꢀ
The epoxidation of octꢀ1ꢀene with tertꢀbutyl
hydroperoxide was performed in a glass reactor
equipped with a temperatureꢀcontrolled jacket to
maintain a constant temperature with an accuracy of
tion rate with time [5–9]. This increase in the reaction
rate in the case of Мо В catalyst is mainly due to the
partial dissolution of the catalyst and the formation of
an active homogeneous form of the catalyst in the
reaction medium [6]; in the case of Мо В, MoB, and
2
5
±
0.1°С and with a magnetic bar, a reflux condenser,
2
and a sampler. The rotational speed of the magnetic
stirrer was over 900 rpm, ensuring that the process is
kinetically controlled (i.e., the reaction rate is indeꢀ
pendent of the stirring speed and reactor volume). The
reactor was loaded with the reactants and solvent, the
mixture was heated under stirring to the preset reacꢀ
tion temperature, and the catalyst was added. The
instant the catalyst was added was taken to be the reacꢀ
tion onset time. The reaction mixture was sampled at
intervals and was analyzed for the hydroperoxide and
epoxide contents.
МоВ2, this is due to the activation of the catalyst surꢀ
face [7–9].
While the epoxidation of some unsaturated comꢀ
pounds (octꢀ1ꢀene and ꢀethylallyl ethylacrylate) in
α
the presence of activated forms of these catalysts has
been investigated in sufficient detail [5, 7–9], the catꢀ
alyst activation step remain obscure.
In the present work, we studied the activation of
molybdenum boride catalysts by the reaction medium
in the oxidation of octꢀ1ꢀene with tertꢀbutyl hydroperꢀ
oxide in toluene.
The hydroperoxide was determined by iodometric
titration with sodium thiosulfate [11]. The epoxide
was quantified on a LKhMꢀ72 (Russia) chromatoꢀ
graph with a thermalꢀconductivity detector. To sepaꢀ
EXPERIMENTAL
Octꢀ1ꢀene (Oct) was obtained by the pyrolytic
decomposition of
nꢀoctyl acetate. The resulting
rate the mixture, a column (3 m
0% Apiezon on Chromatone NꢀAW was used. The
oven temperature was 450 K and the carrier gas
hydrogen) flow rate was 3.6 L/h. Specialꢀpurpose
×
4 mm) packed with
hydrocarbon was dried with calcium chloride and was
distilled using a glassꢀpacked laboratory column
30 theoretical plates). The purity of octꢀ1ꢀene was
checked chromatographically and kineticall (comparꢀ
ing the reproducibility of results for oxidation of difꢀ
ferent portions of the hydrocarbon). The purity of the
resulting product was at least 99%.
1
(
(
experiments showed that the epoxide determination
accuracy is unaffected by the presence of hydroperoxꢀ
ide in the reaction mixture.
tertꢀButyl hydroperoxide (TBHP) was prepared
and purified via a standard procedure [10].
The rates of hydroperoxide consumption and
epoxide formation were determined by drawing tanꢀ
Reagentꢀgrade Мо В and MoB with a specific surꢀ gents to the kinetic curves of tertꢀbutyl hydroperoxide
2
2
face area of 0.220 and 0.3 m /g, respectively, were used consumption and epoxide accumulation with time,
without additional purification.
respectively.
212

ON THE ACTIVATION OF Mo B AND MoB CATALYSTS
213
2
a
0
4
–1 –1
(
a)
w × 10 , mol L s
a
0
4
–1 –1
5
4
3
2
1
w × 10 , mol L s
2
1
1
0
.0
.5
.0
.5
0
2
1
3
1
2
0
500
1000 1500 2000 2500
, s
2
0
0
0
0.2 0.4 0.6 0.8 1.0 1.2 1.4 [Cat] , m /L
0
t
0.2 0.4 0.6 0.8 1.0 [TBHP] , mol/L
0
Fig. 2. Time variation of the tertꢀbutyl hydroperoxide conꢀ
sumption rate for the ( ) Mo B and ( ) MoB catalysts.
[TBHP] = 0.5 mol/L, [Oct] = 2.5 mol/L, [Cat] =
w
1
2
2
0
1
2
3
4 [Oct] , mol/L
0
0
0
2
0
.22 m /L, T = 360 K.
(
b)
a
0
4
–1 –1
w × 10 , mol L s
The reaction in the presence of Мо В is firstꢀorder
2
with respect to octꢀ1ꢀene throughout its concentration
range. In the presence of MoB at olefin concentraꢀ
tions of >2.0 mol/L, there is a region of saturation and
a region where the reaction order with respect to this
reactant decreases from 1 to 0.
5
4
2
1
3
2
1
The saturation of the functions w = f([Cat] ) and
0
3
w = f([Oct] ) (for MoB) is probably due to complexꢀ
0
ation between the catalyst and the olefin. The MoB–
octꢀ1ꢀene complex appears to be more stable than the
Мо2B–octꢀ1ꢀene complex, and, as a consequence,
2
0
0.4 0.8 1.2 1.6 2.0 2.4 [Cat] , m /L
0
the function w = f([Oct] ) in the presence of MoB flatꢀ
tens out even at octꢀ1ꢀene concentrations over
0
0
0
0.4
2
0.8
4
1.2 [TBHP] , mol/L
0
2.0 mol/L.
6 [Oct] , mol/L
0
Similar dependences of the reaction rate on the
hydroperoxide, olefin, and catalyst concentrations
were also observed for the activated forms of these catꢀ
alysts [7, 8].
Fig. 1. Initial hydroperoxide consumption rate wa for the
0
(
a) Mo Band (b) MoB catalysts as a function of (
1
) the
catalyst content of the reaction mixture ([TBHP]0
.5 mol/L, [Oct] = 2.5 mol/L) and ( ) the initial conꢀ
tertꢀbutyl hydroperoxide ([Cat]0
.22 m /L, [Oct] = 2.5 mol/L) and ( ) octꢀ1ꢀene
[TBHP] = 0.5 mol/L, [Cat] = 0.22 m /L). = 360 K.
2
=
Figure 2 illustrates the time variation of the tert
ꢀ
0
2, 3
0
centrations of (
0
(
2
0
)
=
butyl hydroperoxide consumption rate for the Мо В
2
2
3
and MoB catalysts. In both cases, the reaction rate
increases up to a certain value and then begins to
decrease. The time required for the reaction rate to
reach the peak value and the peak value itself depend
on the nature of the catalyst, and, in the presence of
2
T
0
0
RESULTS AND DISCUSSION
Figure 1 shows the initial rate of hydroperoxide MoB, it is almost two times shorted than in the presꢀ
consumption as a function of the initial concentraꢀ ence of Мо В. The initial reaction rates are also in a
2
tions of octꢀ1ꢀene and tertꢀbutyl hydroperoxide and ratio of ~2 : 1, although the maximum rates differ less
the catalyst content in the reaction mixture in the significantly.
presence of the unactivated forms of (a) Мо В and
2
Under the assumption that this change in the reacꢀ
tion rate with time is caused by the change in the catꢀ
As can be seen from Fig. 1, the reaction order with alyst’s activity, as was shown in [12], this process
(
b) MoB.
respect to hydroperoxide in the presence of the unacꢀ appears to be complete in the region of maximum
tivated catalyst is unity in both cases. The reaction reaction rate. If the catalyst activity then remains
order with respect to the catalyst changes from unity unchanged, the observed postꢀmaximum decrease in
2
(
at a catalyst content of 0–0.5 m /L) to zero at higher the reaction rate is due to the decrease in the current
catalyst contents.
hydroperoxide concentration. The data shown in
KINETICS AND CATALYSIS Vol. 55
No. 2
2014

214
NYKYPANCHUK et al.
activated catalyst contents ([Catunact], [Cat ) can be
expressed as follows:
act]
4
–1
(
wt/[TBHP]t) × 10 , s
1
2
0
8
6
4
2
act
unact
⎡
⎣
Cat ⎤ =
[
Cat
]
α
and ⎡⎣ Cat ⎤ =
[
Cat
]
(
1−α .
)
0
⎦
0
1
⎦
y
= 7.182x + 3.529
R2 = 1
In this case, the current rate of hydroperoxide conꢀ
sumption can be related to the degree of catalyst actiꢀ
vation by the linear equation
2
1
y
= 9.994x
= 1
+ 1.645
wt
TBHP t
unact
act
eff
unact
=
keff [Cat] +(k −keff )[Cat] α.
(3)
R
0
0
2
[
]
wt
TBHP]t
Figure 3 plots the
=
f(α) dependence,
[
0
0.2
0.4
0.6
0.8
1.0
α
which is indeed linear with a large correlation coeffiꢀ
cient.
Using the Y axis intercepts of the function
Fig. 3. Specific reaction rate as a function of the degree of
wt
activation of the (
1
) Mo B and (
2) MoB catalysts.
=
f
(
α
)
and the slope ratios of the straight
2
[
0
TBHP] = 0.5 mol/L, [Oct] = 2.5 mol/L, [Cat] =
[TBHP]t
0
0
0
2
.22 m /L,
T
= 360 K.
lines of this function, the kact and
unact
k
values were
eff
eff
unact
10 s–1 and
–4
act
calculated to be
k
= 1.6
×
k
=
eff
–4 –1
eff
–4
–1
B and kunacteff = 3.5
× 10 s
Fig. 2 suggest also that MoB is activated much more 11.6
and kact = 10.7
× 10 s for MoB.
rapidly than Мо В
×
10 s for Mo
2
–4
–1
.
2
eff
The degree of activation of the catalyst ( ) was calꢀ
α
The data obtained suggest that the unactivated
form of MoB has indeed a twice higher activity than
culated from the time dependence of the reaction rate.
Since the reaction rate depends on the change in both
the catalyst activity and the current hydroperoxide
concentration, the following equation was used to calꢀ
the unactivated Мо В, whereas the activated form of
2
MoB is less active than the activated Мо В catalyst.
2
The curves of
α
as a function of the reaction time
culate
α
:
are Sꢀshaped, which is typical of topochemical proꢀ
cesses (Fig. 4). The Avraami–Erofeev equation was
⎛
⎞
⎛
⎞
w
w
−
⎜
⎝
⎟
⎠
⎜
⎝
⎟
⎠
used in the quantitative description of the
function [13–15]
α
= f(t)
[
TBHP
]
[
TBHP
]
t
0
α =
,
(1)
⎛
⎞
⎛
⎞
w
w
n
−
−ktt
⎜
⎝
⎟
⎠
⎜
⎝
⎟
⎠
α = 1 − e
,
(4)
[
TBHP
]
[
TBHP
]
max
0
where kt and
n
are topochemical constants of the proꢀ
⎛
⎜
⎞
⎛
⎞
⎟
⎛
⎞
cess.
w
w
w
where
, ⎜
⎝
, and ⎜
are
⎟
⎠
⎟
⎠
The kt and
n values were determined by a graphoꢀ
[
TBHP
]
[
TBHP ⎠
]
[
⎝
TBHP
]
⎝
0
t
max
analytical method based on the linear relationship
the initial, current, and maximum values of the
reduced rate of hydroperoxide consumption, respecꢀ
tively.
Since the rate of hydroperoxide consumption
involving both the unactivated and activated forms of
catalyst changes linearly with the hydroperoxide conꢀ
centration, the total reaction rate for both catalysts
can be represented as follows:
ln(−ln(1 − α)) = ln k + nln t,
(5)
t
which should be observed when the catalyst activation
is describable in terms of Eq. (4).
For both catalysts, there are two linearꢀlaw regions
describing activation, which are characterized by difꢀ
ferent topochemical constants (Fig. 5). While in the
first region the topochemical constants n1 are less
than 2 and are similar for the two catalysts (n1 = 1.7
unact
unact
act
act
w = keff [TBHP][Cat
] + k [TBHP][Cat ],(2)
eff
for Mo B and n1 = 1.4 for MoB), in the second region
2
unact
act
where keff and keff are effective rate constants of the
reaction involving the unactivated and activated forms
of the catalyst, respectively, which include the octꢀ1ꢀ
ene concentration, which can be regarded as constant
n2 > 2. The n2 value for Мо В is larger (n2 = 3.7) than
2
for MoB ( 2
n = 2.7). The topochemical constants kt are
× ×
–6
–11
Мо В kt1 = 2.96
10 and kt2 = 1.7
10 for Mo B
2
2
–4
–8
and kt1 = 1.03
×
10 and kt2 = 9.5 10 for MoB.
×
unact
under the experimental conditions; [Cat ] and
Note that the main process of catalyst activation
apparently takes place in the second region, characterꢀ
ized by the topochemical constants n2 and kt2. This is
act
[
Cat ] are the unactivated and activated catalyst conꢀ
tents of the reaction medium, respectively.
The correlation between the starting catalyst conꢀ evidenced both by the degree of catalyst activation,
tent ([Cat] ), the value, and the unactivated and which is no larger than 20% in the first region, and by
α
0
KINETICS AND CATALYSIS Vol. 55
No. 2
2014

ON THE ACTIVATION OF Mo B AND MoB CATALYSTS
215
2
ep
4
–1
α
(w /[TBHP] ) × 10 , s
t
t
1
0
0
0
0
.0
2
12
1
.8
.6
.4
.2
1
y
= 11.23x + 0.005
R2 = 0.999
2
6
y
= 9.720
R2 = 0.999
x
+ 0.000
0
200
400
600
800
1000
t, s
0
0.2
0.4
0.6
0.8
1.0
α
Fig. 4. Time variation of the degree of activation of the
) Mo B and ( ) MoB catalysts. [TBHP] = 0.5 mol/L,
(1
2
Fig. 6. Dependence of wep /[TBHP] on the degree of actiꢀ
t
2
0
2
t
[
Oct] = 2.5 mol/L, [Cat] = 0.22 m /L, T = 360 K.
0 0
vation of the (1) Mo B and (2) MoB catalysts. [TBHP] =
2 0
2
0
.5 mol/L, [Oct] = 2.5 mol/L, [Cat] = 0.22 m /L, T =
0 0
360 K.
the conversion of hydroperoxide (5–10%). Apparꢀ
ently, the first region corresponds mainly to nucleꢀ
into account, Eq. (6) can be represented as the followꢀ
ation. In the second region, the
α
value varies between
ing linear equation with respect to :
α
0.2 and 0.95 and the hydroperoxide conversion varies
between 10 and 95%. Apparently, this is the time interꢀ
val in which the formation of a new, more active form
of the catalyst occurs.
ep
wt
unact
eff(ep)
act
eff(ep)
unact
eff(ep) 0
= k
[Cat] + (k
− k
)[Cat] α. (7)
0
[
TBHP
]
t
The reaction order with respect to hydroperoxide
The data shown in Fig. 6 suggest that, on both catꢀ
in the presence of both unactivated and activated catꢀ alysts, the epoxide forms only involving the activated
ep
alyst forms is unity; therefore, the formation rate of
wt
catalyst form, since the straight lines in the
=
the epoxide (if the latter forms involving the unactiꢀ
vated and activated catalyst forms) should be
described by the equation
[
TBHP t
coordinates originate from the origin of coordiꢀ
nates. The effective rate constant of epoxide formation
]
f( )
α
ep
t
unact
eff(ep)
unact
act
eff(ep)
act
act
–4 –1
w = k [TBHP][Cat ] + k [TBHP][Cat ], (6) is
k
= 9.72 10 s in the presence of activated
×
eff(ep)
act
eff(ep)
–4 –1
where kunacteff(ep) and kacteff(ep) are the effective rate conꢀ
stants of epoxide formation involving the unactivated
and activated catalyst forms, respectively.
MoB (curve
2
) and
k
= 11.23
×
10 s in the
presence of activated Mo B (curve 1).
2
A comparison between these values and the effecꢀ
tive rate constants of hydroperoxide consumption
involving the active catalyst form shows that the forꢀ
mation selectivity of epoxide under the reaction conꢀ
ditions during catalyst activation should vary between
With the correlation between the amounts of
unact
the unactivated form ([Cat ]) and activated form
act
(
[Cat ]) and the degree of catalyst activation taken
0
and 95%, which almost coincides with the experiꢀ
mental value (>90%) found for the activated forms of
both catalysts.
ln [–ln(1 –
4
α
)]
1
Thus, the activation of the Mo B and MoB cataꢀ
y
= 2.684
R2 = 0.998
x
– 16.163
2
2
lysts is due to the change in their surfaces under the
action of the reaction medium. This process includes
at least two main steps—nucleation and formation of
a new, more active and selective phase of the cataꢀ
lyst—and follows the regularities of topochemical
reactions, being described by the Avraami–Erofeev
equation. The epoxide formation selectivity of the
unactivated catalyst is close to zero. The activated
0
0
2
4
6
8 lnt [s]
x – 24.776
y
= 3.703
R
2
y
= 1.400
x
– 9.184
–
4
8
= 0.990
R2 = 0.998
y
= 1.722x – 12.730
R2 = 0.996
–
forms of the catalyst based on Mo B and MoB differ
2
–12
slightly in selectivity; however, the activated Mo B is
2
more active, which is evidenced by the effective rate
constants of hydroperoxide consumption and epoxide
formation.
Fig. 5. Graphical solution of the Avraami–Erofeev equaꢀ
tion: ( ) Mo B and ( ) MoB. [TBHP] = 0.5 mol/L,
Oct] = 2.5 mol/L, [Cat] = 0.22 m /L,
1
2
2
0
t
2
[
= 360 K.
No. 2
0
0
KINETICS AND CATALYSIS Vol. 55
2014

2
16
NYKYPANCHUK et al.
10. Surgenor, D.M., J. Am. Chem. Soc., 1946, vol. 68, no. 2,
REFERENCES
p. 205.
1
2
3
4
. US Patent 4110352, 1979.
. US Patent 4203910, 1981.
. US Patent 4101560, 1979.
. Markevich, V.S., Ul’yanova, V.N., and Loginova, V.A.,
1
1. Antonovskii, V.L., Organicheskie perekisnye initsiatory
(Organic Peroxide Initiators), Moscow: Khimiya,
1972.
Neftepererab. Neftekhim., 1979, no. 11, p. 47.
. Trach, Yu., Nykypanchuk, M., and Komarenska, Z.,
Pol. J. Chem., 2002, no. 76, p. 1323.
12. Komarenskaya, Z.M., Vestn. Gos. Univ. L’vov Politekh.,
Khim., Tekhnol. Veshchestv Ikh Primen., 1999, no. 374,
p. 3.
5
6
7
8
9
. Pyrig, I.Yu., Nykypanchuk, M.V., and Chernyak, B.I., 13. Arslanov, V.V., Sheinina, L.S., Bulgakova, R.A., and
Kinet. Katal., 1983, vol. 24, no. 3, p. 600.
. Trach, Yu.B., Nykypanchuk, M.V., and Komarensꢀ
kaya, Z.M., Kinet. Catal., 2004, vol. 45, no. 4, p. 504.
. Trach, Yu.B. and Chernii, M.O., Ukr. Khim. Zh., 2003,
Belomestnykh, A.V., Langmuir, 1995, vol. 11, p. 3953.
1
4. Hardiman, K.M., Cooper, C.G., and Adesina, A.A.,
Chem. Eng. Sci., 2006, vol. 61, no. 8, p. 2565.
1
5. Protashchik, V.A., Novye zakonomernosti v topokhimii
vol. 69, no. 12, p. 112.
(Advanced Topochemistry), Moscow: Znanie, 1974.
. Trach, Yu.B. and Makota, O.I., Theor. Exp. Chem.
002, vol. 38, no. 4, p. 250.
,
2
Translated by K. Utegenov
KINETICS AND CATALYSIS Vol. 55
No. 2
2014