164
DEDOV et al.
silica sol (silica nanoparticles stabilized in water) with it is necessary to take into account many factors,
salt solutions (mixture slurry method (MSM)), and which include the total contact surface accessible to
sol–gel synthesis (SGS, mixing of homogeneous soluꢀ reactants, the fraction of the surface occupied by
tions of silicon, manganese, sodium, and tungsten active sites, the structure and state of these sites, etc.
compounds)—had the same initial phase composiꢀ
tion: Na2WO4, Mn2O3, and cristobalite (with an
admixture of quartz in the case of MSM). However,
the synthesis method had an effect on the results of
In this study, the OCM process catalyzed by Li–
W–Mn–O–SiO2 composite materials prepared by the
different synthesis methods—SPS, SGS, and IM—
using reactants with different chemical nature has
been examined. The OCM results are compared to the
data on the phase composition of Li–W–Mn–O–
SiO2 composites synthesized by the different methods.
The phase composition of the composites after
involvement thereof in the OCM reaction has been
determined. The features of the phase composition of
the Li–W–Mn–O–SiO2 composites that affect the
efficiency of these materials in OCM have been estiꢀ
mated.
OCM (820°C, V
= 30 L g–1 h–1, CH4 : O2 = 4 : 1). The
yields of C2 products were as follows: 20.72 (IM),
19.39 (MSM), and 15.53% (SGS) [6].
The authors of [7] have conducted a comparative
study of the properties of two Mn–Na2WO4/SiO2
composites prepared by a conventional IM and flame
spray pyrolysis (FSP) using a solution of the catalyst
components, respectively. Under longꢀterm tests in
OCM, the catalysts showed stable values of methane
conversion, while the selectivity varied in different
ways: it increased and decreased in the case of the IM
and FSP samples, respectively. The decrease in the
selectivity of the sample was accompanied by a
decrease in the specific surface area and the crystalliꢀ
zation of amorphous silica to cristobalite, as was found
by in situ analysis of XRD patterns.
A composite containing 2 wt % Mn and 5 wt %
Na2WO4–SiO2, prepared by solidꢀphase synthesis
(SPS, annealing of a dry homogenized mixture of solid
salts and silica at 800°C) [8] had the same phase comꢀ
position as its counterpart prepared by MSM disꢀ
cussed in [6]: Na2WO4, Mn2O3, and cristobalite with an
admixture of quartz. During OCM [900–910°C, V =
EXPERIMENTAL
The studied Li–W–Mn–O–SiO2 composite
materials had a gross ratio of cations of nLi
nSi = 2 : 1 : 2.14 : 91.
: nW : nMn :
Sample 1 was prepared by SPS via grinding a mixꢀ
ture of special purity grade amorphous silicon(IV)
oxide (Specifications TU 6ꢀ09ꢀ4901ꢀ80), reagent
grade lithium carbonate Li2CO3 (TU 6ꢀ09ꢀ3728ꢀ83),
analytical grade ammonium tungstate (NH4)10W12O41
⋅
nH2O (TU 6ꢀ09ꢀ3924ꢀ75), and analytical grade manꢀ
ganese(III) oxide Mn2O3 (CAS 1317ꢀ34ꢀ6) and subseꢀ
quently calcining the resulting powder in a muffle furꢀ
nace at 800°С.
53–56 L g–1 h–1, CH4 : O2 = (2.7–3) : 1], the SPS comꢀ
posite showed a C2+ yield of 24%, which decreased to
17% after 40 h. The replacement of Na ions by Li ions
(a Li–W–Mn–O–SiO2 composite prepared by SPS)
significantly increased the stability of the catalytic perꢀ
formance in OCM. At the same molar ratio of cations
of Li(Na) : W : Mn : Si = 2 : 1 : 2.14 : 91, during the
Sample 2 was prepared by SPS using reagent grade
aerosil Aꢀ300 SiO2 (State Standard GOST 14922ꢀ77)
as a source of silicon. Sample 3 was prepared also by
SPS using analytical grade manganese(II) acetate
(CH3COO)2Mn 4H2O (GOST 16538ꢀ79) as a source
⋅
of manganese. Samples 1 and 2 were homogenized in
a ball mill with tungsten carbide balls; sample 3, in a
ball mill with porcelain balls.
100ꢀh OCM (900°C, V =
= 57–60 L g–1 h–1, CH4 : O2
2.1 : 1), the C2+ yield over Li–W–Mn–O–SiO2 preꢀ
served a constant value of about 20% [8]. At the same
time, according to [9–11], for Li–W–Mn–O–SiO2
samples prepared by the IM, the C2+ yield did not
exceed 10–15%. It has been shown [8, 10, 11] that the
Li–W–Mn–O–SiO2 composite differs from composꢀ
ites containing Na, K, Rb, and Cs primarily by the fact
that, in the presence of Li+, silica largely undergoes
crystallization of quartz rather than cristobalite, the
presence of which is commonly considered to be
responsible for the catalytic activity of Na(K, Rb,
Cs)–W–Mn–O–SiO2 composites in OCM; however,
this assumption has been recently disproved [7]. In
addition, note that the polymorphism of Li2WO4 [12]
and the presence of tungstates Li6WO6 [10] and
Li6W2O9 [12, 13] have been found for Li–W–Mn–O–
SiO2.
Samples 4 and 5 were prepared by a method that
can be attributed to SGS because the aerosil used as a
source of silica was dissolved in ammonia water.
Reagent grade aerosil Aꢀ300 (GOST 14922ꢀ77) was
introduced into an aqueous solution of ethanol under
vigorous stirring; after that, preliminarily prepared
solutions of ammonium tungstate in 8% ammonia
water, Li(CO3)2 in special purity grade acetic acid
(GOST 19814ꢀ74), and manganese acetate in distilled
water were sequentially added to the resulting mixture
under stirring. The mixture was alkalified with special
purity grade 25.0% ammonia (CAS no. 1336ꢀ21ꢀ6) to
a pH of 8 to dissolve the aerosil. The resulting solution
was thoroughly mixed, dried at 80
°
С
, and then heated
at 200 for 4 h and calcined at 800°С
°С
for 6 h. The
weight of synthesized samples 4 and 5 was 10 g and
4.5 kg, respectively.
To reveal the causes of changes in the rate of heterꢀ
ogeneously catalyzed reactions, including OCM, durꢀ
Samples 6 and 7 were based on sample 5 prepared
ing an experiment or from one experiment to another, by SGS. Sample 6 was prepared from a powder of samꢀ
PETROLEUM CHEMISTRY Vol. 55
No. 2
2015