Tailoring the physical and catalytic properties of lanthanum oxycarbonate nanoparticles

Article abstract of DOI:10.1016/j.apcata.2017.01.019

The synthesis of lanthanum oxide and its carbonate analogues has been performed by flame spray pyrolysis (FSP). Two different feeds have been studied: an organic solution and an aqueous/organic microemulsion. A key experimental parameter of FSP, the O₂ dispersion, i.e., the flow rate of the dispersing gas in the FSP nozzle, exhibits an effect on the properties of the materials prepared. Increasing the level of O₂ dispersion led to an increase in surface area and a decrease in mean particle size and basicity when a lanthanum containing organic solution was used as FSP feed. Lanthanum can form different phases, such as oxides, hydroxides, oxycarbonates and carbonates. The increase of O₂ dispersion also produced a phase change, going from a mixture of type Ia and type II La₂O₂CO₃ and La₂O₃ to pure La₂O₃. The use of an aqueous/organic microemulsion feed, which had a higher viscosity than the organic feed, resulted in materials with a lower surface area and a higher mean particle size than those prepared using the organic solution at the same O₂ dispersion. In this case a mixture of type II La₂O₂CO₃ and La₂O₃ was obtained. The materials were tested for oxidative coupling of methane (OCM). We were able to demonstrate that the OCM performance of the materials could be modified by changing the synthesis parameters. For example, lower O₂ dispersion produced the highest CH₄ and O₂ conversions. We also demonstrated that on ageing the mean particle size remain stable; however, the phases do not, showing a new phase, La(OH)₃, formation and resulting in an increase in OCM activity. While the OCM performances are modest they do demonstrate the power of this approach for controlled synthesis of lanthanum materials.

Full text of DOI:10.1016/j.apcata.2017.01.019

Accepted Manuscript
Title: Tailoring the physical and catalytic properties of
lanthanum oxycarbonate nanoparticles
Authors: C. Estruch Bosch, M.P. Copley, T. Eralp, E. Bilbe,
J.W. Thybaut, G.B. Marin, P. Collier
PII:
S0926-860X(17)30039-X
DOI:
Reference:
http://dx.doi.org/doi:10.1016/j.apcata.2017.01.019
APCATA 16130
To appear in:
Applied Catalysis A: General
Received date:
Revised date:
Accepted date:
10-10-2016
20-1-2017
26-1-2017
Please cite this article as: C.Estruch Bosch, M.P.Copley, T.Eralp, E.Bilbe,
J.W.Thybaut, G.B.Marin, P.Collier, Tailoring the physical and catalytic properties
of lanthanum oxycarbonate nanoparticles, Applied Catalysis A, General
http://dx.doi.org/10.1016/j.apcata.2017.01.019
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

Tailoring the physical and catalytic properties of lanthanum
oxycarbonate nanoparticles
C. Estruch Bosch*,1, M. P. Copley1, T. Eralp1, E. Bilbe1, J.W. Thybaut*,2 , G.B.
Marin2, P. Collier1
1 Johnson Matthey Technology Centre, Blount’s Court Road, Sonning Common, Reading, UK RG
2 Laboratory for Chemical Technology, Ghent University, Tech Lane Ghent Science Park, Campus 914, B-9052 Ghent,
Belgium
(*) corresponding authors: boschce@matthey.com, Joris.Thybaut@UGent.be
Graphical abstract
Highlights
Flame spray pyrolysis allows tuning of the properties of La based nanoparticles which may
lead to improved catalysts for oxidative coupling of methane (OCM).
Changes in the synthesis parameters not only affected the particle size but also basicity and
phase.
Higher C2 yields were obtained with materials of higher basicity.
A mixture of La2O2CO3 and La2O3 exhibited better OCM performance than La2O3 only.

Abstract
The synthesis of lanthanum oxide and its carbonate analogues has been performed by flame
spray pyrolysis (FSP). Two different feeds have been studied: an organic solution and an
aqueous/organic microemulsion. A key experimental parameter of FSP, the O2 dispersion,
i.e., the flow rate of the dispersing gas in the FSP nozzle, exhibits an effect on the properties
of the materials prepared. Increasing the level of O2 dispersion led to an increase in surface
area and a decrease in mean particle size and basicity when a lanthanum containing organic
solution was used as FSP feed. Lanthanum can form different phases, such as oxides,
hydroxides, oxycarbonates and carbonates. The increase of O2 dispersion also produced a
phase change, going from a mixture of type Ia and type II La2O2CO3 and La2O3 to pure
La2O3. The use of an aqueous/organic microemulsion feed, which had a higher viscosity than
the organic feed, resulted in materials with a lower surface area and a higher mean particle
size than those prepared using the organic solution at the same O2 dispersion. In this case a
mixture of type II La2O2CO3 and La2O3 was obtained. The materials were tested for oxidative
coupling of methane (OCM). We were able to demonstrate that the OCM performance of the
materials could be modified by changing the synthesis parameters. For example, lower O2
dispersion produced the highest CH4 and O2 conversions. We also demonstrated that on
ageing the mean particle size remain stable; however, the phases do not, showing a new
phase, La(OH)3, formation and resulting in an increase in OCM activity. While the OCM
performances are modest they do demonstrate the power of this approach for controlled
synthesis of lanthanum materials.
Keywords: lanthanum oxide, lanthanum carbonate, flame spray pyrolysis, basicity, nanoparticles,
oxidative coupling of methane
1 Introduction
Lanthanides show unique electric, optical, magnetic and catalytic properties due to their
partially filled 4f shell electronic configuration. In particular, lanthanum has been widely
used in different applications and for example, can be found in electrode material for
magnetohydrodynamic (MHD) power generation1, for solid oxide fuel cells interconnects2,
special glasses for optical devices3,4, and as catalysts, either as a promoters5, stabilizers6 or
the active phase7. Depending on the application of interest, different solid state compounds of

lanthanum such as oxide, hydroxide, oxycarbonates and carbonate, are used. For example,
La(OH)3 is used as an intermediate to form the oxides or sulfides; La2O3 is added to prepare
special optical glasses, as it increases density, refractive index and hardness8; La2(CO3)3 is
used in medicine as a phosphate binder9-11; and, lanthanum oxycarbonates have been shown
to be active for the oxidative coupling of methane (OCM)12,13
.
Different techniques have been used to synthesise solid state lanthanum compounds such as
the thermal decomposition of La salts14 and precipitation15. Interest has also been shown in
synthesising lanthanum based materials as nanoparticles16-18. Particle size and morphology
are two parameters of high importance that can potentially influence material properties. In
particular, it is well known that nanoparticles can have different properties thanks to their
particle size. An example of such different behaviour was shown by Balusami et al.19 who
demonstrated that lanthanum oxide nanoparticles exhibited antimicrobial activity, which was
not observed for the bulk material. A wide range of methods have been used to produce
lanthanum based nanoparticles such as, hydrothermal16,20, sol-gel17 and sonification21.
However, the scale up of these materials through these techniques may prove to be difficult
due to elevated reactant prices and efficiency.
Flame spray pyrolysis (FSP) has been used for the single-step production of powders and the
technique allows the production of materials with controlled properties at a high rate22-24
.
This can be achieved by modifying the process parameters and the feed composition.
Therefore, FSP could be a good alternative for the preparation of nanoparticles of La2O3 and
its carbonate analogues, because it may allow the tuning of the phases and their scale-up.
To our knowledge La2O3 and its carbonates have not been prepared by FSP before and in this
work the synthesis of lanthanum based materials by this technique has been performed. We
show that by changing the synthesis parameters, such as O2 dispersion and feed composition,
the particle size and phase composition of lanthanum materials can be tuned, obtaining pure
La2O3 and mixtures of this with type II and type Ia La2O2CO3.
Oxidative coupling of methane to ethylene has attracted significant interest since the early
80s when Keller and Bhasin25 first investigated catalysts for this reaction. In 1985, Lunsford2
showed that Li/MgO could give a 19% yield for C2, with ethylene as the major product,
starting from methane. However, further studies showed catalyst deactivation during time in

stream due to Li loss26-28. Since then many catalyst combinations have been studied for OCM
with the highest obtained C2 yields of around 25-30%29. The most common catalyst
formulations have been oxides based on alkaline earth metals doped with alkali metals and
rare earth metals doped with alkali or alkaline earth metals3. Several studies have shown that
the activity of the OCM catalyst is affected by catalyst structural properties30,31. Also basicity
and oxygen ion-conductivity, which have been identified as key parameters for this reaction,
are influenced by catalyst structural properties32,33. The use of lanthanum based materials in
OCM has been studied extensively34-39. The preparation method has shown to have an effect
on the properties and, therefore, the activity of the materials. Choudhary et al.35 found that the
catalyst precursor and calcination conditions used to prepare La2O3 influenced the surface
properties, basicity and base strength distribution, and activity and selectivity in the oxidative
coupling of methane. A comparison of the reactivity of phases was performed by Taylor et
al.13. They showed that the starting phase had an effect on the activity and selectivity despite
La2O3 being the phase after reaction as the carbonates are not stable under OCM reaction
conditions40. Therefore, OCM has been chosen as a comparative test reaction to test
structure-activity relationships for the lanthanum based FSP materials.
There has been some preliminary work looking at nanoparticles for OCM and Farsi et al.
reported in a preliminary study that Li/MgO nanoparticles showed higher CH4 conversion
and C2 yield compared with the conventional Li/MgO catalyst41,42. However, this study only
compared one sample of each type of material. Our aim in this work was to test the
lanthanum based materials for OCM and to perform a more detailed systematic study into the
effect of particle size.
2 Experimental
Two different lanthanum solutions were used as feed for the flame spray pyrolysis to prepare
two different series of materials and the conditions used for the experiments are summarised
in Table 1:
 An organic solution consisting of lanthanum 2-ethylhexanoate in toluene/2-
ethylhexanoic acid (99 %, Alfa Aesar) (1:1 in volume) designated FSP-01, -02,-03
and -04.
 A microemulsion consisting of cyclohexane, Igepal CO-520 and an aqueous solution
of lanthanum nitrate designated FSP-M1 and -M2.

Lanthanum 2-ethylhexanoate was synthesised using La(NO3)3·6H2O (99.9 %, Alfa Aesar),2-
ethylhexanoic acid (99 %, Alfa Aesar) and NH4OH (35 % (d=0.88), Fisher). A mixture of 2-
ethylhexanoic acid and ammonium hydroxide was added to ~50 ml of deionised water.
Lanthanum nitrate was dissolved in ~75 ml of deionised water and this solution was added
slowly to the first solution. 500 ml of deionised water were added with stirring to the cottage
cheese like precipitate which was formed. The solid was filtered on a vacuum filter and
washed with ethanol several times. The solid was then left to dry at room temperature
overnight in an evaporating basin in the fume cupboard, resulting in lanthanum 2-
ethylhexanoate that could be used to synthesise the materials by FSP.
Flame Spray Pyrolysis was performed using equipment which has been described
previously23 and a diagram of the FSP rig used to prepare the mateials presented in this work
is shown in Figure S1 in the Supporting information (SI). The precursor solution, either the
organic or the microemulsion solution, was fed into the flame using a syringe pump and
dispersed into fine droplets with a nozzle using O2, the latter being referred to as O2
dispersion. The flame was fed with O2 and CH4 jets positioned in a ring around the nozzle
outlet. Additional O2 is provided to surround the flame as a sheath and the lanthanum
containing solid compounds were collected on a glass fibre filter with the help of a vacuum
pump.
BET surface area analysis was performed using a Quanta Chrome Autosorb-1 apparatus
using nitrogen as the adsorbate. Prior to analysis, samples were outgassed at 150 °C under
vacuum for approximately 24 h. X-ray diffraction data (XRD) were obtained with a Bruker
AXS D8 diffractometer and using Cu Kα radiation and the data collected from 10 to 130° 2
with a step size of 0.02°. Ratios of the identified phases and their crystallite sizes were
determined by Rietveld refinements using TOPAS (Total Pattern Analysis Solution)43.
Structural models for La2O3 and the type II polymorph of La2CO3O2 are well documented44-
46. The type Ia structure was determined by combining documented lattice parameters with
suggestions that it is a monoclinic distortion of the tetragonal Bi2O2CO3 structure47,48
.
Samples for transmission electron microscopy (TEM) analysis were ground between two
glass slides and dusted onto a holey carbon coated Cu TEM grid and a Tecnai F20
transmission electron microscope was used to examine the samples at a 200 kV accelerating

voltage. X-ray Photoelectron Spectroscopy (XPS) of the FSP materials was measured with a
Thermo Escalab 250. The radiation used was monochromatised aluminium Kα radiation with
a 650 µm spot size. Charge compensation was provided by the in-lens electron flood gun at a
2 eV setting and the "401" unit for "zero energy" argon ions.
Temperature programmed decomposition of the as prepared samples was carried out with a
Micromeritics AutoChem II 2920 chemisorption analyser. The samples were heated up to
900 °C(10°C/min) under a helium flow in a “U” shaped quartz tube. The effluent was
analysed with a mass spectrometer, MKS Cirrus 2.
Oxidative coupling of methane (OCM) testing was performed with a high throughput reactor
comprising eight quartz reactors (internal diameter = 4 mm, outside diameter = 8 mm). The
reaction mixture consisted of CH4, O2 and N2. The conditions used were: W/FCH4 = 2
kgs/mol, CH4/O2 ratio = 2:1, 10 % N2 (internal standard) and a temperature programme of:
650, 750, 650, 850 and 650 °C. The ramp to the different temperatures was performed under
nitrogen and the results presented in the paper were started to be taken after 2.5 hours. The
temperature was controlled with a thermocouple located in one of the reactors containing
quartz wool, which was used as a blank reactor to account for the gas phase reactions.
Reference to the blank reactor results are given where appropriate in the results section. A
Varian CP4900 Micro-GC was used to analyse nitrogen, methane and upup to C9 products.
However, the discussion of this paper has been based on the C2 products as only traces of C3
were observed, in particular at high O2 conversions. Carbon balance was normally around 90
%, however under extreme conditions (CH4 converions <15 % and complete O2 conversion)
the carbon balance was lower. Drawing conclusions at these extreme conditions was avoided.
3 Results and discussion
3.1 Nanoparticle synthesis
It is well known that La2O3 tends to react at room temperature with water vapour, forming
hydroxides which can then react with the atmospheric CO2 to form carbonates49. This could
affect the shelf stability of these materials and for this reason the stability when stored was
also studied. The characterisation of the fresh samples and samples aged for 24 months stored
in vials is presented.

3.1.1 Particle size and surface area
TEM images of both the organic and microemulsion samples showed sphere like materials,
see Figure 1, with different mean particle size distributions for all the materials. The O2
dispersion had an influence on the mean particle size for the organic solution derived
materials: the higher the O2 dispersion the smaller the droplet size and ,hence, the smaller the
mean particle size23. Using the microemulsion instead of the organic solution with the same
O2 dispersion, 8 l/min, resulted in larger mean particle size, FSP-M1 (40.4 nm) and FSP-M2
(35.6 nm). This difference in size is probably due to the different viscosity of the feed, since
the microemulsion was more viscous than the organic solution. An increase in viscosity
would be expected to slow down the atomization process by increasing the surface stability of
the droplet. The liquid is then capable of withstanding the shear force from the dispersion
gas50 producing bigger droplets and therefore materials with higher mean particle size. The
BET surface area of all materials correlated well with the TEM results, where an increase in
mean particle size showed a decrease in the surface area, see also Figure 2.
TEM analysis performed on the aged samples showed similar mean particle sizes compared
to sizes obtained for the fresh ones; see Table 2, indicating that standing for 24 months has
not resulted in significant particle growth.
3.1.2 Phase
3.1.2.1 Fresh samples
XRD analysis of the organic and microemulsion samples revealed the presence of a number
of different phases such as hexagonal La2O3, and type Ia and type II La2O2CO3, Table 3. The
XRD patterns are presented in Figure S2-7 in the Supporting information. Type Ia and type II
La2O2CO3 are stable intermediates observed in the thermal decomposition of La2(CO3)3 to
form the thermodynamically most stable La2O3 phase. The formation of the oxycarbonate
phases has been reported previously between 450-550 °C21,51, however, higher temperatures
between 700-900 °C are needed to form La2O351,52. There are a number of factors including
the heat transport rate through the droplet, the residence time in the flame, oxygen availability
or phase thermodynamics that may determine which phases are ultimately produced during
FSP. It is noticeable, however, that the O2 dispersion appears to be the most significant
parameter in the control of the phase produced, see Figure 3. At lower O2 dispersions, such as
with samples FSP-O2 and 1, the formation of type Ia La2O2CO3 phase is observed even

including some type II La2O2CO3 at the lowest O2 dispersion; in contrast at higher
dispersions only hexagonal La2O3 is formed. Furthermore, the amount of hexagonal La2O3
phase increases with O2 dispersion. The suggested explanation for this is that increasing O2
dispersion leads to smaller droplet sizes of the liquid as it is injected into the flame which will
lead to higher temperatures during the lanthanum 2-ethylhexanoate oxidation/decomposition
process and a more oxidising environment for oxide formation. This means that as the droplet
size decreases the extent of intermediate phase formation decreases and the formation of the
thermodynamically most stable phase becomes dominant.
XRD shows that the materials prepared using the microemulsion feed comprise type II
La2O2CO3 and hexagonal La2O3 with different ratios. Sample FSP-M1 contained a slightly
higher amount of type II La2O2CO3, 63.6%, than sample FSP-M2, 42.2 %. In addition to the
significant effect of the O2 dispersion seen for the materials prepared with the organic
solution it is clear for the FSP-M that the precursor solution has an effect as well. For
example, compare the FSP-M samples and FSP-O3 which were all prepared using the same
O2 dispersion and yet show a different phase composition. This suggests that the presence of
oxide carbonates in the FSP-M samples is due to an additional factor such as the precursor
composition. The microemulsion samples contain water which will lower the flame
temperature and result in a different thermodynamic equilibrium composition comprising a
mixture of intermediate phases and La2O3. Additionally the presence of water will also allow
more liquid phase interactions to take place during the nucleation stage of particle formation
immediately prior to evaporation and combustion.
3.1.2.2 Aged samples
XRD analysis was repeated on all samples 24 months after being prepared, see Table 4. The
XRD patterns are presented in Figure S2-7 in the Supporting information . These showed
that, for the samples prepared using the organic solution, the La2O3 had reacted with
atmospheric water to form the La(OH)3. However, the samples prepared with the
microemulsion feed, FSP-M1 and FSP-M2, showed no or, at most, a small amount of
La(OH)3. An increase of amorphous phase was observed for the samples initially containing
La2O3 and type Ia La2O2CO3, i.e., FSP-O2, FSP-O3 and FSP-O4. No amorphous phase was
observed for the samples containing type II La2O2CO3.

XPS performed on the aged samples shows the presence of two O1s signals for all of the
samples. The O1s signal at 528 eV corresponds to O2- species from lanthanum oxide and the
one at 531 eV could be assigned to hydroxyl groups or to carbonates39,53. Because of the
presence of a C1s signal at 289 eV, which corresponds to carbonates54, the signal at higher
binding energy from the O1s could be assigned to the carbonate oxygen. Adventitious carbon
can be found at 284.6 eV54. The XPS results suggest the presence of carbonates on the
surface of all the samples. These results differ from the ones obtained with the XRD, where
aged sample FSP-O3 and FSP-O4 showed the presence of only La(OH)3. XRD shows the
composition of the bulk crystalline material with a particle size bigger than 2 nm while XPS
shows the composition of the s-form and is able to also detect amorphous material in addition
to crystalline phases which suggests that the particles have a surface layer of amorphous or
semi crystalline (particles less than 2nm) lanthanum carbonate55. The La3d signal confirms
the presence of La+3. This signal includes the main peak (La 3d cf0), the satellites (La 3d cf1),
the plasmons and the auger peaks (Figure 4). Previous work by Sunding et al.53 reported on
the relative intensities between the different peaks for La2O3 and La(OH)3 which can be used
to aid the interpretation of the phases. For simplification we have looked into the relative
position of the La 3d cf0 and the La 3d cf1 L bonding (BE) for all the samples and also for
bulk La2O3, La2(CO3)3 and La(OH)3. The reported BE value for La2O3 is 4.9 and for
La(OH)3 is 3.9 eV which are similar to the ones obtained for the pure materials measured in
our equipment, see Table 5. The value for the La2(CO3)3 reference material is similar to that
for the hydroxide, making it difficult to differentiate these. When comparing the relative peak
positions for the bulk materials and the FSP materials a clear trend can be observed. The
samples prepared using lower O2 dispersion, FSP-O1 and FSP-O2, have similar relative peak
position to the La2(CO3)3 and La(OH)3 reference materials. Instead, for the ones with higher
O2 dispersion, FSP-O3 and FSP-O4, their values started to be similar to the relative peak
position of the La2O3, being FSP-O4 the closest. The microemulsion samples, FSP-M1 and
FSP-M2, appear to be close to the oxide. These results suggest that on FSP-O4, FSP-M1 and
FSP-M2 the oxide is stabilised on the surface.
3.1.3 Basicity
The results obtained from the decomposition of the aged materials under helium are given in
Figure 5 and show that two types of carbon(ate) are present, surface (<500 °C) and structural

(>500 °C). The amounts observed can be used as an indicator of the strength of the basic sites
present on the material. The decomposition temperature of a carbonate has been used
previously to assess the basicity of its oxide56. More basic oxides will tend to release CO2
only at higher temperatures due to a stronger interaction. The strength of the basic sites can
be classified as a function of the temperature at which CO2 release is observed: 50-250 °C
(weakly basic sites), 250-600 °C (intermediate basic sites) and 600-950 °C (strongly basic
sites). Differences can be observed between the samples in the sites of weak and intermediate
basic strength, where an increase of the O2 dispersion produces samples with higher amounts
of these sites, see Table 6. Interestingly, FSP-O1, which is the sample with the lowest O2
dispersion (2 l/min), only exhibits one peak (254 °C) at the low temperatures, instead of
three. All four samples exhibit a single, pronounced peak at high temperatures which
corresponds to the strongly basic sites. Only FSP-O1 and FSP-O2 showed the presence of
carbonates by XRD. Nevertheless, the presence of carbonates in all the samples is also
observed by XPS. The decomposition temperature of the strong basic sites57 is much higher
on the sample prepared with the lowest O2 dispersion. This sample also has the largest
particles with a mean diameter amounting to 40 nm, suggesting that a smaller particle size
leads to a weaker basicity. This is in agreement with the results reported by Montero et al.58
on MgO, who showed that the basicity of MgO decreased with decreasing particle size.
3.1.4 Summary on the materials stability
The samples prepared using high O2 dispersion, i.e., FSP-O3 and FSP-O4, which presented
the smallest mean particle size and are only composed of a single phase, were not stable over
24 months and transformed from the La2O3 into the hydroxide. The same behaviour was
observed for sample FSP-O2, which only contained a small amount of type Ia La2O2CO3.
However, the La2O3 present in FSP-O1, FSP-M1 and FSP-M2, samples that contained mainly
carbonates, were more stable and even XPS showed it to be stable on the surface. However,
changes of the phase were observed and this did not affect the mean particle size of the
materials.

3.2 OCM testing
3.2.1 Fresh samples
3.2.1.1 Stability of nanomaterials to temperature
Lanthanum based nanomaterials were tested for OCM at different temperatures to address,
among others properties, material thermal stability, see Figure 6. The materials prepared from
the organic solution showed low activity at 650 °C. The effect on the materials to the
exposure of high temperatures can be observed in Figure 7. The thermal history for the
samples prepared with the organic solution is significant even after reaching 750 °C.
However, smaller differences are observed for the samples prepared using a microemulsion.
Surface area was not looked at in the spent catalysts because a limited amount of catalyst was
available. However, when looking at the spent catalyst by TEM, aggregation of the particles
can be observed producing flake like materials, see Figure 8, which were not there on the
catalyst as prepared. The FSP-ME samples also suffer aggregation but it does not have such
an impact to the activity. This suggests that for these materials the aggregation and therefore
change in morphology, between spheres and flakes, does not play a dominant role in defining
OCM activity. As expected, an increase in OCM activity is observed for these materials when
the temperature is increased. However, at 850 °C the C2 yield from the blank reactor, i.e., 7.9
%, exceeds that from the FSP-O samples except for the one prepared with 2 l/min of O2
dispersion.
The materials prepared using an aqueous/organic microemulsion mixture as a feed were
tested under the same conditions, see Figure 6, and for these materials a decrease in the C2
yield is observed with increasing of temperature which is not the expected trend. The reason
for this phenomenon could be that the system has already reached its maximum performance
even at lower temperatures with 100 % O2 conversion. At 650 °C sample FSP-M1 shows a
significant C2 yield, i.e., 11.5 %. However, when the system is cooled down again to this
temperature after reaching 750 and 850 °C a decrease in the C2 yield is observed. These
changes on the activity depending on the thermal history can be observed as well with the
materials prepared using the organic solution but is not observed with the sample FSP-M2.

3.2.1.2 Properties- OCM activity relationship
With increasing O2 dispersion used in catalyst synthesis, i.e., from FSP-O1 to FSP-O4, a
decrease in the C2 yield and CH4 conversion is observed. The O2 conversion also decreases
with increasing O2 dispersion which suggests that the capacity of the catalyst to activate the
O2 is decreasing. At the lowest O2 dispersion the ethylene production exceeds that of ethane
and, additionally, some propylene is formed. The behaviour observed for this catalyst
prepared at the lowest O2 dispersion, FSP-O1, is clearly starting to differentiate from the
other FSP-O samples and getting closer to the performance of the FSP-M samples.
It is well known that an increase in surface area can have a positive effect on the catalyst
activity as more active sites are available. In particular for OCM, previous work has shown
that when increasing the surface area an increase of the activity is observed. Thybaut et al.59
reported simulation data for the catalyst Sn/Li/MgO in which an increased surface area
resulted in an increased CH4 conversion. There will be an optimum surface area to achieve
the highest C2 yield and this is because with increasing surface area the catalyst will also
accelerate the further reaction of C2 products decreasing the yield to C2. Figure 9 represents
the CH4 conversion versus surface area of all the prepared catalysts tested at 750 °C. The
activity of the empty reactor at 750 °C temperature is included for comparison. A decrease in
the CH4 conversion is observed with an increase of surface area which is the opposite trend to
the one expected from literature. Therefore, the observed trend should be due to other factors,
such as, particle size, phase or basicity and these other factors are discussed below.
The particle size has an effect on the OCM activity. The materials with a smaller mean
particle size (FSP-O3 and FSP-O4) have lower activity, 16.8 %, and the ones with the larger
mean particle size (FSP-M1 and FSP-M2) show the higher activity, 44 % (Figure 6).
However, as mentioned before, Farsi et al.41,42 reported that Li/MgO nanoparticles were more
active than a bulk catalyst. Also given the trend found at the microscale by Thybaut et al.59 it
should be expected to reach higher activity at lower particle size. In our case, this is the
opposite because the change in particle size also is accompanied by other structural changes.
The samples containing type II La2O2CO3, FSP-M1 and FSP-M2, demostrate the higher
OCM activity, followed by FSP-O1, which contains traces of type II La2O2CO3 and its major
phase is the type Ia La2O2CO3. FSP-O2 shows also similar OCM activity and this contains
lower amounts of La2O2CO3. The materials showing the lowest OCM activity, FSP-O3 and

FSP-O4, only contain the hexagonal La2O3. Taylor et al.13 reported that OCM activity was
dependant on the phase as follows, even if after reaction La2O3 was the only phase present:
type II La2O2CO3 > type Ia La2O2CO3 > La(OH)3 > La2O3. This suggests that the reason for
the increase in activity is due to the transformation in phase from hexagonal La2O3 to type Ia
La2O2CO3 and then to type II La2O2CO3.
Another important property that materials active for OCM have in common is their basicity
which is known to enhance the OCM activity because previously reported mechanisms
suggest that methane activation occurs on a basic O-2 via hydrogen abstraction60. Montero et
al.32 reported that the basicity of MgO increased with increasing particle size. The same effect
may govern the observations on the materials described in this paper and the results presented
from the decomposition of the carbonates confirm that the materials with lower mean particle
size release CO2 at lower temperatures, which is related to lower basicity. Therefore, a
decrease in activity could be related to a decrease in basicity of the materials obtained which
could also be related to the phase present.
3.2.2 Aged samples
3.2.2.1 Properties- OCM activity relationship
The aged samples showed similar sensitivity to the thermal history as the fresh samples
(Figure 7) and higher C2 yield at lower temperatures compared to the fresh samples. Due to
their high activity, 100 % O2 conversion was observed (Figure 10), making comparison
between the aged materials not possible. In fact, the CH4 conversion at 650 °C is comparable
to that obtained at 850 °C with the fresh samples. As already discussed, this difference in
activity cannot be due to changes in morphology or particle size because TEM shows that
when the samples are aged they remain spherical with a similar mean particle size. The
explanation for the increase of activity can be found, as for the fresh samples, in the phase.
XRD and XPS suggest that the La2O3 from the fresh samples has reacted with the
atmospheric water and CO2 to form the hydroxide/carbonate. This happened even if the
samples were stored in an air tight bottle and had minimum contact with the atmosphere. This
new phase now present in the aged samples is more active than the oxide as also suggested by
Tylor13 in previous studies. Therefore, the preparation of materials containing carbonates by
FSP should be further studied. The aim would be to obtain flame conditions in which
carbonates are formed, so high temperatures in the flame should be avoided.

Conclusions
FSP has produced a series of materials that show a number of changes with increasing O2
dispersion: surface area, phase and basicity. The catalytic properties of the final materials are
dependent on a combination of all three parameters with surface area being less influential
than phase and basicity. Two different feeds have been used to prepare lanthanum oxide and
its carbonates analogues. For these materials a phase change was observed by XRD, from a
mixture of type II and type Ia La2O2CO3 and La2O3 to pure La2O3, which could be due to an
increase in the O2 availability. The use of a more viscous feed, the microemulsion, resulted in
the production of materials with larger mean particle sizes and, therefore, smaller surface
areas. The XRD for these materials was shown to be a mixture of type II La2O2CO3 and
La2O3.
Despite the lower surface area the materials containing a mixture of type II and type Ia
La2O2CO3 and La2O3 exhibited C2 yields of around 12 % at 750 °C, instead the materials
containing only La2O3 presented much lower C2 yields, 4 %. The materials showing the
higher C2 yields were also the materials that presented higher CO2 decomposition
temperatures which implies higher basicity. The same trend was observed for the C2=/C2 ratio
which was higher for the materials with higher amounts of carbonates.
This work has shown that FSP can be used to tune the properties of La2O3 based
nanoparticles and samples anywhere within the three dimensional experimental space defined
by surface area, phase and basicity and this flexibility may lead to improved catalysts for
OCM.
Acknowledgements
The work was undertaken within the context of the project “OCMOL, Oxidative Coupling of
Methane followed by Oligomerization to Liquids”. OCMOL is a Large Scale Collaborative
Project supported by the European Commission in the 7th Framework Programme (GA
n°228953). For further information about OCMOL see: http://www.ocmol.eu.
CEB would like to thank the analytical department at Johnson Matthey for their assistance
during the project.

References
(1)
(2)
(3)
Meadowcroft, D. B. J. Phys. D: Appl. Phys. 1969, 2, 1225.
Furtado, J. G. M.; Oliveira, R. N. Matéria (Rio de Janeiro) 2008, 13, 147.
Song, Q. W.; Wang, X.-M.; Bussjager, R.; Osman, J. Appl. Opt. 1996, 35,
3155.
(4)
Jing, F.; Fu, P.; Wu, Y.; Zu, Y.; Wang, X. Opt. Mater. (Amsterdam, Neth.)
2008, 30, 1867.
(5) Aneggi, E.; Leitenburg, C.; Dolcetti, G.; Trovarelli, A. Top. Catal. 2007, 42-
43, 319.
(6)
(7)
(8)
Kašpar, J.; Fornasiero, P.; Hickey, N. Catal. Today 2003, 77, 419.
Ma, J.; Fang, M.; Lau, N. T. Appl. Catal., A 1997, 150, 253.
Bahari, A.; Anasari, A.; Rahmani, Z. J. Eng. Technol. Res. (Nairobi, Kenya)
2011, 3 203.
(9)
D'Haese, P. C.; Spasovski, G. B.; Sikole, A.; Hutchison, A.; Freemont, T. J.;
Sulkova, S.; Swanepoel, C.; Pejanovic, S.; Djukanovic, L.; Balducci, A.; Coen, G.; Sulowicz,
W.; Ferreira, A.; Torres, A.; Curic, S.; Popovic, M.; Dimkovic, N.; De Broe, M. E. Kidney Int.
2003, 63, S73.
(10) Finn, W. F.; Joy, M. S. Curr. Med. Res. Opin. 2005, 21, 657.
(11) Joy, M. S.; Finn, W. F. Am. J. Kidney Dis. 2003, 42, 96.
(12) Lacombe, S.; Sanchez, J. G.; Delichere, M. P.; Mozzanega, H.; Tatibouet, J.
M.; Mirodatos, C. Catal. Today 1992, 13, 273.
(13) Taylor, R. P.; Schrader, G. L. Ind. Eng. Chem. Res. 1991, 30, 1016.
(14) Hussein, G. A. M.; Gates, B. C. J. Chem. Soc., Faraday Trans. 1996, 92,
2425.
(15) Rosynek, M. P.; Magnuson, D. T. J. Catal. 1977, 46, 402.
(16) Zhang, Y.; Han, K.; Cheng, T.; Fang, Z. Inorg. Chem. 2007, 46, 4713.
(17) Wang, X.; Wang, M.; Song, H.; Ding, B. Mater. Lett. 2006, 60, 2261.
(18) Djerdj, I.; Garnweitner, G.; Sheng Su, D.; Niederberger, M. J. Solid State
Chem. 2007, 180, 2154.
(19) Balusamy, B.; Kandhasamy, Y. G.; Senthamizhan, A.; Chandrasekaran, G.;
Subramanian, M. S.; Kumaravel, T. S. J. Rare Earths 2012, 30, 1298.
(20) Bo, T.; Jiechao, G.; Linhai, Z. Nanotechnology 2004, 15, 1749.
(21) Salavati-Niasari, M.; Hosseinzadeh, G.; Davar, F. J. Alloys Compd. 2011, 509,
134.
(22) Thiebaut, B. Platinum Metals Review 2011, 55, 149.
(23) Camenzind, A.; Caseri, W. R.; Pratsinis, S. E. Nano Today 2010, 5, 48.
(24) Wegner, K.; Schimmoeller, B.; Thiebaut, B.; Fernandez, C.; Rao, T. N. KONA
Powder Part. J. 2011, 29, 251.
(25) Keller, G. E.; Bhasin, M. M. J. Catal. 1982, 73, 9.
(26) Korf, S. J.; Roos, J. A.; De Bruijn, N. A.; Van Ommen, J. G.; Ross, J. R. H.
Applied Catalysis 1990, 58, 131.
(27) Arndt, S.; Laugel, G.; Levchenko, S.; Horn, R.; Baerns, M.; Scheffler, M.;
Schlögl, R.; Schomäcker, R. Catalysis Reviews 2011, 53, 424.
(28) Arndt, S.; Simon, U.; Heitz, S.; Berthold, A.; Beck, B.; Görke, O.; Epping, J.-
D.; Otremba, T.; Aksu, Y.; Irran, E.; Laugel, G.; Driess, M.; Schubert, H.; Schomäcker, R.
Top Catal 2011, 54, 1266.
(29) Machida, K.-i.; Enyo, M. J. Chem. Soc., Chem. Commun. 1987, 1639.
(30) Hargreaves, J. S. J.; Hutchings, G. J.; Joyner, R. W.; Kiely, C. J. J. Catal.
1992, 135, 576.
(31) LeVan, T.; Che, M.; Tatibouët, J.-M. Catal. Lett. 1992, 14, 321.
(32) Montero, J. M.; Gai, P.; Wilson, K.; Lee, A. F. Green Chem. 2009, 11.

(33) Durá, O. J.; López de la Torre, M. A.; Vázquez, L.; Chaboy, J.; Boada, R.;
Rivera-Calzada, A.; Santamaria, J.; Leon, C. Phys. Rev. B 2010, 81, 184301.
(34) Borchert, H.; Baerns, M. Journal of Catalysis 1997, 168, 315.
(35) Choudhary, V. R.; Rane, V. H. Journal of the Chemical Society, Faraday
Transactions 1994, 90, 3357.
(36) Ghose, R.; Hwang, H. T.; Varma, A. Applied Catalysis A: General 2013, 452,
147.
(37) Ilett, D. J.; Islam, M. S. Journal of the Chemical Society, Faraday
Transactions 1993, 89, 3833.
(38) Otsuka, K.; Jinno, K.; Morikawa, A. Journal of Catalysis 1986, 100, 353.
(39) Lacombe, S.; Geantet, C.; Mirodatos, C. J. Catal. 1995, 151, 439.
(40) Levan, T.; Che, M.; Tatibouet, J. M.; Kermarec, M. Journal of Catalysis 1993,
142, 18.
(41) Farsi, A.; Mansouri, S. S. Arabian J. Chem.
(42) Farsi, A.; Moradi, A.; Ghader, S.; Shadravan, V. Chin. J. Chem. Phys. 2011,
24, 70.
(43) Coelho, A. A.; v3.1 ed.; Bruker AXS GmbH: Karlsruhe, Germany, 2003.
(44) Zachariasen, W. H. Z. Phys. Chem. (Leipzig) 1926, 123.
(45) Olafsen, A.; Larsson, A.-K.; Fjellvåg, H.; Hauback, B. C. J. Solid State Chem.
2001, 158, 14.
(46) Attfield, J. P.; Férey, G. J. Solid State Chem. 1989, 82, 132.
(47) Caro.
(48) Grice, J. D. The Canadian Mineralogist 2002, 40, 693.
(49) Klingenberg, B.; Vannice, M. A. Chem. Mater. 1996, 8, 2755.
(50) Rosebrock, C.; Pokhrel, S.; Madler, L.; Fritsching, U.; Noriler, D.; Meier, H. F.
In ILASS - Europe 2011, 24th European Conference on Liquid Atomitzation and Spray
Systems Estoril, Portugal, 2011.
(51) Turcotte, R. P.; Sawyer, J. O.; Eyring, L. Inorg. Chem. 1969, 8, 238.
(52) Squire, G. D.; Luc, H.; Puxley, D. C. Appl. Catal., A 1994, 108, 261.
(53) Sunding, M. F.; Hadidi, K.; Diplas, S.; Løvvik, O. M.; Norby, T. E.; Gunnæs, A.
E. J. Electron Spectrosc. Relat. Phenom. 2011, 184, 399.
(54) Moulder, J. F.; Stickle, W. F.; Sobol, P. E.; Bomben, K. D. Handbook of X Ray
Photoelectron Spectroscopy: A Reference Book of Standard Spectra for Identification and
Interpretation of Xps Data UL VAC-PHI, Inc (Japan) Physical Electronics (USA), 1995.
(55) West, A. R. Solid State Chemistry and Its Applications; Wiley, 1985.
(56) Maitra, A. M.; Campbell, I.; Tyler, R. J. Appl. Catal., A 1992, 85, 27.
(57) Sato, S.; Takahashi, R.; Kobune, M.; Gotoh, H. Appl. Catal., A 2009, 356, 57.
(58) Montero, J. M.; Gai, P.; Wilson, K.; Lee, A. F. Green Chem. 2009, 11, 265.
(59) Thybaut, J. W.; Sun, J.; Olivier, L.; Van Veen, A. C.; Mirodatos, C.; Marin, G.
B. Catal. Today 2011, 159, 29.
(60) Maitra, A. M. Applied Catalysis A: General 1993, 104, 11.

Figure captions
Figure 1. TEM images for the samples prepared using the organic solution and the microemulsion.
Figure 2. O2 dispersion effect on the BET surface area and mean particle size measured by TEM.
Figure 3. Phase distribution depending on the O2 dispersion.
Figure 4. XPS spectra from the La 3d region for the sample FSP-O2.
Figure 5. CO2 measured during decomposition of the materials prepared using an organic solution.
Figure 6. OCM results for the materials prepared by FSP with different O2 dispersion.
Figure 7. C2 yields for fresh (line) and aged (dotted line) samples at 650 °C. 1) First time at 650 °C; 2)
after reaching 750 °C; and, 3) after reaching 850 °C.
Figure 8. TEM images for the materials prepared by FSP after OCM reaction.
Figure 9. CH4 conversion at 750 °C is shown against surface area for the materials prepared using the
organic solution and the microemulsion.
Figure 10. OCM results for the aged samples prepared at different O2 dispersion.

Fig.1

Fig.2

Fig.3

Fig.4

Fig.5

Fig.6

Fig.7

Fig.8

Fig.9

Fig.10

Table 1. Preparation details for the lanthanum nanomaterials.
Precursor O2
Sample
[La]
Solution
Feed Rate Dispersion
Name
(M)
(ml/min)
(l/min)
FSP-O1,
2,5,8 and
10
Organic Solution 2, 3 and
0.2
5
4
Aqueous/Organic FSP-M1
0.06-
0.12
3-5
5
Microemulsion
and 2
*O2 sheath = 5 l/min; flame O2 = 3.2 l/min and flame CH4 = 1.5 l/min

Table 2. Mean particle size of the fresh and aged samples obtained with TEM.
Fresh sample (nm)
38.44 (±22.75, n=64)
14.79 (±3.39, n=25)
9.94 (±3.07, n=43)
11.95(±2.38, n=25)
33.35 (±11.61, n=71)
21.21 (±7.16, n=38)
Aged sample (nm)
-
FSP-O1
FSP-O2
FSP-O3
FSP-O4
FSP-M1
FSP-M2
14.9 (±6.6, n=27)
11.1(±3.1, n=28)
11.6 (±3.1, n=28)
57.9 (±35, n=18)
28.3 (±5.5, n=22)
*n is the number of particles used to calculate the mean particle size.