Sol–gel preparation of alumina stabilized rare earth areo-and xerogels and their use as oxidation
catalysts
Björn Neumann a ,Thorsten M.Gesing b ,Andrii Rednyk c ,Vladimir Matolin c ,Alexander E.Gash d ,⇑,Marcus Bäumer a ,⇑
a
Institute of Applied and Physical Chemistry &Center for Environmental Research and Sustainable Technology,University of Bremen,Leobener Str.,Bremen D-28359,Germany b
Solid State Chemical Crystallography,Institute for Inorganic Chemistry,University of Bremen,Leobener Str.NW2,D-28359Bremen,Germany c
Charles University,Faculty of Mathematics and Physics,Department of Surface and Plasma Science,V Holešovic
ˇkách 2,18000Prague 8,Czech Republic d
Physical and Life Sciences Directorate,Lawrence Livermore National Laboratory,Livermore,CA 94551,USA
a r t i c l e i n f o Article history:
Received 14November 2013Accepted 5February 2014
Available online 13February 2014Keywords:
Sol–gel chemistry Aerogels Xerogels
Rare earth oxides CO oxidation
a b s t r a c t
A new sol–gel synthesis route for rare earth (Ce and Pr)alumina hybrid aero-and xerogels is presented which is based on the so-called epoxide addition method.The resulting materials are characterized by TEM,XRD and nitrogen adsorption.The results reveal a different crystallization behavior for the praseo-dymia/alumina and the ceria/alumina gel.Whereas the first remains amorphous until 875°C,small ceria domains form already after preparation in the second case which grow with increasing calcination tem-perature.The use of the calcined gels as CO oxidation catalysts was studied in a quartz tube (lab)reactor and in a (slit)microreactor and compared to reference catalysts consisting of the pure rare earth oxides.The Ce/Al hybrid gels exhibit a good catalytic activity and a thermal stability against sintering which was superior to the investigated reference catalyst.In contrast,the Pr/Al hybrid gels show lower CO oxidation activity which,due to the formation of PrAlO 3,decreased with increasing calcination temperature.
Ó2014Elsevier Inc.All rights reserved.
1.Introduction
Rare earth oxides (REOs),also called Lanthanide oxides,play an important role in many today’s technological applications,such as automobiles,wind turbines and computers [1].Due to their special electrical,optical and magnetic properties,they are very promising ionic conductors (e.g.for solid oxide fuel cell applications),laser materials,colorants and contrast agents,just to name a few [2,3].In addition,the chemical properties,including acid-base and redox properties,render REOs also interesting as catalysts or catalyst components [4,5].Reactions for which REO catalysts are employed comprise complete and partial oxidation reactions and oxidative coupling of methane (OCM)[6,7].
In order to have high surface to volume ratios which are usually important for catalysis,a high porosity is needed.In particular,mesopores (2–50nm)play a particularly important role as they ensure fast enough transport by diffusion (as compared to microp-ores <2nm)while keeping the specific surface area (SSA)high (as compared to macropores).Interesting synthesis techniques for preparing materials with a suitable porosity for catalytic applica-tions are sol–gel methods [8,9].In a previous work,we reported the so-called epoxide addition method (EAM)to prepare aero-and xerogels of pure REOs.The approach turned out to be quite universal as basically all REOs can be processed in this way [10].In contrast to other routes,using more expensive alkoxides,simple chloride salts can be used as precursors.(Notably,nitrates cannot be used as detailed in Ref.[10].)The gel formation is initiated by the
addition of propylene epoxide which acts as a proton scaven-ger.During the ring opening reaction one equivalent of protons is consumed per equivalent of epoxide so that a slow and homog-enous increase of the pH is achieved in comparison to the addition of a conventional base [10].A problem with the reported synthesis for the pure REOs was the formation of significant amounts of oxy-chlorides (depending on the lanthanide:up to 30wt%),which could not be avoided because of their high stability.
Especially for catalytic applications it is not always necessary (and sometimes not even desired)to use pure materials.Pure com-pounds might be too expensive and dispersing the catalytically ac-tive component by supporting it onto or embedding it into an inert matrix of a cheap oxide is a much more economical way.Using sol–gel chemistry for preparing catalysts,such an option exists when synthesizing a hybrid material,he inert oxide is the main component and the catalytically active oxide is dispersed in this matrix.Of course,it must be ensured that it is exposed at
/10.1016/j.jcis.2014.02.004
0021-9797/Ó2014Elsevier Inc.All rights reserved.
⇑Corresponding authors.Fax:+14942121863188.
E-mail addresses:v (A.E.Gash),mbaeumer@uni-bremen.de (M.Bäumer).
the surface and not incorporated in a mixed oxide which may have no or inferior catalytic activity for the respective reaction.
For several transition metals mixed oxides have already been prepared in this way.Recently,we reported on the successful preparation of alumina promoted Co and Fe catalysts for Fischer–Tropsch synthesis with the EAM approach[11].Here,however, the catalysts werefinally reduced(before the catalytic application) resulting in a phase separation and the formation of metallic Co or Fe domains.The question arises whether in the case of an oxide catalyst the approach is also successful.Therefore,we have extended the EAM method to the preparation of rare earth/Al hybrid aero-and xerogels taking ceria and praseodymia as exam-ples.In the present paper,we report the synthesis,the character-ization by TEM and XRD and their use as oxidation catalysts, taking CO oxidation as an example(testing the ability for total oxidation).
The results show that the Pr/Al hybrid systems remain non-crystalline up to temperatures where pure alumina and REO systems already crystallize.On the contrary,the Ce/Al system be-haves differently and the formation of small ceria domains with diameters in the range of3–12nm is observed.In line with these findings,the ceria catalyst shows catalytic properties similar to pure ceria yet exhibiting a distinctly superior sintering stability. The Pr/Al system,on the other hand,is less suited as a catalyst sin
ce the catalytic activity and sinter stability are inferior as com-pared to a pure praseodymia reference system.
2.Experimental
2.1.Sol–gel preparation
Rare earth–aluminum oxide aero and xerogels were prepared in analogy to the epoxide addition sol–gel method established by Gash et al.for pure REO aero-and xerogels[12].All reactants were reagent grade or better and used as received.The rare earth/alumi-num oxide precursor mixture consisted of aluminum nitrate nona-hydrate(Alfa)and rare earth(III)nitrate hexahydrate(Chempur, Karlsruhe,Germany).The ratio was chosen such that gels with 80at-%Al and20at-%Ce or Pr,respectively,were obtained. 3mmol metal salt were dissolved in5g of ethanol;the molar ratio of metal salt to gelling agent was0.1.In addition to the nitrates, also chlorides were used as precursors.Here,aluminum chloride hexahydrate(Alfa)and rare earth(III)chloride hexahydrate (Chempur,Karlsruhe,Germany)were employed.As in the case of the nitrates,the metal salts were dissolved in absolute ethanol (VWR international,Germany).Propylene oxide(Sigma Aldrich) was added as gelation agent.All samples were prepared in PE vials. Following the addition of the gelling ag
ent,the gels were aged for at least24h under ambient conditions.The resulting materials were then immersed in a bath of absolute ethanol.These alcogels were either processed to aerogels in a BALTEC supercritical point drier or dried under ambient conditions resulting in so-called xero-gels.In the former case the alcohol in the gel was exchanged for li-quid CO2for3days at about10°C,after which the temperature of the vessel was ramped up to about45°C,not exceeding a pressure of$100bar.The vessel was then depressurized slowly.The result-ing aero-and xerogels were calcined at650°C or1000°C,respec-tively(4h in air).
For comparison of the catalytic properties,pure oxides of Ce and Pr were prepared by thermal decomposition of the rare earth ni-trates at650°C and1000°C(4h in air).
2.2.Characterization
Powder X-ray diffraction(PXRD)data were acquired for the xero-and aerogel samples,using a PANalytical X’Pert MPD Pro l diffrac-tometer in Bragg–Brentano geometry.The setup was equipped with a secondary Nifilter,Cu K a1,2radiation and a X’Celerator mul-ti strip detector.The temperature-dependent X-ray powder diffrac-tion data were collected on the same diffractometer using an Anton Paar HTK1200N heating chamber.The sample was placed using acetone on aflat corundum holder ha
ving small evaporation chan-nels that served for optimum space during thermal expansion of the samples.Diffraction was carried out between25°C and 999°C with a ramping slice of25°C.Each diffraction pattern was recorded from5°to100°2h with a step size of0.0167°and a 180s/step total data collection time.For the data evaluation,Riet-veld refinements were carried out.The fundamental parameter ap-proach,where the fundamental parameters werefitted against a LaB6standard material,was applied for the Rietveld refinement using‘‘Diffrac Plus Topas  4.2’’software(Bruker AXS GmbH, Karlsruhe).For this purpose the atomic coordinates of the struc-tures were extracted from the ICSD data base(Fachinformations-zentrum Karlsruhe,Germany)andfixed during the refinements.
Table1
Specific surface area of the hybrid aero-and xerogels and d10and d95according to the BJH evaluation of the samples calcined at650°C.After calcination at1000°C only the BET surface area was measured.
Notation Temperature[°C]BET specific surface area[m2/g]Pore volume[nm]d10[nm]d95a[nm]
Pr/Al-XG NO36501040.15$3.5$7 Pr/Al-AG NO3650204  1.3$15>70 Pr/Al-XG NO31000<1––
Pr/Al-AG NO3100019––
Ce/Al-XG NO36501600.17$3.5$7 Ce/Al-AG NO3650261  1.1$15>70 Ce/Al-XG NO3100033––
Ce/Al-AG NO31000100––
a Additional macropores larger than100nm for the aerogels are likely so that d
10
and d95may be larger.
72  B.Neumann et al./Journal of Colloid and Interface Science422(2014)71–78
The background of the pattern was fitted with two parameters,one 1/x contribution for the primary beam air scattering and a linear parameter for the Bragg–Brentano linear background contribution.From the reflex broadening the average crystallite size L Vol (IB)and the micro-strain e 0were calculated.Diffraction intensities which
could not be described with the structural model were fitted with additional peaks and assumed to be diffuse scattering from amor-phous sample parts.
Transmission electron microscopy (TEM )including energy disper-sive X-ray spectroscopy (EDX)was performed using an FEI Tecnai F20S-TWIN microscope with an operating voltage of 200kV.TEM images were acquired with a slow-scan CCD camera with an integrated Gatan Image Filter Model 2001.For the preparation of the TEM grids,a small amount of each sample was suspended in acetone and ultrasonicated.Finally,a droplet (25l L)was placed on carbon-coated copper grid.
The specific surface areas (SSA)of all aero-and xerogels (after calcination)were determined by nitrogen adsorption at 77K.The powder samples were pretreated at 200°C for a minimum of 2h.The SSA was determined at a relative pressure of 0.3(single point),using a b eta Scientific Corp.instrument
(model 4200).Pore size distribution and full N 2adsorption isotherms were measured using a Quantachrome NOVA instrument.2.3.Catalytic tests
The catalytic tests were carried out in two different setups.The xerogel catalysts and the reference catalysts prepared by calcina-tion of the nitrates were studied in a continuous flow fixed bed lab-oratory reactor.The reactor consisted of a quartz tube which was connected to a controlled gas supply (MFC Bronkhorst Maettig)via PEEK and stainless steel tubing.Gas analysis was carried out with two photometric detectors for CO 2(Hartmann &Braun URAS 3G)and CO (Hartmann &Braun URAS 10E).For the
experiments,
X-ray diffraction patterns of the Ce/Al hybrid xerogels (left)and aerogels (right)calcined at 650°C (upper)and 1000°C (lower)and the results of the respective refinement.CeO 2was identified in all cases.The crystallite size (L Vol (IB))for ceria is <3nm for the samples calcined at 650°C and >12nm after calcination at for the aerogel calcined at 650°C c -Al 2O 3was identified as alumina phase.For all samples a diffuse scattering contribution is observed which is likely to be phase.
950° C
900° C 25° C
925° C 875° C Diffraction angle 2Theta (CuKa1,2) /°
10
20
30
40
50
60
70
80
90
100
High temperature X-ray diffraction patterns of the Pr/Al hybrid xerogel at indicated temperatures.PrAlO 3was identified as the dominating phase at higher temperatures.The reflections from a -Al 2O 3are artifacts from the stallization the volume of the sample changes drastically so that small amounts of the a -Al 2O 3sample holder became visible.
40mg of the different xerogels were ground to powder(200–300l m grains),mixed with180mg of SiC(same grain size)and placed on quartz wool in a tube.For the reference catalyst the same amount of rare earth component(calculated as LnO2)were mixed with SiC.6%carbon monoxide(Linde4.7),20%
oxygen(Linde5.0) were used as reaction gases,helium(Linde5.0)as gas balance. The experiments were performed with a constant gasflow of 50sccm(standard cubic centimeter).
Due to electrostatic charging,the aerogels could not be investi-gated in the same way.In particular,a homogenous distribution within the SiC was not possible.To investigate their catalytic prop-erties,aflow chip microreactor was used.The setup permits reli-able measurements of small amounts of catalyst while ensuring good control of temperature even without dilution of the catalyst. Two milligrams of the respective catalyst were dispersed on the heated10mmÂ10mm SiO2/Si sample holder.The sample was covered by aflat piece of quartz resulting in a slit of100l m through which the gases were directed.CO oxidation was mea-sured by using the same ratio of reactants as in the tube reactor. The totalflow was here5sccm.CO2production was detected by
a differentially pumped quadrupole mass spectrometer.
3.Results and Discussion
3.1.Synthesis and characterization
We based the synthesis of the Ln/Al hybrid aero-and xerogels on the EAM approach recently reported
for pure REOs[10].While in the latter case,only chlorides turned out to be usable,we tried chlorides and nitrates in the present case.Regardless which pre-cursor was used,gelation occurred in less than10min,revealing that for the hybrid gels also nitrates can be used.The addition of water slowed down the gelation process as it was also observed for pure alumina gels[13].
Although both types of metal salts can be used,characteristic differences can be noticed if the preparation is carried out under ambient conditions.While the alcogel obtained from chlorides underwent significant shrinkage(>30%)during the ripening of the alcogels,the samples from nitrates did not shrink to a notice-able extend.Yet,if the samples were cooled during the gel forma-tion,a similar behavior was observed in both cases.Then,also the nitrate derived samples underwent significant shrinkage in anal-ogy to the chloride derived samples.Moreover,their mechanical stability was improved.A possible explanation for this behavior is an insufficient removal of the reaction heat in case of the nitrate approach.This leads to faster ripening which makes the gel less stable.Thus it is obvious that,by cooling the sample during the gelation,the stability of the nitrate derived samples can be in-creased.After ambient drying the obtained xerogels showed the color of the respective rare earth metal salt,which was green in case of praseodymia and yellow in case of cerium.The supercriti-cally dried aerogels,on the other hand,only exhibited a very light color of the respective rare earth elements.
Subsequently,the samples were calcined at650°C to remove residues from the synthesis and to prepare them for high-temper-ature applications(in the present case for
high-temperature electron micrographs for Pr/Al and Ce/Al hybdrid aerogels(AG)and xerogels(XG).Both aerogels show separated ligaments,
catalysis).After calcination,the aerogels remained whereas the xerogels broke into several pieces.The hy-from praseodymium chlorides showed a signif-
chlorine residues after calcination which was
to2at-%)and the silver chloride test.(For the
latter the gel was dissolved in nitric acid and AgNO
On the contrary,for the Ce/Al hybrid aero-and xerogels from chlorides no chloride residues could be detected.
of residual chlorides within the gels was already observed rare earth oxide aerogels obtained from the chlorides 0102030405060708090100 0
3
6
9
c
o
u
n
t
s
praseodymium content [%]
as prepared
102030405060708090100
reactor pressure vesselcontent of praseodymium [%]
annealed sample
an annealed hybrid aerogel powder sample(upper left)and an as prepared sample(upper right).The scale bars are different.These homogeneity of the material via EDX.The histograms in the lower part show the content of praseodymium within different spots of the are not shown.
Conversion of CO as a function of temperature in the tube reactor for xerogels and reference samples pretreated at different calcination temperatures.The activity xerogels decreases much less in comparison to the pure ceria catalyst when the calcination temperature is increased.For praseodymia samples such a
was not observed.
B.Neumann et al./Journal of Colloid and Interface Science422(2014)71–7875

版权声明：本站内容均来自互联网，仅供演示用，请勿用于商业和其他非法用途。如果侵犯了您的权益请与我们联系QQ:729038198，我们将在24小时内删除。