Catalytic reduction of NO x by biomass-derived activated carbon supported metals

Yun Shu*,Fan Zhang,Fan Wang,Hongmei Wang*

Research Center of Air Pollution Control Technology,Chinese Research Academy of Environmental Sciences,Beijing 100012,China

Keywords:Biomass NO x reduction Activated carbon Selectivity

ABSTRACT In this study,to prepare a series of activated carbon-supported metals for the catalytic reduction of NO x to N2 in excess O2,activated carbons derived from lignocellulosic and herbaceous biomasses were selected as the reducing agents,and alkali and transition metals were used as the catalytic active phases.The effects of the type of biomass,carbonization temperature and catalyst composition on NO x reduction efficiency were analyzed in a fixed-bed flow reactor.The results showed that two temperature regimes are present for the NO x-carbon reaction:at temperatures below 250°C,the NO x adsorption process on the carbon surface was predominant,whereas true NO x reduction by carbon occurred at temperatures above 250°C,producing N2,CO2 and CO.The in fluence of the carbonization temperature on carbon reactivity depended on the effect of the carbonization temperature on the carbon surface area and the reduction of the metal species on carbon.All studied metals catalyzed both NO x and O2 reduction by carbon,and potassium could strongly enhance the C-NO x reaction without substantial carbon consumption by O2.Moreover,the potassium supported by sawdust-derived activated carbon exhibited higherselectivity and capacity towards NO x reduction than did its previously reported coal-derived counterparts.These properties were ascribed to the high dispersion of the active potassium species on the carbon surface,as observed through the comparison of X-ray photoelectron spectroscopy and powder X-ray diffraction results for the carbons made from biomass and coal-based precursors.

1.Introduction

Removal of nitrogen oxides(NOx)from both stationary and mobile sources has become increasingly important in recent years because NOxmake a substantial contribution to acid rain,ground-level ozone and photochemical smog.NOxreduction can be achieved using suitable reducing gases such as ammonia[1,2](for example,the most widely available technology selective catalytic reaction),carbon monoxide,hydrogen and hydrocarbons[3-5].However,these processes suffer from many shortcomings owing to the use of the additional gaseous reducing agents.Carbon materials(e.g.,coal chars and activated carbons)have also been proposed as reducing agents for NOxremoval.The C-NO reaction would lead to the formation of N2,CO and CO2as shown below:

The use of carbon for this purpose offers obvious potential advantages over the gaseous reactants,such as low cost and elimination of the environmental problems related to the gaseous reducing agent slip[6,7].According to the literature,the heterogeneous interaction between carbon and NOxinvolves physisorption/chemisorption and gasification.In the lower temperature range,the NOxadsorption process is predominant on the carbon surface,and the amount of adsorbed NOxdecreased with increasing temperature.In the higher temperature range,carbon gasification by NOxoccurs,leading to the formation of N2,CO2and CO[8,9].The NOxreduction by carbon is an environmentally benign process that will not transfer the pollutants NOxto the other environmentalmedia.However,the large carbon consumption by combustion with O2is a major disadvantage ofthe use ofcarbon,because O2concentration in exhaust gases is usually considerably higher than that of NOxand because of the high affinity of carbon for O2.

Reducing the temperature of the C-NOxreaction has been proven to be the most effective method to minimize the undesired carbon combustion by O2.In previous work,an important decrease in the temperature required for NOxreduction using alkali,alkaline-earth and transition metals,such as potassium[9-11],calcium[12],nickel[13]and copper[14]as catalysts was observed.Illάn-Gόmez et al.[13]investigated the catalytic reduction of NO by activated carbon supported transition metals and concluded that Ni was the most suitable metal,showing the best selectivity for NO reduction and the lowest activity for the O2-carbon reaction.The study of metal-containing char briquettes showed that the extremely promising results with respect to activity and selectivity were obtained for NO reduction by K-containing coal briquettes[15].However,due to the relatively high operation temperature(e.g.,300-400°C),the selectivity of these catalysts still needed to be improved.Many studies suggest that increasing the interfacial area of catalyst/carbon contact which,in turn,is related to the catalyst dispersion on carbon surface,can effectively decrease the temperature required for NOxreduction and thus enhance the selective reduction of NOxin oxygen-rich exhausts.

Biomass materials are a promising type of raw materials for activated carbon production owing to their abundance,low cost,as well as nearly CO2-neutral characteristics.Compared to coal,biomass has different organic structure and morphological characteristics and its volatile content is often higherthan 50%.Vallejos-Burgos etal.[16]recently made comparison between biomass-and coal-derived chars using similar methods.They found that biomass-derived chars were more microporous and showed higher reactivity for combustion or gasification than their coalderived counterparts.In addition,Macías-Pérez etal.[17]recently studied the SO2removalactivity using calcium oxide loaded activated carbon prepared from wood and almond shells.A high SO2removal activity was achieved thatcould be attributed to the high calciumdispersion on the activated carbon.Sumathi et al.[18]reported the simultaneous removal of SO2and NOxusing coconut shell based activated carbon prepared by KOHimpregnation.This activated carbon allowed the access ofpotassium into the internalpore surface and the high degree ofsurface reactivity,improving the efficiency of SO2and NOxremoval.Above all,biomass based activated carbon may be a good candidate to improve the selective reduction of NOxby carbon.However,only a limited number of studies were devoted to the catalytic reduction of NO to N2by biomass-derived activated carbon in an O2-rich environment.

A number of the variables related to the preparation conditions,such as the types of carbon and the carbonization temperature,have a great impact on NOxreduction effectiveness.Illάn-Gόmez et al.[15]examined ten different activated carbons and found that the total surface area was proportional to the NO reduction activity excluding any catalytic effects.While in the presence of catalyst,the total surface area effect could be negligible.García-García etal.[10]investigated the effectofpyrolysis temperature on NO reduction.In the higher reaction temperature regime,the reactivity of the low-rank coal chars increased with increasing pyrolysis temperature,which was ascribed to the increase in the turnover frequency.In the lower reaction temperature regime,the bene ficial effect of higher pyrolysis temperature on the turnover frequency was overshadowed by the effect of catalyst sintering.

The purpose ofthe presentstudy is to study the selective behavior of biomass-derived activated carbon supported metals towards NOxreduction against O2combustion.Lignocellulosic and herbaceous biomasses were used to prepare the activated carbons,and alkali and transition metals were selected as the catalytic actives.The in fluences of the type of biomass,the carbonization temperature and the catalyst composition on the NOxreduction by carbons were analyzed.Furthermore,to understand the differences between the behaviors exhibited by biomass-and coal-derived activated carbons,the selectivity and NO reduction over various carbons described in the literatures were compared to those over the carbon developed in this work.

2.Experimental

2.1.Sample preparation

Sawdust,rice husk and corncob were used as the raw materials for the biomass based activated carbon preparation.Table 1 shows the proximate and ultimate analyses of the raw biomasses used in this study.As shown in Table 1,the raw biomasses exhibited high volatile matter and low N and S contents,and their heating values were rangedfrom 15.06 to 17.53 MJ·kg-1.These values indicated thatthe biomass is an ideal renewable energy resource with minor potential for environmental pollution.The ash analysis of the raw biomass samples was performed by X-ray fluorescence spectroscopy(XRF),as listed in Table 2.The main inorganic elements in the raw biomass samples were Na,K,and Si.Sawdust had the highest amount of Na2O and K2O,whereas rice husk had the highest amount of SiO2.To prepare the activated carbons,the raw biomasses were ground and sieved to the size range of 200 to 400 μm.Carbonization(5 °C·min-1)of the raw biomass was carried out in a horizontal furnace using N2as the carrier gas(150 ml·min-1,2 h at maximum temperature).

Table 1 Proximate and ultimate analyses of the raw materials for activated carbon preparation

Table 2 Ash analysis of the raw materials for activated carbon preparation(wt%)

All metals were introduced by excess-solution impregnation.The prepared activated carbon was first demineralized by the standard acid treatment[19]to diminish the catalytic effects ofthe inherentmineral matter on the NOxreduction.Next,the demineralized sample was impregnated in the aqueous solution of HNO3(2 N)for 24 h and later dried at 180°C for 5 h.Subsequently,the oxidized sample was impregnated with the aqueous solution of metal nitrate of appropriate concentration(10 ml of solution per gram of sample).After the impregnation,the sample was dried in air at 120°C for 12 h,and calcined in N2(150 ml·min-1)at 500 °C for 3 h.The final metal content of the samples was approximately 10 wt%as determined by inductively coupled plasma atomic emission spectrometry(ICP-AES).Hereafter,the prepared samples are denoted as x AC-m(x represents the initial raw material for the activated carbon preparation,m=K,Cu,Fe and Ni).For example,SAC-K indicates the sawdust based activated carbon-supported potassium.

2.2.Catalytic reduction test

The removal of NOxwas conducted in a fixed bed reactor with the inner diameter of14 mmunderheating in an electric furnace atambient pressure.The experimental setup is shown in Fig.1.Two types of NOxreduction tests were carried out:(1)temperature programmed reactions(TPRs),where the sample was heated at 5 °C ·min-1to 600 °C in a NO/O2/N2mixture.(2)Isothermalreactions,where the temperature was raised under N2until the selected temperature,and then the gas mixture replaced the inert gas.Isothermal reactions were typically carried out for 2 h.Approximately 500 mg sample(220 mg of catalyst diluted with 280 mg SiC)was charged into the reactor during each experiment.The typical compositions of the reactant gases were:10-3NO,5 vol%O2,and N2as the balance gas.The total flow rate was 500 ml·min-1(refers to 0.1 MPa and 298 K),corresponding to the gashourly space velocity(GHSV)of10000 h-1.Both the inletand outlet compositions of the reactant gas(NO,NO2,CO,CO2)were measured by a Fourier transform infrared(FTIR)analyzer system(GASMET DX-4000).NO reduction efficiency was calculated according to

Fig.1.Schematic diagram of the experimental setup.1-N2,2-NO,3-O2,4-mass flow controller,5-electric furnace,6- fixed bed reactor,7-catalyst,8-temperature controller,9-gas analyzer.

where ηNOxis the NOxreduction efficiency in%,φ(NOx)inis the[NO]+[NO2]concentration at the inlet of the reactor in 10-6,and φ(NOx)outis the[NO]+[NO2]concentration at the outlet of the reactor in 10-6.

The selectivity factor(F),taking a value in the range 0-1,defines the proportion of carbon consumed in the reduction of NOxrelative to its combustion with O2.This value was calculated according to the following expression:

2.3.Characterization

Specific surface area was measured by nitrogen adsorption using the Brunauer-Emmett-Teller(BET)method(Quadrasorb SI).Powder X-ray diffraction(XRD)measurements were carried out on a Rigaku D/MAX-2400 X-ray diffractometer with Cu Kα radiation.The scanning rate was 8(°)·min-1and the scan was performed from 10°to 80°.X-ray photoelectron spectroscopy(XPS)was implemented on a surface analysis system(Thermal ESCALAB 250)using AlKαradiation.The C 1s line at 284.6 eV was considered as the reference for the binding energy calibration.Temperature-programmed desorption(TPD)was carried out in a quartz fix-bed reactorsystemcoupled to a quadrupole mass spectrometer(IPC400,INFICONCo.Ltd.).In the TPDexperiment,the sample washeated from room temperature to 800 °C at the heating rate of 10 °C·min-1in He flow(300 ml·min-1).The amounts of CO and CO2evolved from the sample were monitored.

3.Results and Discussion

3.1.Sample characterization

Table 3 summarizes several characteristics of the samples,including the nitrogen surface area,ash content,and metal loading.As shown inTable 3,the order of activated carbon surface area was SAC N RAC N CAC,which was in accordance with the volatile content of the raw biomass.Illán-Gómez et al.[15]reported that the volatile content of coal had little in fluence on the activated carbon properties,such as surface area and pore size distribution.Whereas Zhong et al.[20]found thatthe greater the amountofvolatiles in a coal,the more spongy the remaining particles would be after devolatilization,and accordingly the specific area of char particles might be greater.So it suggested that under the experimental conditions used in the current work,the amount of the volatiles of biomass might affect the magnitude of the specific surface area of activated carbons.In addition,the surface area showed a decrease after metal impregnation,suggesting a partial pore blocking by the metal particles.

Table 3 Characteristics of activated carbons and catalysts

3.2.Temperature programmed reactions

TPR profiles for SAC,RAC and CAC samples expressed as the NOxconversion percentages are compared in Fig.2.These experiments were carried out under the same experimental conditions allowing direct comparison of the NO conversion for the various samples.The TPR curves exhibited two distinct activity stages:(1)The NOxconversion decreased with increasing reaction temperature in the 25-250°C range,even reaching negative values.(2)The percentage of NOxconverted increased with increasing reaction temperatures for temperatures above 250°C.These two stages of activity behaviour were in accordance with the results in the literature[9,21],where a significant change in the mechanism of the C-NOxreaction was observed to occur at approximately 250°C.The first stage consisted of NOxadsorption,and the second stage involved carbon gasification by NOx.It should be noted that at any temperature,the sample SAC showed the highest reactivity among the three samples,which may be attributed to its high surface area,as well as high potassium and sodium contents(as shown in Tables 1 and 2).

Fig.2.TPR profiles for the SAC,RAC and CAC samples prepared at 800°C.

Fig.3.TPR profiles for the SAC prepared at different carbonization temperatures.

Fig.3 compares the TPR curves of the SAC samples prepared at different carbonization temperatures.It was observed that in the lowtemperature region,the NOxconversion decreased with increasing carbonization temperature,while in the high-temperature region,the opposite effect was observed:as the carbonization temperature increased,the NOxconversion showed an increase during TPR.Hu et al.[22]studied char gasification under a wide variety of temperature conditions.It was reported that increasing degree of pyrolysis would cause the carbon crystallite growth,and thus resultin a decrease in the carbon surface area.This phenomenon was confirmed by our results(as shown in Table 3)that the specific surface areas of the SAC decreased from607 to 413 m2·g-1when the carbonization temperature rose from 600 to 1000°C.Therefore,the observed decrease in NOxconversion with carbonization temperature in the low temperature region should be ascribed to the decrease in the carbon surface area.

On the other hand,the higher carbonization temperature enhances the reduction of the metal species(e.g.,potassium species)presented on the surface of sample,and thus promotes NOxreduction.The reduction of potassiumspecies was analyzed using TPD tests from room temperature to 800°C in N2(as shown in Fig.S1 in Supporting Information).The SAC showed CO2and CO evolution curves,and the CO desorption occurred at higher temperatures than CO2desorption.The intense CO peak approximately 750°C coincided with the reduction temperature range reported in the study of Gilchrist et al.[23],where it was found thatpotassium oxide was reduced at760°C.Itsuggested thatthe higher carbonization temperature was necessary for the transformation of potassium to its active reduced state.Thus,the increase in the NOxconversion with increasing carbonization temperature in the hightemperature region was due to the enhanced redox property of the metal present on the sample.Above all,the effect of the carbonization temperature on carbon reactivity depended on the effect of the carbonization temperature on the carbon surface area and the reduction of the metal species on carbon.For example,at the lower temperature range,the bene ficial effect of higher carbonization temperature on the redox property of metal present on carbon was overshadowed by the effect of the decrease in the carbon surface area.

Fig.4 illustrates the TPR profiles for SAC and SAC samples loaded with different metals.In agreement with the trends observed for the above samples,the metal-loaded SAC samples showed a decrease in NOxconversion with increasing reaction temperature,reaching negative NOxconversions at approximately 200-250°C.Above this threshold,the percentage of NOxconverted increased with reaction temperature.In the low-temperature region,the catalytic activities of metals on NOxconversion could be hardly observed.While in the high-temperature region,the addition of metal to SAC significantly enhanced the NOxconversion,shifting the starting temperature for NOxconversion and the temperature required to reach the highest conversion to lower temperatures.In this region,the catalytic activity of metals increased in the following order:Cu N K N Ni N Fe N none.This order seemed to indicate that the redox property of the metal determined its catalytic activity at the relatively high temperature(with the exception of SAC-K sample).This is in agreement with the results of Moulijn etal.[24]who found thatthe metalactivity for carbon gasification by O2-containing molecules at high temperature depended on the redox property of the metal.

Fig.4.TPR profiles for the SAC and the samples with different metals prepared at 800°C.

The analysis of the gases produced from the reaction during the TPR tests complements the data supplied by the reduction curves.Fig.5 illustrates the gas evolution curves for SAC and SAC-K.From Fig.5a and b,two stages of the TPR curves could be observed:the first stage(at temperatures below 250°C)was accompanied by a NOxadsorption process,and the evolutions of N2,CO2and CO were not observed in this process.In the presence of O2,NO2is the thermodynamically favored nitrogen oxide,allowing its adsorption on carbon at temperatures below 150°C.[15]The NOxevolution increased with increasing reaction temperature,which may be caused by an increase in NOxdesorption.The NOxdesorption justified the negative values found in Figs.2-4.In the second stage(at temperatures above 250°C),N2,CO2and CO were produced,suggesting that true NOxreduction by carbon occurred.In the presence of potassium(as shown in Fig.5b),the appearance of product evolution(N2,CO2and CO)shifted to lower temperatures than for the metal-free sample,indicating the catalytic effect of potassium on the C-NOxreaction.In addition,the evolution of N2O was not significant on SAC and SAC-K samples,which was in contrast to the study of Thomas et al.[25]that showed that a large amount of N2O was formed during the C-NO reaction in the presence of oxygen.

Fig.5.Gas evolution during TPR of SAC(a)and SAC-K(b).

3.3.Isothermal reactions

Mostofthe samples did notreactuntil350°C(as shown in TPR tests),and thus,isothermalreaction tests at350 and 400°C were performed for the SAC and metal-loaded SAC samples.Figs.6 and 7 present the variations of NOx/O2reductions with time at 350 and 400°C.Examination of the figures shows that all metals catalyze NOxand O2reduction by carbon.The order of the catalytic activity for the NOxreduction by carbon was K N Cu N Ni N Fe N none,which was similar to that obtained from the TPR results(at the high-temperature range of TPR tests,as shown in Fig.4).But the activity order of the C-O2reaction showed some differences:the most active Cu was followed by Fe and K,and Ni was least active.This finding indicated that potassium could strongly enhance the C-NOxreaction without significant carbon consumption.The sample SAC-K showed the most stable reduction curves among the tested samples.It should be noted that the NOxand O2reduction profiles showed a decrease with reaction time due to carbon consumption,and the extent of the activity decrease at 350 °C was lower than that at 400 °C.

Fig.6.Isothermal reaction at 350°C with the SAC and the samples with different metals(a)NO reduction,(b)O2 reduction.

To estimate the selectivity for NOxreduction,parameter F was calculated from the Eqs.(2)and its obtained values are compiled in Table 4.This factor reflects the extent of carbon consumed by NO reduction versus total carbon consumed during the isothermal reaction.It was observed that the sample SAC-K showed the highest F factor among these samples,indicating its most selectivity towards NOxreduction.Ni-and Fe-loaded samples showed lower F factors that still indicated certain selectivity.In contrast,the selectivity achieved for SAC-Cu was unacceptable.These results confirmed the conclusion of previous studies[9-11]that the alkali and transition metals catalyzed both NO-carbon and oxygen-carbon reactions and that potassium exhibited a good selectivity towards NOxreduction against oxygen combustion.For all the samples,an increase in the reaction temperature led to an increase in the amount of NOxreduced during the 2 h-experiment,but the F factors showed an obvious decrease(with the exception of SAC).This was because as the temperature increased,oxygen combustion was more favored than NO reduction.

As listed in Table 4,the CO/CO2ratio ofthe emitted products showed an increase from 350 to 400°C for all the samples.The SAC-K sample displayed the lowest CO/CO2ratio,regardless of the reaction temperature.The minimization of this ratio is important for two fundamental reasons:(1)CO is well-known to be toxic,and(2)the use of carbon is maximized if most of the product is CO2.Thus,the low CO/CO2ratio of the emitted products is another bene ficial effect of the use of potassium as the catalyst for the C-NOxreaction.

3.4.Comparison to the literature

Fig.7.Isothermal reaction at 400°C with the SAC and the samples with different metals(a)NO reduction,(b)O2 reduction.

Table 4 Selectivity and reaction data for tested samples during the isothermal reaction at 350 and 400°C

Fig.8.Selectivity and NOreduced over the various activated carbon-supported potassium catalysts.

It is well known that the catalytic active phase,preparation procedure and operating conditions greatly affect the NOxreduction by carbon.Fig.8 compares the selectivity and NOxreduction amount for the sample developed in this work and the catalysts(K content=10%-15%,carbonized at800°C)described in the literatures tested under similar operating conditions.It could be seen that SAC-K displayed higher selectivity and capacity towards NOxreduction than the previously reported potassium-loaded activated carbons based on lignite coal[9],bituminous coal[26]or metallurgical coke breeze[27].According to the literature,the NO reduction effectiveness of metal-loaded carbon depends on the nature and distribution of the metal.To understand the differences between the behaviors exhibited by biomass-and coalderived activated carbon supported metal,lignite coal was selected to prepare the activated carbon-supported potassium and apply it for NOxreduction.From Fig.8,sample SAC-K displayed much higher selectivity and capacity towards NOxreduction than its coal-derived counterpart.X-ray photoelectron spectroscopy and powder X-ray diffraction were used to analyze the state of the potassium atoms(as shown in Table S1 and Fig.S2 in Supporting Information).Interestingly,the SAC-K sample had a much higher surface K/C ratio of 0.61 than that for the LAC-K sample(0.29)even though similar potassium and ash contents are present in both samples.This indicated that the dispersion of potassium species was higher for the former sample.As shown in Fig.S3,for the SAC-K no visible diffraction peaks could be found,while for LAC-K the diffraction peaks at 30.3°and 31.6°assigned to the presence of potassium carbonate were observed[9].These results indicated that the SAC-K sample displayed a good metal dispersion,which was bene ficial for the selective reduction of NOxby carbon in the presence of O2.

3.5.Lifetime tests

Fig.9.Lifetime tests for the SAC and LAC samples.

Studies on the stability of the carbon under oxidizing atmosphere are important from the theoretical and technological points of view.Fig.9 shows the NOxand O2reduction as a function of time for SAC-K and LAC-K samples.As shown in Fig.9,the SAC-Kshowed a stable activity with the NOxconversion maintained at 40%during 20 h continuous running duration,even though an obvious decrease in NOxreduction was observed in the beginning of the test.Whereas,the NOxreduction of the LAC-K sample decreased to 10%after 20 h.This should be attributed to the much higher dispersion degree of active potassium species on the SAC-K compared to that on the LAC-K(as discussed in Section 3.4),which in turn led to the higher selectivity towards NOxreduction ofthe sample SAC-K.Furthermore,there was little change in the state ofthe potassium species on spent SAC-Ksample after the longevity test(as shown in Table S1 and Fig.S3 in Supporting Information).Above all,the sawdust based activated carbon supported potassium shows a fairly good stability and activity under oxidizing atmosphere,predicting an attractive prospectfor the use ofbiomass-derived activated carbon in practical applications.

4.Conclusions

In the present study,different biomasses were used for the production of activated carbons that could be used as the reducing agents,and alkali and transition metals were used as the catalytic active phases to prepare a series of activated carbon-supported metals for the catalytic reduction of NOxto N2in an O2-rich environment.TPR and isothermal reaction experiments of NOxreduction by carbon in the presence of O2allow us to draw the following conclusions:

(1)Using TPR results,two characteristic temperature regimes were identified:at temperatures below 250°C,NOxadsorption on carbon surface was the dominant phenomenon;at temperatures above 250°C,true NO reduction by carbon occurred,producing N2,CO2and CO.SAC showed the highest reactivity among the tested samples,which may be attributed to its high potassium and sodium contents.

(2)In the low-temperature region,NOxconversion decreased with the carbonization temperature,while in the high-temperature region,the opposite effect was observed.The in fluence of the carbonization temperature on carbon reactivity depended on the effectofthe carbonization temperature on the carbon surface area and on the reduction of the metal species on carbon.

(3)All studied metals catalyze both NOxand O2reduction by carbon,and the order of catalytic activity for the NOx-carbon reaction was not the same as that for the oxygen-carbon reaction.Among the tested metals,potassium was the most selective towards NOxreduction against O2combustion.Furthermore,the low CO/CO2ratio of the emitted products was another bene ficial effect of the use of potassium as the catalyst for the NOx-carbon reaction.

(4)SAC-Kshowed much higher selectivity and capacity towards NOxreduction than the coal-based activated carbon-supported potassium catalysts described in the literature,which was ascribed to the high dispersion ofthe active potassium species on the surface of the former.The results of this study emphasize the possible bene fit of the lack of a gaseous reducing agent(such as ammonia or hydrocarbons)because the reduction of NOxis performed by the carbon support itself.

Supplementary material

Characteristics of SAC-K,LAC-K and spent SAC-K samples,CO2(a)and CO(b)TPD patterns for the sample SAC,X-ray diffraction patterns of samples SAC-K and LAC-K and X-ray diffraction patterns of SAC-K and spent SAC-K samples are available free of charge via the Internet.Those supplementary data to this article can be found online at https://doi.org/10.1016/j.cjche.2018.04.019.