1029

Journal of Oleo ScienceCopyright ©2017 by Japan Oil Chemists’ Societydoi : 10.5650/jos.ess17026J. Oleo Sci. 66, (9) 1029-1039 (2017)

Catalytic Ethanol Dehydration to Ethylene over Nanocrystalline χ- and γ-Al2O3 CatalystsJakrapan Janlamool and Bunjerd Jongsomjit＊

Center of Excellence on Catalysis and Catalytic Reaction Engineering, Department of Chemical Engineering, Faculty of Engineering, Chulalongkorn University, Bangkok 10330, THAILAND

1 IntroductionAt present, ethylene as one of the most important

olefins, is increasingly attracting more attention. It has been used as the major feedstock in many petrochemical industries. In general, ethylene is mainly produced by thermal cracking of naphtha. Many efforts have been done to explore an alternative route to produce ethylene from other carbon resources using cleaner technology. The biomass is considered as one of the most important renew-able energy resources. The fermentation of biomass is famous as one of the most powerful technologies that can be applied to produce bio-ethanol. It has been well known that the advantage of renewable biomass is that the biomass can be converted to green bio-ethanol. The use of bio-ethanol to ethylene has greater development potential and broad application prospects. Ethanol can be dehydrat-ed through intramolecular route to produce ethylene or via intermolecular pathway to obtain diethyl ether（DEE）1）. Many studies have been investigated on the correlation between the catalyst properties and catalytic performances in ethanol dehydration. The catalytic ethanol dehydration has been well known that using solid catalysts with acidic

＊Correspondence to: Bunjerd Jongsomjit, Center of Excellence on Catalysis and Catalytic Reaction Engineering, Department of Chemical Engineering, Faculty of Engineering, Chulalongkorn University, Bangkok 10330, THAILANDE-mail: bunjerd.j@chula.ac.thAccepted April 11, 2017 (received for review February 2, 2017)Journal of Oleo Science ISSN 1345-8957 print / ISSN 1347-3352 onlinehttp://www.jstage.jst.go.jp/browse/jos/　　http://mc.manusriptcentral.com/jjocs

character2, 3）. However, the relationship between catalytic activity and the characteristic of acid type, which is evalu-ated by NH3-TPD, is unclear. Nowadays, the most widely used catalyst in this application is still γ-Al2O3

1, 4）.Although, some other catalysts such as zeolite5, 6）, heteropolyacid7）, transition metal oxides8）, acid carbon9）and montmorillonite clays10）are also investigated.

Recently, the advanced solid materials have been attrac-tive due to their structures that are organized into na-noscale. The nanoscale HZSM-5 zeolite catalyst exhibits better catalytic stability and selectivity for ethylene during ethanol dehydration than microscale HZSM-5 zeolite cata-lyst11）. Moreover, the activated alumina-based catalyst has good stability, and the purity of produced ethylene product based on such a catalyst is also high. However, the concen-tration of raw material of ethanol should not be too low, or it can make the ethanol dehydration reaction, which re-quires higher temperature and lower space velocity, that in turn leads to higher energy consumption. In addition, the nanocrystalline γ-Al2O3 has been well known as a bifunc-tional oxide comprising acidic and basic sites12）. It is the most widely used inert carrier for metal catalysts in hetero-

Abstract: The study is aimed to investigate the combination of nanocrystalline γ- and χ- alumina that displays the attractive chemical and physical properties for the catalytic dehydration of ethanol. The correlation between the acid density and ethanol conversion was observed. The high acid density apparently results in high catalytic activity, especially for the equally mixed γ- and χ- phase alumina (G50C50). In order to obtain a better understanding on how different catalysts would affect the ethylene yield, one of the most powerful techniques such as X-ray photoelectron spectroscopy (XPS) was performed. Hence, the different O 1s surface atoms can be identified and divided into three types including lattice oxygen (O, 530.7 eV), surface hydroxyl (OH, 532.1 eV) and lattice water (H2O, 532.9 eV). It was remarkably found that the large amount of O 1s surface atoms in lattice water can result in increased ethylene yield. In summary, the appearance of metastable χ-alumina structure exhibited better catalytic activity and ethylene yield than γ- alumina. Thus, the introduction of metastable χ- alumina structure into γ- alumina enhanced catalytic activity and ethylene yield. As the result, it was found that the G50C50 catalyst exhibits the ethylene yield (80%) at the lowest reaction temperature ca. 250℃ among other catalysts.

Key words: ethanol dehydration, XPS analysis, solvothermal method, mixed phase alumina

J. Janlamool and B. Jongsomjit

J. Oleo Sci. 66, (9) 1029-1039 (2017)

1030

geneous catalysis because of its fine particle size, high surface area, surface catalytic activity, high mechanical re-sistance, excellent thermal stability, and wide range of chemical, physical, and catalytic properties13－15）. Several studies attempt to synthesize alumina that has uniform mesoporous distribution. Besides, the surface area and pore structure also play important roles in the catalytic ac-tivity. The nanocrystalline Al2O3 has been generally synthe-sized by the sol-gel16）, precipitation17）, flame spray pyroly-sis18）and solvothermal13－15）methods. The solvothermal synthesis attracts the most attention because this applica-tion gives products with small uniform morphology, homo-geneous chemical composition, and narrow size distribu-tion19）.

The nanocrystal l ine Al2O3 having mixed γ- and χ-crystalline phases, which is prepared from the thermal decomposition of aluminium isopropoxide（AIP）in organic solvent exhibits high thermal stability. The nanocrystalline χ-Al2O3 is one of the metastable polymorphs based on the hexagonal close packing（hcp）of the oxygen atom20）. The previous studies show that highly stable nanocrystalline Al2O3 with mixed γ- and χ-crystalline phases prepared by the solvothermal method resulted in interesting outcomes. It can be employed as catalyst supports in many catalytic reactions such as CO oxidation21）, and propane oxidation22）.

The goal of this research was to present the outstanding properties of the nanocrystalline alumina comprising of mixed γ- and χ-crystalline phases for catalytic behaviors via vapor phase ethanol dehydration. The mixed γ- and χ-crystalline phase alumina was prepared using the solvo-thermal method. The results obtained from catalytic be-haviors are used to evaluate the role of the different char-acteristics on the overall reaction activity and product selectivity.

2 Experimental2.1 Materials

Chemicals as follows were used; aluminium isopropoxide 98％（Aldrich）, methanol（commercial grad）, 1-butanol（Fischer scientific）, toluene（Fischer scientific）, and 99.9％ ethanol（VWRPROLABO）.

2.2 Nanocrystalline aluminapreparationNanocrystalline alumina having different chi and gamma

phases（Table 1）were prepared by the solvothermal method according to the procedure described by Meephoka et al.21）. 25 g of aluminium isopropoxide（AIP, 25g）was dis-solved in 100 ml with different ratios of mixed solvent（butanol and toluene as shown in Table 1）in a test tube. The autoclave was purged with nitrogen after that it was heated up to the desired temperature（300℃）at the rate of 2.5℃ min－1 and the temperature was held for 2 h. The precipitated powder was collected after the autoclave had been cooled to room temperature. The white powder was repeatedly washed with methanol and dried at room tem-perature. The synthesized alumina was subsequently cal-cined with a heating rate of 10℃/min at the temperature raised to 600℃ and held on that temperature for 6 h using an air flow rate of 100 Ml/min. The obtained alumina cata-lysts consisting of single γ-phase, 30％, 50％ χ-phase and single χ-phase are denoted as G100, G70C30, G50C50 and C100, respectively as confirmed by the XRD measurement.

2.3 Catalyst characterization and reaction testThe prepared alumina catalysts were characterized by

several techniques as follows; 2.3.1 Surface area and porosity

N2 physisorption（N2 adsorption at －196℃ in a Mi-cromeritics ASPS 2020）was performed to determine surface areas, pore size and adsorption isotherm of the dif-ferent alumina catalysts.2.3.2 X-ray diffraction（XRD）

XRD was used to determine the phase composition of the different mixed phase alumina using Bruker D8 Advance X-ray diffractometer with Cu-Kα radiation（λ＝1.54056 Å）with Ni filter in the 2θ range of 20-80 degrees with resolution of 0.04°.2.3.3 Scanning electron microscopy（SEM）and dispersive

X-ray spectroscopy（EDX）SEM（JEOL mode JSM-5800LV）and EDX（Link Isis Series

300）were used to determine the morphology and elemental distribution of the catalyst particles. The morphology of alumina samples was observed using JEOL-JEM 200CX transmission electron microscope operated at 100 kV.

Table 1　Detail of different alumina catalysts and nomenclatures.

CatalystsPhase % Dissolved solvent (ml) the gap between the test

tube and autoclave wall (ml)χ- phases γ- phases Toluene 1-Butanol Toluene 1-Butanol

G100 0 100 0 100 0 30G70C30 30 70 30 70 9 21G50C50 50 50 50 50 15 15

C100 100 0 100 0 30 0

Catalytic Ethanol Dehydration to Ethylene over Nanocrystalline χ- and γ-Al2O3 Catalysts

J. Oleo Sci. 66, (9) 1029-1039 (2017)

1031

2.3.4 Ammonia temperature programmed desorption（NH3-TPD）

The acidity of alumina was investigated by temperature programmed desorption of ammonia using a Micromeritics Chemisorp 2750 with a computer. First, 0.10 g of the alumina sample was placed in a glass tube and pretreated at 500℃ with flowing of helium for 1 h. Then, the sample was saturated with 15％ NH3 in He at room temperature. After saturation, the physisorbed ammonia was desorbed in a He flow for 3 h. Then, the sample was heated from 30 to 500℃ with a heating rate of 10℃/min. The effluent amount of ammonia was measured via TCD signal as a function of time.2.3.5 X-ray photoelectron spectroscopy（XPS）

The XPS analysis was performed originally using an AMICUS spectrometer equipped with a Mg Kα X-ray radia-tion. For a typical analysis, the source was operated at voltage of 15 kV and current of 12 mA. The pressure in the analysis chamber was less than 10－5 Pa.

Ethanol dehydration was performed to determine the overall activity and desired product yield of the catalysts. Typically, 0.05 g of catalyst was packed in the middle of the glass microreactor. The catalyst sample was pre-treated in situ in flowing Ar（50 mL/min）at 200℃ for 1 h prior to ethanol dehydration. Ethanol was introduced into the reactor by bubbling Ar as a carrier gas through the satura-tor, while the temperature was kept at 45℃ to maintain the partial pressure and hence the composition of the feed. A flow rate of Ar was kept at 50 mL/min with WHSV of 8.4［gEthanol×（gcat×h）－1］. The reaction temperature was ranged between 200 to 400℃ at atmospheric pressure. The product gas samples were analyzed by gas chromatography［Flame ionization detector（FID）, where the DB-5 capillary column was used for separation of ethanol（C2H5OH）, ethylene（C2H4）, diethyl ether（C2H5OC2H5）, and acetalde-hyde（C2H4O）］.

3 Results and discussion3.1 Characteristics

The physical properties of the catalysts were obtained from the N2 physisorption and XRD data. The compositions of γ- and χ-phases of nanocrystalline alumina were pro-duced depending on the different types and amounts of solvents, such as 1-butanol and toluene employed during the synthesis. In fact, the mixed γ- and χ-phases of nano-crystalline alumina was controllably obtained by mixing solvents between 1-butanol and toluene with desirable composition as shown in Table 1. The phase identification of all catalysts was carried out on the basis of data obtained from XRD measurement as shown in Fig. 1. The G100 cata-lyst, exhibits the diffraction peak at 32 37, 39, 45, 61 and 66° as proven that G100 catalyst is the single γ-phase.

Moreover, with the use of mixed solvent between toluene and 1-butanol, the characteristic diffraction peak of χ-phase nanocrystalline alumina was remarkably appeared at 43°. Certainly, the higher intensity of peak area at 43° can be attributed to the increase of χ-phase present in the catalysts. In addition, the γ- and χ-phase compositions of alumina catalysts were determined by the area of charac-teristic peak at 43° from the calibration curve obtaining from the XRD patterns of the physical mixture between single γ- and χ-phase with different contents according to Meephoka et al.21）

The effect of phase composition on the pore size and pore size distribution of catalysts measured by the nitrogen adsorption-desorption is shown in Figs. 2a and 2b. The isotherms of all catalysts are in the classical shape of the typical Type IV isotherms that refer to mesoporous materi-als as described by the IUPAC. As seen in Fig. 2a, the iso-therms of the G70C30, G50C50, and C100 catalysts reveal the H1 hysteresis loops occurred at a relative pressure range（P/P0）of 0.55 – 0.95. The H1 hysteresis loop is often associated with porous materials indicating a broad pore size distribution with a uniform cylindrical-like pore. In the contrary, the isotherm of G100 catalyst was quite different indicating that the H3 hysteresis loop does not exhibit any limiting adsorption at high relative pressure. This behavior can be for instance caused by the existence of non-rigid aggregation of wrinkled sheet-like particles or agglomera-tion of slit-shaped pores. The pore size distribution present here confirms isotherm assertion and mesoporous struc-ture for all prepared catalysts as seen in Fig. 2b. The struc-tural parameters of all alumina catalysts derived from these isotherms are summarized in Table 2. The BET surface area of the catalysts decreased from 257 to136 m2/g with the larger quantity of χ-phase being present from 0 to 100％, whereas the average pore diameter was rather similar for the different catalysts.

Fig. 1　XRD patterns of all prepared catalysts.

J. Janlamool and B. Jongsomjit

J. Oleo Sci. 66, (9) 1029-1039 (2017)

1032

The SEM images of prepared catalysts containing the single γ-phase（G100）, the mixed-phase（G50C50）and χ-phase（C100）are shown in Figs. 3a-3c, respectively. It can be observed that the catalyst particles are the second-ary agglomerated catalyst particles of the prepared single γ-phase and single χ-phase alumina resulting in the pres-ence of large micron sized granules.

It is well known that the acidity on the surface of alumina catalysts importantly influences on their catalytic properties. The acidity and acid strength were investigated

using the temperature-programmed desorption of NH3

（NH3-TPD）. The bimodal desorption peaks obviously reveal that the catalyst samples occupy a heterogeneous distribution of two different strength types of acid sites. Two distinguish desorption peaks are according to the temperature range between 40 to 250 and 250 to 500℃ as

Fig. 2　 N2 physisorption profiles of all prepared catalysts（a）adsorption-desorption isotherm and（b）pore size distribution profiles.

(a)

(b)

Table 2　�BET surface area, pore volume and pore diameter of all prepared catalysts.

Catalysts ABET

(m2/g)Vp

(cm3/g)DBJH

(nm)G100 257 0.84 8.27

G70C30 254 0.87 8.19G50C50 168 0.37 5.46

C100 137 0.55 10.42

Fig. 3　 SEM micrographs of prepared pure γ- phase and χ-phase alumina catalysts（a）G100, （b）G50C50, （c）C100.

(a)

(b)

(c)

Catalytic Ethanol Dehydration to Ethylene over Nanocrystalline χ- and γ-Al2O3 Catalysts

J. Oleo Sci. 66, (9) 1029-1039 (2017)

1033

seen in Fig. 4. This can be generally assigned to the overlap of two desorption steps of weak acid sites and moderate to strong acid sites. The weak acid sites are probably defined as Lewis acid site on aluminum oxide at lower tempera-ture1）. It was found that the fraction of weak acid sites for all catalysts was ranged between 22.5 to 36.4％. On the contrary, the moderate to strong mode of interaction is perhaps defined as the Brønsted acid sites at higher tem-perature1）. However, the identity of acid type, which was evaluated by NH3-TPD, was unclear. The total amount of

acid sites determined either per mass of catalysts or per surface area is practically equal for all catalysts as given in Table 3. The total acid sites for all catalysts were ranged from 804 to 1,332 μmol NH3/g. It can be observed that mainly the acid sites distribution was consistent upon the moderate to strong acid sites. In addition, if the NH3 de-sorption data are expressed in per unit surface area, the density of surface acid sites for all alumina catalysts will approximately range between 4.4 to 7.7 μmol NH3/m

2. The G70C30 catalyst exhibited the highest concentration of the acid sites at 1,332 μmol NH3/g, while the G50C50 catalyst displayed the highest acid density on the surface of the catalyst at 7.7 μmol NH3/m

2. This crucial characteristic will be used to explain the catalytic behaviors for all catalysts during ethanol dehydration reaction.

3.2 Ethanol dehydration reactionIn order to determine the catalytic behaviors of all pre-

pared catalysts, the conversion of ethanol and the yield of products such as ethylene, diethyl ether and acetaldehyde as a function of temperature were investigated in vapor phase ethanol dehydration. The experiment was performed with the same catalyst weight. The dehydration activity is already significant at ca. 200℃ and total at 400℃ as dis-played in Table 4. Typically, ethanol conversion increases as the reaction temperature is raised. The obtained cata-lytic activity for all catalysts can be achieved completely for ethanol conversion at 300℃. The resulting conversions

Fig. 4　NH3-TPD profiles of all prepared catalysts.

Table 3　NH3 TPD analysis of all prepared catalysts.

CatalystsAcidity (µmol NH3/g) Acid site density (µmol NH3/m

2)

Weak Moderate to Strong

Total acid site Weak Moderate to

Strong Total

G100 393.82 743.17 1136.99 1.53 2.89 4.42G70C30 533.08 799.31 1332.39 2.09 3.15 5.24G50C50 356.53 936.36 1292.89 2.12 5.58 7.70

C100 180.80 623.22 804.02 1.32 4.55 5.87

Table 4　Conversion and reaction rate of ethanol dehydration for all prepared catalysts.

Temperature (℃)Ethanol Conversion (%) Rate of ethanol consumption

(gethanol×(gcat×h)–1)G100 G70C30 G50C50 C100 G100 G70C30 G50C50 C100

200 29.3 43.4 47.0 40.0 2.5 3.7 3.9 3.4225 50.4 54.2 68.2 67.0 4.2 4.6 5.7 5.6250 70.7 72.6 87.3 83.7 5.9 6.1 7.3 7.0275 86.6 86.5 97.2 95.1 7.3 7.3 8.2 8.0300 97.0 96.2 100.0 99.5 8.2 8.1 8.4 8.4350 100.0 100.0 100.0 100.0 8.4 8.4 8.4 8.4400 100.0 100.0 100.0 100.0 8.4 8.4 8.4 8.4

J. Janlamool and B. Jongsomjit

J. Oleo Sci. 66, (9) 1029-1039 (2017)

1034

were ranged between 96.2 to 100％ in accordance with the ethanol consumption rate of 8.1 to 8.4（gEthanol×（gcat×h）－1）, respectively. The catalytic activity decreased in the order: G50C50＞C100＞G70C30＞G100 during the reaction tem-perature around 200 to 300℃. Moreover, it can be observed that the catalytic activity of the mixed of γ- and χ-alumina（G70C30 and G50C50）and single χ-phase alumina（C100）exhibited remarkably higher catalytic conversion than the single γ-phase alumina（G100）at the temperature of 200℃. During the reaction temperature between 225 to 275℃, it was attractively observed that G50C50 and C100 still present distinctly high catalytic conversion, whereas G70C30 was not significantly different in the catalytic ac-tivity with G100. In a number of researches, they were ob-served that the acidity over solid catalysts predominantly plays an important role on alcoholic dehydration. The presence of strong acid sites on alumina catalysts increases the alcoholic conversion14, 23, 24）. The result indicated that the dehydration activity was independent of the amount of acid sites on the catalyst. Evidently, this is consequence of an effect of phase composition on alumina catalysts as being observed from the result that C100 exhibits higher catalytic activity thanG100 in spite of its lower acidity. Moreover, the effect of different morphologies on the phys-icochemical, surface and catalytic properties of γ-alumina was also explained25）. Interestingly, this catalytic observa-tion displayed the ethanol conversions and reaction rate related to the acid density10）（μmol NH3/m

2）as mentioned in Table 3. The obtained highest activity of G50C50 was in ac-cordance with the largest total acid density（7.7 μmol NH3/m2）, while the lowest activity of G100 was corresponding to the least total acid density（4.4 μmol NH3/m

2）. The im-portant factor is difficult to understand, while the mecha-nistic detail of ethylene formation remains unclear. In a number of experiments, it was demonstrated that the eth-

ylene and diethyl ether formation depends on acidity. The catalytic process may occur via two reaction pathways. One possible pathway includes diethyl ether acts as a reac-tion intermediate, which is subsequently decomposed to ethylene and ethanol26, 27）as seen in Scheme 1. A bimolecu-lar nucleophilic substitution mechanism is preferred with two adjacent ethanol molecules which probably co-adsorb on two nearby neighboring sites and react to form an ethanol dimeric species（diethyl ether）at low temperature. The shorter distances between two nearby neighboring acid sites（high acid density）probably lead to rapidly react to form diethyl ether as seen in Scheme 1. This rationale can be used to clarify the fact that why diethyl ether selec-tivity is dominant at low temperature5, 10）.

Ethanol conversion can be catalyzed by acidic or basic site, therefore the yield of desired products can be easily related to the characteristics of the surface. Besides, the measurement of elemental composition on the catalyst surface such as XPS is famous as one of the most powerful techniques that can be also applied to obtain the informa-tion of the oxidation state and chemical environment of the elements on the surface of the catalyst. The XPS analysis, that has the depth of ca. 10 Å, convinces the external surface elemental concentrations influencing on the cata-lytic activity. The XPS data for all alumina catalysts are given in Table 5. The Al 2p and O 1s peaks were detected at the binding energy between 74.10 to 75.05 eV and 531.00 to 531.95 eV, respectively. As seen from XPS results, the atomic ratio of Al 2p over O 1s is slightly lower than the expected composition of Al2O3, which has the value of 0.67. It was observed that the atomic ratios of Al 2p over O 1s were ranged from 0.478 to 0.530. Moreover, this ratio slightly decreased with increasing the quantity of χ-phase. This reason becomes apparent from the possessed defect spinel lattices, which are slightly tetragonal distort-

Scheme 1　The conceptual pathway for ethanol conversion.

Catalytic Ethanol Dehydration to Ethylene over Nanocrystalline χ- and γ-Al2O3 Catalysts

J. Oleo Sci. 66, (9) 1029-1039 (2017)

1035

ed that can be pronounced for γ-alumina25）. Hexagonal χ-alumina seems to possess a layer structure. The arrange-ment of anions inherited from gibbsite, whereas the alumi-num cations occupy octahedral sites within the hexagonal oxygen layer28）.

In principle, the transition alumina generally from alumi-num monohydroxide and trihydroxide. As the dehydration process of transition alumina in air takes place, the loss of water by desorption of physisorbed water or by condensa-tion of hydroxyl group occurs. The dehydroxylation condi-tion predominantly presents that the condensation of two nearby surface hydroxyls forms water molecule on the surface during the formation of alumina catalysts as seen in Scheme 21）.

As mentioned above, the O 1s spectra show the presence of three different oxygen species. The spectra analysis of XPS core level spectra focusing on O 1s was used to identi-fy the oxygen species on the surface of all alumina catalysts as seen in Figs. 5a-5d. The deconvolution of the O 1s spectra, separated and fitted into three signals can be gen-erally attributed to the different species of atomic oxygen such as H2O, OH and O on the surface. Normally, the crystal structure of corundum has lattice oxygen（O）, which is in the form of aluminium oxide in accordance with binding energy of 530.7 eV29）. Besides, the corundum structure of alumina contains O 1s peak including surface hydroxyl（OH）and lattice water（H2O）corresponded to 532.1 and 532.9 eV, respectively. The atomic concentra-tions of each O 1s species for all catalysts were fitted and determined as mentioned in Table 5. This result presents

that the G100 catalyst exhibited the highest area fraction of the lattice oxygen about 77.4％ from the total atomic oxygen, while the G50C50 catalyst displayed the maximum value of the lattice water ca. 22.6％ from the total atomic oxygen.

The alumina catalysts predominantly lead to the forma-tion of ethylene and diethyl ether either with minor amount of acetaldehyde for ethanol dehydration. The process involves the unimolecular dehydration of ethanol leading to ethylene formation, while the bimolecular reac-tion produces diethyl ether. The yield towards ethylene in-creased with rising reaction temperature, while the yield towards diethyl ether apparently decreased（Table 6）. At low temperature, ethanol dehydration is predominantly a bimolecular reaction whilst the unimolecular reaction route prevails at high temperature30）. On the contrary, the mech-anism of ethylene formation has been expected from the decomposition of diethyl ether at higher temperature31, 32）. The yield towards ethylene reaches around 88.4 to 97.9％ at 300℃. The obtained yield of ethylene for all catalysts was approximately completed at the reaction temperature range between 300 to 350℃. It decreased in the following order: G50C50＞C100＞G70C30＞G100 as observed at the reaction temperature between 200 to 300℃. This result can likely be illustrated as the dehydration of ethanol to ethylene corresponded to the amount of lattice water（Brønsted acid）on the surface, which was evaluated from the O 1s core level spectra as shown in Table 5. The H2O fraction decreased in the following order: G50C50（22.6％）＞ C100（20.5％）＞ G70C30（14.6％）＞ G100（2.3％）. The

Scheme 2　The conceptual pathway for acid site formation on alumina surface by dehydroxylation process1）.

Table 5　XPS analysis of all prepared catalysts.

Catalysts BE for Al2p(eV)

BE for O1s(eV)

Atomic concentration (%)

Atomic ratio

O 1s peak area fraction (%O 1s)

O 1s atomic concentration (%)

Al 2p O 1s Al/O H2O OH O H2O OH OG100 74.10 531.00 34.61 65.39 0.530 0.023 0.203 0.774 1.52 13.28 50.59

G70C30 74.30 531.70 34.37 65.63 0.524 0.146 0.477 0.378 9.56 31.29 24.79G50C50 75.05 531.95 32.95 67.05 0.491 0.226 0.518 0.256 15.14 34.74 17.17

C100 74.90 531.80 32.36 67.64 0.478 0.205 0.457 0.338 13.86 30.93 22.85

J. Janlamool and B. Jongsomjit

J. Oleo Sci. 66, (9) 1029-1039 (2017)

1036

predominant concentration of lattice water on the surface of G50C50 was received while the highest efficient yield of ethylene was detected. This relation is in good agreement with the Brønsted acid, that being active sites as the direct conversion of ethanol into ethylene via elimination reac-tion8）. Recently, the appearance of lattice water in hydrated AgPW salts plays important role in generation of protons, which needed for an acidic type reaction33）. Moreover, a

moderate surface acid site of ZSM-5 seems to be suitable for bioethanol to ethylene reaction34）. In addition, C100 has a higher yield of ethylene than that of G100. This attractive observation is an effect of phase composition on alumina catalysts. Moreover, the possible pathway including diethyl ether formation and decomposition leads to ethylene pro-duction as mentioned before. The higher efficiency of diethyl ether formation probably took place to higher eth-

Fig. 5　XPS analysis for O 1s spectra of all prepared catalysts（a）G100, （b）G70C30, （c）G50C50, （d）C100.

(a) (c)

(b) (d)

Table 6　Yield to ethylene, diethyl ether and acetaldehyde of ethanol dehydration for all prepared catalysts.

Temperature (℃)

Ethylene yield (%) Diethyl ether yield (%) Acetaldehyde yield (%)

G100 G70C30 G50C50 C100 G100 G70C30 G50C50 C100 G100 G70C30 G50C50 C100

200 7.18 18.86 25.12 19.61 18.26 22.94 20.29 19.04 3.86 1.60 1.60 1.35

225 26.50 40.22 54.52 51.71 19.08 11.46 11.48 12.46 4.82 2.53 2.20 2.83

250 54.04 62.52 78.12 75.85 10.28 6.13 5.66 3.83 6.38 3.95 3.52 4.03

275 75.71 78.72 91.08 89.85 2.67 2.66 0.73 0.74 8.22 5.11 5.39 4.51

300 88.35 91.24 97.92 95.93 0.71 0.46 0.00 0.05 7.94 4.50 2.08 3.52

350 97.99 99.09 99.50 98.72 0.00 0.00 0.00 0.00 2.01 0.91 0.50 1.28

400 99.55 99.58 99.71 99.61 0.00 0.00 0.00 0.00 0.45 0.42 0.29 0.39

Catalytic Ethanol Dehydration to Ethylene over Nanocrystalline χ- and γ-Al2O3 Catalysts

J. Oleo Sci. 66, (9) 1029-1039 (2017)

1037

ylene formation rate.The formation of diethyl ether begins around 18.3 to

22.9％ at 200℃（Table 6）. The obtained yield of diethyl ether for all catalysts was absolutely disappeared at the re-action temperature of 300℃. It decreased in the following order: G100＞G70C30＞C100＞G50C50during the reaction temperature between 200 to 300℃. The yield of diethyl ether is disproportional to the yield of ethylene. The highest ethylene yield of G50C50 was obtained, while the lowest efficient yield of diethyl ether was found. The for-mation towards acetaldehyde can be observed at the reac-tion temperature around 200 to 350℃. It decreased in the following order: G100＞G70C30＞C100＞G50C50. More-over, the yield of acetaldehyde from the mixed phase of γ- and χ-alumina（G70C30 and G50C50）and single χ-phases alumina（C100）remarkably displayed the low yield at 1.6％, while the single γ-phase alumina（G100）exhibited 3.9％ at the reaction temperature of 200℃. This evidence can be alternatively used to suggest that the appearance of χ-phases promoted the inhibition of dehydrogenation reac-tion. Nevertheless, the suitable temperature of the acetal-dehyde formation was seen around 275℃. The basic cata-lysts are responsible for the dehydrogenation of ethanol to produce acetaldehyde35）.

It should be noted that the appearance of metastable χ-alumina structure based on hexagonal close packing（hcp）exhibited better catalytic activity and ethylene yield than γ-alumina with the cubic close packing（ccp）. However, to compare the performance for ethylene pro-duction, the ethylene yield at 80％ is considered for this framework alumina catalyst. Thus, the plot of ethylene yield as a function of temperature was shown in Fig. 6. For the reaction temperature at 200℃, the ethylene yield of G50C50 is 3.5, 2.7 and 2.6 times higher than that of G100, C100 and G70C30, respectively. In addition, the tempera-ture at which ethylene yield reaches at 80％ is designated as T80. The obtained T of G50C50, C100, G70C30 and G100 catalysts is 252, 257, 277 and 284℃, respectively. It indi-cated that the G50C50 distinctly exhibits the best perfor-mance in ethylene production, even at the reaction tem-perature lower than 300℃.

4 ConclusionIn summary, the combination of nanocrystalline γ- and

χ-alumina can be attractively used for catalytic dehydra-tion of ethanol with distinct chemical and physical proper-ties from single γ- and χ-phase alumina. The correlation between ethanol conversion and acid density of the cata-lysts suggested that the possible pathway includes diethyl ether as a reaction intermediate, which is continually de-composed to ethylene at low temperature. The relationship between ethylene yield and H2O on surface as observed

from XPS analysis is crucial. The effect of phase composi-tion and crystallographic orientation on alumina displays the distinct activity of acid site. The appearance of meta-stable χ-alumina structure based on hexagonal close packing（hcp）exhibits better catalytic conversion and eth-ylene yield than γ-alumina based on cubic close packing（ccp）, in spite of its lower acidity. The resulting catalytic activity in ethanol dehydration was dependent on the sample acidity. This behavior can be explained by the com-bination of samples acid density and surface adsorbed water. This work highlights on both adsorbed water and acid density that must be considered in the evaluation of the overall activity of the catalysts.

AcknowledgementThe authors thank the Royal Golden Jubilee Ph.D. schol-

arship from the Thailand Research Fund, Grant for Inter-national Research Integration: Chula Research Scholar, Ratchadaphiseksomphot Endowment Fund, Grant for Re-search: Government Budget, Chulalongkorn University（2017）, and the National Research Council of Thailand（NRCT）for the financial support of this project.

References1） Martins, L.; Cardoso, D.; Hammer, P.; Garetto, T.; Pul-

cinelli, S.H.; Santilli, C.V. Efficiency of ethanol conver-sion induced by controlled modification of pore struc-ture and acidic properties of alumina catalysts. Appl. Catal., A: Gen. 398, 59-65（2011）.

2） Takahara, I.; Saito, M.; Inaba, M.; Murata, K. Dehydra-tion of ethanol into ethylene over solid acid catalysts. Catal. Lett. 105, 249-252（2005）.

3） Kamsuwan, T.; Jongsomjit, B. A comparative study of

Fig. 6　Ethylene yield of all prepared alumina catalysts.

J. Janlamool and B. Jongsomjit

J. Oleo Sci. 66, (9) 1029-1039 (2017)

1038

different Al-based solid acid catalysts for catalytic de-hydration of ethanol. Engineering Journal 20, 13（2016）.

4） Phung, T.K.; Herrera, C.; Larrubia, M.Á.; García-Dié-guez, M.; Finocchio, E.; Alemany, L.J.; Busca, G. Sur-face and catalytic properties of some γ-Al2O3 powders. Appl. Catal., A: Gen. 483, 41-51（2014）.

5） Kamsuwan, T.; Praserthdam, P.; Jongsomjit, B. Diethyl ether production during catalytic dehydration of etha-nol over Ru- and Pt- modified H-beta zeolite catalysts. J. Oleo Sci. 66, 199-207（2017）.

6） Madeira, F.F.; Gnep, N.S.; Magnoux, P.; Maury, S.; Cadran, N. Ethanol transformation over HFAU, HBEA and HMFI zeolites presenting similar Brønsted acidity. Appl. Catal., A: Gen. 367, 39-46（2009）.

7） Varisli, D.; Dogu, T.; Dogu, G. Ethylene and diethyl-ether production by dehydration reaction of ethanol over different heteropolyacid catalysts. Chem. Eng. Sci. 62, 5349-5352（2007）.

8） Riad, M.; Sobhi, Z.; Mikhail, S. Catalytic dehydration reaction of ethanol over transition metal catalysts. J. Eng. Appl. Sci. 49, 195-208（2002）.

9） Bedia, J.; Barrionuevo, R.; Rodríguez-Mirasol, J.; Cor-dero, T. Ethanol dehydration to ethylene on acid car-bon catalysts. Appl. Catal., B: Environ. 103, 302-310（2011）.

10） Krutpijit, C.; Jongsomjit, B., Catalytic ethanol dehy-dration over different acidactivated montmorillonite clays. J. Oleo Sci. 65, 347-355（2016）.

11） Bi, J.; Guo, X.; Liu, M.; Wang, X. High effective dehy-dration of bio-ethanol into ethylene over nanoscale HZSM-5 zeolite catalysts. Catal. Today 149, 143-147（2010）.

12） Vít, Z.; Vala, J.; Málek, J. Acid-base properties of alu-minium oxide. Appl. Catal. 7, 159-168（1983）.

13） Janlamool, J.; Jongsomjit, B. Oxidative dehydrogena-tion of ethanol over AgLi–Al2O3 catalysts containing different phases of alumina. Catal. Commun. 70, 49-52（2015）.

14） Inmanee, T.; Pinthong, P.; Jongsomjit, B. Effect of Cal-cination Temperatures and Mo Modification on Nano-crystalline（γ-χ）-Al2O3 Catalysts for Catalytic Ethanol Dehydration. J. Nanomater. 2017, 1-9（2017）.

15） Mekasuwandumrong, O.; Silveston, P.L.; Praserthdam, P.; Inoue, M.; Pavarajarn, V.; Tanakulrungsank, W., Synthesis of thermally stable micro spherical χ-alumina by thermal decomposition of aluminum iso-propoxide in mineral oil. Inorg. Chem. Commun. 6, 930-934（2003）.

16） Keshavarz, A.R.; Rezaei, M.; Yaripour, F. Nanocrystal-line gamma-alumina: A highly active catalyst for di-methyl ether synthesis. Powder Technol. 199, 176-179（2010）.

17） Hosseini, Z.; Taghizadeh, M.; Yaripour, F. Synthesis of

nanocrystalline γ-Al2O3 by sol-gel and precipitation methods for methanol dehydration to dimethyl ether. J. Nat. Gas Chem. 20, 128-134（2011）.

18） Liu, C.; Liu, Y.; Ma, Q.; He, H., Mesoporous transition alumina with uniform pore structure synthesized by alumisol spray pyrolysis. Chem. Eng. J. 163, 133-142（2010）.

19） Inoue, M.; Kominami, H.; Inui, T. Thermal transforma-tion of X-alumina formed by thermal decomposition of aluminum alkoxide in organic media. J. Am. Ceram. Soc. 75, 2597-2598（1992）.

20） Brindley, G.W.; Choe, J.O. The reaction series, gibbsite- ＞chi alumina-＞kappa alumina-＞corundum. Am. Mineral. 46, 15（1961）.

21） Meephoka, C.; Chaisuk, C.; Samparnpiboon, P.; Praser-thdam, P. Effect of phase composition between nano γ- and χ-Al2O3 on Pt/Al2O3 catalyst in CO oxidation. Catal. Commun. 9, 546-550（2008）.

22） Glinrun, T.; Mekasuwandumrong, O.; Panpranot, J.; Chaisuk, C.; Praserthdam, P., Improvement of propane oxidation activity over Pt/Al2O3 by the use of MIXED γ- and χ-Al2O3 supports. React. Kinet. Mech. Catal. 100, 441-448（2010）.

23） Oleg, V.K.; Valery, A.M. Intermediates and mechanisms of the heterogeneous catalytic reactions. The simplest reactions of hydrocarbons, alcohols, and acids. Russ. Chem. Rev. 63, 559（1994）.

24） Zhang, D.; Wang, R.; Yang, X. Effect of P content on the catalytic performance of P-modified HZSM-5 cata-lysts in dehydration of ethanol to ethylene. Catal. Lett. 124, 384-391（2008）.

25） Phung, T.K.; Lagazzo, A.; Rivero Crespo, M.Á.; Sán-chez Escribano, V.; Busca, G. A study of commercial transition aluminas and of their catalytic activity in the dehydration of ethanol. J. Catal. 311, 102-113（2014）.

26） Nguyen, T.M.; Le Van Mao, R. Conversion of ethanol in aqueous solution over ZSM-5 zeolites: Study of the re-action network. Appl. Catal. 58, 119-129（1990）.

27） Zhang, M.; Yu, Y. Dehydration of ethanol to ethylene. Ind. Eng. Chem. Res. 52, 9505-9514（2013）.

28） Chang, P.-L.; Wu, Y.-C.; Lai, S.-J.; Yen, F.-S. Size effects on χ- to α-Al2O3 phase transformation. J. Eur. Ceram. Soc. 29, 3341-3348（2009）.

29） Kloprogge, J.T.; Duong Lv Fau-Wood, B.J.; Wood Bj Fau-Frost, R.L.; Frost, R.L. XPS study of the major minerals in bauxite: gibbsite, bayerite and（pseudo-）boehmite. J. Colloid Interface Sci. 296, 572-576（2005）.

30） Sheng, Q.; Ling, K.; Li, Z.; Zhao, L., Effect of steam treatment on catalytic performance of HZSM-5 cata-lyst for ethanol dehydration to ethylene. Fuel Process. Technol. 110, 73-78（2013）.

31） Ciftci, A.; Varisli, D.; Cem Tokay, K.; Aslı Sezgi, N.; Dogu, T., Dimethyl ether, diethyl ether & ethylene

Catalytic Ethanol Dehydration to Ethylene over Nanocrystalline χ- and γ-Al2O3 Catalysts

J. Oleo Sci. 66, (9) 1029-1039 (2017)

1039

from alcohols over tungstophosphoric acid based mes-oporous catalysts. Chem. Eng. J. 207-208, 85-93（2012）.

32） Roca, F.F.; De Mourgues, L.; Trambouze, Y., Catalytic dehydration of ethanol over silica-alumina. J. Catal. 14, 107-113（1969）.

33） Matachowski, L.; Zimowska, M.; Mucha, D.; Machej, T., Ecofriendly production of ethylene by dehydration of ethanol over Ag3PW12O40 salt in nitrogen and air atmo-spheres. Appl. Catal., B: Environ. 123-124, 448-456（2012）.

34） Le Van Mao, R.; Nguyen, T. M.; Yao, J., Conversion of ethanol in aqueous solution over ZSM-5 zeolites: Influ-ence of Reaction Parameters and Catalyst Acidic Properties as Studied by Ammonia TPD Technique. Appl. Catal. 61, 161-173（1990）.

35） Rahmanian, A.; Ghaziaskar, H.S. Continuous dehydra-tion of ethanol to diethyl ether over aluminum phos-phate–hydroxyapatite catalyst under sub and super-critical condition. J. Supercrit. Fluids 78, 34-41（2013）.