Catalytic Hydrogenation Performance of Methyl Isobutyl Ketone over Ni/γ-Al2O3 Ca

Liu Lixia; Liao Tao; Jin Haibo; He Guangxiang; Yang Suohe; Guo Xiaoyan; Luo Guohua

(Beijing Key Laboratory of Fuels Cleaning and Advanced Catalytic Emission Reduction Technology, School of Chemical Engineering, Beijing Institute of Petrochemical Technology, Beijing 102617)

Abstract: Supported nickel-based catalysts were prepared by the incipient wetness impregnation method for the selective hydrogenation of methyl isobutyl ketone to methyl isobutyl carbinol in a f ixed-bed reactor. The effects of the nickel source, Ni loading, calcination time, and calcination temperature on the hydrogenation performance were studied. The experimental results showed that the Ni/γ-Al2O3 catalyst demonstrated the highest catalytic performance under the preparation conditions by using nickel nitrate as the nickel source with a NiO loading of 20%, followed by calcination at 440 °C for 5 h. In addition, this catalyst showed the largest specific surface area, best crystal structure, highest active component content, smallest particle size, and uniform distribution of NiO on the surface of the carrier. The nickel-based catalyst prepared using the optimized conditions exhibited a 96.1% conversion of methyl isobutyl ketone, with a methyl isobutyl carbinol selectivity of 99.6%. The described procedure is very effective for the preparation of methyl isobutyl carbinol using methyl isobutyl ketone as the feedstock.

Key words: Ni/γ-Al2O3 catalysts; hydrogenation; methyl isobutyl ketone (BIBK); methyl isobutyl carbinol (MIBC)

1 Introduction

Methyl isobutyl carbinol (MIBC), which is a minor byproduct of methyl isobutyl ketone (MIBK) synthesis, is a middle-boiling solvent with good industrial application prospect. MIBC is widely used for manufacture of pesticides, dyes, pharmaceuticals, adhesives, synthetic resins, coatings, and mineral processing, as well as other industrial products. At present, the major method for preparation of MIBC is the catalytic hydrogenation of mesityl oxide (MO), but this reaction is carried out in a small batch reactor which cannot meet the requirements of continuous mass production in industry.

Hydrogenation processes are commonly applied to reduce or saturate the double bond of a molecule by adding pairs of hydrogen atoms[1]and have attracted an increasing research interest[2-3]. Various noble metal (Pt[4-6], Pd[7-9], Rh[10-12], Ag[13-15], and Ru[16-18]) catalysts are used in hydrogenation reactions to obtain high conversion and selectivity at room temperature[19-20]. However, the high price of noble metal catalysts leads to increased production costs compared to Ni catalysts; hence, there is a demand for the adoption of nickel-based catalysts[21-25]. Supported nickel catalysts are one of the most common nickel-based catalysts, and are composed of the active metal Ni, carriers, and additives. These catalyst carriers include Al2O3, SiO2, MgO, ZrO2, and CeO2, while the metal additives include Cu, Mn, Ti, Mo, Co, and other elements. The active catalyst is located at the surface of the carrier in supported catalysts, which can contribute to increased contact between the active component (Ni) and the reactant molecules, thus improving the efficiency of the reaction. As the reaction proceeds, the active components also undergo sintering and loss, so that the catalytic activity is reduced and the service life of the catalyst is shortened. Therefore, the selection of the catalyst carrier and the preparation method is very important to improve the reactivity of the catalyst and prolong its service life. Alumina has been proven to be the best catalyst support for Ni, as compared to some other carriers, because the surface of Al3+ions has much stronger bonding ability, and the metal-support interaction between the Ni and the alumina carrier is stronger, which also can improve the stability of the nickel particles[26]. In most studies, the strength of the interaction between NiO and γ-Al2O3and the existence form of nickel is generally considered to be related with the nickel content and calcination temperature. Kanget, et al. found that the activity of Ni/Al2O3increases with an increasing Ni content; on the other hand, when the amount of Ni is increased, the Ni particles will accumulate on the surface of the carrier, which is not conducive to its high catalytic activity[27]. The activity of a series of Ni/Al2O3catalysts prepared at different calcination temperatures was also studied by Wang, et al.[28], who found that the catalyst activity was the highest when a calcination temperature of 400 °C was used, but the influence of the calcination temperature on the catalyst performance differed depending on the carrier species.

In this study, the effect of the preparation conditions on the activity of Ni/Al2O3was studied, and the best preparation method for the selective hydrogenation of methyl isobutyl ketone was obtained through reactions conducted in a f ixedbed reactor. Moreover, the catalysts were characterized by X-ray diffraction (XRD), BET measurements, and temperature-programmed reduction (TPR) technique.

2 Experimental

2.1 Preparation of the catalysts

The Ni/γ-Al2O3catalysts were prepared by the impregnation method. A required amount of nickel salt was dissolved in distilled water, and then the γ-Al2O3carrier was added to the nickel nitrate and nickel acetate solution separately, followed by stirring in a water bath at 60 °C for 3 h. The precursor was dried at 120 °C for 3 h. The samples were finally calcined at different temperatures and reduced by H2.

2.2 Catalytic tests

The liquid-phase hydrogenation of methyl isobutyl ketone to methyl isobutyl carbinol was tested in a stainlesssteel f ixed-bed reactor with an inner diameter of 22 mm. 10 mL of the catalyst sample (40―60 mesh) were loaded in the constant temperature section of the reactor, and the other parts of the reactor were f illed with inert f illers composed ofinert ceramic balls. The catalysts were reduced in pure H2gas (at a flow rate of 50 mL/min) at 350 °C for 3 h. The temperature was cooled to 170 °C (the reaction temperature), at which the hydrogenation reaction proceeded. The volume ratio of hydrogen to MIBK in the liquid fraction was 400:1, the H2pressure was 4.0 MPa, and the liquid hourly space velocity was 0.80 h-1. Catalytic reaction took place in the kinetic region according to the kinetic modeling of the liquid-phase hydrogenation over

2.3 Analytical methods

Qualitative analysis of the liquid products was conducted using a GC-14C gas chromatograph equipped with a crosslinked capillary HP-5 column (50 m×0.22 mm×0.33 µm) and a FID detector using nitrogen as the carrier gas. The injection port temperature was 220 °C, and the detector temperature was 220 °C.

2.4 Catalyst characterization

The powder X-ray diffraction patterns (XRD) of the Ni/γ-Al2O3catalysts were recorded using a D/max-IIIA diffractometer (Rigaku, Japan) with CuKα (λ=0.1548 nm) radiation in the 2θ range of between 10°―80°, at 40 kV and 100 mA. The grain size of the catalyst was analyzed by XRD and calculated using the Debye-Scherrer equation[29]. H2-TPR detection of the catalyst was carried out in a Micromeritics Chemisorb 2750 system equipped with a thermal conductivity detector (TCD). The volume, surface area, aperture size, and nitrogen adsorptiondesorption curve of the catalyst could be measured by the Autosorb-1 instrument using the BET method.

3 Results and Discussion

3.1 Effect of hydrogen-ketone ratio on the Hydrogenation reaction

The hydrogen/oil volume ratio is also a key factor affecting the activity and stability of catalysts. A suitable hydrogen/ketone ratio is benef icial to the improvement of the MIBK conversion rate. High hydrogen/ketone volume ratio means the increase of hydrogen partial pressure, which is benef icial to reducing the temperature of catalyst bed and inhibiting the production of carbon deposits. Small hydrogen-ketone volume ratio can affect the reaction in the process of heat transfer and mass transfer efficiency. Under the reaction conditions covering a temperature of 170 °C, a pressure of 4.0 MPa, and a liquid space velocity of 1.0 h-1, the effect of different hydrogen/ketone volume ratio on the conversion of MIBK is shown in Figure 1. It can be seen from Figure 1 that when the hydrogen/ketone volume ratio is 400, the conversion rate of MIBK (expressed byX(%)) has a maximum value, with the MIBC selectivity equating to 99.9%, so the hydrogen/ketone volume ratio is selected as 400.

Figure 1 Effect of the hydrogen/ketone ratio on conversion of MIBK

3.2 Effect of different precursors on the Ni/γ-Al2O3 catalysts

The catalyst precursors are the source of the active components in the catalyst, so the selection of the precursor has an important influence on the hydrogenation activity of the catalyst. The interaction between the nickel species and the carrier would also show tremendous differences due to the different anions, resulting in different NiO dispersion behavior; then NiO was loaded onto the catalyst with different loading amount. After reduction by hydrogen, NiO was reduced to active component Ni species. Thus, choosing a suitable source of nickel is beneficial to the hydrogenation performance of the catalyst. Figure 2 shows the XRD patterns of Ni-N and Ni-Ac. For both catalysts, the characteristic diffraction peaks of NiO appeared at 2θ= 37.2°,43.4°, and 62.9°, but the width and intensity of the NiO peaks differed depending on the nickel source of the catalyst. The peak assigned to NiO in Ni-Ac was sharper, and its strength was greater than that of Ni-N. The dispersion capacity of nickel acetate is lower than that of nickel nitrate. The reason can be attributed to the larger size of nickel acetate than nickel nitrate.

Figure 2 XRD patterns of the catalysts prepared from different nickel precursors

The smaller specific surface area of the Ni-Ac catalyst can also be explained by the fact that the surface of the active oxide particles is larger. Figure 3 shows the results of the evaluation of the two catalysts on the hydrogenation performance. When the reaction was carried out for 10 h or more, the conversion rate of the reaction remained basically stable. The conversion of methyl isobutyl ketone reached 90% over the Ni-N catalyst. The formation of NiO species in Ni-Ac catalysts and Ni-N catalysts could be explained by the different ability of solutions of nickel nitrate and nickel acetate to disperse in the Al2O3support[30]. The hydrogenation activity of Ni-Ac was lower than that of Ni-N, so nickel nitrate was usually chosen as the nickel source that would be more conducive to the hydrogenation reaction comprehensively.

Figure 3 Hydrogenation performance using the catalysts prepared from different nickel sources

3.3 Effect of theoretical NiO content on Ni/γ-Al2O3 catalysts

The theoretical NiO content plays one of the most important roles for highly active and selective Ni/γ-Al2O3catalyst. In this experiment, Ni(NO3)2·6H2O was used as the catalyst precursor, the calcination temperature was 480 °C, and the calcination time was 4 h. Then the Ni/γ-Al2O3catalysts loaded with differing NiO mass content, which was equal to 5%, 10%, 15%, 20%, and 25%, respectively, were obtained.

Table 1 shows the results of BET characterization of the Ni/γ-Al2O3catalysts with different NiO loadings. At first, with the gradual increase in the catalyst loading, the active component content increased, and the specific surface area began to decrease due to a small amount of NiO particles that blocked the γ-Al2O3carrier and led to a decrease in the microporous pore volume of the catalyst. However, as the NiO content was further increased, the active NiO particles would possibly form a part of the pores to increase the porous structure of Ni/γ-Al2O3catalyst.

Table 1 Texture properties of the Ni/γ-Al2O3 with various theoretical NiO contents

Figure 4 shows the results of the H2-TPR curves of the Ni/γ-Al2O3catalyst samples with various theoretical NiO contents. All of the hydrogen consumption peaks could be divided into three temperature ranges, viz.: (1) 300―500 °C, (2) 500―750 °C, and (3) 750―900 °C. When the NiO loading content was less than 10%, the hydrogen consumption peak in the temperature region (1) should belong to the α-NiO, which was free NiO with weak interactions with the γ-Al2O3[27]. In the temperature region (2), there was only one major hydrogen consumption peak, which belonged to β1-NiO and was stronger than that of α-NiO. As the catalyst loading was further increased, two obvious hydrogen consumption peaks were present in the temperature range (2); the hydrogen consumption peak on the left still corresponded to β1-NiO, while the hydrogen consumption peak on the right was attributed to β2-NiO. The interaction between β2-NiO and the carrier was slightly stronger than that of β1-NiO. With an increasing catalyst loading, the reduction peak of NiO had a tendency to move towards the low temperature region, and the area of the reduction peak increased. When the NiO loading content was increased to 25%, there was an obvious α-NiO reduction peak in the temperature range (1). This fact demonstrates that the active component of the catalyst surface was less stable at high NiO loading. The α-NiO was further reduced to Ni0particles, and these particles interacted weakly with the γ-Al2O3and were more prone to loss. The peak at around the temperature range (3) was ascribed to NiAl2O4(γ-NiO), known to be the most inert phase of nickel oxide for hydrogenation catalysis[30].

Figure 4 The H2-TPR curves of the Ni/γ-Al2O3 catalysts with various theoretical NiO contents

The results of a series of hydrogenation experiments are shown in Figure 5. It can be seen that the catalysts with various Ni contents showed notable differences in activity. When the NiO loading amount was 5%, the reaction conversion rate was the lowest. With an increasing NiO loading, the conversion rate of methyl isobutyl ketone increased. When the NiO content was 25%, the catalyst was reduced by hydrogen gas at high temperature. There were many large particles of Ni0that could only interact weakly with γ-Al2O3, which would lead to sintering and easy loss of the NiO particles. Therefore, a 20% of NiO was found to be the best nickel loading of the catalysts.

Figure 5 Catalytic performance of the Ni/γ-Al2O3 catalysts with various NiO theoretical contents

3.4 Effect of calcination time on the Ni/γ-Al2O3 catalysts

A suitable calcination time is an important parameter to optimize the catalytic activity of the catalyst. All samples were calcined at 480 °C over different lengths of time. In order to investigate the influence of different calcination time on the specific surface area and pore structure of the catalyst, the BET method was used to characterize the catalyst, with the results shown in Table 2. With an increasing calcination time, the specific surface area of the Ni/γ-Al2O3catalysts f irst showed an increasing trend, and then decreased, while the pore size of the catalyst and pore volume gradually decreased. These results showed that with the increase in calcination time, the Ni particles in the catalyst could increase in size and would lead to blockage of the catalyst.

Table 2 BET textural properties of the Ni/γ-Al2O3 catalysts obtained at different calcination time

It can be seen from the H2-TPR results of the catalysts calcined at different times (Figure 6) that with an increasing calcination time, the reduction peak area f irst decreased and then increased. The reduction peak area was the greatest and the catalyst itself contained the most active Ni element upon being subjected to calcination for 5 h. Figure 7 shows the catalytic performance of catalyst samples which were treated with different calcination times. With an increasing calcination time, the conversion of methyl isobutyl ketone first increased and then decreased. The NiO dispersion over the Al2O3supports can be improved by increasing the calcination time to some extent. Thermal treatment for a long time, 6 h, is possible to cause sintering, since the thermodynamic conditions required for this undesired change are provided. A highest conversion rate of about 95% was obtained at a calcination time of 5 h, and the selectivity toward MIBC remained at 99%. These data and the XRD and H2-TPR spectra demonstrated that a calcination time of 5 h is the ideal one.

Figure 6 The H2-TPR results of the catalysts with different calcination times

Figure 7 Catalytic performance of the catalysts with different calcination times

3.5 Effect of calcination temperature on the Ni/γ-Al2O3 catalysts

Using the optimized nickel loading (20%) and calcination time (5 h), the effects of the calcination temperature were measured. The calcination temperature had an important influence on the pore structure and specific surface area of the catalyst. Table 3 shows the BET characterization results for the Ni/γ-Al2O3catalysts obtained at different calcination temperatures. The reason for this phenomenon may be that the catalyst particles became larger and the active components were sintered when the calcination temperature of the catalyst was increased.

Table 3 BET textural properties of the Ni/γ-Al2O3 catalysts obtained at different calcination temperatures

Figure 8 shows the H2-TPR results of the catalysts obtained at different calcination temperatures. The reducible NiO species are usually divided into four types, viz.: α,β1,β2, and γ[31]. When the calcination temperature was 400 °C, the catalyst contained mainly α-NiO, β1-NiO, and β2-NiO. As the calcination temperature was further increased, the α-NiO related hydrogen consumption peak disappeared and the β1-NiO and β2-NiO peak areas increased. The thermal treatment at 440 °C showed that some of the α type transformed to the β1type which favored the hydrogenation reaction. With an increasing calcination temperature, the reduction peak area in the low temperature region gradually decreased, while the reduction peak area at the high temperature region gradually increased. When the calcination temperature reached 520 °C, the content of β1-NiO was less than the β2-NiO content, because β2-NiO had stronger interaction than β1-NiO, which indicated that the interaction between NiO particles could be enhanced by the increase in calcination temperature. When the calcination temperature reached up to 560 °C, the result of H2-TPR analysis showed that there apparently existed only the peak of γ-NiO. As the most inert phase of nickel oxide, the catalytic activity of γ-NiO would decrease obviously[31-32].Figure 9 shows the catalytic performance of catalysts treated at different calcination temperatures. As the calcination temperature was increased, the conversion of methyl isobutyl ketone f irst increased, and at a calcination temperature of 440 °C, the conversion of methyl isobutyl ketone reached about 96.1%. As the calcination temperature was further increased, the conversion of methyl isobutyl ketone began to decline, reaching 70% at a calcination temperature of 560 °C. The main factors that caused this phenomenon might be related with the following fact: the active Ni component was produced by the catalyst calcined at high temperature, and it showed a weaker interaction with the carrier, resulting in the declining activity of the catalyst. Therefore, the combination of these data and the XRD and H2-TPR results showed that the optimum calcination temperature should be 440 °C.

Figure 8 The H2-TPR results of the catalysts for different calcination temperatures

Figure 9 Catalytic performance of catalyst obtained at different calcination temperatures

4 Conclusions

A series of Ni/γ-Al2O3catalysts were prepared for the liquid-phase hydrogenation of MIBK to MIBC, and the conditions for the catalysts preparation were found to have an important influence on their catalytic activity. The characterization study using H2-TPR and XRD analyses showed that the obtained Ni/γ-Al2O3catalyst demonstrated the best activity when the precursor was prepared using nickel nitrate with a NiO loading amount of 20% at a calcination temperature of 440 °C and over a calcination time of 5 h. A selectivity of 99.6% toward methyl isobutyl carbinol (MIBC) and a methyl isobutyl ketone (MIBK) conversion rate of 96.1% were obtained in a fixed-bed reactor for hydrogenation of methyl isobutyl ketone conducted under conditions covering a reaction time of 22 h, a reaction temperature of 170 °C, a H2f low rate of 50 mL/min, a hydrogen to ketone volume ratio of 400:1, and a liquid space velocity of 0.8 h-1. Therefore, the procedure reported in this study represents a good strategy for the hydrogenation of MIBK to MIBC in industrial production. Compared with the traditional industrial process, this continuous production mode can greatly simplify the production technology of MIBC, which not only can gain a higher quality of methyl isobutyl carbinol, and also can reduce the production cost at the enterprise. This technology can greatly improve the production efficiency in industrial practice.

Acknowledgement:This work is f inancially supported by the National Natural Science Foundation of China (91634101) and the Project on Construction of Innovative Teams and Teacher Career Development for Universities and Colleges under Beijing Municipality (IDHT20180508).