Effect of synthesis and activation methods on the character of CoMo/ultrastable Y-zeolite catalysts

2.1 Material

NH4-Y zeolite (HSZ-300 series: 341NHA) was purchased from Tosoh Inc (Japan). Other materials in p.a grade were purchased from Merck (Germany), i.e., ammonium heptamolybdate, cobalt(ii) nitrate hexahydrate, and ammonia (25%).

2.2 Catalyst preparation and characterization

NH4-Y zeolite was calcinated at 550°C under N₂ for 3 h to obtain H-USY. It was followed by the impregnation of Co and Mo.

An amount of H-USY and ammonia 25% were soaked on 0.111 M of [(NH₄)₆Mo₇O₂₄·4H₂O] solution and refluxed for 2 h at 60°C. Then, the solution was filtered, and the residue was added to 0.053 M of [Co(NO₃)₂·6H₂O] solution. The mixture was refluxed again for 2 h at 60°C, then filtered and dried with a rotary evaporator at 200 mbar and 48°C, until obtained powder product. The sequential impregnation process with the order such as Mo followed by Co was denoted as “Ap.” The powder was activated by calcination for 3 h at 550°C under N₂ gas, and the sample obtained in this step was, namely, “Ac.” The next step, the reduction process, was used on some amount of calcinated powder. It was conducted under the H₂ gas for 2 h at 400°C and labeled as “Ar.”

The impregnation process of metals with reversed order, Co followed by Mo, was denoted as “Bp.” The method refers to “Ac” and “Ar” was repeated for Bp and labeled as Bc and Br, respectively.

The last method, called “Cp,” was prepared using the co-impregnation method. An amount of H-USY was added with ammonia and then immersed in 0.111 M of [(NH₄)₆Mo₇O₂₄·4H₂O] solution simultaneously with 0.053 M of [Co(NO₃)₂·6H₂O). Then, the mixture is refluxed, filtered, and dried. Following the previous calcination and reduction process, the calcined and reduced powder were labeled as Cc and Cr, respectively. All of the activated products are kept in the desiccator.

X-ray diffraction (Philips X’Pert) was used to analyze the crystallinity and phase-type of the catalyst. The functional group’s properties were analyzed using a Shimadzu FT-IR 8201 PC. While the quantity of total acidity of the catalyst was analyzed with the gravimetry method using ammonia as probe molecules [8,9]. The surface area was analyzed using SAA NOVA 1200e. The catalyst morphology was analyzed using SEM JSM-6510 (accelerating voltage of 6 kV) and JEOL/EO JEM 1400 ver 1.0 TEM (accelerating voltage of 120 V). The sample was dispersed on solvent before the TEM observation. The analytical methods for catalyst characterization were performed according to the procedure followed in our previous publication [10,11].

1. Ethical approval: The conducted research is not related to either human or animal use.

3.1 The characterization of the catalyst

The diffraction pattern analysis is shown in Figure 1a–c. Refinement of the catalyst was carried out by fitting with the standard of Zeolite Y (ICSD #31542), cobalt(ii) oxide (ICSD #9865), molybdenum(vi) oxide (ICSD #36167), cobalt(ii) molybdenum(vi) oxide (ICSD #281235), cobalt metal (ICSD #41507), molybdenum metal (ICSD #173127), and an alloy of cobalt molybdenum (ICSD #624215).

Figure 1

(a) Diffraction patterns of Y zeolite ICSD #31542, and catalyst sample that was produced by impregnation Molybdenum followed by Cobalt precursor after (Ap) impregnation, (Ac) calcination, and (Ar) reduction. (b) Diffraction patterns of Y zeolite ICSD #31542, and catalyst sample that was produced by impregnation Cobalt followed by Molybdenum precursor after (Bp) impregnation, (Bc) calcination, (Br) reduction. (c) Diffraction patterns of (A) Standard Y zeolite ICSD #31542, and catalyst sample that was produced by co impregnation methods after (Cp) impregnation, (Cc) calcination, (Cr) reduction.

Figure 1a–c shows that the catalyst has similar diffraction patterns in general but differs only in intensity. The addition of Co and Mo increases the intensity of the observed peak (*), which is 2θ (°) of 10.2°, 15.8°, and 23.8°. These identical diffraction patterns indicate no structural changes in USY with the metal addition. It can be concluded that the impregnation methods were successful, and the structure of the catalysts was not collapsed.

The correspondence between XRD diffraction and refinement of Inorganic Crystal Structure Database (ICSD) was determined by Rietica software using Le Bail methods. The refinement molar weight percentage of particles in the catalyst has resulted in the analysis (Table 1).

The acidity strength of alumina–silica such as zeolite can be studied using FTIR spectroscopy. Theoretically, the typical absorption band of the hydroxyl group – stretching mode indicated the presence of the Brønsted acid site [12,13]. Zeolites have a firm acidity when the bridging OH bond is weakened. It is indicated by the reduced frequency of OH stretching.

Figure 2 shows the OH_stretch by the wavenumber of 3408 cm⁻¹, obtained from the Brønsted acid site USY [12]. According to Niwa et al., the most strongest Brønsted acid site of the zeolite was obtained at the wavenumber 3595 cm⁻¹ [13]. The vibrational shift to lower wavenumber (from 3595 to 3408 cm⁻¹) indicates an increase in the strength of OH Brønsted caused by the synergistic influence of the Brønsted and Lewis acid sites [14].

Figure 2

FTIR spectra of catalyst after impregnation process.

Figure 2 also presented the comparison FTIR spectra of USY and labeled catalysts as Ap, Bp, and Cp. The spectra of Ap, Bp, and Cp show absorption at 3100 and 1400 cm⁻¹, which indicated a bond stretch and a bend of the NH from the amine group. The appearance of the new absorption is due to the interaction between the ammonia precursors and USY. Two bands are observed around 700–800 cm⁻¹, known as the external Si/Al–O.

The appearance of the band at 1650–1600 cm⁻¹ indicates the vibration OH_bend of the adsorbed water [15]. The absorption at the wavenumber 1140–995 cm⁻¹ is derived from the Si–O–Si (Al)_stretch bond [16,17], while the uptake of the internal Si (Al)–O functional group absorption appears at 420–500 cm⁻¹.

The spectra’s redshifts or blueshifts indicate the bond strength change due to the interaction between loaded metals and the USY [14]. The redshift is a peak shift toward a larger wavelength (small wave number) indicating weak bond strength, as the opposite of the blueshift.

In comparison to the absorption spectra of the NH stretch of Ap, Figure 2 confirms a redshift and blueshift for Cp and Bp, respectively. The redshift on Cp indicates a more robust interaction of the Co and O framework than the others. The functional group of NH_bend also shows a firmly bound on the Cp, in contrast with the weakly attached of Co on the Ap and Bp. Figure 2 demonstrates a blueshift of the OH’s absorption on the Cp and redshift on the Ac dan Bc, affirming a more superior bonding of OH groups on the Cp than on the Ac and Bc.

The OH_bend functional group of Bp has a strong bonding, while the Ap is weak. This weak bonding occurs because the CoO is bound to the Oxygen atom of the OH functional group. As a consequence, the bonding between O and H (OH bend) is weakened.

The bonds of Si (Al)–O in the Cp catalyst’s internal and external frameworks are weaker than the Bp catalyst. The Co, which binds to the O via its lone pair electrons, disturbs the framework’s strength. The strength of the Si (Al)–O internal and external bonds of the Ap catalyst is equal to the Bp catalysts, confirmed by the similar pattern of the obtained Si (Al)–O.

The spectra of the catalysts after the calcination processes (Ac, Bc, and Cc) are presented in Figure 3. The spectra’s difference before and after calcination can indicate the catalyst functional group’s changes, for example, a loss of the amine group. However, the shift in the wavenumbers, referring to the stronger or weaker bond, also contributes to it. The comparison is illustrated in Figures 3 and 4, respectively.

Figure 3

FTIR spectra of catalysts after the calcination process.

Figure 4

FTIR spectra of catalysts after the reduction process.

The OH functional group’s stretch shows a blueshift for the Bc and a redshift for the Cc catalyst when compared with Ac and USY (Figure 3). These phenomena were probably caused by the Mo and CoMo metals that are bonded to O atoms on USY. The functional group stretching on the Si–O–Si (Al) is obtained: a blueshift for the Ac catalyst and a redshift for the Bc catalyst. It indicates a strong interaction between metal and the functional group stretching on the Si–O–Si (Al) from the USY framework. The shift to the smaller wavenumber (redshift) for the external Si (Al)–O of Ac and Bc catalysts shows a decrease in bond strength when compared with the Cc. The blueshift wavenumber of Cc indicates that the metal MoO₃ has a strong interaction on the Si (Al)–O internal bond.

The next analysis compares the phenomenon after the reduction process for the catalyst with co-impregnation and sequential impregnation methods. The FTIR absorption spectra of the catalyst is depicted in Figure 4. The shift of the adsorption wave number from the FTIR spectra shows the effect of Co and Mo metals on the bond strength of the catalyst functional groups. The OH bend on the catalyst Ar obtained a redshift when compared with Ac, indicating a decrease in the bond strength of OH. While the blueshift for the Br and Cr confirming a stronger bond. The adsorption of water molecules on the catalyst causes weakening of the OH bonds functional groups

The shift of the wavenumber of Si–O–Si (Al) stretching refers to the blueshift for Ar and Cr, indicating a stronger bond. Meanwhile, the redshift in Br indicates a weak bond, probably a consequence of the strong interaction between metal and USY. The external Si (Al)–O showed a blueshift in Ar, revealing an increase in the bond strength. For Cr, the redshift indicates a weak bond from the interaction between O and Mo metal, following Anderson statements [18]. The bond strength of OH is affected by the interaction between Mo and Oxygen atoms on USY. Robust interaction between Mo and O causes the weakness of the OH-catalyst bond. In contrast, for Br, the solid bond formation of the internal functional group is observed by the blueshift of the Si (Al)–O internal.

Generally, the OH groups of all the catalysts underwent a blueshift compared to USY (Figure 4). Except for the Ar and Br, the OH wavenumber shifted to a weaker bond (redshift) than in USY. Probably, it is caused by the interaction of water molecules with the hydrophilic groups. The blueshift of Cr indicates that such interaction did not occur.

The Si–O–Si (Al) functional groups on the Ar and Cr catalysts have a stronger bond than USY. Contradictive to Br, with the smallest shift on wavenumbers, the Br catalyst’s framework structure has the weakest bond. The weakness of the framework is contributed by the strong interaction of loaded metal and the USY. For a similar reason, the shift of the wavenumber of Si (Al)–O external from Ar, Br, and Cr was smaller than USY. The Si (Al)–O internal cluster in contrast gained a blueshift when compared with USY.

The catalyst after reduction (Figure 4) has the same phenomenon as the catalyst after calcination, i.e., the OH stretch undergoes a blueshift for the Cr catalyst and a redshift for the Br catalyst. The cluster of OH bend undergoes a blueshift for the Cr catalyst and a redshift for the Ar catalyst. This phenomenon was the same as that which occurs on the Cp catalyst, where Lewis sites of Co metals accept free-electron pairs from the O atoms at OH, which causes a weak bond between OH stretch and OH bend.

The Si–O–Si (Al) stretch has a strong bond for the Ar catalyst and a weak bond for the Br catalyst. It is probably due to the weak bonds that are formed by the OH bend of Si (Al)–O internally. Meanwhile, the internal functional groups (Al)–O for Ar and Br obtained a weak bond, and the Cr catalyst obtained a strong bond. The occurrences were allowed because on the Ar and Br catalysts, the presence of Co metals was interfering with the internal Si (Al)–O bond.

The results of total acidity analysis and the specific surface area of the catalyst are listed in Table 2. It was found that the total acidity and specific surface area of the catalysts increase after the calcination process, which indicates that this process is sufficient to remove impurities from the catalyst surface. Further effects include the opening of pores and increasing active sites that are characterized by the absorption of ammonia. An increase in acidity supports the FTIR spectral data, where the absorption of wavenumber in the increasingly large Si–O–Si (Al) stretch functional group due to loss of Al³⁺ cations after calcination indicates that after the calcining process the catalysts are increasingly acidic [19].

After the reduction process, the total acidity and specific surface area of the Ar catalyst are decreased. The closure of the pore and active sites by the metal is responsible for reducing the acidity. The specific surface area of the Br catalyst is equal to Bp, but its total acidity was decreased. The decrease in the intensity of the Si–O–Si (Al) stretching relates to the acidity’s weakness. In contrast, the total acidity of Cr increases, but its specific surface area is decreased. The smaller of the specific surface area can be occurred by the loss of CoO and Mo in the calcination process. Another possibility is pore blockage by CoMoO₄ or CoMo formation on the surface. The presence of acidic CoMo particles enhances the total acidity. The acidity value of all catalysts listed in Table 2 is higher than that of the previous research [20–22] but lower than the results of Anggoro and Co-worker’s study [23]; it is because the addition of metal in this research was much lower than in the Anggoro and Co-worker’s study [23].

The presence of impregnated metals in the catalyst can be observed morphologically by using SEM and TEM. Besides, SEM and TEM can also show the size of a particle in a sample. SEM analysis of USY and catalyst samples after reduction is shown in Figure 5.

Figure 5

Morphology of (a) USY; catalyst after reduction process that produces by (b) co impregnation methods (Ar); sequential impregnation methods where (c) cobalt was impregnated firstly followed by Molybdenum precursor (Br), and vice versa (d) molybdenum was impregnated first followed by cobalt precursor.

In morphology at 10,000× magnification, it can be seen that after the addition of Co and Mo metal, there is a change in the particle size of the catalyst. Figure 5 shows that the Ar (Figure 5b), Br (Figure 5c), and Cr (Figure 5d) catalysts contain heterogeneous particles, whereas USY (Figure 5a) has the size of homogeneous particles. The change in morphology indicates that the metal is impregnated successfully in USY. Also, Figure 5 shows that the Cr catalyst is dominated by particles smaller than those of the Ar and Br catalysts.

Based on Figure 6, it was found that USY has a particle size with a dominant value at 0.61 μm. The Ar catalyst has a small particle size, about 0.21–1.21 μm, with a predominant particle size of 0.41 μm. While Br and Cr catalysts have heterogeneous particle sizes, according to the results of XRD pattern analysis that both types of catalysts contain various types of metal phases, such as pure metals, metal oxides, and metal alloys.

Figure 6

Relationship of average particle size and percentage of particles amount.

The result of characterization using SEM was also analyzed in terms of the dispersive energy of X-ray spectroscopy (EDX). EDX analysis was conducted to determine the percentage weight of the mass of particles contained in Ar, Br, and Cr catalysts. The mass percentage data of the particles is listed in Table 3.

Table 3 lists the obtained elements of O, Al, Si, Co, and Mo in catalysts in detail. The dominant mass percentage of O, Al, and Si elements indicates that the USY composition in the catalyst is more than Co and Mo. The Si/Al ratio is obtained based on the mass percentage ratio of Si and Al, which shows that the Si/Al ratio of all catalysts is equivalent to 3:1. The higher Si/Al ratio suggests that the catalyst has a high degree of acidity. Anderson and coworker’s research showed that the addition of Mo and Co metals on USY, followed by the calcination, oxidation, and reduction process, lead to the ratio was reduced to 2:1 [18]. This research concludes that catalysts with the activation using calcination and reduction process without oxidation step produce a better Si/Al ratio.

The morphological analysis of metals was done in detail using TEM. The shape, type, and composition of metals on USY were figured out. The first phase of analysis using TEM was done on USY samples. The data of the USY sample analysis using TEM is presented in Figure 7. Based on the TEM USY micrograph analysis shown in Figure 7a, the physical form of USY crystals was obtained. Meanwhile, the analysis using EDX (Figure 7b) shows that the sample has an elemental composition of 40.44 wt% oxygen (0.525 keV), 13.17 wt% alumina (1.486 keV), and 46.40 wt% silica (1.739 keV). The element composition is equivalent to the composition of Al₂O₃ and SiO₂, respectively, 22.11 and 77.89 wt%. From further analysis of the diffraction pattern (7C), it was found that the sample had basal spacing (d) that corresponded to the lattice plane 111, 311, 511, 622, and 822 of faujasite. Both the micrograph and the diffraction pattern of the USY sample showed clean crystals, without any impurities.

Figure 7

TEM Images of USY (a) with a magnification of 50,000×, (b) EDX spectra, and (c) diffraction pattern.

The results of the TEM image analysis are in accordance with the results of the analysis using XRD (Figure 1), which states that the carrier sample used is synthetic faujasite with type Y zeolite, which is thermally stable (USY). The absence of impurities apart from being supported by the results of analysis using XRD, the analysis is also supported by the results of the analysis using SEM (Figure 5a).

The distribution of particles over the catalyst can be determined by the white dots present in the catalyst (Figure 8). This analysis was performed on Br catalysts that were prepared with the sequential impregnation method shown in Figure 8. In Figure 8, the white dots show the morphology, which indicate the presence of particles or metal impregnated on the USY. There is also a channel showing the presence of pore channels in the catalyst.

Figure 8

TEM Image of Br catalyst with magnification on (a) 40,000×, (b) 150,000×, and (c) diffraction pattern.

As shown in Figure 8, the results of the analysis showed some dots on USY as same as in a red square. It causes a change in the morphology of USY. The dots are thought to come from particles embedded in the USY pore cavity. The types of particles carried on USY were analyzed using the plotting method between basal spacing data from the diffraction patterns of TEM, JCPDS, and XRD data. It was found that the diffraction pattern (Figure 8c) corresponds to the basal spacing, d, of 8.692, 4.669, and 3.275 Å, which are equivalent to the presence of the d₂₂₀, d₅₁₁, and d₆₄₂ of H-FAU.

USY, when exposed by Co and Mo metals, also showed a diffraction pattern corresponding to the basal spacing, d for metal Co (JCPDS #05-0727) at 1.971 and 1.811 Å, for metal Mo (JCPDS #42-1120) at 2.228 and 1.575 Å, and for CoMo4 (JCPDS #28-1235) at 3.054 and 2.511 Å. The results of the analysis indicated that Co and Mo particles were successfully embedded in USY as in the form of metals with zero oxidation numbers and metal oxides.