Elsevier

Molecular Catalysis

Volume 533, December 2022, 112796
Molecular Catalysis

The role of Zn in the Cu-Zn-Al mixed oxide catalyst and its effect on glycerol hydrogenolysis

https://doi.org/10.1016/j.mcat.2022.112796Get rights and content

Highlights

  • Glycerol hydrogenolysis to propylene glycol on reduced Cu-Zn-Al oxides was studied.

  • Zn presence influenced Cu particles size as well as their surface.

  • Zn effect had its maximum at the molar ratio of Zn/(Cu+Al) = 0.35.

  • Excess of Zn led to the formation of ZnO and deterioration of Cu surface.

  • Acidic centres were not primarily important for the reaction mechanism.

Abstract

A series of Cu-Zn-Al oxide catalysts reduced in H2 was tested in glycerol hydrogenolysis in a stirred batch autoclave at 230°C. The catalysts were prepared by the precipitation – calcination method from layered double hydroxides (LDH) structures. The oxide forms of the catalysts were obtained by calcination of these structures at programmed temperatures, up to 350°C. The catalysts were characterized by the following methods: X-ray diffraction (XRD), temperature-programmed reduction (TPR), thermogravimetric analyses (TGA), N2-physisorption and temperature programmed desorption of ammonia (NH3-TPD). It was found that Zn plays acts as a structure modifier, influencing Cu dispersion and particle size. Consequently, Zn has an impact on the catalyst activity. In particular, the Zn effect had its maximum at the concentration corresponding to a ratio of Zn/(Cu+Al) = 0.35 where conversion of 65% and the selectivity to 1,2-propanediol of 70% was achieved after 60 min of reaction time.

Introduction

The production of 1,2-propanediol (propylene glycol) via selective dehydroxylation (hydrogenolysis) of glycerol is a good example and a prospective possibility for the substitution of traditional petrochemical raw materials for renewables. In recent decades, glycerol has been produced in large quantities as a byproduct in fatty acid methyl ester (FAME) production [1]. Its utilization as a substitute for propylene in propylene glycol production is a promising way of utilizing renewables in the petrochemical industry [1].

Hydrogenolysis of glycerol is a reaction of glycerol with hydrogen usually performed in the liquid phase in the presence of a heterogeneous catalyst, although gas phase reactions have also been studied [2,3]. Many noble and other transition metals have been described as being effective as catalysts in the selective hydrogenolysis of glycerol, where Cu based catalysts represent a cheaper alternative to noble metals with comparable effectiveness [2], [3], [4]. Cu containing catalysts are usually prepared by coprecipitation [5], [6], [7], [8], [9], [10], although impregnated catalysts have also been studied [11,12]. Ƴ-alumina[11] or SiO2 [12] are most frequently used as carriers for the preparation of Cu impregnated catalysts, but the use of MgO for this purpose has also been described [7]. Different methods of coprecipitation have been described as effective in the preparation of Cu based catalysts; the oxalate gel–coprecipitation method [8] and coprecipitation with alkali (NaOH, KOH, Na2CO3, K2CO3) [8,[13], [14], [15], [16], [17], [18]] gare preferential preparation methods.

Under certain conditions, layered double hydroxides (LDH) i.e., hydrotalcite-like structures, are formed [13] by coprecipitation of a solution containing divalent and trivalent metal salts. pH solution control is very important for the layered double hydroxide structure to be formed [13,14] during precipitation. As in natural hydrotalcite, also in synthetic LDH, aluminium usually plays the role of trivalent metal, while different divalent metals can be used. Cu is the most important divalent metal used if hydrotalcite structures are prepared as intermediates in the preparation of oxide catalysts intended for hydrogenolysis [14,15]. The properties of the resulting catalysts can be improved by another divalent metal addition. The use of Zn [13,16], Mg, Ni, Co[13] or Fe [14] for this purpose has been described. The LDH containing Al, Cu and possibly another divalent metal(s) is calcined at high temperature, usually between 400 and 650°C [14], [15], [16], [17], to obtain mixed oxides which are characterized by small dimensions of primary particles [15]. The particles therefore have a large Cu surface area. After being activated, i.e., reduced by hydrogen at higher temperature and elevated pressure, these oxides provide active catalysts [13], [14], [15], [16], [17], [18], usually more active than the catalysts prepared from the oxides obtained by other preparation methods [6], [7], [8], [9], [10], [11]. Compared to other oxide mixtures [19], [20], [21], the mixed oxides from LDH precursors are more homogeneous, having a higher specific surface area, and individual metals are better mixed there, therefore a higher synergistic effect of metals is ensured.

In recent research, special attention has been given to the influence of Zn on the properties of hydrogenolysis catalysts prepared by coprecipitation via hydrotalcites. Cu/Al hydrotalcites with the addition of another metal, i.e., Zn. Co, Ni [13,14], Fe [13], Mg [14] or Ca [18] were studied as precursors for the preparation of catalysts for glycerol hydrogenolysis.Cu/Zn/Al catalysts prepared from hydrotalcites have been found to have both activity and selectivity higher than similar compositions in which another divalent metal mentioned above was used [13,14]. The Cu/Zn/Al catalyst was even active when crude glycerol from FAME production, in which remainders of KOH were neutralized, was used.

Although Zn has been described as an effective modifier of Cu based catalysts several times, no detailed study about its role in Cu/Zn/Al catalytic systems prepared from LDH structures has been published thus far. Zhang et al. [16] prepared a series of Cu/Zn/Al catalysts with different atomic ratios ranging from 1:3:1 to 3:1:1 and tested these catalysts in phenol oxidation. The Cu/Zn ratio had an impact on the internal surface area (BET) and the dispersion of Cu species on the catalyst surface, which was very important for the catalyst activity. The maximum Cu surface and activity were also achieved with the catalyst having a Cu/Zn ratio of 1:1.

Du et. al. [22], who studied the deactivation of Cu-ZnO catalysts, concluded that the Cu and ZnO crystallite morphology played an important role in the catalyst activity and that the aggregation of these particles caused deterioration of the catalyst activity.

A positive effect of Zn on the decrease in the dimensions of Cu primary particles and, at the same time, on the Cu surface increase has also been mentioned in other studies [9,23]. In these studies, an increase in the concentration of acidic sites is also emphasized as being a remarkable effect of zinc addition. However, the role of the surface acidity and its impact on the catalyst activity in glycerol hydrogenolysis have not been fully clarified. Acidic sites are necessary for the first step of the so-called dehydration-hydrogenation mechanism [24] dehydration of glycerol to acetol. On the other hand, they play no role in the dehydrogenation-dehydration-hydrogenation mechanism, which takes place in parallel with the dehydration route [24].

This study was performed with aim of clarifying and analysing in detail how zinc influences the properties and performance of catalysts for glycerol hydrogenolysis obtained by calcination of Cu/Zn/Al layered double hydroxide structures.

Section snippets

Catalyst preparation

The layered double hydroxides, which were intermediates for the mixed oxides, were synthesized by the coprecipitation method. The coprecipitation was carried out in a stirred glass precipitation reactor equipped with two pumps, a thermometer and a pH metre (Syrris Globe). The cation containing solution was prepared by dissolution of Cu, Zn, Al nitrates in deionized water (Cu(NO3)2 ֗ 3H2O; Al(NO3)3 ֗ 9H2O; Zn(NO3)2 ֗ 6H2O; Lach:ner; purity p.a.) . The total concentration of metals in the

ICP-OES

The real metal content in the prepared mixed oxides was determined by inductively coupled plasma atomic emission spectroscopy (ICP-OES - Agilent 725, USA). Before the analysis, 1 g of MO was mineralized with 0.1 g of p-toluenesulfonic acid at 500°C. Then the sample was dissolved in sulfuric acid and diluted with demineralized water. The mass fraction of the metals was converted to a molar fraction being used further for discussion.

XRD

X-ray diffraction (XRD) of powder samples (LDH, MO and aMO) was

Results and discussion

A series of CuZnAl hydrotalcite like compounds with different metal molar ratios were prepared by the coprecipitation method. It was found that the real molar ratio values (Table 1) were actually slightly different from what was expected based on the input mass balance. The real ratio values of divalent v.s. trivalent cations (Cu+II+Zn+II)/Al+III were approximately 8-20% higher than theoretically expected. It is however in agreement with the literature as divalent cations have better ability to

Conclusion

Zn proved to be an efficient modifier of the Cu based hydrogenolysis catalysts prepared by calcination of hydrotalcite-like structures. In a certain range of concentrations, Zn can increase the catalyst activity. It was disclosed that its presence in the catalyst compositions influences the Cu particle size as well as their surface. Zn thus plays a role of a structure modifier which is in principle the reason for its effect on the catalyst activity.

However, only a small amount of Zn, which

Declaration of Competing Interest

The authors declare the following financial interests/personal relationships which may be considered as potential competing interests:

Acknowledgments

The project No. TN01000048 has been financially supported by TA CR. The result was achieved using the infrastructure included in the project Efficient Use of Energy Resources Using Catalytic Processes (LM2018119) which has been financially supported by MEYS within the targeted support of large infrastructures. This work was supported by the University of Pardubice, Czech Republic, project SGS_2022_007.

References (38)

Cited by (7)

View all citing articles on Scopus
View full text