Unexpected events in sulfated zirconia catalyst during glycerol-to-acrolein conversion

doi:10.1016/j.apcata.2011.10.015

Applied Catalysis A: General

Volumes 409–410, 15 December 2011, Pages 267-278

https://doi.org/10.1016/j.apcata.2011.10.015 Get rights and content

Abstract

The aim of the work was the analysis of the phenomena occurring during glycerol dehydration in the presence of sulfated zirconia used as solid acid catalysts; the attention was focused on catalyst deactivation. A close correlation between the sulfate content and catalytic behavior was found. In addition to the accumulation of carbon residues on the catalyst surface, there were other phenomena contributing to catalyst deactivation: (i) the self-reduction of sulfuric into sulfurous groups – an event which occurred, however, only in samples with the higher S content under anaerobic conditions – and (ii) the leaching of S from catalysts, due to the hydrolysis of sulfate groups and the formation of volatile esters. These findings are of general interest in relation to both the transformation of glycerol into acrolein by means of gas-phase dehydration and the use of sulfated zirconia catalysts for high-temperature, acid-catalyzed reactions.

Graphical abstract

Highlights

► Sulfated zirconia catalysts deactivate during glycerol dehydration into acrolein. ► The catalytic behavior of sulfated zirconia is affected by sulfate loading. ► Accumulation of heavy compounds is not the main reason for deactivation. ► Deactivation is mainly due to the self-reduction of S(VI) and to the loss of S. ► With catalysts having high S loading, O₂ co-feeding limits catalyst deactivation.

Introduction

Glycerol will play a crucial role in future biorefineries, because its derivatives can be used in various sectors, such as the fuel, chemical, pharmaceutical, detergent, and building industries [1], [2]. In this context, catalytic dehydration of glycerol to acrolein has attracted great interest [3], [4], [5], [6], [7], [8], [9], [10], [11], [12], [13], [14], [15], [16], [17], [18]; acrolein is mostly used as an intermediate for the synthesis of acrylic acid, this reaction being catalyzed in current industrial processes by Mo/V/W mixed oxides [19]. The best-performing catalysts with Brønsted-type acid properties include WO₃/ZrO₂ [5], [6], supported Keggin polyoxometalates [7], [8], [9], V/P/O [10], [11], [12], [13], rare earth pyrophosphates [11], and H-ZSM5 [12], [14], [18]; total glycerol conversion is achieved, with selectivity to acrolein as high as 70–80%. One peculiarity of this reaction is that with many of the strongly acidic catalysts studied in literature, the co-feed of oxygen helps keep the catalytic surface cleaner, thus limiting the formation of heavy deposits that may lead to catalyst deactivation [4], [20].

Among the various acid systems studied in literature, sulfated zirconia has also been considered as a possible catalyst for glycerol dehydration [5], [15], [20]. In principle, sulfated zirconia may be one suitable catalyst for this reaction, when conducted in the presence of molecular oxygen, since it has both acid and redox properties [21], [22], [23], [24], [25]; the latter may help in accelerating the removal of heavy compounds incipiently formed on the catalyst surface.

In a previous work [26], we reported on how the behavior of a sulfated zirconia catalyst (having 4.4 wt% of sulfate) in glycerol dehydration is affected by the main reaction parameters. In the present work, we discuss the effect of sulfate loading on catalytic behavior, in terms of events that are responsible for catalyst deactivation.

Section snippets

Experimental

Catalysts were prepared as described in our previous work [26], via a conventional precipitation method [27]. Table 1 shows the samples prepared, and their specific surface area after calcination; samples are labeled SZx, where SZ stands for sulfate zirconia and x for the sulfate amount (wt% SO₄) present in the catalysts.

Catalysts were characterized by means of hydrogen-thermal-programmed-reduction (H₂-TPR) and thermal-programmed-desorption (TPD). Hydrogen-thermal-programmed-reduction (TPR)

The effect of sulfate loading on catalytic behavior

Fig. 1(top) compares the glycerol conversion of SZ2, SZ3 and SZ4 samples in relation to the reaction temperature, under anaerobic conditions. With SZ4, glycerol conversion was close to 80% at 290 °C, but then decreased down to 55% when the temperature was increased to 360 °C, possibly because of a deactivation that contrasted the expected increase in conversion. At higher temperatures the significant conversion increase registered was also due to the contribution of homogeneous reactions [26].

Discussion

The main results of the catalytic experiments may be summarized as follows:

1.
We observed, in anaerobic conditions, a rise in the deactivation rate as the sulfur amount in sulfated zirconia increased. However, this was not due to the greater formation of heavy compounds, which may eventual be retained on catalyst surface and form residues that are precursors to coke formation; in fact, heavy products formed with similar selectivity on various catalysts, regardless of the sulfate content.
2.
The

Conclusions

The acid and redox characteristics of sulfated zirconia catalysts led us to investigate the catalytic behavior of sulfated zirconia catalysts in the gas-phase dehydration of glycerol, especially in relation to causes for catalyst deactivation under both aerobic and anaerobic conditions. The catalytic behavior was found to be greatly affected by the amount of sulfate loading in catalysts; the major difference concerned the rate of deactivation, which was very low with catalysts having the lower

Acknowledgements

MIUR (Ministero dell’Università e della Ricerca Scientifica) is acknowledged for the PhD grant to SG (Progetto Giovani). MIPAAF (Ministero delle Politiche Agricole Alimentari e Forestali) is acknowledged for financial support (Progetto Extravalore, PNR 2005–2007).

References (45)

H. Atia et al.
J. Catal.
(2008)
S.-H. Chai et al.
Appl. Catal. A
(2009)
F. Wang et al.
J. Catal.
(2009)
C.-J. Jia et al.
J. Catal.
(2010)
F. Wang et al.
Appl. Catal. A
(2010)
A. Corma et al.
J. Catal.
(2008)
E. Tsukuda et al.
Catal. Commun.
(2007)
L. Ning et al.
Chin. J. Catal.
(2008)
Y.T. Kim et al.
Microp. Mesop. Mater.
(2010)
L. Giebeler et al.
J. Mol. Catal. A
(2006)

X. Li et al.

J. Am. Chem. Soc.

(2005)

C. Breitkopf et al.

Phys. Chem. Chem. Phys.

(2007)

M. Signoretto et al.

Microp. Mesop. Mater.

(2005)

C.A. Emeis

J. Catal.

(1993)

J. Deleplanque et al.

Catal. Today

(2010)

R. Srinivasan et al.

J. Catal.

(1995)

B.S. Klose-Schubert et al.

Top. Catal.

(2011)

M.V. Luzgin et al.

J. Catal.

(2004)

R.A. Nyquist et al.

Infrared Spectra of Inorganic Compounds

(1971)

A.C. Ferrari et al.

Phys. Rev. B

(2000)

E. Tuinstra et al.

J. Chem. Phys.

(1970)

A. Alhanash et al.

Appl. Catal. A

(2010)

Cited by (44)

Unveiling the contribution of Mo, V and W oxides to coking in catalytic glycerol oxidehydration
2021, Molecular Catalysis
Citation Excerpt :
%W-Mt > 2.5 wt. %W-Mt. The glycerol conversion and selectivity to acrylic gradually decreased with an increase in W over the W-Mt. The literature attributes deactivation to the loss of strong acid sites [52]. As a result, the initial rapid decrease in glycerol conversion and selectivity to acrolein and acrylic acid could be attributed to the accumulation of carbon species on strong acid sites of Mo-Mt and W-Mt.
Mo, V and W oxides supported on montmorillonite (Mt) have shown good activity in catalytic glycerol oxidehydration. However, these catalysts deactivate rapidly due to coking. Understanding the role of Mo, V and W oxides in coke formation is essential to overcome the coking problem. Herein, individual and combination of Mo, V and W oxides supported on Mt were investigated to reveal the contribution of Mo, V and W oxides to coke formation. Among single metal oxide catalysts, Mo-Mt and W-Mt suffered more from coke formation than V-Mt. A mixture of Mo⁵⁺ and Mo⁶⁺ in the Mo-Mt catalyst hindered full oxidation during the oxidehydration, thus promoting coking. The oxidation of V⁴⁺ to V⁵⁺ helped diminish coke in mixed Mo-V-Mt and W-V-Mt. This work is expected to provide new information for designing high-performance Mo, V and W oxide catalysts with superior coke resistance.
Mesoporous silica supported phosphotungstic acid catalyst for glycerol dehydration to acrolein
2021, Catalysis Today
Citation Excerpt :
Hydroxyacetone is the main by-product with a yield lower than 10 %. The carbon balance was hard to calculate due to the unanalyzable heavy compounds produced and deposited on the catalyst or reactor [31,32]. Because glycerol conversion is 100 % in all cases, the acrolein yield equals to the selectivity, the highest selectivity to acrolein is obtained on HPW/MSU-x catalyst.
Conversion of glycerol to acrolein is a useful reaction for value-added application of biodiesel-derived glycerol and bioenergy development. The high-performance solid acid catalyst is essential to this dehydration reaction. In this paper, tungsten-based heteropolyacids (HPA) were supported on non-ordered mesoporous silica (MSU-x) to increase their dispersion and used as catalysts for glycerol dehydration to acrolein. Aiming to reveal the surface structure of HPA and resulting acidic properties, as well as the relationship between acidic properties and dehydration activity, different loadings of H₃PW₁₂O₄₀ were supported on MSU-x (10–50 wt%) and the catalysts were characterized by X-ray diffraction (XRD), BET, SEM/TEM, UV–vis diffuse reflectance spectra (DRS), Raman and FT-IR techniques. Their acidic properties were studied by NH₃-Temperature Programmed Desorption (NH₃-TPD) and Pyridine adsorption methods. The molecular structure and dispersion of H₃PW₁₂O₄₀ supported on the catalysts was revealed. The Keggin unit preserved well but with different hydration level for various loadings. The total acid concentration and respective Brønsted/Lewis acid identification were calculated. The acrolein yield increased with H₃PW₁₂O₄₀ loading until 30 wt% and showed less change with higher loadings. Based on the correlation of acrolein formation rate with acidic properties, the active role of Brønsted acid and the cooperative role of Brønsted/Lewis acid sites for glycerol dehydration to acrolein were discussed. This work provides new insight into the structure evolution of heteropolyacids and the catalyst design for the glycerol to acrolein.
Palladium-doped sulfated zirconia: Deactivation behavior in isomerization of n-hexane
2020, Fuel
The deactivation and regeneration behaviors of palladium doped sulfated zirconia was studied in the n-hexane isomerization. Results indicated that sulfate reduction is the major catalyst deactivation reason. Sulfate leakage also contributed to the deactivation but not the predominant cause. Since the appearance of hydrogen, the amount of carbon deposition is very low, and it plays a positive role in the reaction: covering part of high active sites, avoiding cracking. Reduced sulfite can be re-oxidized into sulfate under high temperature at the atmosphere of air, but accompanied by the gasification loss of sulfate and carbon deposition. Although the effect of nitrogen on oxidizing sulfites is not as good as air, nitrogen can inhibit the gasification loss of sulfate and carbon deposition, and finally the catalyst regenerated under nitrogen at 400 °C reflects superior performance even than fresh catalyst.
Solid acids, surface acidity and heterogeneous acid catalysis
2020, Advances in Catalysis
Citation Excerpt :
However, a recent critical review concluded that a lack of a fundamental understanding still exists on the reasons for high acidity of sulphated zirconia and that the application of modern catalysis studies of this important catalytic system are still needed (280). During reactivity, sulfur loss may also be found to the volatilization of sulfur, as volatile sulphate esters, and reduction to sulphites, with consequent deactivation of the catalyst (281). Also sulphated alumina is reported to be a very acidic material with both strong Lewis and strong Brønsted acid sites (282,283), not less acidic but less active than sulphated zirconia in alkylation and isomerization reactions (284), possibly due to the lack of redox properties of the support.
This review begins with the historical development of the theories of acidity and of the preparation of acids. The historical aspects of acid catalysis and catalysts are also briefly surveyed. The fundamentals concerning acidity and acid catalysis in liquid phases and at the solid-liquid interfaces are discussed. The factors determining acidity and activity, as well as practical applicability, of solid catalysts such as protonic zeolites are reviewed in detail. Methods for investigation of surface acidity of solids at both solid/vapor and solid/liquid interfaces are briefly discussed. Finally, most of solid acid families are considered and their properties critically reviewed.
Assessing carbon or tungstates coverage of ZrO<inf>2</inf> nanoparticles supported on MWCNT via NO<inf>x</inf>-TPD
2018, Nano-Structures and Nano-Objects
Multi-walled carbon nanotubes (MWCNT) promoted by oxides, that are far from trivial to characterize, have emerged as promising candidates in several fields such as biosensors and catalysis. In this study the coverage of ZrO₂ nanoparticles supported on MWCNT, by carbonaceous deposits or tungstates was investigated by NO $_{x}$ temperature-programmed desorption (NO $_{x}$ -TPD). ZrO₂ coverage by carbonaceous deposits was found to be about 30 %, whereas that by tungstates varied from 40 to 77 % depending on the W loading of the sample. The size of the ZrO₂ particles estimated by NO $_{x}$ -TPD ( $\sim$ 2.3 nm) was found to be in excellent agreement with that determined by TEM. Remarkably, it was found that W interacted preferentially with the ZrO₂ nanoparticles for a W surface density of 1.9 W/nm $_{ZrO2}^{2}$ (half monolayer coverage of the ZrO₂ nanoparticles by tungstates). Monolayer coverage of ZrO₂ by tungstates was nearly complete when introducing a W loading twice as high as that theoretically needed. These insights into the surface coverage of ZrO₂ by carbon or tungstates could not be obtained by any means other than NO $_{x}$ -TPD, which makes it a unique method to provide advanced characterization of the surface of oxides supported on MWCNT, in particular, and carbon supports, to a broader general interest.
Pd-Ni doped sulfated zirconia: Study of hydrogen spillover and isomerization of N-hexane
2018, Molecular Catalysis
A palladium-nickel doped sulfated zirconia was prepared, and hydrogen spillover and bimetallic synergic effect were confirmed using this model catalyst. Isomerization of n-hexane indicated that palladium and nickel could enhance both the conversion rates and selectivity. The XRD and BET results showed that nickel had a strongly influence on the crystal form and catalyst structure. TG-DSC analysis showed the loaded nickel could decrease the sulfates concentrations. Therefore, the total acidity, especially the strong Brønsted acid sites decreased in the presence of nickel, as analyzed by pyridine IR spectroscopy. H₂ TPR analysis indicated both the nickel and palladium could improve the reducibility of sulfated zirconia. And combined with XPS results, the nickel oxide was found easier to reduce in the presence of palladium, and higher reduction temperature could lead to the formation of NiAl₂O₄. Finally, the mechanisms of hydrogen spillover and reaction process were deduced.

View all citing articles on Scopus

View full text

Unexpected events in sulfated zirconia catalyst during glycerol-to-acrolein conversion

Abstract

Graphical abstract

Highlights

Introduction

Section snippets

Experimental

The effect of sulfate loading on catalytic behavior

Discussion

Conclusions

Acknowledgements

J. Catal.

Appl. Catal. A

J. Catal.

J. Catal.

Appl. Catal. A

J. Catal.

Catal. Commun.

Chin. J. Catal.

Microp. Mesop. Mater.

J. Mol. Catal. A

J. Am. Chem. Soc.

Phys. Chem. Chem. Phys.

Microp. Mesop. Mater.

J. Catal.

Catal. Today

J. Catal.

Top. Catal.

J. Catal.

Phys. Rev. B

J. Chem. Phys.

Appl. Catal. A