Estudio cinético de la oxidación parcial de tolueno a ...bdigital.unal.edu.co/56147/25/Geraldine.Díaz.2016.pdf · Estudio cinético de la oxidación parcial de tolueno a benzaldehído,

Estudio cinético de la oxidación parcial de tolueno a benzaldehído,

empleando como catalizador un óxido mixto de Ce – Zr – Mo

Geraldine Díaz Chaparro

Universidad Nacional de Colombia

Facultad de Ingeniería

Departamento de Ingeniería Química y Ambiental

Bogotá, Colombia

2016

Estudio cinético de la oxidación parcial de tolueno a benzaldehído,

empleando como catalizador un óxido mixto de Ce – Zr – Mo


Trabajo final de grado, presentado como requisito parcial para optar al título de:

Magíster en Ingeniería – Ingeniería Química

Director:

Julio Cesar Vargas Sáenz.

Campo de estudio:

Catálisis heterogénea

Grupo de investigación:

Procesos Químicos y Bioquímicos


Facultad de Ingeniería

Departamento de Ingeniería Química y Ambiental

Bogotá, Colombia

2016

Kinetic study of toluene partial oxidation to benzaldehyde, using a Ce – Zr – Mo mixed oxide as catalyst


A dissertation submitted in partial fulfilment of the requirements for the degree of

Magister en Ingeniería – Ingeniería Química

Advisor:

Julio Cesar Vargas Sáenz.

Research field:

Heterogeneous Catalysis

Research Group:

Chemical and Biochemical Processes


School of Engineering

Department of Chemical and Environmental Engineering

Bogotá, Colombia

2016

Dedicated to

My parents and my sister, who have been the best mentors and trip allies.

Hernán and Juliana, my little angels.

Acknowledgments

While I am responsible for this thesis, it is nonetheless at least as much of a product of

years of interaction with, and inspiration by friends and colleagues as it is my own work.

For this reason, I wish to express my warmest gratitude to all the people whose comments,

questions, criticism, support and encouragement, personal and academic, have left a mark

on this work. I also wish to thank the institutions that have supported me during the

development on this thesis. Regrettably, the following list of names will be incomplete, and

I hope that those who are missing will forgive me, and will accept my sincere appreciation

for their influence on my work.

First, I praise God for providing me with this opportunity and granting me the skills to

succeed.

I would like to thank my advisor, Professor Julio César Vargas, for the support and

persistent help he gave me. Many thanks also to Professor Anne - Cécile Roger, who

provided me with an opportunity to join her team as an intern, and who gave me access to

the ICPEES & CNRS laboratory and research facilities. Without their precious support, it

would not be possible to conduct this research.

I gratefully acknowledge the funding received during my master program from the

Universidad Nacional de Colombia by the scholarship “Asistentes Docentes”. I am also

grateful for the funding received through the program “Mobilité IdEx” from the Université de

Strasbourg.

I would like to say a heartfelt thank you to my Mum, Dad, Ángela, Hernán and Juliana for

always believing in me and encouraging me to follow my dreams. Special thanks goes to

Nicolas Daza and Diana Sánchez, for helping me to see the light at the end of scary tunnels.

Finally, Laura, Luisa, Adriana and Darwin, thank you for helping in whatever way you could

during this challenging period.

Abstract

The kinetic study of Ce-Zr-Mo catalysts on toluene partial oxidation to benzaldehyde in

vapor phase is presented in this work. Three different families of mixed oxides prepared by

pseudo sol gel method were characterized. The effect of the synthesis method on the

catalysts performance was evaluated by changing procedure steps and the interaction

between the active phase and the support of the materials. For the modified pseudo sol gel

method samples, the changes of structural properties evidence the predominant role of the

Mo species on the catalyst surface promoting total oxidation. The synthesis effect was also

evidenced on materials which exhibit ceramic structures, leading an irregular surface for

deposited catalysts and dispersed grains for integrate catalysts. The theoretical

composition proposed, as well as the synthesis methodology lead to solids, which exhibit

multiple phases and polycrystallinity. The kinetic study was made for the most selective

catalyst evaluated (Ce2.65Zr0.35Mo3O14). The mathematical regression fitting for Mars & Van

Krevelen kinetic model present less relative error with respect to the experimental data,

compared with three LHHW kinetic models and power law rate model evaluated.

Keywords: Mixed oxides, catalysts, toluene, benzaldehyde, kinetic study.

Resumen El estudio cinético de la oxidación parcial de tolueno a benzaldehído en fase de vapor

usando catalizadores de Ce-Zr-Mo es presentado en este trabajo. Tres diferentes familias

de óxidos mixtos preparados por el método pseudo sol gel fueron caracterizadas. El efecto

del método de síntesis en el desempeño de los catalizadores fue evaluado mediante el

cambio de pasos de la metodología y la interacción entre la fase activa y el soporte de los

materiales. Para los catalizadores obtenidos por el método pseudo sol gel modificado, los

cambios en las propiedades estructurales evidencian el papel predominante de las

especies de molibdeno en la superficie del material, promoviendo la oxidación total. El

efecto del método de síntesis fue también evidenciado en materiales que mostraron

estructuras cerámicas, obteniendo como resultado una superficie irregular para los

catalizadores depositados y granos dispersos para los catalizadores integrados. La

composición teórica propuesta, al igual que el método de síntesis usado permitieron la

formación de sólidos que exibieron múltiples fases y policristalidad. El estudio cinético fue

hecho para el catalizador más selectivo entre los evaluados (Ce2.65Zr0.35Mo3O14). La

regresión matemática ajustó para el modelo cinético Mars & Van Krevelen, presentando

menor error relativo con respecto al conjunto de datos experimentales, comparado con tres

diferentes modelos cinéticos LHHW y el modelo de ley de potencias que también se

evaluaron.

Palabras claves: Óxidos mixtos, catalizadores, tolueno, benzaldehído, estudio cinético.

Contents

Page

Acknowledgments ................................................................................................................. ix

Abstract ................................................................................................................................. xi

List of figures ...................................................................................................................... xvii

List of tables ........................................................................................................................ xix

Introduction ......................................................................................................................... xxi

1. Toluene oxidation: a value alternative for the crude oil market ..................................... 23

1.1. Heavy oil: an option to dwindling conventional oil supplies .................................... 23

1.2. Heavy oils market challenge ................................................................................... 24

1.3. Adding value to refinery by-products ....................................................................... 26

1.4. Benzaldehyde as a compound of major interest ..................................................... 27

1.5. Benzaldehyde production processes ...................................................................... 30

1.6. Benzaldehyde production by toluene partial oxidation ........................................... 31

1.7. Mixed oxides as selective oxidation catalysts ......................................................... 33

1.8. Cerium, zirconium and molybdenum mixed oxides ................................................ 34

1.9. Modelling the reaction ............................................................................................. 36

1.10. Scope ..................................................................................................................... 40

2. Vapor phase toluene partial oxidation: the reaction at laboratory scale ........................ 43

2.1. Thermodynamic study ............................................................................................. 43

2.2. Reaction system ...................................................................................................... 45

2.3. Ce–Zr–Mo oxydes as catalysts ............................................................................... 47

2.3.1. Modified pseudo sol gel method catalysts ....................................................... 49

2.3.2 Pseudo sol gel method catalysts ....................................................................... 56

2.3.3 Influence of cation integration vs deposition ..................................................... 60

3. Kinetic study .................................................................................................................... 67

3.1. Data collection ......................................................................................................... 67

3.1.1. Experimental procedure ................................................................................... 67

3.1.2 Experimental results .......................................................................................... 68

3.2 Determination of kinetic parameters ........................................................................ 69

3.2.1 Power law rate expression ................................................................................ 70

3.2.2 LHHW kinetic expression with two active sites ................................................. 72

3.2.3 LHHW kinetic expression with dissociative oxygen adsorption ........................ 73

Contents xvi

3.2.4 LLHW kinetic expression with non-dissociative oxygen adsorption ................. 74

3.2.5 Mars and Van Krevelen kinetic expression ....................................................... 75

4. Conclusions and findings ................................................................................................ 81

References .......................................................................................................................... 83

Annexes .............................................................................................................................. 91

A. Chromatographic methods ................................................................................... 91

i. Non-condensable products analysis .................................................................... 91

ii. Condensable products analysis ........................................................................... 92

B. Catalytic tests results ........................................................................................... 93

C. Kinetic experimental results ................................................................................. 94

xvii

List of figures

Figure 1. Proven vs recoverable and unconventional world oil reserves of crude oil. ...... 24

Figure 2. Global capacity requirements by process type 2011 – 2035 ............................. 25

Figure 3. Exports of aldehydes-ethers, aldehydes-phenols and aldehydes with oxygen

radicals ................................................................................................................................ 28

Figure 4. Imports of aldehydes-ethers, aldehydes-phenols and aldehydes with oxygen

radicals ................................................................................................................................ 28

Figure 5. Aldehydes-eters, aldehydes-phenols and aldehydes with oxygen radicals’

imports and exports in Colombia ........................................................................................ 29

Figure 6. Hydrogen transfer mechanism for the initial steps in catalytic oxidation of

toluene ................................................................................................................................ 37

Figure 7. Electron transfer mechanism for the initial steps in catalytic oxidation of toluene

............................................................................................................................................ 38

Figure 8. Toluene molar conversion as function of reaction conditions. ........................... 43

Figure 9. CO molar selectivity as function of reaction conditions ...................................... 44

Figure 10. CO2 molar selectivity as function of reaction conditions ................................... 44

Figure 11. Experimental arrangement for toluene oxidation .............................................. 45

Figure 12. Recovered volume fraction of the main outlet condensable compounds in

function of the trap temperature ......................................................................................... 46

Figure 13. Global catalyst preparation procedure following the pseudo sol-gel method .. 49

Figure 14. Catalyst preparation procedure following the modified pseudo sol-gel method

............................................................................................................................................ 49

Figure 15. Synthetized CZM2(I), CZM3

(I), CZM4(I) catalysts samples TPR analysis results 50

Figure 16. Synthetized catalysts samples TPO analysis results. ...................................... 52

Figure 17. Diffractograms of fresh mixed oxides: CZM2(I), CZM3

(I), CZM4(I) ....................... 53

Figure 18. Raman spectra of fresh mixed oxides: CZM2(I), CZM3

(I), CZM4(I) ...................... 53

Figure 19. SEM micrographs of fresh mixed oxides: a. CZM2(I), b. CZM3

(I), c. CZM4(I) ...... 54

Figure 20. Catalytic test temperature profile vs time.......................................................... 55

Figure 21. Diffractograms of fresh mixed oxides: CZM2(II), CZM3

(II), CZM4(II). .................... 57

Figure 22. Electron diffraction of fresh mixed oxides: a. CZM2(I), b. CZM2

(II) ..................... 58

Figure 23. TEM micrographs of fresh mixed oxides: a. CZM2(I), b. CZM2

(II) ....................... 58


(II) ...................... 59

Figure 25. Global catalyst preparation procedure for deposited catalysts, following the

pseudo sol gel method ........................................................................................................ 61

Figure 26. Diffractograms of fresh mixed oxides: CM/CZ(I), CM/CZ(II). .............................. 62

Figure 27. Electron diffraction of fresh mixed oxides: a.CM/CZ(I), b. CM/CZ(II). ................ 62

Figure 28. TEM micrographs of fresh mixed oxides: a. .CM/CZ(I), b. CM/CZ(II). ................ 63

Figure 29. SEM micrographs of fresh mixed oxides: a. .CM/CZ(I), b. CM/CZ(II). ................ 64

Figure 30. Test methodology schema. ............................................................................... 68

Figure 31 Kinetic rate’s toluene partial pressure dependency evaluated at 7mL/min

oxygen feed rate. ................................................................................................................ 68

Contents xviii

Figure 32 Kinetic rate’s oxygen partial pressure dependency at 30L/min toluene feed

rate. ..................................................................................................................................... 69

Figure 33. Parity diagram for power law kinetic model ...................................................... 70

Figure 34. Parity diagram for modified power law kinetic model ....................................... 71

Figure 35. Parity diagram for two active sites LHHW kinetic model .................................. 73

Figure 36. Parity diagram for oxygen dissociative adsorption LHHW kinetic model ......... 74

Figure 37. Parity diagram for non- dissociative oxygen adsorption LHHW kinetic model. 75

Figure 38. Parity diagram for M&VK kinetic model ............................................................ 77

Figure 39. Parity diagram for modified M&VK kinetic model ............................................. 78

Figure 40. Parity diagram for M&VK kinetic regression for experiments made at 475°C . 78

Figure 41. M&VK fitting model for experiments made at 7mL/min oxygen feed rate. ....... 79

Figure 42. M&VK fitting model for experiments made at 30L/min toluene feed rate....... 79

xix

List of tables

Table 1. Catalyst proposal development ............................................................................ 48

Table 2. Solids synthetized by the modified pseudo sol-gel method ................................. 50

Table 3. Catalytic test results of: CZM2(I), CZM3

(I), CZM4(I) samples. ................................. 55

Table 4. Samples synthetized by the pseudo sol-gel method ........................................... 56

Table 5. Catalytic test results of: ZM2(I), CZM3

(I), CZM4(I) samples ..................................... 59

Table 6. Third catalyst family synthetized .......................................................................... 61

Table 7. Catalytic test results of .CM/CZ(I) and CM/CZ(II)samples ..................................... 64

Table 8. Power law model kinetic parameters obtained .................................................... 70

Table 9. Modified Power law model kinetic parameters obtained ..................................... 71

Table 10. Two active sites LHHW model kinetic parameters obtained ............................. 72

Table 11 Dissociative oxygen adsorption LHHW kinetic parameters obtained ................. 74

Table 12. Non-dissociative oxygen adsorption LHHW kinetic parameters obtained ........ 75

Table 13. M&VK kinetic parameters obtained .................................................................... 76

Table 14. Modified M&VK kinetic parameters obtained ..................................................... 77

Table 15. Retention times and response factors utilised for the non-condensable

products. ............................................................................................................................. 91

Table 16. Retention times and response factors utilised for the condensable products. .. 92

Table 17. Catalytic tests results summary ......................................................................... 93

Table 18. Experimental data results. .................................................................................. 94

Introduction

Toluene represent a very important feedstock to petrochemical industry. This hydrocarbon

may be use as solvent or as a precursor to other chemicals, e.g. polyurethane foam, the

explosive trinitrotoluene (TNT) and numerous synthetic drugs. By oxidation of toluene, a

wide range of products may be generated, such as benzyl alcohol, benzaldehyde, benzoic

acid, benzyl benzoate, and phenyl benzene. All of the above listed products have more

commercial value than toluene. The most versatile of them is benzaldehyde, which is not

produced in Colombia and could satisfy a continuous growing market.

The oxidation of toluene to benzaldehyde could take place in vapor phase using mixed

oxides as catalyst. However, the synthesized solids had low surface area and show various

phases that may not be the most active or may become unstable. Moreover, in catalytic

partial oxidation, several variables affect toluene conversion and benzaldehyde selectivity,

such as the oxygen to toluene feed ratio, the reaction temperature and the space velocity

(W/F). Therefore it is necessary to study the influence of different conditions in the behavior

of the reactants and the products involved. This study is related to the reaction rate and

could be modelled by mathematical expressions in order to fit the experimental data.

This work aims to contribute with the characterization of a cerium- zirconium and

molybdenum mixed oxide as an active material for the partial oxidation of toluene to

benzaldehyde, and the adjustment of reaction phenomena by a kinetic expression. In order

to present the main results, this document has been divided into three main chapters:

The first one is a brief introduction to the state of the art. This section comprises an overview

about the current scenario of the global crude oil market, the partial oxidation as an

alternative for adding value to toluene, a refinery by-product; and the generalities about the

catalysts used in the toluene to benzaldehyde vapor phase reaction, being the Ce-Zr-Mo

mixed oxide a striking choice.

The second chapter exhibits the results of the experimental work performed during the initial

stage of this project. It includes the formulation, characterization and catalytic test of the

materials produced, as well as the assessments of catalyst obtained by different synthesis

methods and compositions.

Introduction xxii

The chapter three states the behavior of the most selective and stable catalyst evaluated.

Mathematical adjust of different parameters such as temperature and partial pressure of

the reactants related to reaction rates is given by kinetic expressions.

Finally, some conclusions and findings for further research were summarized.

1. Toluene oxidation: a value alternative for the crude oil market

1.1. Heavy oil: an option to dwindling conventional oil supplies

Crude oil is a mix of alkanes, cycloparaffins, aromatic compounds and non-hydrocarbon

heterocompounds. Appreciable property differences appear between crude oils as result of

the variable ratios of the crude oil components. In order to identify the behavior of the

different kind of crude oils, the American Petroleum Institute stablishes a way to express

the crude’s relative gravity (°API). A low API gravity indicates a higher content of heavy

fractions such as cycloparaffins and aromatic compounds, thus it means a heavier crude

oil, while a higher API gravity leads a lighter crude oil [1].

The crude oil found in conventional reserves is mostly light and intermediate crude oils,

with high market value and whose production and processing consist of technically well-

established methods [2]. However, the continuous increase in world energy demand, driven

by the economic development and the population growth recorded in recent decades, has

caused the scarcity of traditional reserves and nowadays new discoveries of conventional

oil are insufficient to supply the estimated future energy demand. Since most of the oil

reserves are situated on the Middle East, it generates a wide gap in the worldwide energy

supply with global economic impacts, such as the one caused by the Organization of the

Petroleum Exporting Countries (OPEC) later in 2015, when its crude oil production growth

and global prices fell sharply [3].

The crude oil global distribution is presented in Figure 1. The darker bars represent the

proven crude oil reserves, most of them traditional reserves with more than 90% of

probability of being productive. The lighter bars represent the recoverable crude oil

reserves, it means heavy crude oil very difficult to recover by traditional methods, so the

probability of being productive is lower. And finally, the bigger bar shows the global

unconventional crude oil reserves, which includes tar sands, shade oil and bitumen, among

others [4].

24 Kinetic study of toluene partial oxidation to benzaldehyde,

using a Ce – Zr – Mo mixed oxide as a catalyst

Figure 1. Proven vs recoverable and unconventional world oil reserves of crude oil [4].

As it is shown, recoverable crude oil reserves present more homogeneous distribution

comparing with proven reserves. On a commercial basis, the exploitation of these reserves

would be of enormous benefit to different countries, contributing with the energy

independence in the intermediate term and providing secure energy supplies to the

Western Hemisphere for the near future. Achieve a better scenario depends on the oil

industry’s capacity to transform potential resources into commercial exploitable reserves

developing more technologies as Enhanced Oil Recovery (EOR), mining or fracking [2].

1.2. Heavy oils market challenge

The global current tendency is to obtain more product on the heavy fractions in the refining

of recovered oils [3], it implies higher content of nitrogen, oxygen, sulphur and heavy metals

hetero-compounds and a wider quantity of naphthenic and aromatic compounds. Therefore

the process requires even more heating and dilution strategies during recovery, which

increases the production costs, compared to light and intermediate oils [5]. The high contain

1. Toluene oxidation: a value alternative for the crude oil market 25

of heavy fractions changes the conditions of refining processes, so, unconventional oils

refining requires great specificity and more severe conditions of pressure and temperature

to obtain high added value products, such as liquefied petroleum gas (LPG), gasoline,

kerosene, and diesel oil [2].

Agreeing with this situation, the OPEC estimated the refining sector would have to increase

its complexity in order to guarantee the products specifications, so this organization

presented the crude oil industry forecast to 2035, given by the Figure 2. Major refinery

process are collected by four groups, showing the current projects and the additional

requirements taking into account the composition changes of the raw materials. In this way,

desulphurization requirements comprises the largest volume of capacity additions to 2035,

nearly 1.5 times those for distillation [6].

Figure 2. Global capacity requirements by process type 2011 – 2035 [6]

In this context, the development of new technologies to handle the increment of sub

products becomes crucial for the economical processing of unconventional resources such

as heavy and extra-heavy oils.



1.3. Adding value to refinery by-products

Crude oil refining includes distillation, coking and thermal units, catalytic cracking and

hydrocracking, hydro treating, reforming, isomerization, alkylation and polymerization

processes [7]. The complexity of each arrangement depends on the market to supply and

the crude oil characteristics. When part of the products are fuels such as gasoline, it is

necessary to produce other fractions with higher octane number to improve their properties,

for example aromatic compounds [1].

Benzene’s derivate hydrocarbons or aromatic compounds are refinery products since early

40’s, before so, their production was by dry distillation of coal [8]. These derivatives

represent a very important feedstock to petrochemical industry, as the main obtained

compounds are benzene, toluene and xylenes (BTX), all of them platform hydrocarbons.

Nevertheless, most of the BTX products were used to improve the fuel’s octane rating, so

since the restrictions against health hazardous components contained in formulated

gasoline are rising, the aromatic compounds are becoming low-value by products in places

where the petrochemical industry is not well developed.

Taking into account the progressive crude oil composition changes, the production of BTX

will increase because of the growth on fuels demand, not because of the complexity of

producing it, as it is shown in the Figure 2: The requirement of octane units is the smallest,

and because of aromatic compounds contained in the current crude oil are increasing, so

its overproduction. This scenario is the one to focus on add value to BTX products as

toluene, a platform compound.

Toluene could be used to obtain valuable products through partial oxidation. In that case,

catalysts are needed to prioritize the wanted reaction with respect to total oxidation and

other possible interactions. Furthermore, heterogeneous gas phase – solid catalysis is

preferred since those kind of processes avoid the need for solvents and complex

procedures to purify products. Liquid – solid heterogeneous catalysis or liquid phase

homogeneous catalysis are the second choice processes, also depending on the oxidant

agent used [9].


Nowadays, the interest to improve the oxidation of different compounds is higher than ever.

The study of different reactions and alternative mechanisms for traditional processes have

increased recently. This particular change is taking place mainly because of the

popularization of the green chemistry concept and the emergent concern on processes

intensification and energy safety [10]. These particular interests affect different areas,

among them, chemistry, engineering and biology, because selective oxidation could be a

solution to different problems, such as the deficiency of particular enzymes, the inability of

natural synthesis of specific compounds or the growing demand of a limited product.

By selective oxidation of toluene a wide range of products such as benzyl alcohol,

benzaldehyde, benzoic acid, benzyl benzoate and phenyl benzene may be generate. All of

the listed products have more commercial value than toluene and most of them are used

in cosmetic and polymer industries. The selectivity depends on the catalyst, the solvent,

reaction temperature, and pressure [11].

The most versatile product of the toluene’s selective oxidation is benzaldehyde. This

compound, best known for having the characteristic odor of essential oil of almonds, is used

in several applications of food, cosmetic and pharmaceutical industries [12].Nevertheless,

it is not produced in Colombia and could represent an alternative of management to the

increasing refinery by-products and an option to a growing market.

1.4. Benzaldehyde as a compound of major interest

Globally, the aldehydes are compounds of greater economic attention and have a high

mobility between countries, reporting an increase of over 179% in exports for Germany,

176% in the case of India and 45% for China in the period 2005-2011 [13]. Figure 3 presents

the trades of the bigger aldehyde exporter countries during the named period.



Figure 3. Exports of aldehydes-ethers, aldehydes-phenols and aldehydes with oxygen radicals [13]

The main importers are Switzerland, UK and India (marketer country). They reported an

increment in their imports higher than 2%, 127% and 150% respectively for the period 2005-

2011 [14]. The main trades of these countries are shown in Figure 4.

Figure 4. Imports of aldehydes-ethers, aldehydes-phenols and aldehydes with oxygen radicals [14]

For most Latin American countries, including Colombia, the trend is to import the necessary

amount of functionalized aldehydes to meet domestic needs and thus be the final

consumers, supplied mostly by the United States. The Colombian case, shows an import

0

5

10

15

20

25

30

35

2 0 0 4 2 0 0 5 2 0 0 6 2 0 0 7 2 0 0 8 2 0 0 9 2 0 1 0 2 0 1 1 2 0 1 2

EXP

OR

TS (

MM

USD

$)

YEAR

China USA Germany India

0

5

10

15

20

25

30

2 0 0 4 2 0 0 5 2 0 0 6 2 0 0 7 2 0 0 8 2 0 0 9 2 0 1 0 2 0 1 1 2 0 1 2

IMP

OR

TS (

MM

USD

$)

YEAR

Switzerland UK France India


growth of 30% in the period 2005-2010, furthermore, on average only 0.8% of the material

arrived to the country came out [13, 14]. Figure 5 present the major Colombian money

trades referring to aldehydes and oxygen radicals for the named period.

Figure 5. Aldehydes-ethers, aldehydes-phenols and aldehydes with oxygen radicals’ imports and exports in Colombia [13, 14]

Those tendencies show the importance of compounds such as benzaldehyde, whose

global production for 2009 was estimated at 90,000 tons per year [15]. Benzaldehyde has

been used in organic synthesis as feedstock for several products, including compounds

such as mandelic acid, benzylic acid, cinnamic acid and benzyl alcohol, various aldehydes

(methyl, butyl, amyl and hexyl cinnamic aldehyde) and some amino acids such as

phenylglycine [12]. Recent developments in the use of benzaldehyde have occurred in the

healthcare industry, for the production of blood pressure regulators and influenza drugs,

also in the agriculture, as part of the formulation of insecticides and in the malachite green

pigment synthesis [16].

Benzaldehyde, which is found naturally in several foods such as the tonsillar glucoside,

which is a compound characteristic of bitter almonds, can also be gotten by different ways,

for example, with the partial oxidation of toluene, as it has been discussed before. In

countries with undeveloped petrochemical industry such as Colombia, toluene is imported,

even when it has been reported residual toluene productions higher than 4% in the last

0

50

100

150

200

250

300

350

400

2004 2005 2006 2007 2008 2009 2010 2011 2012

MO

NEY

TR

AD

E (M

USD

$)

YEAR

Imports Exports



years (quantified as a product of commercial interest) [17], and there is not a market

established to use when it is obtained as by-product.

When this scenario is contextualized in a country directly affected by the crude oil dynamic,

like Colombia, the study of valuable products production processes from oil refining and

chemical industry to strengthen the country’s economy becomes a priority. For the specific

case arises, the benzaldehyde production from oxidation of toluene.

1.5. Benzaldehyde production processes

Benzaldehyde can be extracted naturally from almonds, apples and cherries kernels, since

these fruits contain amygdalin-glucoside. Under enzymatic catalysis this glycoside

produces benzaldehyde, hydrogen cyanide and two glucose molecules [12]. However, in

areas with limited production of the fruits mentioned, benzaldehyde is synthesized by

several ways, including alkaline hydrolysis of cinnamaldehyde, the partial oxidation of

benzyl alcohol, the alkaline hydrolysis of benzyl - chloride and benzene carbonylation [18].

Among the major reactive mechanisms of scientific interest, there are highlighted the

reaction between benzene with dichloromethyl methyl ether in the presence of AlCl3, the

formylation of benzene with formyl fluoride, using as catalyst boron trifluoride, and the

oxidation of benzyl alcohol in the presence of Cu(NO3)2, having as intermediary the

chloromethylbenzene. Benzaldehyde production by these chemical routes is generally

carried at laboratory scale due to the low conversions, the high costs associated with

purification of the benzaldehyde obtained and the environmental implications involving the

effluent treatment [19]. Industrially, the production is carried by the Etard reaction [11],

which includes chlorine compounds as by-products, involving high costs associated with

the purification of the benzaldehyde. Also, it is environmentally unfavorable because metal

chlorinated complexes derivate of CrO2Cl2 are generated, and the end uses of produced

benzaldehyde are limited [20].

However, a chemical pathway that has attracted great interest in recent years is the partial

oxidation of toluene. This reaction involves only two reagents, neither of them halogenated,

and occurs in the presence of a selective catalyst to prevent complete oxidation of the

hydrocarbons [1, 20].


1.6. Benzaldehyde production by toluene partial oxidation

The oxidation of toluene can take place in liquid phase, involving the use of solvents, low

conversions and high reaction times, or in vapor phase, which promotes the contact

between the reactants. However, the last one demands higher temperatures than liquid

phase reaction and may generate mixtures of toluene and oxygen near to the explosion

point, entailing the develop of nontrivial control systems [21].

Considering the limitations outlined for the current process carried out at industrial scale

and the interest on produce benzaldehyde without impurities, which restrict their use and

increase the purification costs associated; recent studies have focused on the partial

oxidation of toluene in liquid phase or vapor phase, particularly focusing on the

development of selective catalysts and strategies to improve yield and selectivity. The main

problem with this kind of reactions is that each successive oxidation is faster than the

previous one, so the tendency is to total oxidation, which is related to formation of acidic

hydrogens after the insertion of the oxygen’s heteroatom or carbon oxides [1].

The selective oxidation of toluene to benzaldehyde in the liquid phase is generally carried

in the presence of organic solvents, liquid acid catalysts and in most of the cases using as

oxidizing agent hydrogen peroxide. In order to promote the conversion and selectivity of

the reaction, it has been suggested to change the oxidizing agent for compounds, such as

tert-butyl hydroperoxide [22].

Toluene liquid phase oxidation implies significant reaction time, leading to higher operating

costs. Concerned with this problem, research have directed their efforts to improve the

contact between the reactants using oxygen as oxidizing agent and a variety of catalysts,

such as ionic liquids [23], tetra hydrate cobalt acetate [24], and different metal oxides as

Fe-V-O [25], MnOx over SBA-15 [11], and vanadium oxide supported on SBA-15 [26].

Considering the results reported by several studies, the economic viability of production of

benzaldehyde by partial oxidation of toluene in liquid phase is still a matter of discussion

because it is essential to use solvents, which hinders the transport mass, promotes the

generation of by-products [27], and leads to low reaction conversion and selectivity.



Recently research have focused on the vapor phase reaction, although the development of

catalysts to prevent complete oxidation is a key point [1, 28]. The oxidation have been

developed using vanadium oxide as active phase without support [29], supported on

different materials as metal oxides [30] or silica [31, 32], enriched by cations as potassium

[33], iron and cerium [34] and mixed with other metals as silver [35] . Vanadium oxide is

used because of its redox properties, but it shows low conversions and selectivity to

benzaldehyde. It seems the selectivity of toluene transformations into benzaldehyde

decrease as the vanadium containing catalysts ability to generate the singlet form of

molecular oxygen grows [36].

Moreover, the feasibility of using oxides of lanthanum [37], cerium [38] and iron [39]

separately and supported on MoO4 [40] where studied. Molybdenum trioxide has been

suggested to be a selective, if not very active, catalyst because the only active chemisorbed

oxygen entity appears to be O2-: other diatomic species that are known to exist on the

catalyst are not involved in the catalytic oxidation [41]. In addition, it was proven structure

the sensitivity of molybdenum oxides for oxidation reactions. Studies clearly demonstrated

how the geometry of the reacting molecule and atomic arrangements at the oxide surface

are important factors in oxidation reactions. E. g. it was shown that propene gives almost

exclusively acrolein on the (1 0 0) lateral face and CO2 on the apical (1 0 1), (1̅ 0 1) and

basal (0 1 0) planes [42]. However, at high concentrations of molybdenum, increasingly

sized MoO3 crystals preferentially expose non-selective (0 1 0) crystal faces, responsible

for undesired C-C bond cleavage and total oxidation reactions [43]

While the presence of stronger Lewis acid sites in vanadium-based catalysts active the

carbon atom of the carbonyl groups, thus facilitate its nucleophilic attack to form benzoate

species which further degrade to carbon oxides [44]; molybdenum based catalysts present

high performance, even when the preparation conditions and the catalyst structure exert

great influences on the catalytic selectivity [40].


1.7. Mixed oxides as selective oxidation catalysts

Catalytic hydrocarbons oxidation determining rate step is the hydrocarbon adsorption on

the active phase, since it is necessary to include a new type of atom in the structure. The

compounds that facilitate this process are metal oxides, due to the bonding force between

carbon and metal oxide [45]. Additionally, oxygen mobility in such solids assists the

reaction, preventing further interaction between the hydrocarbon and molecular oxygen [46,

47].

Oxygen mobility is related with metallic ions estate of valence. If the metallic ion has several

state of valence, different oxides could be present and therefore, compensation effect

promotes the generation of vacancies [48]. Cerium oxides are frequently desired, because

its fluorite type structure enables cycle types like: [CeO2↔ CeO2- + (/2)O2 for =0 to 0.5].

Its oxygen storage capacity (OSC) improves by substitution of cations, doping or changing

the meso and nano solid architecture [49]. Among those reported are Mo - Ce supported

oxides [50, 51], Mo-Zr-Ce oxides [52], Co-Ce-Zr mixed oxides [53], Ce-Zr-Y-Mn oxides [54]

Pr-Zr-Ce oxides [55], and Mn-Ni-Co-Ce-Zr oxides [56].

One of the most popular modifications has been the inclusion of zirconium cation in the

cerium solid matrix (CeO2- ZrO2, Ce1-xZrxO2, or CZ) because this solid solution present

higher OSC and thermal stability, making it more versatile at the inclusion of cations to

provide other characteristics to the oxide [49].

The use of ternary oxides based on Ce-Zr-Mo for the synthesis of benzaldehyde in gas

phase has been proposed [57], in order to achieve better catalytic behavior by improve the

thermal stability of the structures changing their redox properties and oxygen mobility [58].

However, the synthesized catalysts had low surface area and shows various phases that

may not be the most active or may become unstable [55, 56]. Still, a well understand

structure giving by cerium, zirconium and molybdenum mixed oxides is not reported so far

and however represent an option as oxidation catalyst.



1.8. Cerium, zirconium and molybdenum mixed oxides

The main problem for plurimetallic mixed oxides, is to find the components ratio to obtain a

stable solid phase, because within one structure type the size and crystal structure can

modify some of their functional properties therefore the size effect and morphological

characteristics may serve as factors determining the practical use of one or another

structural form [59].

Cerium is the second element of the rare earth series with electronic configuration: (Xe) 6s2

5d1 4f1. In such configuration, the volume of 6s and 5d orbitals are greater than that of the

4f orbital and, therefore, the three 6s2 and 5d1 electrons are the only ones participating for

chemical bonds. The valence of the lanthanide elements in their configurations is habitually

3+. In the particular case of Ce, of which the electronic configuration is (Xe) 4f1, the f

electron is easily released, acquiring the more stable structure of the tetravalent ion Ce4+.

Effectively, cerium dioxide (CeO2) is the thermodynamic stable phase in nature. Its crystal

structure is cubic containing four formula units per unit cell. Each cation is coordinated to

eight oxygen atoms and each anion is coordinated to four cerium atoms [60].

There are three well-defined polymorphs of pure ZrO2, namely, the monoclinic, tetragonal,

and cubic phases. The monoclinic phase is stable up to about 1100°C temperature range

to the tetragonal phase. At approximately 2370°C, the compound adopts the cubic structure

[61].

The structural units of all МоО3 phases are [МоО6] octahedral linked by vertices into layered

(α-МоО3) or framework (β-МоО3) structures. Among them, the hexagonal metastable h-

МоО3 phase has a unique tunnel structure and consists of zigzag chains of the octahedral

sharing vertices and edges. The tunnels are 1D cavities with a diameter of ∼3 Å formed by

the adherence of 12 octahedral towards the с axis [59].

In the case of the CeO2, the reason to mix with other metallic cations is its poor thermal

stability, then support CeO2 based catalyst on ZrO2 is presented as a simple and effective

method to improve the properties of the oxide and its mixtures, especially for the exothermic

oxidation reactions. Thus the ZrO2 acts as a carrier who stabilize the surface active

structure of CeO2 - based oxides by formation of Zr0.88Ce0.12O2 solid solution in the interface,


also it can enhance mobility of active oxygen and improve catalytic performance of total

oxidation [62].

Although most conventional CeO2-ZrO2 systems which exhibit cubic fluorite or tetragonal

zirconia structures have a disordered arrangement of Ce and Zr within the crystal structure,

CeO2–ZrO2 pyrochlore-type compounds have been discovered which arrange these two

atoms in an ordered manner along the [1 1 0] direction in Ce2Zr2O7+ crystallites [63].

Pyrochlore structure is generally obtained through high-temperature reduction, resulting in

a decrease in specific surface area due to sintering and a less functional material for

catalyst. Even so, there are several preparation methods proposed in order to improve the

thermal stability of the compound [64].

CeO2-MoO3 system have also many applications such as methanation catalysts [65], most

of them to improve the thermal behavior of the catalyst and the poison resistance. The

highly dispersed molybdate, crystalline Ce8Mo12O49 and Ce2Mo4O15 are formed in the

MoO3-CeO2 complex oxide calcined, increasing MoO3 content [66]. Nevertheless, the

stability of the compounds is limited by the sublimation of the Mo compounds [67].

Binary compounds of those oxides have been studied in other reactions [68], some of them

similar reactions, as the case of Mo-Zr oxide to promote the selective oxidation of methane

to formaldehyde [69], the results showed the direct relation between the catalytic

performance of the solids and their physicochemical properties with the content of Mo in

the bulk of ZrO2. At the end, the Zr(MoO4)2 structure was the only one stable and the one

with better results.

The behavior of Ce-Zr binary oxides on ternary systems have also been studied. CeO2–

DyO1.5–ZrO2 system [70], present the different phases between the CeO2–ZrO2 binary

mixtures (Fluorite type cubic, fluorite type cubic and tetragonal, monoclinic and tetragonal),

similar with the results reported for the CeO2–ThO2–ZrO2 system [71] and the HfO2–ZrO2–

CeO2 system [72].

The Ba-Zr-Mo oxides ternary system shows that at reducing conditions, the Mo compounds

are reduced into Mo0 in Mo-rich compositions and as a function of composition and

temperature, Mo0, Mo4+ and Mo6+ compounds can thus coexist at these conditions. At



oxidizing conditions all the identified Mo compounds show that the stable Mo valence is

Mo6+ then, solid solutions between Zr4+ and Mo6+ compounds are practically non-existent

[73]. Related studies shows that isolate MoO42- species is the primary molybdenum species

at low loading amount of Mo in Ce-Zr oxides matrix, but polymeric molybdate species

predominates when the Mo6+ loading amount is increased. The molybdena is preferentially

dispersed on the surface of CeO2 at lower loading because the base strength of CeO2 is

stronger than that of ZrO2 [74].

Even without a single phase, the expectations for the well development of the ternary

catalyst in partial oxidations is increase taking into account the good results reported for

another reactions, as the selective reduction of NOX [75]. The addition of Mo controlled the

growth of CeO2 particle size in a mixed matrix of Ce – Zr oxides, improved the redox ability

and increased the amount of surface acidity, especially the Lewis acidity [76]. In fact,

studies conclude that solids with the presence of the three metallic cations exhibits excellent

catalytic activity for the condensation of various aromatic and hetero aromatic anilines

which are similar compounds to the aromatic hydrocarbons [77].

1.9. Modelling the reaction

In catalytic oxidation of toluene several variables affect the conversion of the hydrocarbon

and the benzaldehyde selectivity, as the oxygen to toluene feed ratio, the reaction

temperature and reciprocal of the space velocity (W/F) [27]. Therefore it is necessary to

study at different conditions the behavior of the reactants and the products involved. This

kind of analysis is related to the reaction rate and may be modelled by different

mathematical expressions in order to represent the reaction mechanism which is taken

place in the catalyst surface.

However, thorough discussions of complete reaction schemes seldom appear, precise rate

expressions are difficult to obtain because of the existence of reaction networks [29]. As

example, the simplified initial steps in toluene oxidation by hydrogen transfer mechanism

are show in Figure 6. In the first step, loosely bound, physically adsorbed or hydrogen-

bonded toluene is subjected to a rate-determining hydrogen abstraction forming an OH

group and m-complexed benzyl species of radical character. If the benzyl radicals react


with neutral toluene molecules the coupling products of Route 2 will be obtained. A high

concentration of the benzyl radicals would favoring the formation of dibenzyl by

recombination of two radicals. The same reactions will also occur to some extent in the

vapor phase due to desorption of benzyl radicals. As a parallel path it is found Route 3

since the benzylic species will form resonance structures with the charge or radical

character to some extent delocalized to the ring system in preferentially ortho and para

positions. A direct hydrogen abstraction in the nucleus requires a much higher energy than

in the methyl position and the total process is thought to be unfavorable. Nevertheless,

route 3 is almost negligible [78].

Figure 6. Hydrogen transfer mechanism for the initial steps in catalytic oxidation of toluene [78]

Some of the more reasonable paths are: (1) electron transfer, (2) electrophilic substitution,

(3) nucleophilic substitution within the coordination sphere of “soft” metal ions. These

mechanisms depend on the nature of the metal compound and the ionization potential of

the aromatic compound. The electron transfer mechanism is show in Figure 7. In process



(1), the first step is a reversible transfer of one electron to the metal oxidant with the

formation of a cation radical. This is considered to be a slow and rate-determining step

which is followed by a faster one in which the cation radical subsequently releases one

proton forming the benzyl radical. There is a certain probability for a direct substitution

reaction with nucleophilic oxygen, which will lead to route 3. A proton release (vide supra)

from the nucleus as a competing process to methyl proton release is also feasible. The

methyl-phenyl radical thus formed will also lead to route 3 [78].

Figure 7. Electron transfer mechanism for the initial steps in catalytic oxidation of toluene [78]


Results of steady-state and transient kinetics studies have helped to propose several

mechanisms, i.e. a parallel-consecutive reaction scheme for partial toluene oxidation over

vanadia/titania catalyst [79].

Kinetic expressions are usually of the power-rate law type and are applicable within limited

experimental ranges [80]. The oxidation of toluene on copper catalysts doped with

molybdenum and tungsten oxides on silt between 350 and 450 °C showed that reaction

was zero order with respect to the hydrocarbon (at concentrations from 5% to 30%) and

first order with respect to oxygen (at concentrations from 5% to 25%) [81]

Mars & Van Krevelen redox mechanism is very often used for selective and total oxidation

reaction. In their study utilizing vanadium oxide catalysts, Mars and Van Krevelen

investigated the oxidation of benzene to benzoquinone, maleic anhydride, CO2, CO and

H2O, toluene to benzaldehyde and benzoic acid, naphthalene to naphthoquinone and

phthalic anhydride, and anthracene to anthraquinone and phthalic anhydride. Also applied

their rate equation to data reported in the literature for SO2 oxidation to SO3 over various

vanadium oxide-based catalysts [82]. Their reaction mechanism can be written as:

I. Aromatic compound + oxidized catalyst 𝑘1→ oxidation products + reduced catalyst

II. Reduced catalyst + oxygen 𝑘2→ oxidized catalyst

The following assumptions were stablished [82]:

(1) The reaction in step I is first order with respect to the reactant and the fraction of sites

covered by oxygen ();

(2) Certain lattice O ions at the surface are involved in the oxidation reaction in step I;

(3) Only O ions (atoms) are assumed to exist on these sites (no reactant or products);

(4) The rate of surface re-oxidation (step II), i.e., O2 adsorption, is proportional to 𝑃𝑂2𝑛 and

to the concentration of active sites not covered by oxygen (1-).

The reaction network for toluene oxidation have been modelled assuming the redox

mechanism proposed by Mars and van Krevelen, the approach is oversimplified with



respect to the true reaction mechanism, but provides a better description of the

experimental data than the empirical approaches (power-law equations) [81]. Kinetic

measurements in different kind of oxides supports this affirmation, i. e. Co –Ni - Cu- Zn

oxides catalysts [83], WO3-MoO3 catalyst [84], and V2O5 catalyst, which showed that the

reaction was second order with respect to the hydrocarbon and zero order with respect to

oxygen. The mechanism fitted the experimental data, but the authors did not take into

account the existence of secondary reactions [81].

The oxidation of toluene by air over pure bismuth molybdate catalyst, considering a

complex reaction system and indicated the named mechanism fitted the experimental data.

The reaction was found to be first order in the hydrocarbon partial pressure and first order

in oxygen partial pressure [85].

In this way, kinetic studies of the oxidation of unsubstituted and substituted toluene over

molybdenum trioxide showed that the reaction was controlled by the rate of the reduction

of the catalyst, and the initial rate was first order in toluene and was independent of the

concentration of oxygen. The results were interpreted in terms of a kinetic expression of

the power-rate law type [41].

1.10. Scope

As it was mentioned, the current oil global market tends are pushing the refinery facilities

to change in order to improve their processes and to manage different raw materials. This

scenario leads new tasks to accomplish, for instance to handle toxic sub-products steams.

Such a case is the toluene, a platform molecule which could be oxidize to obtain

benzaldehyde, a simple hydrocarbon well known and used in fine chemistry.

Nevertheless, there are many factors affecting the toluene partial oxidation to

benzaldehyde when this is performed in vapor phase so it avoid mass and heat problems.

Most of the issues in this reaction are related to the catalyst, leading vanadia and molibdena

catalyst as better choices to achieve high selectivity and conversions.

Since vanadium compounds are related to hydrogen transfer mechanism which leads to

total oxidation, molybdenum compounds are preferred because of the great selectivity,


especially mixed with others oxides in order to stabilize the material and control its reduction

rate.

This work aims to contribute with the characterization of a cerium- zirconium - molybdenum

mixed oxide as an active material for the partial oxidation of toluene to benzaldehyde, and

the adjust of reaction phenomena by a kinetic expression.

2. Vapor phase toluene partial oxidation: the reaction at laboratory scale

The partial oxidation of toluene in vapor phase involves different considerations to take into

account not only the type of catalyst to use but the reaction conditions in order to prevent

possible oxygen – toluene explosive mixtures and to improve reaction selectivity and yield.

In this chapter, different considerations are selected to study the reaction and the

equipment to guarantee them are present.

2.1. Thermodynamic study

Duarte [57] presented the equilibrium species distribution for reactions given between

200°C and 800°C, changing the toluene to oxygen molar ratio between 10:1 to 1:10,

concluding that the most selective to benzaldehyde temperature up from 400°C until 600°C

using a toluene to oxygen feeding ratio of 1:1.

Figure 8. Toluene molar conversion as function of temperature and Oxygen/Toluene molar feed.

Using Aspen Plus ® this information was corroborate simulating the equilibrium state by

minimizing the formation free Gibbs energy of the different reactants between 330°C to

330

415

500

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.30 0.40 0.79 0.991.32

1.98

Temperature (°C)

Tolu

ene

con

vers

ion

Oxygen/Toluene molar feed ratio



500°C (this range was set taking into account the results reported by Duarte [57] and

avoiding possible Mo species sublimation [84]). Results confirm the increasing toluene

conversion at higher temperatures and oxygen feeding concentration, present by figure 8.

Figure 9. CO molar selectivity as function of reaction conditions

Also, the results confirm the high selectivity to CO at high temperatures and great selectivity

to CO2 at low temperatures, as it is present in figures 9 and 10.

Figure 10. CO2 molar selectivity as function of reaction conditions

0.3

0.4

0.8

1.01.32.0

0

0.2

0.4

0.6

0.8

330373

415458

500


CO

mo

lar

sele

ctiv

ity

Temperature

0.3

0.4

0.81.0

1.32.0

0.00

0.20

0.40

0.60

0.80

1.00

330373

415458

500


CO

2m

ola

r se

lect

ivit

y

Temperature (°C)

2. Vapour phase toluene partial oxidation: the reaction at laboratory scale 45

CO, CO2 and water are the main products at equilibrium conditions, the benzaldehyde

production increase at high temperatures and small oxygen to toluene feeding ratios, but it

is almost negligible. This information is very important, because it implies the great

importance of the selectivity to benzaldehyde that should improve the catalyst. Also,

according with this brief analysis, the possible conditions to study the reaction were fixed,

and will be present later on.

2.2. Reaction system

The reaction arrangement were carefully chosen, including the feeding system, the reactor

and the analytical equipment, as it is shown in the Figure 11.

At normal conditions (P=1 atm, T=25°C) toluene is liquid, so it is necessary to evaporate it

by changing pressure or temperature. Oxygen may be used pure or mixed (i.e. air) [84,85],

nevertheless, the employment of pure oxygen lead others compounds as possible intern

standards to quantify the outflows. Since not only toluene and oxygen should be injected in

the system, inert compounds whom dilute the reactants, prevent risky mixtures and stabilize

the inlet flow pressure should be part of the feed. Therefore, three individual lines for the

Toluene

Pump

Mixer

Ar

O2

N2

F

F

F

GC

Atmosphere

Reactor

Condensable

products

Non- condensable

products

Cold trap Isopropanol

trap

Catalytic bed

Quartz

wool

Quartz

chips

Oven

Gas lines

Liquid introduction

Products

Temperature

control

Figure 11. Experimental arrangement for toluene oxidation



gases were arranged: the first one for the introduction of carrier gas (argon), the second

one for the oxygen and a third line for the internal standard (nitrogen).

To guarantee stable conditions the flow rate was controlled using mass flow controllers

(Brooks Instrument). The different gases were mixed before the reaction zone in a ¼ ID

tube packed with crushed quartz (pink lines). The toluene (blue line) was pumped using a

305 Gilson micro pump and introduced into the reactor by a needle. Before the reactor, a

“T” connection was disposed to allow the introduction of liquid and gases.

The feeding system was coupled to the reactor, the one used was a quartz straight reactor

with a length of 30 cm and an internal diameter (ID) of 7 mm. A cylindrical oven of 20 cm

of length was used coupled to a k-type thermocouple.

After the reactor, the line was heated and controlled at 200 °C. The reaction products (red

line) were separated into non-condensable and condensable products. The condensable

products were recovered in liquid phase using a dry trap maintained at 0 °C. Recovery of

condensable products giving by temperature only, leads to toluene lost, as it was calculate

using the software Aspen Plus ®. Figure 12 presents the recovered volume fraction of the

main outlet condensable compounds in function of the trap temperature.

Figure 12. Recovered volume fraction of the main outlet condensable compounds in function of the trap temperature

0

0.2

0.4

0.6

0.8

1

1.2

-100 -80 -60 -40 -20 0 20 40 60 80

Rec

ove

r v

olu

me

frac

tio

n

Trap Temperature (°C)

Benzaldehyde Toluene


Therefore, it was placed a second trap which contained isopropanol so the affinity of the

hydrocarbons reduce the mass losses [86].

The gas permanent products were analyzed by on-line gas phase chromatography. The

condensable products were stocked (4°C) and analyzed later on.

The analysis of products was performed according to its recuperation after reaction. The

non-condensable products present in gaseous phase were characterized online by gas

phase chromatography (GPC). The analysis was performed every 25 min using a Network

GC system Agilent Technologies 6890N with TCD detector. The column used was a

Carbosieve II. The products analyzed were H2, O2, CO, CO2, and N2 (internal standard).

The chromatographic method is presented in Annex A.

The analysis of condensable products recuperated in liquid phase, was performed by gas

chromatography using a Network GC system Agilent Technologies 6890N with a FID

detector. The column was a SGE Sol gel Wax (dimensions: 60m length, 0.25mm internal

diameter and 0.25m film thickness). The products analyzed were toluene, benzene and

benzaldehyde. The external standard used was isopropanol. The chromatographic method

is presented in annex A.

2.3. Ce–Zr–Mo oxydes as catalysts

Not only the reaction setup was carefully chosen, the catalysts to evaluate were proposed

in order to obtain better structural stability, keeping the catalytic performance at reaction

conditions. This is not a simple target considering the behavior of each metal oxide to mix,

but there is information to account in order to formulate possible catalytic solids.

As it was present before, the different phases between the CeO2–ZrO2 binary mixes (fluorite

type cubic, fluorite type cubic and tetragonal, monoclinic and tetragonal) are known [70-

72]. At low loading amount of Mo in Ce-Zr oxides matrix, the isolate MoO42- species is the

primary molybdena species, but polymeric molybdate species predominates when the Mo6+

loading amount is increased. The molybdena is preferentially dispersed on the surface of

CeO2 at lower loading because the base strength of CeO2 is stronger than that of ZrO2 [74].



This information is very important, but it didn’t clarify much the scene. Other way to

formulate the catalysts is to start by known stable compounds and try to include the third

metal compound, assuming the oxidation state to make the oxygen approach. In this way,

three formulations were proposed to synthetize, starting by Ce0.5Zr0.5O2. The first one

substituting half of the oxygen by molybdenum oxide, the second one substituting three

quarters of the oxygen and finally by the substitution of all the oxygen, as follows:

Table 1. Catalyst proposal development

4(𝐶𝑒𝑂2)

4(𝐶𝑒0.5𝑍𝑟0.5𝑂2)

𝑪𝒆𝟐𝒁𝒓𝟐𝑶𝟖

𝐶𝑒2𝑍𝑟2𝑂4(𝑀𝑜𝑂4)4

𝐶𝑒2𝑍𝑟2𝑀𝑜4𝑂20

𝑪𝒆𝒁𝒓𝑴𝒐𝟐𝑶𝟏𝟎

𝐶𝑒2𝑍𝑟2𝑂2(𝑀𝑜𝑂4)6


𝑪𝒆𝒁𝒓𝑴𝒐𝟑𝑶𝟏𝟑

𝐶𝑒2𝑍𝑟2(𝑀𝑜𝑂4)8


𝑪𝒆𝒁𝒓𝑴𝒐𝟒𝑶𝟏𝟔

There are several ways to synthetize plurimetallic oxides, e. g. co-precipitation, electrostatic

adsorption and impregnation [87], like those, there is the pseudo sol-gel method, a

derivation of the standard sol-gel scheme, where the reactivity of the departure precursors

is affected before reaction in order to form a polymeric mixed precursor in solution. This

solution is later transformed into a solid resin by controlled evaporation, which after drying

and calcination leads to a mixed oxide with a well-defined structure and homogeneity

composition [88].

This method was the one selected for synthetize the catalysts, the precursor salts used

were cerium (III) acetate hydrate, zirconium (IV) acetylacetonate and bis-acetylacetonate

of molybdenum dioxide. The precursor salts were dissolved in propionic acid, with a cation

concentration of 0.12 mol.L-1 [89], but for the molybdenum salt the solution the cation

concentration was 0.006 mol.L-1. The different solutions formed were then mixed and kept

under total reflux for 1h. This procedure ensured the effective formation of the polymeric

precursor in solution, and, a homogeneous mixture, as it was probed [88]. The elimination

of the organic solvent was carried out by vacuum distillation. After distillation the gel

obtained was calcined at 550°C to prevent Mo-compounds sublimation [84] for 6 h, using

a heating rate of 2 °C.min-1. The global procedure is presented in the Figure 13.


time (h)

Task 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Dissolve precursor salts

Mix the propionates

Mix Ce propionate

Mix Zr propionate

Mix Mo propionate

Vacuum distillation

Calcination

Figure 13. Global catalyst preparation procedure following the pseudo sol-gel method [89]

Since, considerable amounts of propionic acid should be used during each synthesis

procedure, mainly related to the molybdenum salt dilution, a modification in the methods is

implemented. The catalyst proposed were done as it is show below, adding the

molybdenum salt by equal batches during the vacuum distillation (formation of the resin)

and using the evaporated and recuperate solvent of each batch to dilute the next one.

time (h)

Task 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Dissolve precursor salts

Mix the propionates

Mix Ce propionate

Mix Zr propionate

Mix Mo propionate

Vacuum distillation

Calcination

Figure 14. Catalyst preparation procedure following the modified pseudo sol-gel

method

2.3.1. Modified pseudo sol gel method catalysts

Once it was synthetized the catalysts and knowing their expected composition and their

real composition, calculated with the quantity and purity of the primary salts, which is

presented in Table 2.

.



Table 2. Solids synthetized by the modified pseudo sol-gel method

The catalytic behavior of mixed metal oxides depend on the species of the surface, which

are related to the nature of the solid, the content of each metal oxide and the calcination

temperature [66]. Then, it is very important to characterize the synthetized catalysts by

different methods in order to correlate the catalytic performance with the chemical

properties.

The redox properties of the catalyst are of great importance thus they have been proven to

affect the catalytic activity and stability of metal catalysts. These properties may be studied

by temperature programmed analysis, like Temperature Programmed Reduction (TPR) and

Temperature Programmed Oxidation (TPO).

Figure 15. Synthetized CZM2(I), CZM3

(I), CZM4(I) catalysts samples TPR analysis

results

0

100

200

300

400

500

600

700

800

900

1000

0

0.05

0.1

0.15

0.2

0.25

0.3

0 20 40 60 80 100 120 140

Tem

per

atu

re (

°C)

TCD

sig

nal

(a.

u.)

time (mins)

Theoretical molecular formula

Name

Theoretical metal ratio

(M mol/Zr mol)

Expected metal ratio

(M mol/Zr mol)

Ce Zr Mo Ce Zr Mo

CeZrMo2O10 CZM2I 1 1 2 1.04 1.00 2.00

CeZrMo3O13 CZM3I 1 1 3 1.04 1.00 2.99

CeZrMo4O16 CZM4I 1 1 4 1.05 1.00 4.01

CZM2(I)

CZM4(I)

CZM3(I)

Temperature


There are four principal picks expected in a TPR analysis of a solid composed of the metal

oxides involved in the presented formulations, the 500°C peak and the 852°C peak related

with the Ce4+ to Ce3+ change (surface and bulk, respectively) [90]; as well as the 700-770°C

peak and the 900°C peak, related with the Mo6+ to Mo4+ change and the Mo4+ to Mo0

change, respectively [91].

Certainly, the zirconia has also changes, at 610°C and 690°C approx.(related with the Zr4+

to Zr3+ in the surface and in the bulk, respectively), but the signal usually is weak [92].

Samples of the three formulations were analyzed and the results are showed in the Figure

15, the TCD signal profiles present only two principal peaks, which may be related to the

mixed of the different peaks already listed.

Different authors noticed changes in the reduction behaviour of a single metallic oxide with

the presence of a metal or a guest metallic cation in the oxide structure, resulting in the

shift in the temperature of the main H2 consumptions [90]. The profiles of temperature

reduction change as the molybdenum content increase, making every catalyst more stable

each time. And that is the main effect present by the TPR profiles, the CZM2I sample, whom

contain of Mo is the lowest, may exhibit changes in the work operation range temperature

(represent in Figure 15 by the red shape), but the conditions of the reduction are different.

Its reduction begins at 450°C, meanwhile CZM3I sample starts to reduce at 500°C and

CZM4I sample reduces from 650°C.

The presence of oxygen vacancies in the structure of metal oxides of transition metals is

decisive in the redox reactions in which they participate, aside from being fundamental in

the reaction mechanism in catalytic oxidation processes [93]. It has been reported that the

activity of the catalysts is influenced by the presence of the type of oxygen species on the

solid surface, which are consumed in the course of the reaction and can be regenerated

through the presence of an oxidative atmosphere [93]. Then, it is very important to stablish

if the metal species change dramatically because of the excess of oxygen and that

information is giving by TPO analysis results.

Figure 16 present the TPO results for CZM2(I), CZM3

(I) and CZM4(I) samples. As it is present,

there is a small weight loss before 100°C for the CMZ2I sample that may be associate with

the presence of water on the surface. The interaction between and oxidative atmosphere

and the selected samples did not change the weight of any of them, so, there is not reasons

to assume changes in the initial structures giving by the metal- oxygen interaction or the



temperature changes (it has been report sublimation of molybdenum species up to 600°C

[28, 59]).

Figure 16. Synthetized catalysts samples TPO analysis results: (blue) CZM2(I), (red)

CZM3(I), (green) CZM4

(I).

Characterization using X-ray Diffraction (XRD), Raman spectroscopy and Scanning

Electron Microscopy (SEM) were performed to study the some physicochemical properties

of the mixed oxides such as crystallinity, surface morphology, among others. The XRD

analysis results for the three fresh samples, CMZ2(I), CMZ3

(I) and CMZ4(I) are present in

Figure 17. No single oxides corresponding to diffraction peaks were detected (CeO2, ZrO2,

MoO3) then, it may suggest that preparation method allows the crystallization of the mixed

oxides without any single phases rejection. Nevertheless, only corresponding peaks to one

cerium-molybdenum oxide triclinic structure (Ce2(MoO4)2(Mo2O7) SS-VVV-PPP: 01-076-

1040) fit the two main peaks, but the matching is not precise. Peaks related to Ce8Mo12O49

are also present and fit one of the main peaks but their match is also not precise and

predominate at low Mo load, contrary to Ce(MoO4)2 related peaks. Binary Zr – Mo

compounds are evidenced at low Mo load, exhibiting monoclinic structures.

According with these X-ray diffraction patterns, peaks related to Ce2(MoO4)2(Mo2O7)

structure are present but the peak shift at 24.2° and the 26.5° peak lack may occur as effect

of Zr insertion in the structure. Then, it is possible to suggest the integration of the three

metal oxides.

95

96

97

98

99

100

101

102

0 100 200 300 400 500 600 700 800

Wei

ght

per

cen

taje

Temperature (°C)


Figure 17. Diffractograms of fresh mixed oxides: CZM2(I), CZM3

(I), CZM4(I).

■Ce(MoO4)2 ♦Ce8Mo12O49 ●Ce2(MoO4)2(Mo2O7) ▲Zr(MoO4)2

Figure 18 present the Raman analysis result of the three fresh samples, CMZ2(I), CMZ3

(I)

and CMZ4(I). The predominant bands associated to MoO3 are 285, 335, 378, 667, 818, 995

cm-1, their intensity decrease when the sample Mo content increases.

Figure 18. Raman spectra of fresh mixed oxides: CZM2(I), CZM3

(I), CZM4(I),

15 25 35 45 55 65 75 85

Inte

nsi

ty (

a.u

.)

2 (°)

0 100 200 300 400 500 600 700 800 900 1000 1100 1200

Inte

nsi

ty (

a.u

.)

Raman shift (cm-1)

CZM2(I)

CZM4(I)

CZM3(I)

CZM2(I)

CZM4(I)

CZM3(I)



Meanwhile, the highly dispersed molybdate species became more important as the bands

at 332, 856, 946 cm-1 [66] turn out to be the main bands when the sample Mo content

increases.

The bands attributed to Ce8Mo12O49 are present but shifted, although normally appear at

280, 410, 680, 850 and 946 cm-1 [66]. That result may be attributed to the effect of the

zirconium inclusion on the Ce-Mo oxide structure or to the generation of oxygen deficiency,

which changes the molybdate species presented on the surface.

CeO2–ZrO2 mixed oxides bands are no present in the Raman spectra of the different

samples, these bands are usually centered at 130, 300, 460, 540 and 620 cm-1 [94].

The Raman spectra suggest that molybdenum oxide species variably exist as highly

dispersed molybdate species and MoO3 with increasing of Mo content.


(I), c. CZM4(I)

CZM2(I), CZM3

(I) and CZM4(I) samples morphology was evidenced by Scanning Electron

Microscopy, Figure 19 present the micrographs.

These images exhibit the preference for define structure as the Mo content grows, forming

grains of 0.25m average diameter. Meanwhile at low Mo – content, the structure is not

defined and present high porosity.

According to the analysis made from the results of X-ray Diffraction (XRD), Raman

spectroscopy and Scanning Electron Microscopy (SEM), the changes of structural

properties evidence the predominant role of the Mo species and the presence of the

modified Ce8Mo12O49 structure.

The catalytic behavior of the CZM2(I), CZM3

(I) and CZM4(I) samples was also evaluated at

three temperatures (450°C, 475°C and 500°C) using the setup presented in Figure 11. The

toluene’s feed rate was 30L/min in liquid phase and oxygen’s feed rate was 5mL/min in

gas phase, nitrogen as internal standard and argon as balance gas were also feed. Taking

a b c


into account the stabilization of the pump, it was started at the same time as the reactor

oven, but the reactants where mixed after the reactor reach desired temperature. Once the

reactants mixture stabilized, the tramps were connected and every 30 minutes the liquid

products where collected. The other two temperatures were evaluated by the same

procedure but the reactor temperature changes was performed without toluene feed. The

reactants were feeding and mixed until stable regime after reach the desirable conditions.

Test methodology schema is resumed in figure 20.

Figure 20. Catalytic test temperature profile vs time

For each test 0.1 g of powder sample of solid catalyst was load. W/F ratio was keep at

0.162 kgh-1m-3, and the constant GHSV relation was fixed at 20360 h-1. The catalytic

behavior results of the different mixed oxides are summarize in Table 3.

Table 3. Catalytic test results of: CZM2(I), CZM3

(I), CZM4(I) samples.

Catalyst Conversion (%) Selectivity (%)

450°C 475°C 500°C 450°C 475°C 500°C

CZM2(I) 31 35 39 05 05 05

CZM3(I) 32 32 36 13 12 11

CZM4(I) 36 37 37 12 12 10

0

100

200

300

400

500

600

0 100 200 300 400 500

Tem

per

atu

re (

°C)

Time (minutes)

O2, C7H8, N2 and Ar feeding





As it is present, the conversion variation between the different samples is less than 10%,

then the selectivity may be the key result to take into account to choose the catalyst whit

best performance.

Catalytic test results present small variations related to cerium, zirconium or molybdenum

load of the samples, but it is not conclude. It does not exhibit a direct relation between metal

composition and selectivity, neither to benzaldehyde nor CO and CO2, as it is present in

annex B.

By the other hand, selectivity results are not favorable comparing with reported results for

other catalysts (20-80% [57]). As it was state by the characterization analysis, the

molybdenum species are predominant on the catalyst surface, promoting total oxidation,

then the change during the materials synthesis may affect the catalysts performance.

2.3.2 Pseudo sol gel method catalysts

A second family of catalyst of three different Ce-Zr-Mo oxides with the same theoretical

molecular formula, but using the pseudo sol gel method, presented in Figure 13, was

synthetized. The information of the samples is summarized in Table 4.

Table 4. Samples synthetized by the pseudo sol-gel method

Theoretical molecular formula

Name


(Mmol/Zrmol)

Real metal ratio (M mol/Zr mol)

Ce Zr Mo Ce Zr Mo

CeZrMo2O10 CZM2II 1 1 2 1.04 1.00 2.01

CeZrMo3O12 CZM3II 1 1 3 1.05 1.00 3.01

CeZrMo4O16 CZM4II 1 1 4 1.0 1.0 3.90

The three new catalysts were analyzed by X-ray Diffraction (XRD). The CMZ2(II), CMZ3

(II)

and CMZ4(II) XRD diffraction patterns are present in figure 21.


Figure 21. Diffractograms of fresh mixed oxides: CZM2(II), CZM3

(II), CZM4(II).

–MoO3 ■Ce(MoO4)2 ♦Ce8Mo12O49 ●Ce2(MoO4)2(Mo2O7) ▲Zr(MoO4)2

The preparation method allows the crystallization of the mixed oxide with phase rejection,

as MoO3 is detected in CMZ4(II) X-ray diffraction pattern. By the other hand, the

diffractograms change significantly comparing with the presented for the samples, CMZ2(I),

CMZ3(I) and CMZ4

(I). In this case, peaks change drastically as the composition variate. For

low Mo content (sample CMZ2(II)) X-ray diffraction patterns evidence the presence of binary

compounds of the different oxides. As the Mo content grows, Ce - Mo binary oxides are

preferred. For CMZ3(II) sample, Ce(MoO4)2 is predominant, but as Mo load increase for

CMZ4(II) its diffractogram became more likely to the CMZX

(I) samples X-ray patterns.

The main difference between the X - ray diffraction patterns of the family samples lead a

structural effect as result of the synthesis method. This assumption was confirmed by

electron diffraction pattern of two samples with the same theoretical formula, but different

synthesis procedure. The images, presented in Figure 22 where taken on CZM2(I) and

CZM2(II) samples. These micrographics confirm the poly crystallinity of both materials,

although CZM2(I) sample present less phases than CZM2

(II) sample.

15 25 35 45 55 65 75 85

Inte

nsi

tiy

(a.u

.)

2 (°)

CZM4(II)

CZM3(II)

CZM2(II)



Figure 22. Electron diffraction of fresh mixed oxides: a. CZM2(I), b. CZM2

(II)

The same result was obtain by Transmission electron microscopy (TEM). The micrographs

of CZM2(I) and CZM2

(II) samples are present in Figure 23. The images corroborate the poly

crystallinity of both materials, although CZM2(I) sample present less different kind of crystals

with 20 nm average diameter. .

Figure 23. TEM micrographs of fresh mixed oxides: a. CZM2(I), b. CZM2

(II)

The synthesis method effect was also evidenced by Scanning Electron Microscopy, Figure

24 present the micrographs of the CZM2(I).and CZM2

(II) samples with equivalent theoretical

a b

a b


formula, but different synthesis procedure, these images exhibit structures completely

different, with more dispersed grains (internal diameter of 0.3m) for the traditional

synthesis methodology.


(II)

Finally, at the same conditions, given by figure 20, the catalytic tests were made,

maintained the catalyst load at 0.1 g for each test (particle diameter between 315m and

500m), W/F ratio was keep at 0.162 kg.h-1.m-3, and GHSV relation constant at 20360 h-1.

The general results of the different catalytic test are summarize in Table 5.

Table 5. Catalytic test results of: ZM2(I), CZM3

(I), CZM4(I) samples


450°C 475°C 500°C 450°C 475°C 500°C

CZM2(II) 33 40 41 23 17 13

CZM3(II) 36 38 39 24 24 18

CZM4(II) 35 38 41 26 16 06

The conversion present for the different samples had not change dramatically, leading once

again the selectivity as the key factor to evaluate the catalysts performance. The results

clearly change comparing with the obtained for CMZ2(I), CMZ3

(I) and CMZ4(I)samples. For

a b



these cases is evidenced an inverse relation between selectivity and temperature, but it is

not precise the role of each metal on the reaction selectivity.

Once again none of the evaluated metal oxides reach high selectivity comparing with similar

catalyst (20-80%) [57]. The main reason for this situation may be the metal ratio proposed

for the formulated catalyst, previous studies referred Mo/(Ce+Zr)=1 as a suitable relation

for toluene partial oxidation catalyst [57]. However, only CMZ2(I) and CMZ2

(II) samples have

this metal ratio, but both of them exhibit the lowest selectivity comparing with CMZ3(x) and

CMZ4(x) samples. Then, even when the catalyst metal ratio match this relationship, the

solids formulations proposed so far may not have the most selective metal oxides

proportions.

It is important to notice that not only the metal content of the solids affects the catalytic

performance, but also the species remaining on the surface of the materials as it was

showed. For CMZx(I) samples, the synthesis method lead Mo- compounds on the surface

which improve the CO and CO2 selectivity comparing with CMZx(II) samples.

2.3.3 Influence of cation integration vs deposition

Ce-Mo mixed oxides deposits on Ce – Zr mixed oxides supports have been proposed [57].

Among the different formulations valuated, the best results obtained was for

Ce2(Mo3O4)3/(Ce0.65Zr0.35O2). Giving the great influence of the synthesis method on the

catalyst performance, two samples of this formulation were prepared:

The first one was deposits, the Ce-Zr oxide support was previously synthetized using the

pseudo sol like method and subsequently the active phase was introduced on the support

by the inclusion of the Ce-Mo propionate resins, as it is present by figure 26. The second

sample, with the same Ce-Mo-Zr oxides content was prepared following the pseudo sol gel

method (Figure 13).

For both cases the precursor salts used were cerium (III) acetate hydrate, zirconium (IV)

acetylacetonate and bis-acetylacetonate of molybdenum dioxide, the elimination of the

organic solvent was carried-out by vacuum distillation and after so the gel obtained was


calcined at 550°C to prevent Mo-compounds sublimation [84] for 6 h, using a heating rate

of 2 °C.min-1.

time (h)

Task 1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

Precursor salts

disolution

Mix of the propionates

Ce propianate

mix

Zr propianate

mix

Mo propianate

mix

Vacuum distillation

Mix of support and

active phase

Calcination

Figure 25. Global catalyst preparation procedure for deposited catalysts, following the pseudo sol gel method

Table 6 presents the expected composition of the samples and their real composition

calculated with the quantity and purity of the primary salts.

Table 6. Third catalyst family synthetized

Theoretical molecular formula Name

Theoretical metal ratio

(Mmol/Zrmol)


(M mol/Zr mol)

Ce Zr Mo Ce Zr Mo

Ce2(Mo3O4)3/(Ce0.65Zr0.35O2) CM/CZI 7.6 1 8.6 7.78 1.00 8.50

Ce2.65Zr0.35Mo3O14 CM/CZII 7.6 1 8.6 7.75 1.0 8.55

The CM/CZI and CM/CZII catalysts were analyzed by X-ray Diffraction (XRD) (see Figure

26). The preparation method allows the crystallization of the mixed oxides without any



phase rejection and no single oxides are detected (CeO2, ZrO2,MoO3). For both samples it

is present peaks related with binary compound of the three metals, for the CM/CZ(I) sample

the Mo based compounds diffraction pattern are preferred, meanwhile for the CM/CZ(II)

sample, present less well defined peaks, related with less crystallinity.

Figure 26. Diffractograms of fresh mixed oxides: CM/CZ(I), CM/CZ(II). ■Ce(MoO4)2 ♦Ce8Mo12O49 ●Ce2(MoO4)2(Mo2O7) ▲Zr(MoO4)2

The difference between the X - ray diffraction patterns of the samples lead a structural effect

as result of the synthesis method. This correlation is also confirm by electron diffraction

made to both samples, showed in Figure 27.

Figure 27. Electron diffraction of fresh mixed oxides: a.CM/CZ(I), b. CM/CZ(II).

15 25 35 45 55 65 75 85

Inte

nsi

tiy

(a.u

.)

2 (°)

a b

CM/CZ(II)

CM/CZ(I)


The electron diffraction of fresh CM/CZ(I) and CM/CZ(II) samples, confirm the poly

crystallinity of both materials. Although CM/CZ(II) sample present more intense peaks,

perhaps related to binary well crystalize compounds.

Transmission electron microscopy (TEM) analysis images of fresh CM/CZ(I) and CM/CZ(II)

samples are present in Figure 28.

Figure 28. TEM micrographs of fresh mixed oxides: a. .CM/CZ(I), b. CM/CZ(II).

The micrographs confirms the presence of different kind of crystals. For both materials, well

faceted crystals are evidenced. In the images trapezoidal well defined grains are present

(particle dimensions: 10nm to 20 nm average) related to Mo-compounds. For the CM/CZ(II)

sample another well-defined kind of grain is evidence (internal diameter: 20nm average)

related to fluorite type cubic Ce-Zr compounds.

Scanning Electron Microscopy (Figure 29) present the micrographs of fresh CM/CZ(I) and

CM/CZ(II) samples, which exhibit ceramic structures completely different, leading

heterogeneous surface for CM/CZ(I) sample and dispersed grains for CM/CZ(II) sample.

Therefore, it is possible to state that synthesis method affect the final oxide obtained.

a b



Figure 29. SEM micrographs of fresh mixed oxides: a. .CM/CZ(I), b. CM/CZ(II).

Finally, at the same conditions, given by figure 20, the catalytic tests were made,

maintained the same stipulations of catalyst load at 0.1 g for each test (particle diameter

between 315m and 500m), W/F ratio was keep at 0.162 kg.h-1.m-3, and GHSV relation

constant at 20360 h-1. The general results of the different catalytic test are summarize in

table 7 and annex B.

Table 7. Catalytic test results of .CM/CZ(I) and CM/CZ(II)samples


450°C 475°C 500°C 450°C 475°C 500°C

CM/CZ(I) 41 45 47 12 11 07

CM/CZ(II) 37 37 40 28 24 20

When the formulated catalysts were evaluated at the same conditions as the others

catalysts, the results remained in similar ranges. This behavior implies that the GHSV

condition affects directly the conversion reach by the evaluated catalyst, as Duarte stated

[57].

The catalytic test results are very promising, because the highest selectivity was reach by

mixture of the three metal oxides, not by the deposited catalyst. It is also important to notice

the similar results obtained by CMZ3(I) sample and CM/CZ(I) sample, which may suggest a

different relationship between the metal content and the selectivity to benzaldehyde.

As it was state before, the synthesis method lead different surface species which affect the

catalytic performance of the materials. For instance, the effect of cation deposition leads

a b


Mo-Ce compounds on the solid surface as it was proved by XRD, SEM and TEM analysis.

These species improve the CO and CO2 selectivity, unlike the cation integrate samples.

As the best catalytic results were obtained with CM/CZ(II) sample, this one was select to do

the kinetic study.

3. Kinetic study

The design of a catalytic reactor involves the need for a correlation relating the reaction

rates and the partial pressures of the reactants and products, for a particular type of

catalyst. The most widely used rate expressions are those based on the extended Langmuir

Hinschelwood theory, which are often referred to as Hougen and Watson rate expressions

(LHHW) [95]. The selection of the most appropriate rate expression is a lengthy process,

involving postulation of reaction mechanisms, isothermal regression for discarding

obviously inappropriate mechanisms, nonisothermal regression, and the use of

physicochemical criteria for selecting the most appropriate one from among competing

mechanisms [95].

Global expressions takes into account mass and heat transfer phenomena, which is related

to extrinsic effects such as diffusional limitations. Also, it considers temperature and

interactions between reactants and catalytic surface, among others effects which influences

directly the reaction rate and are associate to the intrinsic phenomena [96]. When reactor

and physical effects are not include in the mathematical equation, the expression is related

to the intrinsic rate [24]. In this chapter, an intrinsic kinetic correlation is developed.

3.1. Data collection

3.1.1. Experimental procedure

To obtain the kinetic model, three experiments were carried-out using the setup presented

in Figure 11 using 0.1g of catalyst (particle diameter between 315m and 500m), at three

different temperatures (450°C, 475°C and 500°C). For each experiment liquid toluene feed

was fixed (30, 40 or 50 L), and oxygen feed was increasing from 5 to 9 mL, nitrogen as

internal standard and argon as balance gas were also feed to the reactor, keeping constant

the gas flow (25mL/min) . Each arrangement was carried at three different temperatures

(450°C, 475°C and 500°C)

Taking into account the stabilization of the pump, it was started at the same time as the

heating procedure of the reactor oven using a by-pass. The reactants where mixed after

the reactor reach the desired temperature. Once the reactants mixture stabilized, the

tramps were connected and every 50 minutes the liquid products where collected. At each

condition two samples of recollected liquid products were taken.



The other two oxygen partial pressure were evaluated by the same procedure: for 20

minutes the oxygen inlet changes without toluene feeding, after, the reactants were feeding

and mixed until stable regime and then it was the data taken. Test methodology schema is

resumed in figure 30.

Figure 30. Test methodology schema.

The information related for each experiment is summarized in annex C.

3.1.2 Experimental results

The different experiments lead the oxygen and toluene dependency for the kinetic rate, as

it is present in figure 31, which summarize the results obtained for experiments made at

fixed oxygen feed rate of 7mL/min.

Figure 31 Kinetic rate’s toluene partial pressure dependency evaluated at 7mL/min oxygen feed rate. ▲450° ●475°C ♦500°C

0

1

2

3

4

5

6

7

8

9

10

0 100 200 300 400 500

Oxy

gen

fee

din

g (m

L)

time (min)

0

5

10

15

20

25

0 0.05 0.1 0.15 0.2 0.25 0.3

r (

mo

l/s*

g) (

E-6)

Toluene partial pressure (atm)




3..Kinetic study 69

Figure 32 presents the results for experiments made at fixed toluene feed rate of 30L/min.

Figure 32 Kinetic rate’s oxygen partial pressure dependency at 30L/min toluene feed rate. ♦450°C ■475°C●500°C

It is important to notice the different effect that implies the increment of each

reactant in the kinetic rate. The direct relationship implies an important role in the

kinetic mechanism for each case.

3.2 Determination of kinetic parameters

Different models may be evaluated using the method of Least Squares, which is a

procedure to determine the best fit line to data. The proof uses simple calculus and linear

algebra. The basic problem is to find the best fit straight line y = ax + b given that, for n ∈

{1, . . . , N}, the pairs (xn, yn) are observed.

The method easily generalizes to finding the best fit of the form:

𝑦 = 𝑎1𝑓1(𝑥) + ⋯+ 𝑐𝑘𝑓𝑘(𝑥);

It is not necessary for the functions fk to be linearly in x – all that is needed is that y is to be

a linear combination of these functions.

The parameters for each model were obtained minimizing the relative error between the

calculated rate and the experimental rate.

0

2

4

6

8

10

12

14

16

18

20

0.00 0.05 0.10 0.15 0.20 0.25 0.30

r (m

ol/

s*g)

(E-

6)

Oxygen partial pressure (atm)



3.2.1 Power law rate expression

The power law rate model is the simplest expression, which links the reaction rate and the

reactants pressure by reaction orders. The relation is giving by equation 1:

𝑟𝑇 = 𝐾𝑃𝑂2𝛼 𝑃𝑇

𝛽 Eq.1

In this case K is the specific reaction rate constant and is given by the Arrhenius Equation

[96]. The reaction orders are and . These number may be enters but it is not mandatory.

Table 8 summarize the parameters obtained by mathematical regression.

Table 8. Power law model kinetic parameters obtained

Temperature (°C)

Kinetic parameter

K

450 6.91E-4 0.1157 2.5329 475 4.33E-4 0.11088 2.0429 500 7.66E-4 0.1136 2.3202

The fitting of the model may be evaluated by parity diagram. The adjust of the calculated

rates and the experimental data is giving by the coefficient of determination, denoted R2

which is a number between 0 to 1 that indicates the proportion of the variance in the

regression. The results are present in figure 33.

Figure 33. Parity diagram for power law kinetic model

3..Kinetic study 71

As the coefficient of determination is not near to 1, which implies good accuracy of the

model, the orders were fixed to entire numbers as it was proposed for others studies which

showed that the reaction was second order with respect to the hydrocarbon and zero order

with respect to oxygen [81] and the obtained numbers were near to those ciphers. The new

calculated parameters are present in table 9.

Table 9. Modified Power law model kinetic parameters obtained

Temperature (°C)

Kinetic parameter

K

450 2.58E-4 0 2 475 4.01E-4 0 2 500 4.83E-4 0 2

The parity diagram present a higher determination coefficient, nevertheless, the correlation

is very small, which implies the assumptions to simplify the model are not accurate

Figure 34. Parity diagram for modified power law kinetic model

R² = 0.6603

5.0E-06

1.0E-05

1.5E-05

2.0E-05

2.5E-05

3.0E-05

3.5E-05

4.0E-05

5.0E-06 1.0E-05 1.5E-05 2.0E-05 2.5E-05

calc

ula

ted

rat

e (m

ol/

s*g)

observate rate (mol/s*g)



3.2.2 LHHW kinetic expression with two active sites

The Langmuir – Hinshelwood – Hougen –Watson (LHHW) model is based on a continuum

model, in which the surface of the catalyst is described as an array of equivalent sites which

do interact neither before nor after chemisorption. Furthermore, the derivation of rate

equations assumes that both reactants and products are equilibrated with surface species

and that a rate-determining step can be identified. Surface coverages are correlated with

partial pressures in the fluid phase by means of Langmuir adsorption isotherms.[96, 97].

As the model leads several hypotheses, the first approach will be taking into account the

following assumptions:

1. Two active site type

2. Non dissociative toluene adsorption.

3. Dissociative oxygen adsorption

4. Surface reaction as controlling step

The LLHW kinetic expression associate to these statements is represented by equation 2:

𝑟𝑇 =𝐾𝑘𝑇𝑃𝑇𝑘𝑂2𝑃𝑂2

(1+𝑘𝑇𝑃𝑇)∗(1+√𝑘𝑂2𝑃𝑂2) Eq.2

Table 10 summarize the parameters obtained by mathematical regression.

Table 10. Two active sites LHHW model kinetic parameters obtained

The parity diagram presented by figure 35 exhibit low determination coefficient, as the fitting

is not good enough, then the assumptions should be changed.

Temperature (°C)

Kinetic parameter K kT KO2

450 1.18E-2 1.21E-2 2.23E-2 475 7.87E-3 2.37E-1 2.37E-1 500 1.57E-2 1.85E-1 1.77E-1

3..Kinetic study 73

Figure 35. Parity diagram for two active sites LHHW kinetic model

3.2.3 LHHW kinetic expression with dissociative oxygen adsorption

Several theories may be consider in order to obtain a kinetic expression, nevertheless,

taking into account the probable reaction mechanism, the proposed assumptions for the

second LHHW model evaluated are:

1. Only one active site type


3. Dissociative oxygen adsorption

4. Competitive toluene and oxygen adsorption.


The LLHW kinetic expression associate to these assumptions is giving by equation 3:


(1+𝑘𝑇𝑃𝑇+√𝑘𝑂2𝑃𝑂2)2 Eq.3

The mathematical treatment of the data leads the regression of the different constants and

as result, the parity diagram presented by figure 36 exhibit higher determination coefficient,

but the fitting may improve, then the assumptions should be changed.

R² = 0.7447

5.0E-06

1.0E-05

1.5E-05

2.0E-05

2.5E-05

3.0E-05

5.0E-06 7.0E-06 9.0E-06 1.1E-05 1.3E-05 1.5E-05 1.7E-05 1.9E-05 2.1E-05 2.3E-05 2.5E-05

calc

ula

ted

rat

e (m

ol/

s*g)




Figure 36. Parity diagram for oxygen dissociative adsorption LHHW kinetic model

The kinetic parameters for the proposed model are present in Table 11 11.

Table 11 Dissociative oxygen adsorption LHHW kinetic parameters obtained

Temperature (°C)

Kinetic parameter K kT KO2

450 7.50E-3 1.43E-2 32.95 475 1.02E-2 1.99E-2 11.18 500 1.06E-2 2.05E-2 14.50

3.2.4 LLHW kinetic expression with non-dissociative oxygen adsorption

The last LHHW model evaluated includes the following assumptions:

1. Only one active site type


3. Non dissociative oxygen adsorption

4. Competitive toluene and oxygen adsorption.


The LLHW kinetic expression associate to these assumptions is:


(1+𝑘𝑇𝑃𝑇+𝑘𝑂2𝑃𝑂2)2 Eq.4

R² = 0.9517

1.0E-05

1.2E-05

1.4E-05

1.6E-05

1.8E-05

2.0E-05

2.2E-05

2.4E-05

2.6E-05

1.0E-05 1.2E-05 1.4E-05 1.6E-05 1.8E-05 2.0E-05 2.2E-05 2.4E-05

r ca

lc (

mo

l/s*

g)


3..Kinetic study 75

Table 12 summarize the parameters obtained by Least Squares mathematical regression

of non-dissociative oxygen adsorption LHHW kinetic model.

Table 12. Non-dissociative oxygen adsorption LHHW kinetic parameters obtained

Temperature (°C)

Kinetic parameter

K kT kO2

450 6.77E-2 3.69E-2 34.01 475 1.55E-2 1.86E-1 27.50 500 6.76E-3 6.84E-1 40.14

The parity diagram for the non-dissociative oxygen adsorption LHHW kinetic model,

presented by figure 37 have almost the same determination coefficient than dissociative

oxygen adsorption LHHW kinetic model parity diagram. Although, only the specific reaction

rate constant “K” present a direct relationship with the temperature change and its variations

exhibit a clear tendency. The other two parameters kT and kO2 do not show any tendency

and may be product of mathematical processes but not reveal physical meaning.

Figure 37. Parity diagram for non- dissociative oxygen adsorption LHHW kinetic model

3.2.5 Mars and Van Krevelen kinetic expression

For M&VK mechanism the assumptions [82] were already listed:

R² = 0.9579

5.0E-06

1.0E-05

1.5E-05

2.0E-05

2.5E-05

3.0E-05

5.0E-06 1.0E-05 1.5E-05 2.0E-05 2.5E-05

calc

ula

ted

rat

e (m

ol/

s*g)




(1) The reaction in step I is first order with respect to the reactant and the fraction of sites

covered by oxygen ();

(2) Certain lattice O ions at the surface are involved in the oxidation reaction in step I;

(3) Only O ions (atoms) are assumed to exist on these sites (no reactant or products);

(4) The rate of surface re-oxidation (step II), i.e., O2 adsorption, is proportional to 𝑃𝑂2𝑛 and

to the concentration of active sites not covered by oxygen (1-).

The M&VK kinetic expression associate to these assumptions is represented by the

equation 5:

𝑟𝑇 =1

1

𝑘1𝑃𝑇+

𝛽

𝑘2𝑃𝑂2𝑛

=𝑘1𝑘2𝑃𝑇𝑃𝑂2

𝑛

𝑘1𝑃𝑇+𝑘2𝑃𝑂2𝑛 Eq.5

The mathematical regression for M&VK kinetic model leads the parameters present in

Table 13.

Table 13. M&VK kinetic parameters obtained

Temperature (°C)

Kinetic parameter n k1 k2

450 2.2184 6.10E-05 6.04E-3 475 2.4672 8.81E-05 5.85E-3 500 1.9727 9.53E-05 6.37E-3

The parity diagram for the M&VK kinetic model, presented by figure 38. The regression

leads second order respect the oxygen, then it is fixed to an entire number to recalculate

the parameters.

3..Kinetic study 77

Figure 38. Parity diagram for M&VK kinetic model

The new results are summarized in table 14:

Table 14. Modified M&VK kinetic parameters obtained

Temperature (°C)

Kinetic parameter n k1 k2

450 2 6.47E-05 2.33E-3 475 2 8.87E-05 2.62E-3 500 2 1.01E-04 4.10E-3

The parity diagram for the modified M&VK kinetic model, presented by figure 39 have

almost the same determination coefficient than non-dissociative oxygen adsorption LHHW

kinetic model and M&VK model without modifications, nevertheless, the new parameter

obtained present a clear tendency with the increment of the temperature, which is a regular

behavior for kinetic constants (Arrhenius theory).



Figure 39. Parity diagram for modified M&VK kinetic model

As long, as the kinetic parameters where regressed separately for each temperature

evaluated, Figure 40 presents the parity diagram for experiments made at 475°C.

Figure 40. Parity diagram for M&VK kinetic regression for experiments made at 475°C

The good accuracy of the model is also evidenced in figure 41 and 42.

R² = 0.967

9.0E-06

1.1E-05

1.3E-05

1.5E-05

1.7E-05

1.9E-05

2.1E-05

2.3E-05

2.5E-05

2.7E-05

1.0E-05 1.2E-05 1.4E-05 1.6E-05 1.8E-05 2.0E-05 2.2E-05 2.4E-05

calc

ula

ted

rat

e (m

ol/

s*g)


R² = 0.9987

9.0E-06

1.1E-05

1.3E-05

1.5E-05

1.7E-05

1.9E-05

2.1E-05

2.3E-05

1.0E-05 1.2E-05 1.4E-05 1.6E-05 1.8E-05 2.0E-05 2.2E-05

cula

ted

rat

e (m

ol/

s*g

)


3..Kinetic study 79

Figure 41 presents the experimental data related to experiments made at oxygen feed rate fixed and the fitting for M&VK model.

Figure 41. M&VK fitting model for experiments made at 7mL/min oxygen feed rate. ▲450° ●475°C ♦500°C

Figure 42 presents the experimental data related to experiments made at toluene feed rate fixed and the fitting for M&VK model.

Figure 42. M&VK fitting model for experiments made at 30L/min toluene feed rate. ♦450°C ■475°C●500°C

0.0E+00

5.0E-06

1.0E-05

1.5E-05

2.0E-05

2.5E-05

0 0.05 0.1 0.15 0.2 0.25 0.3 0.35

r (

mo

l/s*

g)

Toluene partial pressure (atm)

0.0E+00

5.0E-06

1.0E-05

1.5E-05

2.0E-05

2.5E-05

0.00 0.10 0.20 0.30

r (m

ol/

s*g)

Oxygen partial pressure (atm)

4. Conclusions and findings

It was synthetized three different families of Ce-Zr-Mo mixed oxides, two by pseudo sol gel

like method and a third one by a modified pseudo sol gel procedure. For all the cases the

calcination temperature was 550°C.

The theoretical composition proposed, as well as the synthesis methodology lead to stable

oxides at reaction conditions as it was proved by TPO and TPR analysis. Nevertheless, the

solids exhibit different phases and polycrystallinity.

After the characterization and catalytic evaluation of the Ce-Zr-Mo catalyst it is possible to

state that synthesis method affect the final oxide obtained and its catalytic performance.

The synthesis method lead different surface species which affect the catalytic performance

of the materials. The modified pseudo sol gel method and the cation deposited strategy

leads Mo-Ce compounds on the solid’s surface as it was proved by XRD, SEM and TEM

analysis. These species on the catalyst surface promoting total oxidation.

For further researches it is suggested to evaluate the influence of each metal load.as the

catalytic test results present small variations related to cerium, zirconium or molybdenum

load of the samples, but it is not conclude. It does not exhibit a direct relation between metal

composition and conversion or selectivity, neither to benzaldehyde nor CO and CO2.

The kinetic study was made for the most selective catalyst evaluated (Ce2.65Zr0.35Mo3O14).

The mathematical regression for Mars & Van Krevelen kinetic model present less relative

error respect with the experimental data, than three LHHW kinetic models proposed and

power law rate model. The M&VK parameters obtained present a clear tendency with the

increment of the temperature, which is a regular behavior for kinetic constants (Arrhenius

theory).

Determination coefficient for Mars & Van Krevelen kinetic model obtained was 0.96, which

implies low correlation with the model proposed. Nevertheless, the kinetic expression only

consider the reactants effect. It is suggest for further research to include the influence of

the main products (benzaldehyde, CO, CO2 and water) on the kinetic study.

References

[1] S. Matar, "Chemistry of petrochemical processes", 2nd ed., vol. 1995, no. 345. Houston, Texas: Gulf Publishing Company, 2000.

[2] R. G. Santos, W. Loh, A. C. Bannwart and O. V Trevisan, “An overview of heavy oil properties and its recovery and transportation Methods,” Brazilian J. Chem. Eng., vol. 31, no. 03, pp. 571–590, 2014.

[3] British Petroleum, “BP Statistical Review of World Energy,” June, 2013.

[4] B. Kovarik, “The oil reserve fallacy” The environmental history, 2002. [Online]. Available: http://www.environmentalhistory.org/billkovarik/about-bk/research/oil-fallacy1/. [Accessed: 20-Aug-2016].

[5] M. Saajanlehto and V. Alopaeus, “Heavy oil characterization method for PC-SAFT,” Fuel, vol. 133, pp. 216–223, 2014.

[6] J. Melorose, R. Perroy, and S. Careas, “World Oil Outlook”, vol. 1. Vienna, Austria: OPEC Secretariat, 2012.

[7] J. H. Gary and G. E. Handwerk, “Petroleum Refining”, 4th ed. New York: Marcel Dekker Inc., 2001.

[8] J. M. López, “La industria petroquímica de los aromáticos en el siglo XXI” Ing. Química, pp. 99–102, 1999.

[9] G. Busca, “Oxidation Catalysts,” in “Heterogeneous Catalytic Materials”, 2014, pp. 375–419.

[10] G. I. Panov, “Advances in oxidation catalysis,” Cattech, vol. 4, no. 1, pp. 18–31, 2000.

[11] W. Zhong, S. R. Kirk, D. Yin, Y. Li, R. Zou, L. Mao, and G. Zou, “Solvent-free selective oxidation of toluene by oxygen over MnOx/SBA-15 catalysts: Relationship between catalytic behavior and surface structure” Chem. Eng. J., vol. 280, pp. 737–747, 2015.

[12] Marketsandmarkets.com, “Benzaldehyde Market by Applications and Geography - Global Trends & Forecasts up to 2018”. [Online]. Available: http://www.marketsandmarkets.com/Market-Reports/benzaldehyde-market-1093.html. [Accessed: 20-Jul-2016].

[13] IndexMundi, “Aldehyde-ethers, aldehyde-phenols & aldehydes with oth. oxygen function (Exports by Country in US Dollars,” United Nations Commodity Trade Statistics Database, 2010. [Online]. Available: http://www.indexmundi.com/trade/exports/?commodity=291249. [Accessed: 20-Jul-2016].



[14] IndexMundi, “Aldehyde-ethers, aldehyde-phenols & aldehydes with oth. oxygen function,”

United Nations Commodity Trade Statistics Database, 2010. [Online]. Available: http://www.indexmundi.com/trade/imports/?commodity=291249. [Accessed: 20-Jul-2016].

[15] Eurasian Chemical Market International Magazine, “Global Benzaldehyde Production and Market Overview,” 2015. [Online]. Available: http://www.chemmarket.info/en/home/article/1995/.[Accessed: 22-Aug-2016].

[16] Emerald Performance Materials, “Benzaldehyde, Product information Sheet,” 2015. [Online]. Available: http://www.emeraldmaterials.com/cms/kalama/fis_ftp.downloadPublicDoc?p_filename=BZALDFC_TDS_ENG.pdf&p_doc_type=TDS. [Accessed: 22-Aug-2016].

[17] DANE, “Colombia. produccion y ventas de articulos durante el año y existencias de productos terminados a 31 de diciembre total nacional” 2012.

[18] J. B. Satrio and L. K. Doraiswamy, “Production of benzaldehyde: A case study in a possible industrial application of phase-transfer catalysis” Chem. Eng. J., vol. 82, pp. 43–56, 2001.

[19] C. W. R. Alan R. Katritzky, G. Pattenden and O. Meth-Cohn, “Comprehensive Organic Functional Group Transformations: Synthesis: carbon with one heteroatom attached by a multiple bond”. Elsevier Science Ltda., 1995.

[20] E. Primo Yúfera, “Química orgánica básica y aplicada: de la molécula a la industria”. Valencia. Reverté S.A., 1994.

[21] D. Chelliah, “United States Patent 4,390,728,” 1983.

[22] W. Zhou, K. Huang, M. Cao, F. Sun, M. He, and Z. Chen, “Selective oxidation of toluene to benzaldehyde in liquid phase over CoAl oxides prepared from hydrotalcite-like precursors,” React. Kinet. Mech. Catal., 2015.

[23] Y. Meng, B. Liang, and S. Tang, “A study on the liquid-phase oxidation of toluene in ionic liquids,” Appl. Catal. A Gen., vol. 440, pp. 1–7, 2012.

[24] P. Tan, S. Tang, and B. Liang, “Kinetic models for liquid-phase catalytic oxidation of toluene to benzoic acid with pure oxygen” Chem. Eng. Commun., vol. 197, pp. 953–962, 2010.

[25] G. Q. Zhang, X. Zhang, T. Lin, T. Gong, and M. Qi, “Synergetic effect of FeVO4 and α-Fe2O3 in Fe–V–O catalysts for liquid phase oxidation of toluene to benzaldehyde” Chinese Chem. Lett., vol. 23, pp. 145–148, 2012.

[26] X. Chen, W. Zhao, F. Wang and J. Xu, “Preparation and characterization of vanadium(IV) oxide supported on SBA-15 and its catalytic performance in benzene hydroxylation to phenol using molecular oxygen” J. Nat. Gas Chem., vol. 21, pp. 481–487, 2012.

References 85

[27] J. Hoorn, J. Soolingen and G.Versteeg, “Modeling toluene oxidation: Incorporation of Mass Transfer Phenomena” Chem. Eng. Res. Des., vol. 83, pp. 187–195, 2005.

[28] M. Xue, J. Yu, H. Chen and J. Shen, “Surface acidic and redox properties of V-Ag-O/TiO2 catalysts for the selective oxidation of toluene to benzaldehyde,” Catal. Letters, vol. 128, pp. 373–378, 2009.

[29] S. Lars, “Reaction network and kinetics for the catalytic oxidation of toluene over V205,” vol. 100, pp. 92–100, 1990.

[30] A. Aguilar and V. Cortés, “Selective oxidation of toluene on V2O5/ TiO2/SiO2 catalysts modified with Te, Al, Mg, and K2SO4” Catal. Today, vol. 32, pp. 265–272, 1996.

[31] B. Jonson, B. Rebenstorf, R. Larsson, S. L. Andersson and S. T. Lundin, “Activity measurements and spectroscopic studies of the catalytic oxidation of toluene over V2O5/Al2O3-C catalysts” J. Chem. Soc. Faraday Trans. 1 Phys. Chem. Condens. Phases, vol. 82, pp. 767–783, 1986.

[32] K. Mori, “Activity and selectivity in toluene oxidation on well characterized vanadium oxide catalysts” vol. 83, pp. 3303–3315, 1987.

[33] J. Zhu and S. Lars, “Influence of potassium on the catalytic properties of V2O5/ TiO2 Catalysts for Toluene Oxidation,” J. Chem. Soc. Faraday Trans., vol. 85, pp. 3629–3644, 1989.

[34] V. Kumar and D. Grover, “Catalyst Evaluation for the Vapor-Phase Oxidation of p-Xylene,” Ind. Eng. Chem. Res., vol. 30, pp. 1139–1141, 1991.

[35] Z. Hui-Liang, Z. Wei, D. Xiang, and F. Xian-Cai “A study of catalytic activity, constituent, and structure of V-Ag catalyst for selective oxidation of toluene to benzaldehyde” J. Catal., vol. 129, pp. 426–437, 1991.

[36] N. V. Boikov, E.V. Vishnetskaya, M.V Emel’yanov, A.N. Tomskii and I.S. Shcherbakov “The selective catalytic oxidation of toluene” Russ. J. Phys. Chem., vol. 82, pp. 2233–2237, 2008.

[37] W. Kuang, Y. Fan, J. Qiu and Y. Chen “Ultrafine La–Mo and Ce–Mo complex oxide particle catalysts for selective oxidation of toluene” J. Mater. Chem., vol. 8, pp. 19–20, 1998.

[38] W. Kuang, Y. Fan and Y. Chen, “Structure and reactivity of ultrafine Ce-Mo oxide particles” Catal. Today, vol. 68, pp. 75–82, 2001.

[39] W. Kuang, Y. Fan, K. Chen and Y. Chen, “Selective Oxidation of Toluene over Complex Fe–Mo Oxides in the Absence of Molecular Oxygen” J. Chem. Res., vol. 2, pp. 366, 1997.

[40] W. Kuang, Y. Fan, K. Chen and Y. Chen, “Partial Oxidation of Toluene over Ultrafine Mixed Mo-Based Oxide Particles” vol. 317, pp. 310–317, 1999.



[41] D. L. Trimm and M. Irshad “The influence of electron directing effects on the catalytic

oxidation of toluenes and xylenes” J. Catal., vol. 18, pp. 142–153, 1970.

[42] J. C. Védrine, G. J. Hutchings and C. J. Kiely “Molybdenum oxide model catalysts and vanadium phosphates as actual catalysts for understanding heterogeneous catalytic partial oxidation reactions: A contribution by Jean-Claude Volta” Catal. Today, vol. 217, pp. 57–64, 2013.

[43] B. Stuyven, J. Emmerich, P. Eloy, J. Van Humbeeck, C. Kirschhock, P. Jacobs, J. Martens and E. Breynaert “Molybdenum – vanadium – antimony mixed oxide catalyst for isobutane partial oxidation synthesized using magneto hydrodynamic forces” Appl. Catal. A Gen., vol. 474, pp. 18–25, 2014.

[44] G. Centi, S. Perathoner and S. Tonini “In situ DRIFT study of the reactivity and reaction mechanism of catalysts based on iron-molybdenum oxides encapsulated in Boralite for the selective oxidation of alkylaromatics” Catal. Today, vol. 61, pp. 211–221, 2000.

[45] D. L. Trimm “Design of industrial Catalysts” Elsevier Sci., vol. 1, p. 326, 1980.

[46] D. M. Gómez, V. V Galvita, J. M. Gatica, H. Vidal and G. B. Marin “TAP study of toluene total oxidation over a Co3O4/La-CeO2 catalyst with an application as a washcoat of cordierite honeycomb monoliths.,” Phys. Chem. Chem. Phys., vol. 16, pp. 11447–55, 2014.

[47] Y. Zhiqiang, M. Dongsen, G. Xiaoming and L. Guanzhong “CO oxidation over CuO catalysts supported on CeO2 -ZrO2 prepared by microwave-assisted co-precipitation : The influence of CuO content” vol. 32, no. 2, pp. 117–123, 2014.

[48] A. Eltayeb, R. K. Vijayaraghavan, A. McCoy, A. Venkatanarayanan, A. Yaremchenko, R. Surendran, E. McGlynn and S. Daniels “Control and enhancement of the oxygen storage capacity of ceria films by variation of the deposition gas atmosphere during pulsed DC magnetron sputtering” J. Power Sources, vol. 279, pp. 94–99, 2015.

[49] S. H. Song, J. Moon, J. H. Kim, J. Hong, J.-H. Lee, H.-W. Lee, B.-K. Kim and H. Kim “Panoscopic alloying of cobalt in CeO2–ZrO2 solid solutions for superior oxygen-storage capacity,” Acta Mater., vol. 113, pp. 206–212, 2016.

[50] N. Al-Yassir and R. Le Van Mao “Physico-chemical properties of mixed molybdenum and cerium oxides supported on silica-alumina and their use as catalysts in the thermal-catalytic cracking (TCC) of n-hexane” Appl. Catal. A Gen., vol. 305, pp. 130–139, 2006.

[51] M. P. Yeste, J. C. Hernández, S. Bernal, G. Blanco, J. J. Calvino, J. Pérez-Omil, and J. M. Pintado, “Comparative study of the reducibility under H2 and CO of two thermally aged Ce0.62Zr0.38O2 mixed oxide samples” Catal. Today, vol. 141, pp. 409–414, 2009.

References 87

[52] V. Sánchez Escribano, J. M. Gallardo Amores, E. Fernández López, C. del Hoyo Martínez, G. Busca and C. Resini “Surface and structural characterization of ZrO2-CeO2-supported molybdenum oxide catalysts” Mater. Chem. Phys., vol. 114, pp. 848–853, 2009.

[53] J. Wang, M. Shen, J. Wang, J. Gao, J. Ma and S. Liu “Effect of cobalt doping on ceria-zirconia mixed oxide: Structural characteristics, oxygen storage/release capacity and three-way catalytic performance” J. Rare Earths, vol. 30, pp. 878–883, 2012.

[54] M. Zhao, Y. Yao, H. Wang, Z. Liu, M. Gong, and Y. Chen, “Preparation and investigation of Ce0.4Zr0.5–xY0.1MnxO2 oxygen-storage materials” J. Rare Earths, vol. 28, pp. 260–264, 2010.

[55] R. Ran, D. Weng, X. Wu, J. Fan, L. Wang and X. Wu “Structure and oxygen storage capacity of Pr-doped Ce0.26Zr0.74O2 mixed oxides” J. Rare Earths, vol. 29, pp. 1053–1059, 2011.

[56] P. Pham, M. Le, T. Nguyen, E. Bruneel and I. Van Driessche “The influence of deposition methods of support layer on cordierite substrate on the characteristics of a MNO2–NiO–CO3O4/Ce0.2Zr0.8O2/cordierite three way catalyst” Materials (Basel)., vol. 7, pp. 6237–6253, 2014.

[57] J. C. Duarte Triana “Estudio de la producción en fase gas de benzaldehído a partir de tolueno sobre catalizadores de Ce-Zr-Mo” Tesis de Maestría en Ingeniería, Universidad Nacional de Colombia, 2008.

[58] P. Huang, H. Jiang and M. Zhang “Structures and oxygen storage capacities of CeO2-ZrO2-Al2O3 ternary oxides prepared by a green route: supercritical anti-solvent precipitation” J. Rare Earths, vol. 30, pp. 524–528, 2012.

[59] I. B. Troitskaia, T. Gavrilova, P. Zubareva, D. Y. Troitskii and S. a. Gromilov “Thermal transformations of the composition and structure of hexagonal molybdenum oxide” J. Struct. Chem., vol. 56, pp. 289–296, 2015.

[60] P. Duran, M. Gonzalez, C. Moure, J. R. Jurado and C. Pascual “A new tentative phase equilibrium diagram for the ZrO2-CeO2 system in air” J. Mater. Sci., vol. 25, no. 1990, pp. 5001–5006, 1990.

[61] A. Alper “High temperature oxides part II”. New York. Academic Press Inc., 1970.

[62] H.-F. Lu, Y. Zhou, W.-F. Han, H.-F. Huang and Y.-F. Chen “Promoting effect of ZrO2 carrier on activity and thermal stability of CeO2-based oxides catalysts for toluene combustion” Appl. Catal. A Gen., vol. 464–465, pp. 101–108, 2013.

[63] L. Li, O. Van Der Biest, P. L. Wang, J. Vleugels, W. W. Chen and S. G. Huang “Estimation of the phase diagram for the ZrO2– Y2O3–CeO2 system” J. Eur. Ceram. Soc., vol. 21, pp. 2903–2910, 2001.



[64] A. Morikawa, K. Yamamura, A. Suda, T. Tanabe, T. Nobukawa, A. Chiba and N. Takahashi

“Phase transformation behavior of a pyrochlore-type CeO2 -ZrO2 binary compound” J. Am. Ceram. Soc., vol. 98, pp. 659–662, 2015.

[65] B. Wang, Y. Yao, S. Liu, Z. Hu, Z. Li and X. Ma “Effects of MoO3 loading and calcination temperature on the catalytic performance of MoO3/CeO2 toward sulfur-resistant methanation” vol. 138, pp. 263–270, 2015.

[66] Y. Wu, G. Hu, Y. Xie, M. Guo and M. Luo “Solid state reaction of MoO3–CeO2 complex oxide studied by Raman spectroscopy” Solid State Sci., vol. 13, pp. 2096–2099, 2011.

[67] A. Zaitsev, N. Korotkov and E. M. Lazarev “Oxidation of molybdenum and molybdenum-tungsten alloys” Met. Sci. Heat Treat., vol. 18, pp. 873–876, 1976.

[68] Q. Zhang, H. Li, X. Liu, W. Qin, Y. Zhang, W. Xue and S. Yang “Modified Porous Zr-Mo mixed oxides as strong acid catalysts for biodiesel production” Energy Technol., vol. 1, pp. 735–742, 2013.

[69] X. I. N. Zhang, D. He, Q. Zhang, Q. Ye and B. Xu “Characterization and catalytic performance of Zr-Mo-oxide catalysts for selective oxidation” Compos. Interfaces, vol. 11, pp. 153–167, 2004.

[70] V. Grover and K. Tyagi “Ternary phase relations in CeO2–DyO1.5–ZrO2 system” Ceram. Int., vol. 39, pp. 7563–7569, 2013.

[71] V. Grover and a. K. Tyagi “Subsolidus phase equilibria in the CeO2-ThO2-ZrO2 system” J. Nucl. Mater., vol. 305, pp. 83–89, 2002.

[72] E. R. Andrievskaya, G. I. Gerasimyuk, O. Kornienko, V Samelyuk, L. M. Lopato and V. P. Redko “Phase equilibria in the system HfO2 − ZrO2 − CeO2 at 1500 °C” Powder Metall. Met. Ceram., vol. 45, pp. 448–456, 2006.

[73] K. Mwamba and P.-H. Duvigneaud, “Phase relationships in the pseudoternary BaO-ZrO2-MoO3; system” Key Eng. Mater., vol. 206–213, pp. 791–794, 2002.

[74] H. Wan, D. Li, H. Zhu, Y. Zhang, L. Dong, Y. Hu, B. Liu, K. Sun, L. Dong and Y. Chen “A comparative study on the dispersion behaviors and surface acid properties of molybdena on CeO2 and ZrO2” J. Colloid Interface Sci., vol. 326, pp. 28–34, 2008.

[75] S. Ding, F. Liu, X. Shi, K. Liu, Z. Lian, L. Xie and H. He “Significant promotion effect of Mo additive on a novel Ce–Zr mixed oxide catalyst for the selective catalytic reduction of NOx

with NH3” ACS Appl. Mater. Interfaces, vol. 7, pp. 9497–9506, 2015.

[76] Z. Liu, H. Su, J. Li and Y. Li “Novel MoO3/ CeO2–ZrO2 catalyst for the selective catalytic reduction of NOx by NH3” Cat. Commun. vol. 65, pp. 51–54, 2015.

References 89

[77] S. Rathod, B. Arbad and M. Lande “Preparation, characterization and catalytic activity of MoO3/CeO2–ZrO2 solid heterogeneous catalyst for the synthesis of b-enaminonesa” Arab.J.of Chem. vol. 9837, pp. 253–260, 2010.

[78] S. L. Andersson “Reaction networks in the catalytic vapor-phase oxidation of toluene and xylenes” J. Catal., vol. 98, pp. 138–149, 1986.

[79] D. a Bulushev and Renken “Vanadia / titania catalysts for gas phase partial toluene oxidation Spectroscopic characterisation and transient kinetics study” Catal. Today, vol. 57, pp. 231–239, 2000.

[80] S. Larrondo, A. Barbaro, B. Irigoyen and N. Amadeo “Oxidation of toluene to benzaldehyde over VSb0.8Ti0.2O4: Effect of the operating conditions”Catal. Today, vol. 64, pp. 179–187, 2001.

[81] G. Giindiiz and O. Akpolat “Catalytic vapor-phase oxidation of toluene to benzaldehyde” Ind. Eng. Chem. Res. pp. 45–48, 1990.

[82] M. A. Vannice “An analysis of the Mars-van Krevelen rate expression” Catal. Today, vol. 123, pp. 18–22, 2007.

[83] K. Papadatos, “Catalyst screening using a Stone DTA Apparatus: Oxidation of toluene over Cobalt-Metal Oxide Catalysts” J. Catal. vol. 123, pp. 116–123, 1973.

[84] G. P. Wirtz, L. B. Sis and J. S. Wheeler “Sublimation of MoO3 from WO3MoO3 catalysts during the oxidation of toluene” J. Catal., vol. 38, pp. 196–205, 1975.

[85] K. van der Wiele and P. J. van den Berg “Oxidation of toluene over bismuth molybdate catalysts” J. Catal., vol. 39, pp. 437–448, 1975.

[86] P. E. Keefe, F. R. Ahlers, T. William, W. H. Magidson, I. Allen and S. Charles “Non-catalytic pyrolysis of residues from aromatic carboxylic acid manufacture,” United States Patent 4,266,084”, 1981.

[87] J. Regalbuto “Catalyst science and engineering” CRC Press. London, UK. vol. 1, 2007. .

[88] J. Livage and C. Sanchez “Sol-gel chemistry” J. Non. Cryst. Solids, vol. 145, pp. 11–19, 1992.

[89] J. C. Vargas, S. Libs, A.-C. Roger and A. Kiennemann “Study of Ce-Zr-Co fluorite-type oxide as catalysts for hydrogen production by steam reforming of bioethanol” Catal. Today, vol. 107–108, pp. 417–425, 2005.

[90] M. Boaro, M. Vicario, C. De Leitenburg, G. Dolcetti and A. Trovarelli “The use of temperature-programmed and dynamic/transient methods in catalysis: Characterization of ceria-based, model three-way catalysts” Catal. Today, vol. 77, pp. 407–417, 2003.



[91] M. Xue, X. Gu, J. Chen, H. Zhang and J. Shen “Characterization of acidic and redox

properties of Ce-Mo-O catalysts for the selective oxidation of toluene,” Thermochim. Acta, vol. 434, pp. 50–54, 2005.

[92] J. Ge, M. Xue, Q. Sun, A. Auroux and J. Shen “Surface acidic and redox properties of V-Zr-O catalysts for the selective oxidation of toluene to benzaldehyde” J. Mol. Catal. A Chem., vol. 278, pp. 209–214, 2007.

[93] C. Perdomo, A. Pérez, R. Molina and S. Moreno “Storage capacity and oxygen mobility in mixed oxides from transition metals promoted by cerium” Appl. Surf. Sci., vol. 383, pp. 42–48, 2016.

[94] C. Cau, Y. Guari, T. Chave, J. Larionova and S. I. Nikitenko “Thermal and sonochemical

synthesis of porous (Ce,Zr)O2 mixed oxides from metal -diketonate precursors and their catalytic activity in wet air oxidation process of formic acid” Ultrason. Sonochem., vol. 21, pp. 1366–1373, 2014.

[95] M. Shacham and N. Brauner “Using the power-law rate expression for assessment of rate data and detection of infeasible mechanisms for reversible reactions” Ind. Eng. Chem. Res., vol. 35, pp. 2790–2794, 1996.

[96] H. S. Fogler, “Elements of chemical reaction engineering.” Third Edition. Prentince Hall India. New Delhi. p. 769, 2004.

[97] J. M. Smith, Chemical Engineering Kinetics, Third. Mcgraw-Hill Chemical Engineering Series, New York 1981.

Annexes 91

Annexes

A. Chromatographic methods

The non-condensable and condensable products were analysed by gas phase

chromatography (GPC), using in both cases Network GC system Agilent Technologies

6890N. For the non-condensable products a TCD detector was used; while for the

condensable products a FID detector was employed. Next are presented the

chromatographic method, retention times and response factors of the different products

analysed for each case.

i. Non-condensable products analysis

The non-condensable products analysed were H2, O2, CO, CO2, and N2 (internal standard).

The column used was a Carbosieve II of dimensions: 2m, i.d. 1/8".

The chromatographic method was the following:

Injection: 150 °C

Carrier Gas: Ar, flow: 15.3 ml min-1

Oven program: 90 °C for 5 min

from 90°C to 210 °C at 10 °C min.-1

210 °C for 3 min

Detector: TCD 250 °C

In Table 15 are reported the retention times according the method above described. The

molar response factors with N2 as internal standard are also presented.

Table 15. Retention times and response factors utilised for the non-condensable products.

Product Retention time

(tR) Response factor

(molar)

H2 1.3 0.099 O2 4.7 0.88 N2 5.0 1 CO 5.9 1.104 CO2 11.0 0.078



ii. Condensable products analysis

The condensable products analysed were benzene, toluene and benzaldehyde isopropanol

(internal standard). The column was a SGE Sol gel Wax of dimensions: 60m x 0.25mm,

i.d.0.25m.

The chromatographic method was the following:

Injection: 250 °C; split ratio: 120:1; Flow: 109 ml min-1

Carrier Gas: He, flow: 0.9 ml min-1

Oven program: 100 °C for 2 min

from 100 °C to 130 °C at 20 °C min-1

from 130 °C to 280 °C at 60 °C min-1

280 °C for 4 min

Detector: FID 250 °C; 40 ml min-1 H2

In table 11 are reported the retention times of the condensable products, according the

method above described. The molar response factors with n-propanol as internal standard

are also included.

Table 16. Retention times and response factors utilised for the condensable products.

Product Retention time

(tR) Response factor

(molar)

Isopropanol 3.4 1 Benzene 3.46 0.313 Toluene 3.6 0.173

Benzaldehyde 5.5 0.219

Annexes 93

B. Catalytic tests results

Complement information about the catalytic tests made to the different evaluated materials is summarized in table 12.

Table 17. Catalytic tests results summary

Conversion Selectivity to Benzaldehyde Selectivity to CO Selectivity to CO2

Catalyst 450°C 475°C 500°C 450°C 475°C 500°C 450°C 475°C 500°C 450°C 475°C 500°C

CZM2(I) 31 35 39 05 05 05 29 28 29 17 15 16

CZM3(I) 32 32 36 13 12 11 24 24 24 13 13 14

CZM4(I) 36 37 37 12 12 10 16 21 21 17 20 20

CZM2(II) 33 40 41 23 17 13 20 21 23 9 10 14

CZM3(II) 36 38 39 24 24 18 26 16 20 5 11 9

CZM4(II) 35 38 41 26 16 06 18 25 33 8 4 4

CM/CZ(I) 41 45 47 12 11 07 26 27 27 18 18 17

CM/CZ(II) 37 37 40 28 24 20 19 17 19 5 8 12



C. Kinetic experimental results

The information taken for each experiment is summarize in Table 13:

Table 18. Experimental data results.

Experiment number

Temperature Oxygen feeding

Toluene feeding

W/FA0 X toluene rate

°C mL/min L/min kg*h/m3 mol/s.g (E-06)

1 450 5 30 0.31593 0.23 9.10

2 450 7 30 0.31593 0.27 10.64

3 450 9 30 0.31593 0.28 10.85

4 450 5 40 0.24294 0.22 11.08

5 450 7 40 0.24294 0.25 12.97

6 450 9 40 0.24294 0.23 11.56

7 450 5 50 0.1863 0.20 13.54

8 450 7 50 0.1863 0.21 13.89

9 450 9 50 0.1863 0.22 14.34

10 475 5 30 0.31593 0.33 13.00

11 475 7 30 0.31593 0.36 14.18

12 475 9 30 0.31593 0.38 14.85

13 475 5 40 0.24294 0.29 15.02

14 475 7 40 0.24294 0.30 15.40

15 475 9 40 0.24294 0.35 18.03

16 475 5 50 0.1863 0.24 16.30

17 475 7 50 0.1863 0.27 17.86

18 475 9 50 0.1863 0.31 20.67

19 500 5 30 0.31593 0.37 14.52

20 500 7 30 0.31593 0.46 18.21

21 500 9 30 0.31593 0.56 21.91

22 500 5 40 0.24294 0.35 17.96

23 500 7 40 0.24294 0.36 18.61

24 500 9 40 0.24294 0.38 19.42

25 500 5 50 0.1863 0.32 21.13

26 500 7 50 0.1863 0.33 21.72

27 500 9 50 0.1863 0.35 22.99

Documents

Estudio cinético de la oxidación parcial de tolueno a ...bdigital.unal.edu.co/56147/25/Geraldine.Díaz.2016.pdf · Estudio cinético de la oxidación parcial de tolueno a benzaldehído,