catalytic conversion of biomass-derived synthesis...

Catalytic conversion of

biomass-derived synthesis gas

to liquid fuels

Rodrigo Suárez París

Doctoral Thesis in Chemical Engineering KTH Royal Institute of Technology School of Chemical Science and Engineering Department of Chemical Engineering and Technology Stockholm, Sweden 2016

Catalytic conversion of biomass-derived synthesis gas to liquid fuels RODRIGO SUÁREZ PARÍS TRITA-CHE Report 2016:9 ISSN 1654-1081 ISBN 978-91-7595-862-0 Akademisk avhandling som med tillstånd av Kungliga Tekniska högskolan i Stockholm framlägges till offentlig granskning för avläggande av teknologie doktorsexamen, fredagen den 15 april 2016 klockan 10:00 i Kollegiesalen, Brinellvägen 8, Kungliga Tekniska högskolan, Stockholm. Fakultetsopponent: Professor Agustín Martínez Feliu, UPV-CSIC, Valencia, Spain. © Rodrigo Suárez París 2016 Tryck: Universitetsservice US-AB

“If we knew what it was we were doing,

it would not be called research, would it?”

Albert Einstein

i

Abstract

Climate change is one of the biggest global threats of the 21st century.

Fossil fuels constitute by far the most important energy source for

transportation and the different governments are starting to take action to

promote the use of cleaner fuels.

Biomass-derived fuels are a promising alternative for diversifying fuel

sources, reducing fossil fuel dependency and abating greenhouse gas

emissions. The research interest has quickly shifted from first-generation

biofuels, obtained from food commodities, to second-generation biofuels,

produced from non-food resources.

The subject of this PhD thesis is the production of second-generation

biofuels via thermochemical conversion: biomass is first gasified to

synthesis gas, a mixture of mainly H2 and CO; synthesis gas can then be

catalytically converted to different fuels. This work summarizes six

publications, which are focused on the synthesis gas conversion step.

Two processes are principally examined in this summary. The first part of

the PhD thesis is devoted to the synthesis of ethanol and higher alcohols,

which can be used as fuel or fuel additives. The microemulsion technique

is applied in the synthesis of molybdenum-based catalysts, achieving a

yield enhancement. Methanol cofeeding is also studied as a way of boosting

the production of longer alcohols, but a negative effect is obtained: the

main outcome of methanol addition is an increase in methane production.

The second part of the PhD thesis addresses wax hydroconversion, an

essential upgrading step in the production of middle-distillate fuels via

Fischer-Tropsch. Bifunctional catalysts consisting of noble metals

supported on silica-alumina are considered. The deactivation of a

platinum-based catalyst is investigated, sintering and coking being the

main causes of decay. A comparison of platinum and palladium as catalyst

metal function is also carried out, obtaining a fairly different catalytic

performance of the materials in terms of conversion and selectivity, very

likely due to dissimilar hydrogenation power of the metals. Finally, a

kinetic model based on the Langmuir-Hinshelwood-Hougen-Watson

formalism is proposed to describe the hydroconversion reactions, attaining

a good fitting of the experimental data.

Keywords: biofuels, Fischer-Tropsch wax, higher alcohols,

hydroconversion, kinetic modelling, methanol cofeeding, microemulsion,

MoS2, noble metal, syngas

ii

Sammanfattning

Klimatförändringarna är ett av de största globala hoten under det

tjugoförsta århundradet. Fossila bränslen utgör den helt dominerande

energikällan för transporter och många länder börjar stödja användning av

renare bränslen.

Bränslen baserade på biomassa är ett lovande alternativ för att diversifiera

råvarorna, reducera beroendet av fossila råvaror och undvika

växthusgaser. Forskningsintresset har snabbt skiftat från första

generationens biobränslen som erhölls från mat-råvaror till andra

generationens biobränslen producerade från icke ätbara-råvaror.

Ämnet för denna doktorsavhandling är produktion av andra generationens

biobränslen via termokemisk omvandling. Biomassa förgasas först till

syntesgas, en blandning av i huvudsak vätgas och kolmoxid; syntesgasen

kan sedan katalytiskt omvandlas till olika bränslen. Detta arbete

sammanfattar sex publikationer som fokuserar på steget för

syntesgasomvandling.

Två processer är i huvudsak undersökta i denna sammanfattning. Den

första delen av doktorsavhandlingen ägnas åt syntes av etanol och högre

alkoholer som kan användas som bränsle eller bränsletillsatser.

Mikroemulsionstekniken har använts vid framställningen av molybden-

baserade katalysatorer, vilket gav en höjning av utbytet. Tillsatsen av

metanol har också studerats som ett sätt att försöka få en högre

koncentration av högre alkoholer, men en negativ effekt erhölls:

huvudeffekten av metanoltillsatsen är en ökad metanproduktion.

Den andra delen av doktorsavhandlingen handlar om vätebehandling av

vaxer som ett viktigt upparbetningssteg vid framställning av

mellandestillat från Fischer-Tropsch processen. Bifunktionella

katalysatorer som består av ädelmetaller deponerade på silica-alumina

valdes. Deaktiveringen av en platinabaserad katalysator undersöktes.

Sintring och koksning var huvudorsakerna till deaktiveringen. En

jämförelse mellan platina och palladium som funktionella metaller

genomfördes också med resultatet att det var en ganska stor skillnad

mellan materialens katalytiska egenskaper vilket gav olika omsättning och

selektivitet, mycket sannolikt beroende på olika reaktionsmönster hos

metallerna vid vätebehandling. Slutligen föreslås en kinetisk modell

baserad på en Langmuir-Hinshelwood-Hougen-Watson modell för att

beskriva reaktionerna vid vätebehandling. Denna modell ger en god

anpassning till experimentella data.

iii

Resumen

El cambio climático es una de las mayores amenazas del siglo XXI. Los

combustibles fósiles constituyen actualmente la fuente de energía más

importante para el transporte, por lo que los diferentes gobiernos están

empezando a tomar medidas para promover el uso de combustibles

más limpios.

Los combustibles derivados de biomasa son una alternativa

prometedora para diversificar las fuentes de energía, reducir la

dependencia de los combustibles fósiles y disminuir las emisiones de

efecto invernadero. Los esfuerzos de los investigadores se han dirigido

en los últimos años a los biocombustibles de segunda generación,

producidos a partir de recursos no alimenticios.

El tema de esta tesis de doctorado es la producción de biocombustibles

de segunda generación mediante conversión termoquímica: en primer

lugar, la biomasa se gasifica y convierte en gas de síntesis, una mezcla

formada mayoritariamente por hidrógeno y monóxido de carbono; a

continuación, el gas de síntesis puede transformarse en diversos

biocombustibles. Este trabajo resume seis publicaciones, centradas en

la etapa de conversión del gas de síntesis. Dos procesos se estudian con

mayor detalle.

En la primera parte de la tesis se investiga la producción de etanol y

alcoholes largos, que pueden ser usados como combustible o como aditivos

para combustible. La técnica de microemulsión se aplica en la síntesis de

catalizadores basados en molibdeno, consiguiendo un incremento del

rendimiento. Además, se introduce metanol en el sistema de reacción para

intentar aumentar la producción de alcoholes más largos, pero los efectos

obtenidos son negativos: la principal consecuencia es el incremento de la

producción de metano.

La segunda parte de la tesis estudia la hidroconversión de cera, una etapa

esencial en la producción de destilados medios mediante Fischer-Tropsch.

Los catalizadores estudiados son bifuncionales y consisten en metales

nobles soportados en sílice-alúmina. La desactivación de un catalizador de

platino se investiga, siendo la sinterización y la coquización las principales

causas del problema. El uso de platino y paladio como componente

metálico se compara, obteniendo resultados catalíticos bastante diferentes,

tanto en conversión como en selectividad, probablemente debido a su

diferente capacidad de hidrogenación. Finalmente, se propone un modelo

Resumen iv

cinético, basado en el formalismo de Langmuir-Hinshelwood-Hougen-

Watson, que consigue un ajuste satisfactorio de los datos experimentales.

v

List of appended papers

Paper I

Catalytic conversion of biomass-derived synthesis gas to fuels

R. Suárez París, L. Lopez, J. Barrientos, F. Pardo, M. Boutonnet, S.

Järås

Catalysis: Volume 27 [J.J. Spivey, Y.-F. Han, K.M. Dooley (eds.)], The Royal Society of Chemistry, 2015, 62-143

Paper II

Higher alcohol synthesis over nickel-modified alkali-doped

molybdenum sulfide catalysts prepared by conventional

coprecipitation and coprecipitation in microemulsions

R. Suárez París, V. Montes, M. Boutonnet, S. Järås

Catalysis Today 258 (2015), 294-303

Paper III

Effect of methanol addition on higher alcohol synthesis over modified

molybdenum sulfide catalysts

R. Suárez París, M. Boutonnet, S. Järås

Catalysis Communications 67 (2015), 103-107

Paper IV

Deactivation of a Pt/silica-alumina catalyst and effect on selectivity

in the hydrocracking of n-hexadecane

F. Regali, R. Suárez París, A. Aho, M. Boutonnet, S. Järås

Topics in Catalysis 56 (2013), 594-601

List of appended papers vi

Paper V

Hydroconversion of paraffinic wax over platinum and palladium

catalysts supported on silica-alumina

R. Suárez París, M. E. L'Abbate, L. F. Liotta, V. Montes, J. Barrientos, F.

Regali, A. Aho, M. Boutonnet, S. Järås

Catalysis Today (2015), http://dx.doi.org/10.1016/j.cattod.2015.11.026

(article in press)

Paper VI

Kinetic modelling of paraffinic wax hydroconversion over platinum- and

palladium-based catalysts

R. Suárez París, M. E. L'Abbate, M. Boutonnet, S. Järås

Manuscript submitted to Computers & Chemical Engineering (2016)

All papers are reproduced with permission from the copyright holders.

vii

Contribution to the appended papers

Paper I

I had the responsibility for coordinating this review. I wrote the following

sections: Introduction; Hydrocracking of FT wax; Other fuels: methanol

and DME; Conclusions.

Paper II

I had the main responsibility for writing this paper. I also performed most

of the experimental work. V. Montes carried out the XPS, SEM-EDX and

TEM measurements and helped with the analysis of the data.

Paper III

I had the main responsibility for writing this paper. I also performed all the

experimental work.

Paper IV

F. Regali had the main responsibility for writing this paper. I performed

part of the experimental work under his supervision and actively

participated in the analysis of the data. A. Aho carried out the FTIR

measurements of adsorbed pyridine.

Paper V

I had the main responsibility for writing this paper. The experimental work

was mainly a joint effort between myself and M.E. L’Abbate, under my

supervision. L.F. Liotta carried out the NH3-TPD and TGA measurements

and helped with the analysis of the data. V. Montes carried out the TEM

measurements and helped with the analysis of the data. A. Aho carried out

the FTIR measurements of adsorbed pyridine.

Contribution to the appended papers viii

Paper VI

I had the main responsibility for writing this paper. The modelling work

was a joint effort between myself and M.E. L’Abbate, under my

supervision.

ix

Other publications and conference contributions

Publications in peer-reviewed journals

1. CO methanation over TiO2-supported nickel catalysts: A carbon

formation study

J. Barrientos, M. Lualdi, R. Suárez París, V. Montes, M. Boutonnet, S.

Järås

Applied Catalysis A: General 502 (2015), 276–286

Book chapters

2. Nanocatalysts: synthesis in nanostructured liquid media and their

application in energy and production of chemicals

M. Boutonnet, A. Marinas, V. Montes, R. Suárez París, M. Sánchez-

Domínguez

Manuscript submitted to Nanocolloids: A Meeting Point for Scientists

and Technologists [M. Sánchez-Domínguez, C. Rodriguez Abreu

(eds.)]. Expected release date: 25th March 2016

Oral presentations 1

3. Hydrocracking of Fischer-Tropsch Wax over Noble Metal / Silica

– Alumina Catalysts

R. Suárez París, M. E. L'Abbate, L. F. Liotta, J. Barrientos, F. Regali, M.

Boutonnet, S. Järås

24th North American Catalysis Society Meeting , 14-19 June 2015,

Pittsburgh, USA

1 Presenting author: underlined

Other publications and conference contributions x

4. Hydroconversion of FT wax over noble metal / silica – alumina

catalysts

R. Suárez París, M. E. L'Abbate, L. F. Liotta, J. Barrientos, F. Regali, M.

Boutonnet, S. Järås

Syngas Convention 2, 28 March – 1 April 2015, Cape Town, South

Africa

5. Synthesis of mixed alcohols over K-Ni-MoS2 catalysts prepared by

conventional coprecipitation and by microemulsion


8th International Conference on Environmental Catalysis , 24-27

August 2014, Asheville, USA

6. Higher alcohol synthesis over K-Ni-MoS2 catalysts prepared by

conventional coprecipitation and by a novel microemulsion technique


16th Nordic Symposium on Catalysis, 15-17 June 2014, Oslo, Norway

7. Behavior of platinum and palladium on silica-alumina catalysts in

the hydrocracking of n-hexadecane

F. Regali, R. Suárez París, M. Boutonnet, S. Järås

Syngas Convention, 1-4 April 2012, Cape Town, South Africa

Poster presentations 2

8. Facing climate change: 100 % renewable energy by 2050

R. Suárez París, P. Haghighi Moud, K. Engvall, M. Boutonnet

2 Presenting author(s): underlined

Other publications and conference contributions xi

KTH Energy Dialogue 2014, 20 November 2014, Stockholm, Sweden

9. Catalysis for renewable fuels: Fischer-Tropsch fuels

R. Suárez París, J. Barrientos, M. Boutonnet, S. Järås

KTH Energy Dialogue 2013, 7 November 2013, Stockholm, Sweden

10. Ni-modified ADM catalyst prepared by microemulsion for mixed

alcohol synthesis

R. Suárez París, R. Andersson, M. Boutonnet, S. Järås,

UBIOCHEM IV: 4th International Workshop of COST Action CM0903,

14-16 October 2013, Valencia, Spain

11. Optimization of ethanol production from syngas in a novel Hollow

Fiber Membrane bioreactor

R. Suárez París, J. Rogut, M. Boutonnet, S. Järås

XIth European Congress on Catalysis, 1-6 September 2013, Lyon, France

12. Ni-modified alkali-doped molybdenum sulfide catalyst prepared by

microemulsion for mixed alcohol synthesis

R. Suárez París, R. Andersson, M. Boutonnet, S. Järås

CRS-2: Catalysis for Renewable Sources: fuel, energy, chemicals, 22-28

July 2013, Lund, Sweden

13. Behavior of platinum on silica-alumina catalysts in the

hydrocracking of n-hexadecane

F. Regali, R. Suárez París, M. Boutonnet, S. Järås,

15th Nordic Symposium on Catalysis, 10-12 June 2012, Mariehamn,

Åland, Finland

xii

Table of contents

1 INTRODUCTION ______________________________ 1

1.1 Setting the scene ____________________________________ 1

1.2 Objective of the work _________________________________ 1

1.3 Thesis outline _______________________________________ 3

2 SECOND-GENERATION BIOFUELS ______________ 5

2.1 The role of second-generation biofuels ___________________ 5

2.2 Production routes of second-generation biofuels ___________ 7 2.2.1 Syngas generation _______________________________ 8 2.2.2 Syngas conditioning ______________________________ 9 2.2.3 Catalytic conversion of syngas to fuels (Paper I) _______ 10

2.3 Production plants of second-generation biofuels ___________ 12

3 HIGHER ALCOHOL SYNTHESIS _______________ 13

3.1 Higher alcohol synthesis from a historical and industrial

perspective _____________________________________________ 13

3.2 Properties of alcohol fuels ____________________________ 13 3.2.1 Main advantages of alcohol fuels ___________________ 14 3.2.2 Main limitations of alcohol fuels ____________________ 15

3.3 Catalysts _________________________________________ 16 3.3.1 Molybdenum-based catalysts ______________________ 16 3.3.2 Catalyst preparation techniques ____________________ 17 3.3.3 Catalyst characterization techniques ________________ 21

3.4 Reaction mechanism ________________________________ 23

3.5 Reaction and analysis system _________________________ 23 3.5.1 Reactor setup __________________________________ 23 3.5.2 Catalyst loading and pretreatment __________________ 24

Table of contents xiii

3.5.3 Operating conditions _____________________________ 26 3.5.4 Analysis system ________________________________ 26 3.5.5 Data treatment _________________________________ 27

3.6 Main results from the experimental work _________________ 27 3.6.1 Conventional precipitation vs coprecipitation in

microemulsions: catalyst performance (Paper II) ______________ 27 3.6.2 Catalyst characterization (Paper II) _________________ 30 3.6.3 Effect of methanol cofeeding (Paper III) ______________ 31 3.6.4 Effect of ethanol cofeeding ________________________ 35

4 HYDROCONVERSION OF FISCHER-TROPSCH WAX

37

4.1 The role of hydroconversion __________________________ 37

4.2 Hydroconversion from a historical and industrial perspective _ 38

4.3 Properties of low-temperature Fischer-Tropsch fuels _______ 39

4.4 Catalysts _________________________________________ 40 4.4.1 Bifunctional catalysts ____________________________ 40 4.4.2 Catalyst preparation techniques ____________________ 41 4.4.3 Catalyst characterization techniques ________________ 42 4.4.4 Catalyst deactivation ____________________________ 45

4.5 Reaction mechanism ________________________________ 45 4.5.1 Classical bifunctional mechanism & ideal hydrocracking _ 45 4.5.2 Concurring mechanisms __________________________ 47

4.6 Reaction and analysis system _________________________ 48 4.6.1 The transition from n-hexadecane to paraffinic wax ____ 48 4.6.2 Reactor setup __________________________________ 48 4.6.3 Catalyst loading and pretreatment __________________ 49 4.6.4 Operating conditions _____________________________ 51 4.6.5 Analysis system ________________________________ 51 4.6.6 Product lumping & data treatment __________________ 52

4.7 Main results from the experimental work _________________ 53 4.7.1 Comparison between platinum and palladium as the metal

function: catalyst performance (Paper V) ____________________ 53 4.7.2 Catalyst characterization (Paper V) _________________ 56 4.7.3 Catalyst deactivation (Papers IV and V) ______________ 59

Table of contents xiv

4.8 Kinetic modelling (Paper VI) __________________________ 61 4.8.1 Overview of modelling strategies ___________________ 61 4.8.2 Description of the proposed model __________________ 62 4.8.3 Main results from the modelling work ________________ 64

5 CONCLUDING REMARKS AND FUTURE WORK __ 69

5.1 Concluding remarks _________________________________ 69

5.2 Future work _______________________________________ 70

Acknowledgements _____________________________ 73

Nomenclature _________________________________ 75

References ____________________________________ 77

1

1 INTRODUCTION

1.1 Setting the scene

The use of energy in the transportation sector accounts for about 28 % of

the world’s total energy consumption today [1]. In addition, the fuel

demand is forecast to continue growing in the coming years; for instance,

the car fleet is expected to triple by 2050, with the largest increase

corresponding to developing countries [2].

Fossil fuels (oil, natural gas and coal) made up the most important energy

source for transportation during the last century. However, the finite

nature of these raw materials and especially the environmental problems

related to their use (CO2, SO2, NOx and particulate emissions) make a

diversification of the fuel mix and a gradual shift to cleaner fuels essential.

More stringent legislations are being promulgated and will also contribute

to accelerating this transition.

In this context, biomass is expected to play a prominent role in the future

energy system. Unlike other renewable resources such as wind or solar

energy, biomass is available on-demand, providing a reliable alternative to

fossil fuels. Thermochemical conversion of biomass generates synthesis

gas (syngas), which can be further converted to various fuels. This route

has a high flexibility because syngas can also be obtained from other carbon

sources, such as natural gas and coal.

Regardless of the raw material, syngas is an attractive building block for

producing liquid or gaseous fuels: synthetic natural gas (SNG); methanol;

dimethyl ether (DME); ethanol and higher alcohols; and Fischer-Tropsch

(FT) fuels. In all these syngas conversion processes, choosing the proper

catalyst is of paramount importance.

1.2 Objective of the work

The objective of this work has been to gain insight into the production of

second-generation biofuels, specifically into the syngas conversion step.

This PhD thesis is based on six appended papers.

Paper I broadly reviews the topic, presenting the latest advances in

catalytic conversion of syngas to fuels, devoting special attention to the use

of biomass as primary raw material.

1.INTRODUCTION 2

The other papers are focused on two particular reactions: higher alcohol

synthesis (HAS) (Papers II and III) and hydroconversion of FT wax

(Papers IV-VI). Different aspects of a heterogeneous catalytic reaction are

addressed in the publications:

Paper II - catalyst preparation:

A novel K-Ni-MoS2 catalyst is synthetized by coprecipitation in

microemulsions (MEs). Higher yields of ethanol and higher

alcohols are obtained over this catalyst, as compared to a

conventionally prepared sample.

Paper III - process optimization:

The possibility of increasing higher alcohol yield by recycling one

of the by-products (methanol) is evaluated.

Paper IV - catalyst deactivation:

The deactivation behavior of a bifunctional platinum/silica-

alumina catalyst is studied in the hydroconversion of n-

hexadecane. This work also served as a starting point for Papers V

and VI, where paraffinic wax is used in the experiments (instead

of a model compound).

Paper V - catalyst formulation:

The differences between platinum and palladium as catalyst metal

function in paraffinic wax hydroconversion are evaluated.

Paper VI - kinetic modelling:

A lumped kinetic model is proposed to describe the

hydroconversion of FT wax. The model is used to further analyze

the differences between the catalysts presented in Paper V.

The work presented in this PhD thesis was performed during the years

2012-2015, mainly at the Department of Chemical Engineering and

Technology - KTH, Stockholm, Sweden. A small part of the catalyst

characterization was conducted at Åbo Akademi University (Finland),

University of Córdoba (Spain) and the Institute for the Study of

Nanostructured Materials (Italy).

1.INTRODUCTION 3

1.3 Thesis outline

Following this introduction, Chapter 2 gives an overview of second-

generation biofuels. Chapter 3 is focused on HAS and summarizes the

results from Papers II and III. Chapter 4 is devoted to the hydroconversion

of FT wax and condenses the experimental and modelling studies from

Papers IV-VI. Finally, Chapter 5 highlights the most important findings of

the PhD thesis and gives recommendations for future work.

5

2 SECOND-GENERATION BIOFUELS

Paper I

2.1 The role of second-generation biofuels

The fossil fuel price clearly shows a decreasing trend over the last 5 years

[3]. Moreover, unconventional fossil fuel reserves have become

exploitable, eliminating the need for a fast shift towards renewable fuels.

However, urgent action is still needed to fight climate change and preserve

the world for future generations.

CO2 is the main anthropogenic contributor to the greenhouse effect.

Although natural emissions of CO2 are much greater than anthropogenic

emissions (about 20 times), they were counterbalanced during the pre-

industrial era by the uptake by oceans and the terrestrial biosphere [4].

However, the human activities initiated in 1750 with the Industrial

Revolution have caused an alteration of the so-called carbon cycle. As a

result, the levels of CO2 in the atmosphere are higher today than at any

point in our history (Figure 2-1): in 2013, the concentration reached 396

ppmv, with a growth of 2 ppmv per year in the last decade [5].

Figure 2-1. Atmospheric concentrations of CO2 from Mauna Loa (19°32’N, 155°34’W – red) and the South Pole (89°59’S, 24°48’W – black) since 1958. Reprinted from [6] with permission from the Intergovernmental Panel on Climate Change.

2.SECOND-GENERATION BIOFUELS 6

The energy sector accounts for almost 70 % of the total anthropogenic

greenhouse gas (GHG) emissions [5], mainly due to the combustion of

fossil fuels. In 2012 fossil fuels represented more than 80 % of the primary

energy supply (Figure 2-2). If only the transportation sector is considered,

the numbers are even more alarming: 93 % of the energy consumed in 2012

was provided by oil-derived products [1].

Figure 2-2. World total primary energy supply in 2012: fuel shares [the total energy supply is indicated below the chart; the category others includes geothermal, solar and wind energy]. Author's own elaboration with data from [1].

The share of renewable energy in transport still remains very limited: in

the EU, it reached 5.4 % in 2013 and it is expected to increase up to 5.7 %

in 2014 [7]. Significant progress has to be made to meet the target set in

the 2009/28/EC Directive: all EU Member States must ensure that at least

10 % of their transportation fuels come from renewable sources by 2020

[8]. A set of sustainability criteria have also been defined to guarantee that

the use of biofuels generate substantial GHG emission savings and, at the

same time, protect biodiversity [9].

First-generation biofuels are obtained from commodities that are also food

resources, such as starch or sugar. Several drawbacks have limited the

development of biofuels of first generation, mainly: competition with food

crops, contributing to higher food prices; controversial GHG emission

reduction; and negative environmental impact on biodiversity or water

resources [10].


In this context, there are high expectations regarding second-generation

biofuels. They are defined by the International Energy Agency (IEA) as

biofuels produced from cellulose, hemicellulose or lignin. They are

obtained from non-food resources such as various by-products (e.g. cereal

straw, sugar cane bagasse or forest residues), waste (organic municipal

solid waste) or dedicated energy crops [10].

Second-generation biofuels tackle most of the drawbacks present in first-

generation biofuels. The GHG balance is still a controversial issue,

especially when considering the effect of indirect land use change [10, 11].

However, the IEA has recently performed a comprehensive life cycle

assessment (LCA) of different biofuels obtained from agricultural residues

or woody biomass; the results show a clear GHG emission reduction, as

compared to conventional fossil-derived fuels or first-generation ethanol

obtained from corn [12].

2.2 Production routes of second-generation biofuels

The production of biofuels from lignocellulosic feedstocks can be

accomplished through two different routes:

Biochemical route:

In a first step, the cellulose and hemicellulose contained in the

biomass are converted into sugars by means of acid or enzymatic

hydrolysis. Thereafter, the sugars are fermented to alcohols,

mainly ethanol [13].

Thermochemical route:

There are three thermochemical options for converting biomass

into fuels. The first alternative involves the production of syngas

via gasification and its subsequent catalytic conversion. A

schematic representation of this alternative is shown in Figure 2-3.

Since it is the approach followed in this PhD work, it will be

thoroughly described in the subsequent sections.

The other two options are biomass pyrolysis and liquefaction,

which yield a bio-oil that can serve, after upgrading, as fuel [14-

16].


Figure 2-3. Production of second-generation biofuels via gasification and catalytic conversion (thermochemical route). Author's own elaboration.

2.2.1 Syngas generation

Gasification is the conversion of a carbonaceous material (such as biomass

or coal) into a gas mixture, known as syngas, by using high temperature

and a reducing (oxygen-deficient) environment.

Mature gasification technology is available for coal conversion [17].

Although biomass can be considered a young coal, there are some key

differences between both feedstocks and research is still needed to

optimize biomass gasification at commercial scale [18, 19]. Particularly,

biomass has a smaller heating value; the ash has a lower melting point and

it is very aggressive in molten state; and the tar content in the product gas

is higher when gasifying biomass, especially at relatively low temperatures

(800-900 °C).

Gasifiers are classified according to different criteria: biomass feeding

section (top, side); gas-solid contacting mode (fixed or moving bed,

fluidized bed, entrained-flow bed); gasifying agent (oxygen, steam, air);

temperature range; heating mode (direct, indirect or allothermal); and

pressure (atmospheric, pressurized) [20-22]. Considering these criteria,

seven different types of gasifiers are mainly identified: updraft fixed bed,

downdraft fixed bed, bubbling fluidized bed (BFB), circulating fluidized

bed (CFB), dual fluidized bed, entrained flow (EF) and plasma gasifiers.

The different technologies have the potential to be used for fuel

applications and from the available data it is not clear which option is the

best in terms of syngas production cost [22].

BFB and CFB gasifiers are promising solutions. They have been widely

used for heat and power applications and research efforts are now focused

on developing pressurized systems using oxygen or steam, more suitable

for fuel production processes [22]. BFB gasifiers operate at relatively low


gas velocities (typically below 1 m/s). On the other hand, CFB gasifiers

work at higher velocities, dragging the solid particles upwards with the gas.

The particles are then separated from the gas in a cyclone and recycled to

the bottom of the bed [23].

Nevertheless, EF gasification is the option most likely to succeed in the

near future due to its current status of development. Even though this

technology may involve high pretreatment costs, it also has the greatest

potential to be scaled up and benefit from economies of scale [22, 24]. In

EF gasifiers, the feedstock is fed co-currently with the gasifying agent by

means of a burner. The operating temperatures are much higher than in

the fluidized gasifiers, allowing the conversion of tars and methane and

producing a high-quality syngas [23].

A scheme of the three aforementioned gasifiers is shown in Figure 2-4.

Figure 2-4. Schematic representation of bubbling fluidized bed, circulating fluidized bed and entrained flow gasifiers. Adapted from [22] with permission from E4tech.

2.2.2 Syngas conditioning

Depending on the biomass feedstock and the gasification technology,

different syngas compositions and impurity levels are obtained. The main

syngas components are H2, CO, CO2, CH4 and N2 (if air is used as gasifying

agent); the potential impurities are tars, particulates, sulfur compounds,

chlorine compounds, nitrogen compounds, heavy metals, alkali metals,

unsaturated hydrocarbons and polyaromatic hydrocarbons [17, 25].

The syngas composition obtained with different technologies is shown in

Table 2-1. The presented values should not be seen as representative


figures for each gasifier, but only as an illustration of the variety of syngas

compositions that may be achieved.

Table 2-1. Syngas composition obtained with different gasification technologies. Author's own elaboration with data from [22].

Technology BFB

(Enerkem)

CFB

(ECN BIVKIN)

EF

(KIT)

Gasifying agent Air Air O2

Heating mode Direct Direct Direct

Composition

(%vol, dry)

H2 6-12 18 40

CO 14-15 16 39

CO2 16-17 16 20

CH4 3-4 6 0.1

C2+ hydrocarbons 3-4 2 -

N2 36-58 42 0.1

Biomass-derived syngas is usually H2-deficient. Since most of the

subsequent catalytic conversion processes demand a H2-rich syngas,

adjustment of the H2/CO ratio (water-gas shift reaction) may be required

[26, 27]. In addition, CO2 and CH4 may need to be separated from the gas

stream [17, 26].

Concerning the removal of impurities, comprehensive reviews on the topic

have recently appeared [28, 29]. Cold gas cleanup technologies are mature,

but they are energy inefficient and generate waste water streams that need

to be treated. Warm and hot gas cleanup processes focus instead on

impurity removal at high temperature. However, in this case, the lifetime

of catalysts and sorbents is still a challenge to be overcome.

2.2.3 Catalytic conversion of syngas to fuels (Paper I)

Syngas can be converted to different fuels, depending on the catalyst and

operating conditions. Although the Fischer-Tropsch synthesis (FTS) is the

most known pathway, there are other routes, as shown in Figure 2-5.


Figure 2-5. Potential syngas-to-fuel pathways. Author's own elaboration.

Research on catalytic conversion of syngas to fuels has been conducted

since the beginning of the 20th century. In 1902 Sabatier and Senderens

discovered that syngas could be converted into methane over a cobalt or

nickel catalyst at atmospheric pressure [30]. In 1913 BASF patented the

synthesis of higher hydrocarbons and oxygenates using cobalt and osmium

catalysts at high pressure [31]. But it was in 1923 when Franz Fischer and

Hans Tropsch at the Kaiser Wilhelm Institute for Coal Research achieved

the greatest breakthrough: they discovered that alkalized iron catalysts at

high temperature and pressure (400-450 °C, 100-150 atm) could convert

syngas into an oily mixture of oxygenated compounds (known as synthol)

[32]. This mixture was further upgraded and successfully used as

transportation fuel in a NSU motorbike, something that marked the

beginning of a new industry: producing transportation fuels from syngas

was indeed possible [33]. In parallel, also in 1923, Alwin Mittasch and

Mathias Pier, working for BASF, developed a process to convert syngas into

methanol over mixed zinc and chromium oxides [34, 35].

From that date, catalyst formulations, reactor systems and process

conditions have been optimized. The development that took place over the

last 100 years is not the subject of this PhD thesis, but the current status of

https://en.wikipedia.org/wiki/Alwin_Mittasch


the technology and the latest advances have been recently reviewed in

Paper I.

2.3 Production plants of second-generation biofuels

Academic institutions, industry and governments have made important

research efforts over the last 10-15 years in order to convert second-

generation biofuels into an industrial reality. There are several commercial

plants in operation following the biochemical approach, mainly located in

the USA, Brazil and Europe [36].

As concerns the thermochemical route, the only commercial facility in

operation is Enerkem Alberta Biofuels (Edmonton, Canada) [37]. In this

plant, municipal solid waste is converted to syngas in a BFB gasifier.

Subsequently the syngas is catalytically transformed to methanol, ethanol

and various chemicals. The project was successfully started up in 2014,

producing only methanol. The ethanol module is expected to be ready for

operation in the second half of 2016. Even though the capacity of the plant

is 38 million liters of biofuel per year, it is not clear whether or not this

capacity has been reached.

Many other pilot, demonstration and commercial plants are planned,

under construction or starting to operate. However, it must be noted that

the available information is usually scarce and/or not updated.

13

3 HIGHER ALCOHOL SYNTHESIS

Papers II and III

3.1 Higher alcohol synthesis from a historical and industrial

perspective

Oxygenates and particularly alcohols were products of the first FT

processes. The discovery in 1923 of zinc and chromium oxide catalysts for

methanol synthesis was essential for the future development of the

synthesis of alcohols from syngas. Shortly after, it was discovered that

modification of those catalysts with alkaline metals enhanced higher

alcohol yields [38]. As a matter of fact, HAS achieved considerable

importance in the 1930-1940s, with several plants in operation in the USA

and Germany [38, 39]. The plants were dismantled after World War II due

to the increasing petroleum availability and the need for more pure

alcohols [40].

The oil crisis in the 1970s renewed the interest in the synthesis of ethanol

and higher alcohols, to be used in gasoline blends. Numerous patents on a

wide range of catalysts were filed at that time [41]. The low oil prices in the

late 1980s again reduced the attention devoted to this process [40, 41].

Nevertheless, the increasing concerns about environment and local energy

security in the recent years have prompted a period of extensive research

on the topic (see Paper I). The current increased use of alcohols has been

driven by mainly legislation, for instance the 2009/28/EC Directive in

Europe [8] and the Energy Independence and Security Act in the USA [42].

However, the catalytic systems developed so far for HAS suffer from low

activity, poor product distribution or severe reaction conditions [41, 43].

Therefore, no commercial plants exist solely for the synthesis of ethanol

and higher alcohols from syngas, although several companies have

pursued commercialization [44]. Range Fuels, who appeared to be the

closest to achieve it [44], closed down its plant in 2011 after running into

technical problems [45].

3.2 Properties of alcohol fuels

The use of alcohols as neat fuel or in blends with gasoline is not a new thing.

Both Henry Ford (founder of the Ford Motor Company) and Charles F.

https://en.wikipedia.org/wiki/Ford_Motor_Company

3.HIGHER ALCOHOL SYNTHESIS 14

Kettering (head of research at General Motors) claimed in the 1920s that

alcohol made from farm products and cellulosic materials was the “fuel of

the future” [46]. Ethanol was indeed used to fuel cars in the 1920s and

1930s, but it was thereafter replaced by fossil-derived products [47].

The aim of this section is to give a general overview of the main advantages

and disadvantages of using alcohol fuels. Due to their high octane numbers

and low cetane numbers, alcohols are more suitable for use in spark-

ignition engines [48], so special attention will be devoted to gasoline-

alcohol blends. In any case, alcohols may also be used in compression-

ignition engines, together with another fuel that is more autoignitable or

adding an ignition improver [48-50].

Some key properties of gasoline and various alcohols are summarized in

Table 3-1; they will be discussed in Sections 3.2.1 and 3.2.2.

Table 3-1. Key properties of gasoline, methanol, ethanol and 1-butanol. Author's own elaboration with data from [49, 51, 52].

Gasoline Methanol Ethanol 1-Butanol

Density at 20 °C 0.74-0.80 0.79 0.79 0.81

Research octane

number 91-99 136 129 96

Motor octane

number 81-89 104 102 78

Heat of vaporization

(MJ/kg) 0.36 1.2 0.92 0.43

Energy density

(MJ/L) 32 16 20 27-29

Reid vapor pressure

(kPa) 65 32 16 2

Boiling point (°C) 27-225 65 78 118

3.2.1 Main advantages of alcohol fuels

The octane number is a measure of the anti-knock quality of a fuel and two

methods are normally used for its determination: the research octane

number (RON) test and the motor octane number (MON) test [53]. Since


the data reported in the literature show considerable scatter, the values

presented in Table 3-1 should only be seen as a trend indicator. Methanol

and ethanol act as octane boosters when mixed with gasoline. The resulting

fuels, with higher octane numbers, enable the use of higher compression

ratios, resulting in an increase of the thermal efficiency [54]. On the other

hand, higher alcohols lead to a smaller improvement or even a reduction

in the octane number.

The greater heat of vaporization of alcohols, as compared to gasoline (Table

3-1), results in an additional evaporative cooling of the air–fuel mixture in

the cylinder prior to ignition. This feature can, to a greater extent, prevent

knock and increase the compression ratio in specifically-designed engines,

leading to a further increase of the thermal efficiency [54].

Alcohols are also attractive in terms of GHG emissions. Spark-ignition

engines fueled with gasoline-alcohol blends in most cases show lower CO,

CO2, unburnt hydrocarbon and particle number emissions [48, 55-58]. In

the case of NOx, results are more controversial [59].

3.2.2 Main limitations of alcohol fuels

Despite the benefits previously described, adding alcohols to gasoline

involves some problems that need to be addressed. One of the obvious

limitations of alcohol fuels is the lower energy density, as compared to

gasoline (Table 3-1). This results in a mileage decrease, more noticeable in

the case of alcohol-rich blends.

The vapor pressure is another crucial property of automotive fuels,

affecting for instance the starting properties of the engine in cold

environments (if the vapor pressure is too low) or the evaporative

emissions from the vehicle (if the vapor pressure is too high) [52]. The Reid

vapor pressure (RVP) is the vapor pressure at 37.8 °C, measured following

a specific method [51], and is a commonly used parameter for fuel

evaluation. Although neat alcohols have significantly lower vapor

pressures than gasoline (Table 3-1), gasoline-alcohol blends form near-

azeotropic mixtures, having different RVPs than expected for ideal

mixtures. For example, addition of small amounts of methanol or ethanol

to gasoline lead to a significant increase of the vapor pressure. To handle

this problem, Andersen et al. [51] have proposed blends using more than

one alcohol; in this way, it is possible to achieve a RVP indistinguishable

from that of the base gasoline.


High vapor pressures and the low boiling point of alcohols (Table 3-1) may

also give rise to vapor-lock problems [60]: in hot environments, the fuel

vaporizes while it is in the fuel line, causing a disruption of the fuel pump

operation [61].

Phase separation problems may also occur when methanol and ethanol are

blended with gasoline, especially at low temperature and/or in the

presence of water, even in small amounts. The resulting mixture of alcohol

and water settles at the bottom of the fuel tank and can cause corrosion

[60]. In general, alcohols can be highly corrosive for some engine parts,

particularly methanol [60].

3.3 Catalysts

3.3.1 Molybdenum-based catalysts

There is a wide variety of catalyst formulations that can be used to convert

syngas into ethanol and higher alcohols. They are normally grouped into

four categories, namely: rhodium-based catalysts; modified methanol

synthesis catalysts; modified FTS catalysts; and modified molybdenum-

based catalysts. For a description of the different catalytic options, the

reader is referred to Paper I or to comprehensive reviews on the topic [41,

60, 62].

Different forms of molybdenum have been studied, including carbides,

oxides, phosphide, sulfides and the metallic form [63]. However,

molybdenum sulfide materials (MoS2) are preferred [64] and, therefore,

they were particularly studied in this PhD thesis. They are sulfur resistant,

display high activity for the water-gas shift reaction and deactivate slowly

by coke deposition [41, 60]. Moreover, they are suitable for processing

syngas with low H2/CO ratios, typically obtained from biomass gasification

processes [26].

MoS2 alone is a methanation catalyst, so modification with promoters is

essential. In the 1980s, The Dow Chemical Company [64] and Union

Carbide Corporation [65] discovered that alkali (Li, Na, K, Rb, Cs)

promoters shift the selectivity from hydrocarbons to alcohols. It is

generally accepted that the heavier alkalis are preferred over lithium and

sodium. Among the others, potassium is by far the most widely used [41,

60, 63]; it has been reported to be the best promoter [66-68], although

there are discrepancies in this ranking [69].


The product distribution over alkali-promoted molybdenum sulfide

catalysts generally follows the Anderson-Schulz-Flory (ASF) distribution,

which limits the production of ethanol and higher alcohols [62]. In order

to enhance C2+-alcohol selectivity, a second promoter, a Group VIII metal,

is usually added [60, 62, 63]. Cobalt was the preferred option in the patent

literature [64, 68] and it has been the most studied metal. Nickel-modified

alkali-doped molybdenum sulfide catalysts have also been examined [70-

74], showing higher activities than the cobalt-modified catalysts, but also

higher hydrocarbon selectivities. Nickel was studied in this work due to the

smaller amount of published information on this promoter, as compared

to cobalt.

The aforementioned patent literature [64, 68] also states the preference of

unsupported catalysts over supported materials. Nevertheless, a great

variety of supports have been studied: Al2O3, CeO2, SiO2, mixed MgAl

oxide, activated carbon or multi-walled carbon nanotubes. In this PhD

thesis, only unsupported catalysts were evaluated and the reader is

directed to specific publications for more information on supported

materials [75-81].

3.3.2 Catalyst preparation techniques

As just mentioned, unsupported nickel-modified potassium-doped

molybdenum sulfide catalysts were studied in this work. Two different

molybdenum precursors are generally used in the synthesis: ammonium

tetrathiomolybdate [(NH4)2MoS4], that contains over-stoichiometric

amounts of sulfur; or ammonium molybdate tetrahydrate

[(NH4)6Mo7O24·4H2O], that instead needs to be sulfidized [82].

The first precursor is chosen in this PhD thesis and, therefore, the catalyst

preparation method involves three main steps: i) coprecipitation of the

nickel and molybdenum precursors; ii) subsequent doping of the

precipitate with the potassium precursor; iii) and, ultimately, a thermal

treatment [70, 71, 73, 74].

The coprecipitation of the nickel and molybdenum precursors is

normally carried out in aqueous solution (conventional

coprecipitation). In this PhD thesis, a novel preparation route is

also proposed (coprecipitation in microemulsions). The details of

both preparation methods are shown below.

Potassium doping can be achieved by physical mixing or incipient

wetness impregnation. Ferrari et al. have recently shown that


physically mixing the materials produces better results than the

impregnation method [83].

The thermal treatment leads to the decomposition of the

molybdenum precursor (see Equation 3-1). A first decomposition

takes place at about 150-200 °C, yielding molybdenum trisulfide

(MoS3,). Further heating up to 300-400 °C is required to obtain

MoS2 [84].

(NH4)2MoS4 ∆→ 2NH3 + H2S + MoS3

∆→MoS2

Equation 3-1

A total of four catalysts will be discussed in this section. The novel ME

catalyst will be compared with an analogous conventional catalyst. Two

additional conventional catalysts, promoted only with nickel or potassium,

have also been studied, in order to have a better understanding of the

individual effect of each promoter and explain the differences between the

ME catalyst and the conventional one. The nominal Ni/Mo and K/Mo

ratios of the catalysts are shown in Table 3-2.

Table 3-2. Nominal HAS catalyst composition. Adapted from Paper II with permission from Elsevier.

Catalyst Ratio (mol/mol)

Ni/Mo K/Mo

K-Ni-MoS2 0.50 1.50

ME K-Ni-MoS2 0.50 1.50

Ni-MoS2 0.50 0

K-MoS2 0 1.50

Conventional coprecipitation

The bipromoted conventional catalyst (K-Ni-MoS2) was prepared by

adding dropwise, under continuous stirring, an aqueous solution of

Ni(CH3COO)2·4H2O to an aqueous solution of (NH4)2MoS4. The resulting

black precipitate was aged for 24 h and then recovered by centrifugation.

The recovered precipitate was first washed with deionized water (Milli-Q)

and, subsequently, with a mixture of ethanol and deionized water. After


each wash, the powder was recovered by centrifugation. The washed

powder was dried at 50 °C and then crushed and sieved to a pellet size of

45–250 µm. Alkali doping was achieved by mechanically mixing the dried

catalyst precursor with K2CO3 (45–250 µm). The final catalyst was

obtained after a thermal treatment at 450 °C for 90 min (heating ramp of

2 °C/min), under H2 flow. A second sieving was performed to discard

particles with a pellet size above 250 µm that could be formed during the

heat treatment. The final sample was kept in a tightly closed container

before being characterized and tested.

The monopromoted catalysts were synthetized using the same procedure,

but excluding the doping step with nickel (K-MoS2) or potassium (Ni-

MoS2).

Coprecipitation in microemulsions

A ME is a three-component solution (water, oil and surfactant)

characterized as being optically isotropic and thermodynamically stable

[85].The droplet size in these colloidal systems is in the order of 2–100 nm

[85-87].

There are three types of ME structures: oil-in-water MEs, which consist of

oil droplets stabilized by a monolayer of surfactant and dispersed in a

continuous water phase; water-in-oil MEs, composed of water droplets and

dispersed in a continuous oil phase; and bicontinuous MEs, in which both

oil and water form continuous domains as interconnected sponge-like

channels [87]. The type of ME that is formed depends on the oil/water ratio

and on the hydrophilic-lipophilic balance of the surfactant [87].

The synthesis of metal nanoparticles in MEs was described for the first

time by Boutonnet et al. [88] in 1982. They used water-in-oil MEs, which

have been the most studied systems for catalyst preparation. Catalysts

prepared using MEs have shown improved properties in many applications

when compared with conventional catalysts [86].

The novel catalyst proposed in this work (ME K-Ni-MoS2) was synthetized

through a coprecipitation route, but using MEs instead of aqueous

solutions. Two analogous water-in-oil ME systems were prepared: the first

ME (ME1) contained the nickel salt and acetic acid; the second ME (ME2)

contained the molybdenum salt.

A quaternary system was used; the detailed composition is shown in Table

3-3. MEs with similar composition were successfully used before for

catalyst preparation [89-91]. Since the chosen surfactant (CTAB) is ionic

and contains a single hydrocarbon chain, a co-surfactant (1-butanol) is

required to form the MEs [92].


Table 3-3. Composition of the ME systems used for the preparation of ME K-Ni-MoS2. Adapted from Paper II with permission from Elsevier.

Phase Compound (s) Composition (wt%)

Oil i-octane 53

Surfactant CTAB 15

Co-surfactant 1-butanol 12

Water ME1: water, Ni(CH3COO)2·4H20, acetic acid ME2: water, (NH4)2MoS4

20

The ME catalyst was prepared by adding dropwise, under continuous

stirring, ME1 to ME2. The black precipitate (see Figure 3-1) was aged for

24 h and then recovered by centrifugation. In order to remove surfactant

and hydrocarbon residues, the washing procedure was modified, following

the work of Nassos et al. [91]: the precipitate was first washed with a

mixture of chloroform and methanol and, then, twice with methanol. The

rest of the preparation procedure is analogous to that used for the

conventional catalyst: drying, crushing and sieving, alkali doping and,

finally, thermal decomposition (the temperature ramp was set to 0.5

°C/min in this case, so as to ensure complete elimination of the catalyst

precursor residues not removed during the washing steps).

Figure 3-1. Coprecipitation in MEs. Reprinted from Paper II with permission from Elsevier.


3.3.3 Catalyst characterization techniques

Catalyst characterization is defined by Bartholomew and Farrauto as the

“measurement of the physical and chemical properties of a catalyst

assumed to be responsible for its performance in a given reaction” [93]. In

this part of the PhD thesis, catalyst characterization aimed to identify the

differences between the conventionally prepared catalyst and the novel

sample prepared using the ME technique.

Catalysts were characterized by means of thermogravimetric analysis

(TGA), inductively coupled plasma (ICP), X-ray photoelectron

spectroscopy (XPS), nitrogen adsorption, X-ray diffraction (XRD),

scanning electron microscopy (SEM) and transmission electron

microscopy (TEM). An overview of the different techniques is given below.

For a deep understanding of the characterization methods, the reader is

referred to specialized literature (e.g. [93]); for detailed information on

the instruments, analysis conditions or sample preparation, Paper II is

advised.

Thermogravimetric analysis

TGA is an analytical technique used to determine the thermal stability of a

sample by monitoring the weight change that occurs when subjecting the

sample to a temperature program. The measurement can be carried out

under different atmospheres [94].

The use of TGA had two main purposes:

Verify the suitability of the temperature used for the thermal

decomposition in H2

Verify that surfactant and oil residues were completely eliminated

from the ME catalyst

Inductively coupled plasma

ICP is a technique for chemical analysis of a sample by vaporizing it in a

plasma heater [93]. ICP analyses were used for evaluating bulk catalyst

composition.

X-ray photoelectron spectroscopy

XPS is a spectroscopic technique based on the excitation of the sample

surface with a beam of X-rays and the detection of the emitted

photoelectrons [93]. It is the most commonly used surface analysis method

in catalyst characterization. The information depth in XPS (i.e. maximum

depth from which useful information can be obtained) is in the order of a

few nanometers [95]. Although the technique can provide various kinds of

information (elemental composition of the surface, chemical or electronic


state, dispersion of supported catalysts) [95, 96], it was used here to

evaluate catalyst surface composition.

Nitrogen adsorption

Nitrogen adsorption at 77 K (liquid nitrogen boiling temperature) is the

most used technique for assessing the surface area and pore texture of a

sample. The volume of adsorbed nitrogen at different pressures

(adsorption isotherm) is recorded and used in the calculations [97].

The model developed by Brunauer, Emmer and Teller (BET model) in 1938

[98] is still the most widely applied for determining surface areas.

Although it is an over-simplified model [99], it provides reliable results at

relative pressures between 0.05 and 0.30 [93]. In this work, data collected

at relative pressures between 0.06 and 0.2 was used to calculate surface

areas.

X-ray diffraction

XRD is a technique used to determine the bulk structure and composition

of samples with crystalline structures [100]. The X-rays interfere with the

sample and the intensity of the diffracted rays is recorded at different

incident angles. The technique presents some limitations when applied in

catalyst characterization: the material needs to be sufficiently crystalline to

diffract X-rays (crystallites larger than 3-5 nm) and the concentration of

the crystalline phases in the sample needs to be reasonably high to be

detected (greater than ~ 1 %) [93, 100].

Microscopy: scanning and transmission electron microscopy

Electron microscopy is a technique used to study the surface texture and

morphology of a sample [93].

In SEM, the sample is probed with a focused electron beam and

the yield of either secondary or back-scattered electrons is

recorded as function of the position of the primary electron beam.

Nowadays SEM instruments can perfectly reach resolutions of

about 5 nm, but the technique is normally used for characterizing

surfaces of micrometer dimensions. SEM can be combined with

energy dispersive X-ray spectroscopy (SEM-EDX) to perform

elemental analysis [100] .

In TEM, the electrons transmitted through a thin sample on a

conducting grid are used to form a high resolution image [93]. This

technique has a higher resolution than SEM (down to 0.1 nm) and

it is generally used for characterizing surfaces of nanometer

dimensions [100].


3.4 Reaction mechanism

Only few works have specifically studied the reaction mechanism taking

place on alkali-doped molybdenum sulfide catalysts [63]. It has been

generally accepted over the years that the reaction occurs via a stepwise CO

insertion, as proposed by Santiesteban et al. in the late 1980s [101]. Their

isotopic studies showed that CO is inserted into an alkyl group adsorbed

on the surface (RCH2*), forming an acyl precursor (RCH2CO*). This acyl

species can be hydrogenated to a longer alkyl group (RCH2CH2*) or to the

corresponding alcohol (RCH2CH2OH). The commonly accepted CO-

addition mechanism is depicted in Figure 3-2.

Figure 3-2. CO-addition mechanism for the conversion of syngas to alcohols and hydrocarbons. Reprinted from [102] with permission from Elsevier.

However, Christensen et al. [103] have proposed an additional chain

growth mechanism. The authors observed that cofeeding ethanol with

syngas over a K-Co-MoS2/C catalyst enhanced the production of 1-butanol

to a greater extent than that of 1-propanol; 1-butanol was also formed by

ethanol conversion in a nitrogen atmosphere. In light of these results, it

was suggested that coupling of alcohols, probably occurring via a classical

aldol condensation, may be an important reaction route over molybdenum

sulfide catalysts. Santos et al. [104] have recently supported this

observation.

3.5 Reaction and analysis system

3.5.1 Reactor setup


The synthetized catalysts were tested in a high-pressure rig. A fixed-bed

tubular reactor (9/16-in stainless steel pipe, 9.12 mm inner diameter and

30 cm length) was employed, operated in down-flow mode. The reactor

was heated by an electrical furnace and the temperature was regulated by

a cascade control, with one sliding thermocouple in the catalyst bed and a

second thermocouple in the oven. To ensure an almost isothermal

operation (SP ± 0.5 °C), this control architecture was complemented with

an external aluminium jacket and the catalyst was diluted with silicon

carbide. The pressure was controlled by a valve placed on the reactor outlet

line and connected to a transducer that measured the pressure prior to the

reactor.

The gases (syngas, He) were fed to the reactor system by means of thermal

mass flow controllers (Bronkhorst EL-FLOW). A premixed syngas with

H2/CO ratio = 1 and 4 mol% N2 as internal standard was utilized in the

experiments. Helium was used during leak testing, to create an inert

atmosphere when needed and to keep the syngas partial pressure constant

when methanol was cofed.

Methanol was cofed to the system, together with the syngas, in the

experiments reported in Paper III. Methanol was added by means of a

HPLC dosing pump (Gilson 307) and the liquid was evaporated before

being mixed with the inlet gaseous stream, immediately prior to the

reactor.

The reactor setup was placed in a hot box, kept at 180 °C, which served two

main purposes: preheating of the inlet gases and avoiding condensation of

the outlet gases. The latter is essential since reaction products are analyzed

on-line, in the gas phase.

A piping and instrumentation diagram (P&ID) of the reactor setup is shown in Figure 3-3.

3.5.2 Catalyst loading and pretreatment

The reactor was loaded with ca. 0.7 g of catalyst in the experiments

reported in Paper II and ca. 1.8 g of catalyst in the experiments reported in

Paper III. Higher amounts were needed in the latter experiments due to

the minimum flow rate limitations of the methanol pump. As mentioned

before, the catalyst was diluted with silicon carbide to minimize

temperature gradients (ratio catalyst/SiC ~ 1/4.5). The catalyst bed was

enclosed by quartz wool and supported on a porous plate.


Figure 3-3. P&ID of the HAS reactor setup. Author's own elaboration.


The catalysts were subjected to thermal treatment ex situ, as described in

Section 3.3.2, and no additional treatment in situ was carried out.

3.5.3 Operating conditions

HAS over molybdenum sulfide catalysts is usually carried out at

temperatures and pressures in the range of 250–350 °C and 50–100 bar,

respectively [60, 63, 105].

The operating conditions used in this study were chosen taking into

consideration previous work by our research group [70]: T = 340/370 °C,

P= 71/ 91 bar, gas hourly space velocity (GHSV) = 2 000 – 14 000 NmL

h-1 gcatalyst-1. In the experiments with methanol cofeeding, concentrations of

methanol ranging from 1.5 % to 10 % were evaluated. More detailed

information can be found in the corresponding publications (Papers II and

III).

It is important to mention that all tested catalysts were first stabilized, until

reaching a pseudo steady-state. The duration of the stabilization period

varied, depending on the catalyst and operating conditions, from 20 h to

more than 100 h. After this period, the catalysts were periodically tested at

reference conditions to check for deactivation, obtaining a fairly stable

performance. Therefore, the activity and selectivity data presented in this

work are not substantially influenced by catalyst decay.

3.5.4 Analysis system

Product analysis was carried out using an on-line GC, equipped with a

thermal conductivity detector (TCD) and two flame ionization detectors

(FIDs). As mentioned before, product condensation should be avoided, so

all piping from the reactor to the GC and all the GC parts in contact with

product samples were kept at 180 °C (see Figure 3-3).

Inorganic gases (H2, N2, CO, CO2) and methane were separated by means

of packed columns and quantified with the TCD.

Organic products (alcohols, hydrocarbons, esters, etc.) were analyzed

through a two-dimensional system with Deans switch. In this system, a

high-polarity column (HP-FFAP) was used to separate hydrocarbons from

oxygenates and, additionally, to separate individual oxygenates (first

dimension); a HP-PLOT Al2O3 S column was used to separate individual

hydrocarbons (second dimension). Depending on the position of the Deans


switch, the compounds eluting from the primary column were directed

either to the primary FID (oxygenates) or to the secondary column and,

subsequently, to the secondary FID (hydrocarbons).

This online analysis system was previously developed by our research

group. It provides excellent quantitative results with carbon mass balance

closures (carbon in/carbon out) typically higher than 98 %. For more

comprehensive details of the system, the reader is referred to [106].

3.5.5 Data treatment

CO conversion was calculated from the corresponding molar flows (Fi),

which can be obtained from the TCD measurements, taking into account

the internal standard (4 mol% N2). The calculations were performed

according to Equation 3-2:

𝐶𝑂 𝑐𝑜𝑛𝑣𝑒𝑟𝑠𝑖𝑜𝑛 (%) = (𝐹 𝐶𝑂

𝑖𝑛 − 𝐹 𝐶𝑂 𝑜𝑢𝑡

𝐹 𝐶𝑂 𝑖𝑛

) ∙ 100 Equation 3-2

All organic material is detected in either the primary or the secondary FID.

Therefore, selectivity to the different compounds can be obtained using the

measured FID areas (A) and the corresponding FID response factors (R),

as shown in Equation 3-3. In the HAS part of the PhD thesis (Section 3),

selectivities are expressed on a CO2-free basis.

𝑆𝑒𝑙𝑒𝑐𝑡𝑖𝑣𝑖𝑡𝑦 𝑡𝑜 𝑝𝑟𝑜𝑑𝑢𝑐𝑡 𝑖 (%, 𝐶𝑂2 − 𝑓𝑟𝑒𝑒) =𝐴𝑖 ∙ 𝑅 𝑖

∑ 𝐴𝑗 ∙ 𝑅 𝑗

∙ 100 Equation 3-3

3.6 Main results from the experimental work

3.6.1 Conventional precipitation vs coprecipitation in

microemulsions: catalyst performance (Paper II)

One of the main problems preventing HAS commercialization is the

inadequacy of the catalytic systems developed so far. In an attempt to

achieve an improvement, two know-hows of our research group were


merged in Paper II: the ME technique [85, 86] and molybdenum sulfide

catalysts for HAS [70, 106-108].

The catalytic tests confirmed the synergetic effect of potassium and nickel

on molybdenum sulfide catalysts. Although the conventional Ni-MoS2

catalyst was the most active in CO hydrogenation, the reaction products

were mainly hydrocarbons. Potassium promotion was essential to shift the

selectivity from hydrocarbons to alcohols (Figure 3-4). When potassium

was combined with nickel, methanol space-time yield (STY) was clearly

decreased (Figure 3-5), enhancing higher alcohol production.

When the bipromoted catalysts were compared, the ME catalyst

outperformed the conventional one. The novel material resulted in higher

productivities of ethanol and higher alcohols, as shown in Figure 3-6 and

Figure 3-7. The most abundant high alcohols are, apart from ethanol, 1-

propanol and i-butanol.

Figure 3-4. Product distribution over molybdenum sulfide catalysts. Conditions: P=91 bar; T=340 °C; GHSV = 6000 NmL h-1 gcatalyst

-1. Reprinted from Paper II with permission from Elsevier.


Figure 3-5. Methanol STY over potassium-promoted molybdenum sulfide catalysts, as function of GHSV. Conditions: P=91 bar; T=340 °C. Reprinted from Paper II with permission from Elsevier.

Figure 3-6. Ethanol STY over the bipromoted catalysts, as function of GHSV. Conditions: P=91 bar; T=340 °C. Adapted from Paper II with permission from Elsevier.


Figure 3-7. 1-Propanol and i-butanol STY over the bipromoted catalysts, as function of GHSV. Conditions: P=91 bar; T=340 °C. Reprinted from Paper II with permission from Elsevier.

3.6.2 Catalyst characterization (Paper II)

In order to understand the reasons behinds the improved performance of

the novel ME catalyst, the materials were extensively characterized. A

summary of the results is presented below; the reader is referred to Paper

II for more details.

Catalyst composition was studied by means of different techniques: ICP

and SEM-EDX were used for evaluating bulk catalyst composition; XPS

was used for analyzing surface composition.

ICP and SEM-EDX confirmed the successful incorporation of the

promoters during catalyst preparation. However, during the reaction,

there was an important removal of nickel that we attributed to the

formation of nickel carbonyls, thermodynamically stable under the chosen

reaction conditions [109].

XPS was the technique that revealed the largest differences between the

bipromoted samples. The concentrations of nickel and potassium were

higher on the surface of the ME catalyst, as compared to the conventional

one. This phenomenon was observed in the fresh catalysts, but also in the

spent samples, as summarized in Table 3-4.


Table 3-4. Surface catalyst composition (bipromoted catalysts), as determined by XPS. Adapted from Paper II with permission from Elsevier.

Catalyst Ratio (mol/mol)

Ni/Mo K/Mo

K-Ni-MoS2 0.05 1.50

ME K-Ni-MoS2 0.18 3.49

spent K-Ni-MoS2 0.06 2.90

spent ME K-Ni-MoS2 0.26 5.27

The other remarkable difference between K-Ni-MoS2 and ME K-Ni-MoS2

concerned the crystallinity. Ferrari et al. [83] have claimed that small

crystallites may have a positive effect on the selectivity in HAS. XRD

analyses evidenced a much lower degree of crystallinity in the ME catalyst.

TEM micrographs qualitatively validated this observation; a quantitative

analysis of the nano-scale morphology could not be performed due to the

limited resolution of the microscope. Representative images of the

bipromoted catalysts are presented in Figure 3-8, showing the typical

layered structure of MoS2.

Figure 3-8. Representative TEM micrographs of the fresh bipromoted catalysts. Reprinted from Paper II with permission from Elsevier.

3.6.3 Effect of methanol cofeeding (Paper III)

The rate-determining step in HAS is the carbon-carbon bond formation to

transform C1 into C2 species. Cofeeding of methanol, together with the

syngas, has been proposed to increase the production of ethanol and higher


alcohols [41]. It is generally agreed that methanol cofeeding has a positive

effect on HAS over Cu/ZnO catalysts [110-112].

Regarding molybdenum sulfide catalysts, there are only few studies

investigating methanol cofeeding [101, 113, 114] and they normally lack a

detailed product distribution or information concerning catalyst stability.

Moreover, as far as this author knows, there is no published information

about the effect of methanol cofeeding on nickel-modified potassium-

doped molybdenum sulfide catalysts.

Consequently, the aim of Paper III was to comprehensively study this

effect on K-MoS2 and K-Ni-MoS2. Methane, ethanol and 1-propanol STY

are used in this summary as an indicator of the catalyst performance; for

thorough information on conversion or detailed selectivity values, the

reader is directed to the publication.

Although isotope labelling experiments showed that methanol can be

converted to ethanol over molybdenum sulfide catalysts [101], our results

mostly show a negative effect of methanol cofeeding. Ethanol and higher

alcohol yield was clearly decreased upon methanol addition over the

monopromoted catalyst (K-MoS2) (Figure 3-9). When nickel was included

in the catalyst formulation (K-Ni-MoS2), high methanol concentrations

enhanced alcohol yield; however, for low concentrations, the effect

remained negative (Figure 3-10).

A simplified scheme of the reaction pathways for the conversion of syngas

to alcohols and hydrocarbons is shown in Figure 3-11. Chemisorption of

CO and H2 and their reaction to obtain formyl species (pathway 1 in Figure

3-11) is reported to be faster than methanol chemisorption and

dehydrogenation (pathway 2 in Figure 3-11) [115]. The methanol added to

the system would compete for adsorption sites with other reaction

intermediates and, since the route for HAS involving methanol would be

slow, the final effect of methanol addition would be a decrease in ethanol

and higher alcohol yield.

On the other hand, methanol cofeeding enhanced hydrocarbon production

(mainly methane) over both catalysts, this effect being more pronounced

with increasing methanol concentrations (see Figure 3-9 and Figure 3-10).

It can then be speculated that, under the studied reaction conditions (71

bar, 340 °C), methanol conversion to methane is the most favorable

reaction pathway. The isotopic studies of Santiesteban et al [101], carried

out at similar conditions, demonstrated that methane could indeed be

formed from intermediates derived from methanol.


Figure 3-9. Effect of methanol addition (K-MoS2) on methane, ethanol and 1-propanol STY 1. Author's own elaboration.

Figure 3-10. Effect of methanol addition (K-Ni-MoS2) on methane, ethanol and 1-propanol STY. Author's own elaboration.

1 The enrichment factor is the ratio STY (with MeOH cofeeding)

STY (without MeOH cofeeding) . A value higher than 1

indicates that the STY is enhanced upon methanol addition.


Figure 3-11. Simplified scheme of the reaction pathways for the conversion of syngas to alcohols and hydrocarbons. Reprinted from Paper III with permission from Elsevier.

Even though, as highlighted before, the topic of methanol cofeeding has

not received much attention in the literature, two publication have

appeared while this thesis was being written.

Portillo Crespo et al. [116] carried out a systematic study, reporting

complete product distributions for different methanol concentrations in

the feed; the catalyst was an alkali-doped molybdenum sulfide material

based on a patent, so the specific composition is unknown [117]. The

authors drew conclusions that are partly equivalent to the outcomes of

Paper III: they observed that hydrocarbon productivity increases

exponentially upon methanol addition, while alcohol productivity

increases to a lower extent (linearly); therefore, they suggested the need for

an economic analysis to determine the optimum amount of methanol

recycle in an industrial process.

Taborga Claure et al. [118] studied methanol cofeeding over a K-MoS2

catalyst and they reported an increase in methane but also ethanol

production. The authors suggested that the divergences from our study

may be due to a different reaction temperature: they carried out the

reaction at 310 °C (T=340 °C in our study), which would favor the

production of alcohols.


3.6.4 Effect of ethanol cofeeding

Following a strategy similar to that of methanol cofeeding, ethanol

addition was studied as a way of boosting higher alcohol production. Only

preliminary experiments with the bipromoted catalyst (K-Ni-MoS2) could

be performed, but the results were fairly promising.

Introduction of ethanol into the system resulted in an increased production

of hydrocarbons, mainly ethane and ethene. However, higher alcohol yield

was also clearly enhanced, mostly 1-propanol and 1-butanol; the boost in

1-butanol yield was even more pronounced than for 1-propanol. These

results are in agreement with the work of Christensen et al. [103] on a K-

Co-MoS2/C catalyst. As explained in Section 3.4, the authors suggested

that alcohol coupling reactions could take place, in addition to the

commonly accepted CO-addition mechanism. This theory could be

extended to our K-Ni-MoS2 catalyst, although more experiments are

needed to confirm the preliminary observations.

Nevertheless, the recent study of Taborga Claure et al. [118] asserts that

ethanol cofeeding does not produce any significant variation in the product

distribution over a K-MoS2 catalyst, supporting the need for further

investigations on the subject.

37

4 HYDROCONVERSION OF FISCHER-TROPSCH

WAX

Papers IV-VI

4.1 The role of hydroconversion

As mentioned in the Introduction, one of the potential routes for producing

second-generation biofuels from syngas is the FTS. However, FT syncrude

requires further upgrading in order to yield transportation fuels. Although

the FTS can be operated at different temperatures, only low-temperature

Fischer-Tropsch (LTFT) will be considered in this summary. For more

details about the different operating modes, the reader is referred to Paper

I.

LTFT mainly produces saturated linear hydrocarbons, but with a wide

range of carbon numbers. The product distribution is generally described

by the ASF polymerization model, with -CH2- as monomer [119]. According

to this model, the product distribution is a function of the so-called chain

growth probability (α) at the surface of the catalyst, which is independent

of chain length. As seen in Figure 4-1, the maximum straight-run middle-

distillate yield achievable (considered as the C10-C22 fraction) is about 45

wt%; significant amounts of lighter, less valuable, hydrocarbons are also

produced. It is generally accepted that the highest middle-distillate yield

can be achieved by maximizing wax production via LTFT, followed by a

selective hydrocracking to the desired product range [120-124].

Hydrocracking is typically defined as the conversion of heavy feedstocks

into lighter products by adding hydrogen [125]. Even though the term

hydrocracking is often used to describe the upgrading of FT wax, terms

like hydroprocessing or hydroconversion are more appropriate and,

therefore, will be utilized throughout this PhD thesis. The explanation lies

in the fact that hydroconversion of FT wax serves two main purposes: i)

selective cracking of the heavy paraffins; ii) increasing the isomer content,

in order to improve the poor cold flow properties of FT-derived middle

distillates. The great dependence of the cold flow properties (e.g. pour

point) on the isomer content is illustrated in Figure 4-2.

4. HYDROCONVERSION OF FISCHER-TROPSCH WAX 38

Figure 4-1. Product distribution as function of chain growth probability assuming ideal Andersson-Shulz-Flory kinetics. Author's own elaboration.

Figure 4-2. Relationship between pour point and isomer content in FT diesel (C15-C22 fraction). Adapted from [126] with permission from Elsevier.

4.2 Hydroconversion from a historical and industrial perspective

Hydrocracking is one of the oldest hydrocarbon conversion processes.

Strategic reasons explain the great development of the technology at the

beginning of the 20th century, mainly in Germany, with the objective of


achieving liquid fuel security using domestic coal. At the end of World War

II, half of the hydrocarbons consumed in Germany were produced by

means of coal hydrocracking processes. After the war, the interest in coal

conversion declined and so did the interest in hydrocracking [125].

After that, in the late 1950s and early 1960s, the demand for high-octane

gasoline resurrected the interest in hydrocracking and new catalytic

processes appeared, operating at lower temperatures and pressures [123].

In 1959 Chevron developed the first modern hydrocracking process and,

by the 1970s, hydrocracking was already a mature technology [125].

However, specific research on hydroconversion of FT wax is less extensive

and more recent. In the 1970s, Sasol investigated selective hydrocracking

of LTFT wax, achieving middle-distillate yields of about 80 wt%; but the

process was not industrially implemented at that time [123]. Universal Oil

Products also completed a research program in 1988, reporting

comparable yields [127].

It was not until the 1990s that the technology was applied at commercial

scale. In 1993 Shell implemented a hydroconversion process at their FT

plant in Bintulu, Malaysia [128]. The hydrocracker uses a noble metal

catalyst and operates at 300–350 °C and 30–50 bar. The product

composition can be varied from 15 % fuel gas and naphtha, 25 % kerosene,

and 60 % diesel (diesel mode) to 25 % fuel gas and naphtha, 50 % kerosene,

and 25 % diesel (kerosene mode) [129, 130].

The Oryx FT plant in Qatar started operations in 2007, incorporating also

a hydroconversion unit with technology from ChevronTexaco. In this case

the catalyst consists of sulfided base metals on an acidic support, thereby

requiring the addition of a sulfiding agent to maintain catalyst

performance. The operating conditions are more severe, as compared to

the former hydrocracker: > 350 °C and 70 bar. The product composition

after distillation is 3–7 % fuel gas, 20–30 % naphtha and 65–75 % middle

distillates [131].

Albeit slowly and gradually, hydroconversion of FT wax is becoming an

industrial reality.

4.3 Properties of low-temperature Fischer-Tropsch fuels

The diesel obtained through hydroconversion of LTFT syncrude is a high-

performance fuel characterized by a high cetane number, low aromatic

content and ultra-low sulfur content [132]. Emission tests reported by


Calemma et al. [126] showed that this synthetic diesel led to a significant

reduction of CO, hydrocarbon, particulate matter and polyaromatic

hydrocarbon emissions.

However, the low aromatic content results in a lower density, as compared

to conventional diesel: LTFT diesel has a density of about 780 kg/m3, below

the minimum permitted for the European diesel EN590 (820 kg/m3) [132].

Even though the density specification might decrease in the future, it is not

likely that it will encompass the values for LTFT diesel density in the

coming years [133].

There is a trade-off between three parameters: density, cetane number and

product yield (the so-called “density-cetane-yield triangle”); meeting the

requirements for the three parameters becomes fairly difficult. De Klerk

[134, 135] discussed this problem and proposed different refinery designs

in order to synthesize compliant diesel fuel. Blending LTFT distillate with

other fuels can also be a solution to meet the specifications.

4.4 Catalysts

4.4.1 Bifunctional catalysts

FT syncrude is very different from the traditional refinery feedstocks and,

therefore, specific catalyst formulations are needed. FT wax

hydroconversion catalysts are bifunctional, comprising a

dehydrogenation/hydrogenation function (metal sites) and an

isomerization/cracking function (acid sites) [136].

For the dehydrogenation/hydrogenation function, different metals have

been studied: noble metals (mainly platinum and palladium); and

transition metals from group VIA (Mo, W), modified by group VIIIA

metals (Co, Ni) and usually in sulfide form (CoMoS, NiMoS, NiWS) [136].

Noble metals have a higher hydrogenation power and are usually

preferred, despite their higher cost, when hydroprocessing FT wax; they

result in higher middle-distillate yields and isomer contents [137-139]. The

stronger hydrogenation function also improves catalyst stability by

keeping a low concentration of coke precursors [140]. If a sulfided catalyst

is used, a sulfiding agent needs to be added during the process, which

seems unreasonable due to the sulfur-free character of FT syncrude.

For the isomerization/cracking function, a number of acidic supports have

been reported: zeolites, silica-aluminas or aluminas with different


formulations, for instance [136, 138]. In order to minimize overcracking,

solids with mild acidity, such as silica-alumina, are usually preferred [136].

If only the published information is considered, it seems that there has

been limited research on catalysts for FT wax hydroconversion. Most of the

work comes from the group of Calemma (Eni S.p.A.), who investigated

platinum-based catalysts supported on silica-alumina [126, 138, 139, 141-

144]. The work of Leckel (Sasol) should also be mentioned; he studied

unsulfided noble metal and sulfided base metal catalysts, supported on

silica-alumina [123, 145-148].

4.4.2 Catalyst preparation techniques

The deposition of the metal component on the acidic support can be

achieved through different methods, mainly impregnation and

precipitation-based.

Impregnation is the most common method for preparing supported

catalysts. A solution of the metal precursor is contacted with the support

and thereafter aged, dried and calcined. Depending on the amount of

solution used, two techniques can be identified [149, 150]:

Dry or incipient wetness impregnation: the volume of solution

equals the pore volume of the support. It is a simple, economic and

reproducible technique.

Wet impregnation: the support is suspended in a volume of

solution larger than the pore volume of the support. This technique

is normally used for preparation of low-loading catalysts (e.g.

expensive noble metal catalysts) requiring high metal dispersions.

The distribution of the metal component will be affected by the

concentration of adsorption sites on the surface of the support.

Precipitation methods are the second traditional alternative for catalyst

preparation. Depending on whether the support is already synthetized or

not, two techniques can be identified [149, 150]:

Coprecipitation: the metal component and the support are formed

at the same time. A solution of the metal salt and a second solution

containing a salt that will be converted into the support are

precipitated as hydroxides or carbonates by adding a base under

stirring. The catalyst precursor is subsequently transformed to an


oxide. This technique is usually applied for preparation of high-

loading catalysts (>10-15 %).

Deposition-precipitation: analogous to the coprecipitation

method, but in the case of an already existing support. The metal

salt (hydroxide or carbonate) is precipitated onto a support

suspended in a solution, by slowly adding a base under stirring.

The catalytic materials evaluated in the hydroconversion part of this thesis

were prepared by incipient wetness impregnation. Noble metals (platinum,

palladium) were used for the metal function. The acid function consisted

of a commercial amorphous silica-alumina support, Siral 40 (Sasol

Germany GmbH) [151]. Siral 40, containing 40 wt% silica and 60 wt%

alumina, was chosen because it has the highest acidity of the Siral series

[152].

Prior to the impregnation, the support was sieved to the desired fraction

(90-200 µm), dried at 120 °C and finally calcined in air at 500 °C for 4 h

(heating ramp of 2 °C/min). The calcined support was impregnated with

the corresponding aqueous solution of platinum nitrate [Pt(NH3)4(NO3)2]

or palladium nitrate [Pd(NO3)2·2H2O]. The volume of water used was equal

to the total pore volume of the support, previously measured by nitrogen

adsorption. The impregnated catalysts were dried and calcined applying

the same temperature programs as for the support. The final samples were

kept in tightly closed containers and, additionally, dried overnight at 120

°C before the catalytic tests.

4.4.3 Catalyst characterization techniques

In this part of the PhD thesis, the main objective of the characterization

techniques was to study the metallic and acidic properties of the catalysts,

since the metal-acid balance will to a great extent affect catalyst

performance, as will be discussed in Section 4.5.

The metallic properties were evaluated by means of volumetric

chemisorption (using hydrogen and carbon monoxide as probe molecules)

and TEM. The acidic properties were studied by Fourier transform infrared

spectroscopy (FTIR) of adsorbed pyridine, temperature-programmed

desorption (TPD) of ammonia and thermogravimetric analysis (TGA).

In addition, ICP and nitrogen adsorption measurements were carried out

to study, respectively, the elemental composition and textural properties

of the samples.


The different techniques are discussed below. The reader is directed to

specialized literature (e.g. [93]) for a deeper understanding of the

characterization methods; for detailed information on the instruments,

analysis conditions or sample preparation, Papers IV and V are advised.

Chemisorption

Selective gas chemisorption is the most utilized technique for

characterizing metallic catalysts [153]. XRD cannot be applied to materials

that are amorphous or contain small crystallites (< 3-5 nm); TEM can be

used even for highly-dispersed catalysts, but it is a more tedious and

expensive method [93].

In this PhD thesis, a static volumetric technique was used [154]: the volume

of gas that is selectively adsorbed on a metal (monolayer coverage) is

measured, as function of the pressure and at a fixed temperature. That

volume, if the stoichiometry of the chemisorption reaction is known, can

be used for estimating metal dispersion, metal surface area and average

metal particle size [153].

Chemisorption analyses were carried out using a dual isotherm method:

the measurements were repeated after a 30 min evacuation and the

calculations were performed by extrapolating to zero pressure the

difference between both isotherms. Specifically, hydrogen and carbon

monoxide were used as probe molecules.

The stoichiometry of the chemisorption reaction was chosen following

previous studies. It is generally assumed that hydrogen adsorbs

dissociatively on supported noble metals [155]. Consequently, a

stoichiometry of one hydrogen atom per surface metal atom (H:M=1) was

considered in this study, although greater and smaller values have also

been reported [155]. Specifically, palladium can adsorb large amounts of

hydrogen and formβ-palladium hydride at high hydrogen pressures.

Therefore, chemisorption measurements on the palladium samples were

performed at pressures between 1 and 12 mmHg [156, 157].

Different stoichiometry factors were considered for carbon monoxide

chemisorption. In the case of platinum, a CO:Pt stoichiometry equal to 1:1

was used [158]. For palladium, a CO:Pd stoichiometry equal to 1:2 was

found to be the most accepted according to various studies [159, 160].

Transmission electron microscopy

TEM, already described in Section 3.3.3, was used to determine metal

particle size distributions. This technique examines only a very small

fraction of the catalyst; a large number of images on different areas needs


to be analyzed to obtain data with high accuracy [93]. In this PhD thesis, a

minimum of 100 particles were examined for each catalyst.

Fourier transform infrared spectroscopy of adsorbed pyridine

Pyridine is extensively used as probe molecule in infrared spectroscopy

studies of surface acidity of solids such as amorphous silica-alumina [161,

162]. This technique can analyze three different aspects: nature, density

and strength of the acid sites.

It is possible to distinguish between Brønsted (BAS) and Lewis

(LAS) acid sites. The FTIR spectra of adsorbed pyridine display a

band at ca. 1540 cm-1 corresponding to BAS and a band at ca. 1450

cm-1 corresponding to LAS [161].

Acid site densities (concentration) can be calculated from the

intensity of the signals, by using extinction coefficients obtained

through calibration; the constants of Emeis were used in this work

[162].

Acid strength can be evaluated by desorbing the pyridine at

different temperatures and studying the corresponding FTIR

spectra [163].

Temperature-programmed desorption of ammonia

NH3-TPD is one of the most used methods for evaluating solid acidity [93],

because it is simple and inexpensive [164]. The sample is first saturated

with ammonia at low temperature; subsequently, the temperature is

increased and the amount of desorbed ammonia is monitored [165].

However, it has some important limitations: the results strongly depend

on the experimental setup and the common configuration of NH3-TPD

does not allow distinguishing between BAS and LAS [93, 165]. Taking this

into account, the technique was only used to validate the results obtained

using FTIR of adsorbed pyridine.

Thermogravimetric analysis

The nature and concentration of surface hydroxyl groups play an

important role in the acidity of metal oxides [166-168]. TGA

measurements were used to determine OH surface density and verify the

conclusions obtained with FTIR of adsorbed pyridine and NH3-TPD.

OH surface density was calculated on the basis of the weight loss

undergone by the samples during a thermal treatment (from 120 °C to 800


°C), corresponding to the removal of surface silanol groups. The procedure

described by Mueller et al. [169] was used in the calculations.

Inductively coupled plasma

ICP, already described in Section 3.3.3, was used to determine the content

of platinum or palladium in the catalysts.

Nitrogen adsorption

The surface area of the materials was calculated using the BET model (see

Section 3.3.3). Additionally, pore volume and average pore diameter were

estimated applying the BJH (Barrett-Joyner-Halenda) model to the

adsorption data.

4.4.4 Catalyst deactivation

If the literature addressing catalyst development for FT wax

hydroconversion is scarce, specific studies on catalyst deactivation are even

more uncommon. The main causes of deactivation appear to be the

formation of carbonaceous deposits and the sintering of metal particles

[170].

Catalyst coking occurs slowly in this reaction [171], since FT wax contains

mostly n-paraffins, which need to go through numerous transformation

steps in order to form coke [172]. However, if the hydrogen partial pressure

is too low, dehydrocyclization of the paraffins may occur, leading to coke

formation [122]. A low metal/acid ratio also accelerates catalyst

deactivation due to coking [140].

Sintering is favored in the presence of water and increases exponentially

with temperature [173]. It has been reported as the main cause of

deactivation on a palladium-based catalyst used in hydroconversion of FT

wax [174].

4.5 Reaction mechanism

4.5.1 Classical bifunctional mechanism & ideal hydrocracking

The mechanism for hydroconversion of n-paraffins over bifunctional

catalysts has been the subject of numerous studies. In 1964, Coonradt and

Garwood [175] developed a mechanism that is still commonly accepted

today. Alternative mechanisms have been proposed, but a comprehensive


discussion on the topic is outside the scope of this work. For detailed

information, the reader is referred to [170].

A simplified scheme of the classical bifunctional mechanism is depicted in

Figure 4-3. After physisorption, the n-paraffin is dehydrogenated to the

corresponding n-olefin on a metal site. The olefin diffuses to a Brønsted

acid site, where it is protonated and transformed into an alkylcarbenium

ion. This carbocation is isomerized (skeletal rearrangement) and/or

cracked (β-scission). The produced ions are subsequently deprotonated

and olefins diffuse back to a metal site, where they are hydrogenated.

Hydrogenated products can finally desorb from the catalyst surface [136].

According to this mechanism, the relative concentration of metal and acid

sites (i.e. metal-acid balance or metal/acid ratio) has a strong influence on

catalyst performance [176, 177].

Figure 4-3. Classical bifunctional hydroconversion mechanism. Adapted from [136] with permission from OGST - Revue d'IFP Energies nouvelles.

Weitkamp [178] introduced in 1975 the term ideal hydrocracking to

describe a catalytic system in which the dehydrogenation/hydrogenation

reactions are at quasi-equilibrium, i.e. the metal function is strong enough

to feed the acid sites with the greatest possible amount of alkenes

(determined by thermodynamics) and, simultaneously, quickly

hydrogenate the cracked intermediate alkenes. Therefore, skeletal

rearrangement and β-scission reactions become rate- and selectivity-

controlling [136, 179].


In contrast to conventional catalysts for fluid catalytic cracking, ideal

bifunctional hydroconversion catalysts offer a great product flexibility

[179]. They result in high isomerization selectivities and a purely primary

cracking (secondary cracking reactions can be avoided, minimizing the

production of unwanted light compounds).

Besides the metal-acid balance, the operating conditions also determine

the occurrence of ideal hydrocracking. There are four main factors that

promote ideality: i) increasing total pressure; ii) decreasing temperature;

iii) decreasing molar hydrogen-to-hydrocarbon inlet ratio; iv) lower

reactant carbon numbers [177].

4.5.2 Concurring mechanisms

In additional to the bifunctional mechanism, hydroconversion can proceed

by means of three additional mechanisms [179].

Hydrogenolysis

It is a monofunctional mechanism catalyzed by metals. Hydrogenolysis

occurs via adsorbed hydrocarbon radical intermediates which undergo a

non-selective C-C scission. Hydrogenolysis products are mostly

unbranched, unlike bifunctional hydroconversion products [120] .

The reaction occurs to a larger extent over base-metal oxides and sulfides,

as compared to noble metals [180].

Some metals, such as nickel or cobalt, result in the preferential cleavage of

the terminal C-C bond of the adsorbed radical, leading to high methane

selectivities [181]. This variant of hydrogenolysis is known as methanolysis

[120].

Catalytic cracking

It is the generic term used for the monofunctional reactions occurring on

the acid sites [179]. The classical mechanism of catalytic cracking is

bimolecular and involves carbenium ions. In 1984, Haag and Dessau

demonstrated the occurrence of a second mechanism, a monomolecular

pathway proceeding via carbonium ions [182]. The reader is referred to

more specialized publications for further information [183, 184].

Thermal cracking

Thermal cracking reactions take place even in the absence of a catalyst at

temperatures of about 500 °C, significantly higher than for the bifunctional

hydrocracking [179, 185]. Thermal conversion proceeds through a radical


chain mechanism [186]. Extensive research on the topic was carried out in

the 1940s [186-188] and it was recently reconsidered as a potential FT wax

upgrading process [189]. Although it does not require hydrogen and it

produces low amounts of light compounds, thermal cracking proved to be

less efficient than conventional hydrocracking for fuel production [189].

4.6 Reaction and analysis system

4.6.1 The transition from n-hexadecane to paraffinic wax

The wax obtained via LTFT synthesis is mainly composed of n-paraffins,

with small amounts of i-paraffins, olefins, oxygenates and aromatics. In

addition, it is virtually free of sulfur and nitrogen [134, 171].

At the beginning of the work on this PhD thesis, n-hexadecane was using

as a model compound in the hydroconversion tests (Paper IV). The main

reason behind this choice is that n-hexadecane is a liquid at room

temperature and, therefore, easy to handle. In addition, the hydrocarbon

chain is long enough to allow for the study of the different hydrocracking

mechanisms. The analysis of the reaction products also becomes faster and

simpler, it being possible to easily distinguish between monobranched and

multibranched isomers of n-hexadecane. In general, the use of short model

compounds is justified when fundamental hydroconversion studies are to

be carried out, e.g. mechanistic studies.

Nevertheless, when the objective is to obtain information more relevant to

industry, such as optimal operating conditions or product yields, the use of

wax becomes necessary. Therefore, in a later stage of this work (Papers V-

VI), n-hexadecane was replaced by wax, which added some complexity to

the reaction and analysis system, as will be detailed below. Paraffinic wax

(Sigma Aldrich: melting point 53-57 °C, ASTM D87) was chosen for this

purpose. It consisted of hydrocarbons in the range C20- C40, mainly linear

(approx. 85 wt% of n-paraffins and 15 wt% of i-paraffins). The choice of

this feed was motivated by its similarity to FT wax.

4.6.2 Reactor setup

The synthetized catalysts were tested in a high-pressure rig. A fixed-bed

tubular reactor (3/4-in stainless steel pipe, 15.7 mm inner diameter and 55

cm length) was employed, operated in co-current down-flow trickle-bed

mode. The reactor was heated by an electrical furnace and the temperature

was regulated by a cascade control, with one sliding thermocouple in the


catalyst bed and a second thermocouple on the outer wall of the reactor.

This control architecture, together with an external aluminum jacket,

allowed for a temperature profile along the bed within ±0.5 °C of the set

point. The pressure was controlled by a valve placed on the gas outlet line

and connected to a transducer that measured the pressure on the reactor

head.

The gases (H2, He, O2/He) were fed to the reactor system by means of

thermal mass flow controllers (Bronkhorst EL-FLOW): H2 was used for

catalyst reduction and, evidently, as reactant in the hydroconversion

reactions; He was mainly used during leak testing and to create an inert

atmosphere when needed; O2/He was used for catalyst regeneration.

In the case of the hydrocarbon feedstock, different feeding systems were

employed depending on whether hexadecane or wax was hydroconverted.

Hexadecane was fed by means of a HPLC pump (Gilson 307), whereas the

wax was introduced using a syringe pump (Vinci Technologies, BTSP

1000-2); this special pump, which can operate at high temperatures (up to

150 °C), was required for handling the wax. Moreover, to ensure that wax

did not solidify in the system, all the lines before and after the reactor

containing wax were heated up to a temperature above the wax melting

point. A number of temperature controllers and heating tapes were placed

in the system for this purpose.

Unconverted wax (only in the wax experiments) and liquid products (in

both wax and hexadecane experiments) were separated from the gas

stream through traps. Two parallel sets of traps with different capacities

were available and they were chosen depending on the expected product

volume. The sampling was performed manually through needle valves

located at the bottom of the traps. The collected products were analyzed

off-line. The gaseous products were analyzed on-line instead, after passing

through a gas flow meter.

A P&ID of the reactor setup is shown in Figure 4-4.

4.6.3 Catalyst loading and pretreatment

The reactor was loaded with ca. 3 g of catalyst in each test. The catalyst bed

was held in place by quartz wool and enclosed by quartz spheres: a layer of

1 mm quartz spheres on top of the catalyst bed ensured preheating and

mixing of the reactants; a layer of 3 mm quartz spheres at the bottom

supported the catalyst bed.


Figure 4-4. P&ID of the hydroconversion reactor setup. Author's own elaboration.


Catalysts were reduced in situ at atmospheric pressure, in flowing

hydrogen (50 NmL min-1 gcat-1), at 400 °C for 2 h (heating ramp of 5 °C

/min). The system was then pressurized in hydrogen and stabilized at the

desired reaction temperature and pressure.

4.6.4 Operating conditions

Hydroconversion of FT wax can be obtained under mild conditions, as

compared to conventional hydrocracking. As already mentioned in Section

4.2, the commercial Shell Middle Distillate Synthesis process operates at

temperatures and pressures in the range of 300–350 °C and 30-50 bar,

respectively [129, 130]. Analogous operating conditions were used in the

experiments with wax: T=300–330 °C and P=35 bar; the H2/wax ratio was

set to 0.1 wt/wt and the weight hourly space velocity (WHSV) was varied

(1-4 h-1) to obtain data at different conversion levels.

In the case of hexadecane, the experiments were performed at a

temperature of 310 °C and a pressure 30 bar. Since the main goal here was

to study deactivation, forced deactivation and regeneration treatments

were carried out. For specific details on the procedures, the reader is

referred to Paper IV.

4.6.5 Analysis system

Two gas chromatographs (GC) were used to analyze the three phases

obtained as reaction products (when hexadecane was used as feed, only a

liquid and a gas phase were produced).

The wax phase was analyzed off-line with an Agilent 6890 GC. The

samples were dissolved in carbon disulfide (CS2) prior to analysis;

CS2 was chosen because it is a good solvent and does not interfere

with the GC analysis. A Cool On-Column inlet was employed for

injecting the wax phase, followed by an HP-5 column [(5 %-

Phenyl)-methylpolysiloxane] and a FID.

The liquid phase was also analyzed off-line in the aforementioned

Agilent 6890 GC. In this case, no dissolution is needed and the

samples were injected by means of a Split/Splitless inlet. The

liquid phase analysis system included also an HP-5 column and a

FID.


The gas phase was analyzed on-line using a Perkin Elmer Clarus

500 GC, equipped with an Alumina PLOT column and a FID.

The results of the wax, liquid and gas analyses were merged in order to

obtain a full product distribution. The carbon mass balance closures were

typically higher than 95 %.

4.6.6 Product lumping & data treatment

The study of hydroconversion reactions is complex when it comes to

product analysis. Even when using hexadecane as feed, the number of

distinct structural isomers reaches 10 359 for hexadecane, 4 347 for

pentadecane or 1 858 for tetradecane [190]. This makes it essential to

follow a lumping strategy: reaction products are grouped into lumps in

order to draw meaningful conclusions within a reasonable analysis time.

In the case of hexadecane, the analysis system permitted the separation of

cracking products with carbon number 14 or less into each linear alkane

and its isomers, which are lumped together. The production of n-

pentadecane could also be estimated, although with lower accuracy; i-

pentadecanes could not be identified due to the overlap of their GC peaks

with those of i-hexadecanes. From a practical standpoint, reaction

products were basically divided into three groups, following the work of

Girgis and Tsao [176]: mono-branched hexadecanes; di-branched or more

highly branched hexadecanes; and cracking products. Selectivities to the

three main product lumps were usually plotted against n-hexadecane

conversion. Conversion was calculated from the corresponding molar

flows (Fi), according to Equation 4-1; selectivities were defined on a carbon

atom basis, according to Equation 4-2.

𝐶𝑜𝑛𝑣𝑒𝑟𝑠𝑖𝑜𝑛 (%) = ( 𝐹

𝑛−𝐶16

𝑖𝑛 − 𝐹 𝑛−𝐶16

𝑜𝑢𝑡

𝐹

𝑛−𝐶16

𝑖𝑛 ) ∙ 100

Equation 4-1

𝑆𝑒𝑙𝑒𝑐𝑡𝑖𝑣𝑖𝑡𝑦 𝑡𝑜 𝑝𝑟𝑜𝑑𝑢𝑐𝑡 𝑖 (%) = (𝑚𝑜𝑙 𝑜𝑓 𝐶 𝑝𝑟𝑜𝑑𝑢𝑐𝑡 𝑖

𝑚𝑜𝑙 𝑜𝑓 𝐶 𝑐𝑜𝑛𝑣𝑒𝑟𝑡𝑒𝑑 𝑛−𝐶16

) ∙ 100 Equation 4-2

When wax was used as feed, the complexity increased markedly, due to the

increased amount of reactants and thus potential products. The analysis

system also made possible the identification of each linear alkane and the

corresponding lump of isomers. However, to facilitate the interpretation of


the results, hydrocracking products were further lumped into the following

cuts, as in the work of Pellegrini et al. [191]: fuel gas (C1-C4), naphtha (C5-

C9), kerosene (C10-C14), gasoil (C15-C22) and unconverted wax (C22+). The

middle-distillate (MD) fraction includes both kerosene and gasoil. In this

case, conversion and selectivity were calculated according to Equation 4-3

and Equation 4-4.

𝐶𝑜𝑛𝑣𝑒𝑟𝑠𝑖𝑜𝑛 (%) = (𝑤𝑡% 𝐶22+ 𝐼𝑁 − 𝑤𝑡% 𝐶22+ 𝑂𝑈𝑇

𝑤𝑡% 𝐶22+ 𝐼𝑁) ∙ 100 Equation 4-3

𝑆𝑒𝑙𝑒𝑐𝑡𝑖𝑣𝑖𝑡𝑦 𝑡𝑜 𝑝𝑟𝑜𝑑𝑢𝑐𝑡 𝑖 (%) = (𝑤𝑡% product i 𝑂𝑈𝑇 − 𝑤𝑡% product i 𝐼𝑁

𝑤𝑡% 𝐶22+ 𝐼𝑁 − 𝑤𝑡% 𝐶22+ 𝑂𝑈𝑇) ∙ 100 Equation 4-4

4.7 Main results from the experimental work

4.7.1 Comparison between platinum and palladium as the metal

function: catalyst performance (Paper V)

Previous work from our research group compared platinum and palladium

as the dehydrogenation/hydrogenation component of bifunctional

hydroconversion catalysts [192]. N-hexadecane was used as model

compound in the tests. The main conclusions of that study were that the

platinum-based catalyst showed higher activity and exhibited a

monofunctional hydrogenolysis mechanism, in addition to the commonly

accepted bifunctional route.

The idea behind Paper V was to extend the aforementioned work, using

wax in the catalytic tests. Two catalysts with the same molar metal loading

(31 µmol / gcat) were examined. They are named Pt/S40 and Pd/S40

throughout Sections 4.7.1 and 4.7.2, depending on the noble metal used.

The experiments showed clear differences between the platinum and

palladium samples. The use of platinum resulted in higher conversions

(Figure 4-5) and also higher selectivities to the desired middle distillate

fraction (Figure 4-6). The palladium-based catalyst exhibited more

secondary cracking, producing greater amounts of lighter compounds

(naphtha and fuel gas).


Figure 4-5. Catalyst activity as function of temperature and space velocity for Pt/S40 and Pd/S40. Conditions: P=35 bar; H2/wax=0.1 wt/wt. Reprinted from Paper V with permission from Elsevier.

The production of methane and ethane was fairly low over both Pt/S40 and

Pd/S40 (< 1 wt%), as expected for noble metal catalysts. However, there is

a clear difference between both metals. The platinum catalyst exhibited

much higher selectivities to the aforementioned compounds (about two

orders of magnitude higher). This would indicate the occurrence of

hydrogenolysis to a much greater extent on the platinum particles, as

compared to palladium. This observation validates previous results from

our research group [192].

As described in Section 4.1, hydroconversion of FT wax not only involves

hydrocracking of the heavy molecules, but also isomerization reactions.

Pt/S40 and Pd/S40 showed also a different isomerization performance, as

shown in Figure 4-7. The isomer content of the middle distillate fraction

was significantly higher when using palladium, even though at high

conversion both metals behave similarly.

To sum up, platinum shows higher hydrocracking activity and a greater

contribution of the hydrogenolysis mechanism. Palladium exhibits higher

isomerization activity, but also a higher overcracking tendency, which

limits the production of middle distillates.


Figure 4-6. Catalyst selectivity as function of conversion for: Pt/S40 (top) and Pd/S40 (bottom). Conditions: P=35 bar; T=300-330 °C; H2/wax=0.1 wt/wt; WHSV=1-4 h-1. Reprinted from Paper V with permission from Elsevier.


Figure 4-7. Isomer content in the kerosene and gasoil fractions for Pt/S40 and Pd/S40, as function of conversion. Conditions: P=35 bar; T=315/330 °C; H2/wax=0.1 wt/wt; WHSV=1-4 h-1. Reprinted from Paper V with permission from Elsevier.

4.7.2 Catalyst characterization (Paper V)

In order to understand the reasons behind the different performances of

Pt/40 and Pd/S40, extensive catalyst characterization was carried out.

ICP analyses confirmed the successful incorporation of the noble metals

onto the Siral 40 support. N2 adsorption measurements showed a slight

decrease in BET surface area after impregnation and verified the

mesoporous character of the materials.

However, the most interesting results concern the characterization of

metal and acid sites. Regarding the metallic properties, both

chemisorption and TEM analyses evidenced that the fresh platinum-based

catalyst had a higher metal dispersion than the corresponding palladium

sample, i.e. higher metal surface area and smaller crystallites. The

estimated metal particle sizes are summarized in Table 4-1.

Estimation of the mean metal particle size by TEM imaging resulted in

higher values for both metals, as compared to H2 or CO chemisorption.

This was attributed to the limited resolution of the microscope, which

would not detect the smallest particles, especially those with sizes

significantly lower than 1 nm, resulting in an overestimation of the particle

size.


Table 4-1. Average metal particle size of fresh Pt/S40 and Pd/S40 (after calcination at 500 °C for 4 h and activation in H2 at 400 °C for 2 h). Author's own elaboration.

Catalyst Average metal particle size (nm)

H2 chemisorption CO chemisorption TEM

Pt/S40 1.1 1.5 2.5

Pd/S40 5.3 4.5 6.1

However, despite being quite different before reaction, particle size

distributions for the spent samples became fairly similar, as shown in

Figure 4-8. It is evident that platinum particles underwent sintering to a

greater extent than palladium particles.

Figure 4-8. TEM particle size distributions for fresh and spent catalysts: Pt/S40 (left) and Pd/S40 (right). Reprinted from Paper V with permission from Elsevier.

As concerns the acidic properties, FTIR of adsorbed pyridine was used for

determining the nature, density and strength of the sites. The results are

summarized in Table 4-2. The acidity of Siral 40 changed after metal

impregnation: the strongest BAS and LAS in the bare support were

removed, but there was an increase in the number of sites of lower

strength.


Table 4-2. Acid site distributions for the support, fresh catalysts (after calcination at 500 °C for 4 h) and spent catalysts, as determined by FTIR of adsorbed pyridine. Reprinted from Paper V with permission from Elsevier.

Support /

catalyst

Brønsted acid sites

(µmol/g)

Lewis acid sites

(µmol/g)

150 °C 250 °C 350 °C 150 °C 250 °C 350 °C

Siral 40 44 31 4 114 54 12

Pt/S40 fresh 47 30 0 161 55 0

Pd/S40 fresh 51 35 0 170 56 0

Pt/S40 spent 47 29 0 158 54 0

Pd/S40 spent 40 29 0 147 47 0

When the fresh materials were compared, the palladium-based catalyst

was somewhat more acidic than the platinum sample, but the differences

were not significant. After reaction, these differences remained small,

although both materials experienced a slight decrease in acidity (more

pronounced for the palladium sample).

NH3-TPD analyses cannot distinguish between BAS and LAS. However, the

desorption profiles obtained for Pt/S40 and Pd/S40 were virtually

equivalent, confirming that the samples are of comparable acidity (see

Figure 4 in Paper V). TGA analyses gave OH surface density values of 4.5

OH/nm2 for Siral 40, 5.4 OH/nm2 for Pt/S40 and 6.4 OH/nm2 for Pd/S40.

The highest OH surface density observed for Pd/S40 is in agreement with

the low dispersion of palladium on the silica-alumina surface. It is likely

that large Pd2+ particles are stabilized by two OH groups through a bridging

coordination, while small Pt2+ particles are linearly coordinated with OH.

According to the view that metal particles act as Lewis acids and polarize

the OH groups [193], a higher amount of OH groups polarized by Pd2+

particles would result in a higher amount of OH removed during TGA .

To summarize, the metal/acid ratio was very different for the fresh

catalysts since platinum was better dispersed than palladium. However,

platinum particles underwent sintering to a greater extent than palladium

particles during the stabilization period (first two days of operation, see

Section 4.7.3) and the catalysts had similar crystallite sizes at steady state;

the differences in acidity were also fairly low. Therefore, the spent catalysts

had comparable metal/acid ratios.


Nevertheless, the catalytic tests showed clear differences between the

platinum- and palladium-based materials (see Section 4.7.1). Assuming

that the reaction follows the commonly accepted bifunctional mechanism,

the different performances of the catalysts were attributed to the different

hydrogenation capability of the metals. The higher hydrogenation power

of platinum would easily hydrogenate unsaturated reaction intermediates;

palladium, instead, would have a lower hydrogenation power and,

therefore, intermediates are more likely to react further on the acid sites,

resulting in further branching and/or secondary cracking reactions.

4.7.3 Catalyst deactivation (Papers IV and V)

In Paper IV, the deactivation behavior of a 0.33-wt% platinum catalyst

supported on silica-alumina was studied, using n-hexadecane as a model

compound. Many of the findings of that work were verified in Paper V,

which studied analogous catalysts for hydroconversion of real wax; this

confirmed the suitability of using short model compounds to obtain

information that can then be applied to real wax.

Going back to Paper IV, catalyst deactivation was studied at reference

conditions (initial deactivation) and also exposing the catalyst to a low

H2/n-C16 inlet ratio (forced deactivation); thereafter, catalyst regeneration

was carried out by means of an oxidation treatment.

Initial deactivation experiments were performed at a temperature of 310

°C, a pressure of 30 bar and a H2/n-C16 molar ratio of 10. The deactivation

was fairly fast at the beginning of the reaction; both activity and selectivity

to cracking products greatly decreased, leveling out after 30-40 h of

operation (see Figure 4-9). Catalyst characterization (Table 4-3) showed

that the acidic function remained unaltered, but there was an important

loss of metal surface area. This particularly reduced the formation of

cracking products through hydrogenolysis, since the occurrence of this

mechanism depends only on the metal sites.

In order to accelerate deactivation by coking, the catalyst was subjected to

a period with a H2/n-C16 molar ratio of 1. After this treatment, activity and

specially cracking selectivity were decreased. The hydrogenolytic activity

was almost completely removed, but the bifunctional mechanism was also

affected after this forced deactivation. The characterization results (Table

4-3) showed a decrease in BET surface area and acidity, probably due to

the formation of carbonaceous deposits blocking the pores. The metal

surface area was further decreased, a fact that is in agreement with the

aforementioned negligible contribution of the hydrogenolytic pathway.


Figure 4-9. Catalyst performance over time on stream for a 0.33-wt% Pt / Siral 40 catalyst. Conditions: P=30 bar; T=310 °C; H2/n-C16=10 mol/mol; WHSV=3 h-1. Author's own elaboration.

Table 4-3. Physicochemical properties of a 0.33-wt% Pt / Siral 40 catalyst at different extents of deactivation. Adapted from Paper IV with permission from Springer.

Time on stream (h)

BET surface area (m2/g)

Metal surface area

(m2/g)

BAS concentration (*)

(µmol/g) fresh 387 0.172 52

6 384 0.097 53 24 384 0.037 n/a

53 (1) 346 0.024 43 53 (2) 381 0.135 n/a 24 (3) 386 0.031 54

(1) After forced deactivation (H2/n-C16=1 mol/mol, 16 h) (2) After regeneration ex situ in 5 % O2 / 95 % He (500 °C, 1 h) (3) After regeneration in situ in 5 % O2 / 95 % He (500 °C, 1 h) (*) Determined by FTIR of adsorbed pyridine (150 °C)

Catalyst regeneration was carried out by flowing 5 % O2 / 95 % He at 500

°C during 1 h; subsequently, the catalyst was again reduced in H2 at 400

°C. The activity of the acidic function could be completely recovered, while

the activity of the metal function was only partially recovered (Table 4-3).


Although studying catalyst deactivation was not the aim of Paper V, some

insights could be gained by looking at the results. The catalysts experienced

a stabilization period of about 2 days; after that period, catalyst

performance was fairly stable, a fact that was expected taking into account

the high hydrogen partial pressure used in the experiments [145]. The

characterization results reported in Paper V confirmed the conclusions

drawn in Paper IV. After several days of operation under industrial

operating conditions (see Section 4.6.4) the acidic function remained

virtually unaltered. However, sintering of the metal particles was evident,

resulting in a significant loss of metal surface area.

4.8 Kinetic modelling (Paper VI)

4.8.1 Overview of modelling strategies

Models are valuable tools for performing preliminary process design.

Experiments are certainly highly useful, but they are more expensive and

time consuming. For instance, a reliable hydroconversion model could be

used to predict the effect that the operating conditions have on conversion

and product selectivity, t0 avoid running costly experiments.

Three main approaches for modelling hydroconversion of FT products (or

hydrocracking in general) are identified [194].

Single-event kinetic modelling has been proposed by some authors [195-

201]. However, FT wax is a complex mixture and, therefore, these models

need to take into account an enormous amount of elementary reaction

steps; the knowledge of a large number of kinetic constants is required

when the schemes are applied without any simplification, demanding

plenty of experimental data. The main advantage in this case is that single-

event rate parameters are independent of the feed composition.

To simplify the aforementioned models, two lumping strategies have been

proposed: continuous and discrete lumping. In the continuous approach,

the models are based on continuous mixtures, where individual chemical

species are not distinguished from each other. The properties of each

individual component (e.g. reactivity) are described through suitable

component indexes (e.g. boiling point or molecular weight) [202-206].

In the discrete approach, the models are based on the interaction of

pseudocomponents (lumps). Discrete lumped models were initially

applied to single n-paraffins [207, 208] and they have been recently

extended to FT wax.


The first model describing FT wax hydroconversion was proposed by

Pellegrini et al. in 2004 [191]. They followed the Langmuir-Hinshelwood-

Hougen-Watson (LHHW) formalism, based on the work published by

Froment in 1987 [209]. The aforementioned model can be simulated with

short computational times, but it presents a large number of drawbacks,

mainly: it considers only 9 lumps; it assumes that all reacting species are

present in the vapor phase; it contemplates that cracking reactions produce

only isomerized fragments, which are generated by the breakage of the C–

C bond placed in the middle of the hydrocarbon chain.

This first model ([191]) has been greatly improved in the subsequent years,

mainly by Eni S.p.A. and Politecnico di Milano. The introduction of an “all

components” kinetics was accomplished by Pellegrini et al. [210] (2007).

Subsequently, the same group proposed a detailed strategy for calculating

the vapor-liquid equilibrium (VLE) [211] (2008). Afterwards, Gamba et al.

enhanced the predictive power of the model by considering a cracking

probability (CP) for the C–C bonds [212] (2009). Finally, Gambaro et al.

discussed the possibility of using the complete form of the reaction rate

expressions, according to the LHHW approach, and they obtained an

improved model (2011) [213].

According to this author’s knowledge, the latter model ([213]) is the most

advanced discrete lumped description, available in literature, applied to FT

wax hydroconversion. It has therefore been used as starting point for this

work; a modified version is proposed here. Earlier modelling studies [191,

210-213] are based on experimental data obtained with platinum catalysts

supported on amorphous silica-alumina. However, two different noble

metal catalysts, based on platinum or palladium (Pt/S40 and Pd/S40),

were considered in this work (the molar metal loading was the same in both

cases: 31 µmol / gcat).

4.8.2 Description of the proposed model

A brief description of the proposed model is made in this summary,

devoting special attention to the differences from the model of Gambaro et

al. [213] . A more thorough description, including the complete system of

differential equations, is given in Paper VI.

Simplified reaction scheme

It is commonly accepted that hydroconversion of n-paraffins takes place according to the bifunctional mechanism explained in Section 4.5, involving adsorption, dehydrogenation, protonation, isomerization and/or cracking, deprotonation, hydrogenation and desorption steps [136].


However, a kinetic model applied to a complex feedstock such as FT wax needs to be based on a more simplified reaction mechanism. Therefore, a scheme where linear paraffins are firstly isomerized and then cracked was adopted to develop the rate expressions (see Equation 4-5). All the i-paraffins with the same number of carbon atoms were indistinctly grouped in a single lump.

𝑛 − 𝑝𝑎𝑟𝑎𝑓𝑓𝑖𝑛 ↔ 𝑖 − 𝑝𝑎𝑟𝑎𝑓𝑓𝑖𝑛 → 𝑐𝑟𝑎𝑐𝑘𝑒𝑑 𝑝𝑟𝑜𝑑𝑢𝑐𝑡𝑠 Equation 4-5

The formation of methane and ethane in this reaction cannot be explained by the classical bifunctional hydroconversion mechanism. Following the work of Gambaro et al. [213], the proposed model considered that methane and ethane are produced by cracking of i-hexane; all bond breakages were considered equally probable.

LHHW approach

Isomerization and cracking rates were calculated following the LHHW

formalism. The adsorption, dehydrogenation, hydrogenation and

desorption steps were considered to be in quasi-equilibrium; each step

involving the presence of a carbenium ion may be the rate-determining

step. The Langmuir isotherm was used to describe paraffin physisorption.

Vapor-liquid equilibrium

The VLE was considered by writing the kinetic equations in terms of

fugacity. Fugacities were calculated using the Soave-Redlich-Kwong

equation of state.

C-C cracking probability

A breakage probability function for the C-C bonds was considered. The

following assumptions were made for paraffins with eight or more carbon

atoms (nC): all bonds placed between C4 and CnC-3 are cracked with the same

probability; the probability of cracking the bonds located in the 3rd and

(nC-3)th positions is half of that value; the four terminal bonds cannot be

cracked.

Gamma

Experimental observations show that cracking products can be either

linear or branched [212, 214]. In order to take this into account, the so-

called parameter ɣ, i.e. fraction of i-paraffins obtained after a cracking

reaction, was introduced in the kinetic model. The model presented in 2011

by Gambaro et al. [213] considered ɣ a constant, not being dependent on

the hydrocarbon chain length. However, there is evidence that this


parameter increases with increasing chain length of the cracked paraffin

[212, 214, 215]. Therefore, in the proposed model, ɣ was correlated with the

carbon number by means of a power law, as shown in the next section.

Correlation of the model parameters with the carbon number

In order to minimize the total number of variables, different functions were

employed to correlate the kinetic and thermodynamic constants with the

carbon number. Table 4-4 summarizes the various functions used in the

proposed model. They were basically chosen following the work of

Gambaro et al. [213]. The main differences with that work concerned the

isomerization equilibrium constant (Kequ) and the parameter ɣ. Regarding

Kequ, the fitting was done with an exponential law, as suggested by

Pellegrini et al. [216]. Concerning the parameter ɣ, different functions were

considered (power, logarithmic and exponential) and a power law was

selected since it gave the lowest sum of squared error (SSE) values.

Table 4-4. Functions used to correlate the kinetic and thermodynamic constants with the carbon number. Author's own elaboration.

Constant Type of function

Langmuir adsorption constants (KL) Hyperbolic tangent

Dehydrogenation-protonation constants (KDHP) Power law

Isomerization equilibrium constants (Kequ) Exponential

Frequency factors (k0) Power law

Activation energies (Ea) Power law

Gamma (ɣ) Power law

Numerical computation of the model

In order to solve the resulting system of differential equations, a home-

made code applying a Runge-Kutta 4th-order method was implemented in

Matlab. The optimization of the model parameters was carried out by

minimizing the SSE values for the mass fractions.

4.8.3 Main results from the modelling work

In order to achieve a high goodness of fit of the kinetic model, it is essential to choose appropriate functions to correlate the model parameters with the carbon number, i.e. the physical meaning of the kinetic and


thermodynamic constants should be respected [213]. For a detailed assessment of the optimal model parameters, the reader is referred to Paper VI. Only some insights will be given in this summary.

Activation energies are probably the most reported constants in previous studies and, therefore, it is easier to assess their suitability. As an example of optimized model constants, the values computed for Pt/S40 and Pd/S40 are presented in Figure 4-10. They are in the range of 120-220 kJ/mol, in line with previous lumped models [138, 207, 208, 210, 213, 217], single-event microkinetic models [195, 218] and also experimental data [219, 220].

Figure 4-10. Isomerization (left) and cracking (right) activation energies computed with the kinetic model for Pt/S40 and Pd/S40, as function of the carbon number. Author's own elaboration.

One of the main novelties of the proposed kinetic model, as compared to the model of Gambaro et al. from 2011 [213], was the dependence of the parameter ɣ on the chain length. The values of ɣ obtained using the model are presented in Figure 4-11. As expected from previous studies [212, 214, 215], ɣ increased with carbon number. The values were always higher than 0.5, which indicates that i-paraffins are predominant among the cracking products. Moreover, the results are in line with the constant figures (0.86, 0.90) reported by Gambaro et al. [213].

When comparing both catalysts, palladium presented higher values of ɣ for

the whole range of carbon numbers (see Figure 4-11). The model described

satisfactorily the experimental results, which showed a higher degree of

isomerization of the products when choosing palladium instead of

platinum.


Figure 4-11. ɣ values computed using the kinetic model for Pt/S40 and Pd/S40, as function of the carbon number. Author's own elaboration.

The developed kinetic model had a high predictive power, as evidenced by the values of the coefficient of determination (R2=0.975 for Pt/S40 and R2=0.991 for Pd/S40). The parity plots for the mass fractions (Figure 4-12) also indicated a good agreement between experimental and modelling data.

Figure 4-12. Parity plot for the mass fractions, for Pt/S40 (left) and Pd/S40 (right). Author's own elaboration.


Experimental carbon number distributions were well fitted by the model. As an example, the predicted and experimental distributions of i- and n-paraffins for Pd/S40 are shown in Figure 4-13; the values correspond to a conversion of 35 %, but the fitting was also good for the whole range of conversions studied within this work.

Figure 4-13. Predicted and experimental carbon number distributions of i-paraffins (left) and n-paraffins (right) for Pd/S40, at C22+ conversion=35.0 %. Conditions: P=35 bar; T=330 °C; H2/wax=0.1 wt/wt; WHSV=2 h-1. Author's own elaboration.

Once the parameters were optimized, the model could be used to predict the effect of the operating conditions on conversion and product selectivity. Since the hydroconversion experiments in this PhD thesis were carried out at a fixed pressure and H2/wax ratio (see Paper V), only the effect of WHSV and temperature could be studied.

The effect of the aforementioned variables on conversion is well foreseen by the model for both Pt/S40 and Pd/S40. However, in the case of selectivity, the trends computed for the palladium-based catalyst deviated from the experimental values, especially at low conversion (i.e. high WHSV and/or low temperature): the model underestimated naphtha selectivity and overestimated middle-distillate selectivity. The results for Pt/S40 and Pd/S40 are presented in Figure 4-14.


Figure 4-14. Effect of WHSV and temperature on selectivity, for Pt/S40 (top) and Pd/S40 (bottom). Author's own elaboration.

69

5 CONCLUDING REMARKS AND FUTURE

WORK

5.1 Concluding remarks

This PhD thesis has addressed the production of second-generation

biofuels, which are a promising solution for diversifying fuel supply. The

thermochemical conversion of biomass resources generates synthesis gas,

which can be converted into liquid and gaseous fuels through

heterogeneous catalytic processes. Although some economic and technical

challenges still need to be overcome, a great development has taken place

in recent years.

Two processes were mainly studied within this work, namely: higher

alcohol synthesis over molybdenum sulfide catalysts and hydroconversion

of Fischer-Tropsch wax over noble metal catalysts supported on silica-

alumina.

Higher alcohol synthesis yields ethanol and longer alcohols, which can be

used as fuel or fuel additives. A novel K-Ni-MoS2 catalyst prepared by

coprecipitation of two microemulsion systems was proposed,

outperforming an analogous conventional material. The higher activity

and selectivity of the novel catalyst was attributed to a higher concentration

of promoters on the catalyst surface, together with a lower degree of

crystallinity.

Cofeeding of methanol, together with the synthesis gas, was studied as a

way of enhancing higher alcohol production. Various concentrations of

methanol were evaluated and produced basically a negative result when

the reaction was performed at industrial conditions. The main outcome

was an increase in methane yield, while the effect on higher alcohols was

negative or not significant. On the other hand, preliminary experiments

with ethanol cofeeding showed a clear increase in the production of longer

alcohols.

Hydroconversion of wax is an essential upgrading step for producing

middle-distillate fuels via Fischer-Tropsch. The deactivation of a platinum

/ silica-alumina catalyst was studied using n-hexadecane as model

compound. Sintering and coking were found to be the main causes of

deactivation. Moreover, it was clear that catalyst performance at different

extents of deactivation could be related to the metal/acid ratio. The

findings obtained using n-hexadecane as feedstock were verified with the

5. CONCLUDING REMARKS AND FUTURE WORK 70

experiments using paraffinic wax, confirming the possibility of utilizing

short model compounds to draw conclusions that can then be correctly

applied to real wax.

Platinum and palladium catalysts with the same molar metal loading were

evaluated in hydroconversion of paraffinic wax, showing clearly different

performances. Platinum exhibited higher hydrocracking activity and a

greater contribution of the hydrogenolysis mechanism; palladium showed

higher isomerization activity, but also a higher overcracking tendency.

Higher yields of middle distillates were thus obtained over the platinum

material, a fact that was attributed to the higher hydrogenation power of

the metal.

Finally, a lumped kinetic model to describe the hydroconversion of long-chain paraffins was proposed. The model was based on previous works, but included some modifications: the fraction of isomers obtained after a cracking reaction was considered to be dependent on the carbon number, unlike in previous studies; the numerical method used to find the optimal values of the model parameters was also modified. The model was applied to both platinum and palladium catalysts, showing a high predictive power. However, further improvement is still needed, especially concerning palladium.

5.2 Future work

There are many possibilities that could be explored as a continuation of

this PhD thesis and some of them are outlined below.

Regarding higher alcohol synthesis, the promising results obtained with

the novel K-Ni-MoS2 material suggest the need for a deeper study on

catalysts prepared using microemulsions. Catalyst optimization by varying

the promoters and their concentration deserves to be considered, since one

of the factors limiting higher alcohol synthesis commercialization is the

ineffective catalytic systems developed so far.

The good preliminary results of ethanol cofeeding also call for continued

investigations. Only a few studies on the topic have been published and

they are contradictory.

As concerns hydroconversion of Fischer-Tropsch wax, it would be

interesting to study catalysts with various metal loadings to verify the

drawn conclusions. Moreover, since both platinum and palladium

materials display interesting features (e.g. higher activity for platinum and

less hydrogenolytic products for palladium), evaluation of bimetallic PtPd

5. CONCLUDING REMARKS AND FUTURE WORK 71

catalysts is of interest. Analogous materials have been successfully used in

other reactions, but studies on wax hydroconversion are very scarce.

Finally, the proposed lumped hydroconversion model could be further

improved. First of all, the model should be tested with a larger dataset. It

would also be important to find an explanation (and then a solution) for

the deviations of the model when used to predict the effect of the operating

conditions on the selectivity of the palladium catalyst. Another interesting

point concerns the production of light hydrocarbons (mainly methane and

ethane); the occurrence of hydrogenolysis should be considered in order to

improve the predictive power of the model.

73

Acknowledgements

This PhD thesis would not have been possible without the help of

numerous people.

Thank you to my supervisors, Associate Professor Magali Boutonnet and

Professor Emeritus Sven Järås, for giving me the opportunity to do a PhD

at KTH and for their continuous confidence in my work.

Thank you, a great one, to Francesco Regali. Thank you for being with me

since the first day I arrived to KTH and being still with me, in a completely

different place, in 2016. You have taught me many things, and not only

about catalysis.

Thank you to my Master Thesis student Mario Enrico L’Abbate. Because

you are one of the main contributors to the hydroconversion part of this

work.

Thank you to my office mate Javier Barrientos, Barri. Thank you for your

friendship and generous help in fixing the problems that unceasingly

appeared in our daily routine, mainly in the lab! Tighten nuts to work at 90

bar is easier if you are not alone.

Thank you to Vicente Montes and Leonarda Francesca Liotta, Nadia, for

your valuable help with the papers and the fruitful discussions about

catalysis.

Thank you to Iraj Bavarian for your altruistic help with my hydrocracking

GC. And thank you to Christina Hörnell for improving the English of this

work and finding even the most hidden mistakes!

Thank you to my BFF Francesco Montecchio. Because the thing I miss the

most when I arrive to work now is that you are not waiting for me to have

a coffee. Mi manchi molto. Se me ne vado, vieni con me?

Thank you to Professor Lars J. Pettersson, Lasse, for your always positive

attitude and great sense of humor, for your help and of course for

introducing me to the POKE life.

Thank you to the Bolivian office – Luis, Jorge and Fátima - for making me

feel (almost) like if I were in the warm Spain. Special thanks to Fátima for

being my partner in many moments and adventures throughout these

years.

Thank you to all the (changing) members of the ACI team, especially to

Master Chef Roberto Lanza.

74 Acknowledgements

Thank you of course to all the present and former colleagues at the division

of Chemical Technology. Thank you for making KT a wonderful place to

work. I will miss it a lot! Thank you to Zari and Henrik for always having a

smile for me! Special thanks also to Alagar, Jerry, Jonas, Klas, Lina,

Matteo, Moa, Pouya, Robert, Sandra, Sara Lögdberg and Yohannes!

Thank you to all the people that passed by or were around KT during these

years, in particular Baldesca, Helen, Josefine, Raquel and Sabine. Thank

you for the nice talks and the endless laughter! And of course thank you to

Silvia, because you will always be a very important part of KT and, the most

important, of my life.

Thank you to my Stockholm family, for being with me not only in the sunny

summers, but also in the dark winters. Thank you especially to Athina,

Carlos, Diego, Eirini, Foto and Gonçalo. For the parties, the dinners, the

coffees, the advices, the walks and your love!

Thank you also to all those who have been by my side, even in the distance.

To Iván, for being permanently with me. To my USC friends, especially

Alejandra and Clara, for hosting me in Barcelona every time I was flying

home, and for continuously trying to persuade me to migrate south again.

To my Erasmus friends, especially my sister Paula, for being available

whenever. And to my friends from Negreira, for welcoming me with open

arms every summer and every Christmas.

Y, por supuesto, gracias papá y mamá. Por tantas cosas.

75

Nomenclature

ASF Andersson-Shulz-Flory

BAS Brønsted acid sites

BET Brunauer-Emmett-Teller

BFB Bubbling fluidized bed

BJH Barrett-Joyner-Halenda

CFB Circulating fluidized bed

CP Cracking probability

DME Dimethyl ether

EF Entrained flow

EU European Union

FID Flame ionization detector

FT Fischer-Tropsch

FTIR Fourier transform infrared spectroscopy

FTS Fischer-Tropsch synthesis

GC Gas chromatograph

GHG Greenhouse gas

GHSV Gas hourly space velocity

HAS Higher alcohol synthesis

ICP Inductively coupled plasma

IEA International Energy Agency

LAS Lewis acid sites

LCA Life cycle assessment (or analysis)

LHHW Langmuir-Hinshelwood-Hougen-Watson

LTFT Low-temperature Fischer-Tropsch

ME Microemulsion

Nomenclature 76

MD Middle distillate

MON Motor octane number

P&ID Piping and instrumentation diagram

RON Research octane number

RVP Reid vapor pressure

SEM Scanning electron microscopy

SNG Synthetic natural gas

SSE Sum of squared error

STY Space-time yield

TCD Thermal conductivity detector

TEM Transmission electron microscopy

TGA Thermogravimetric analysis

TPD Temperature-programmed desorption

VLE Vapor-liquid equilibrium

WHSV Weight hourly space velocity

XPS X-ray photoelectron spectroscopy

XRD X-ray diffraction

77

References

[1] Key World Energy Statistics, International Energy Agency, 2014.

[2] 50by50: Global Fuel Economy Initiative, FIA Foundation / International Energy Agency / International Transport Forum /United Nations Environment Programme, 2009.

[3] http://www.nasdaq.com/markets/, accessed 2015-10-15.

[4] B. Dawson, M. Spannagle, The Complete Guide to Climate Change, Routledge, 2008.

[5] CO2 Emissions from Fuel Combustion, International Energy Agency, 2014.

[6] IPCC, 2013: Summary for Policymakers. In: Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change [T.F. Stocker, D. Qin, G.-K. Plattner, M. Tignor, S.K. Allen, J. Boschung, A. Nauels, Y. Xia, V. Bex, P.M. Midgley (eds.)], Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA.

[7] Renewable energy progress report, European Commission, 2015.

[8] Directive 2009/28/EC of the European Parliament and of the Council of 23 April 2009 on the promotion of the use of energy from renewable sources and amending and subsequently repealing Directives 2001/77/EC and 2003/30/EC

[9] https://ec.europa.eu/energy/node/73, accessed 2015-10-13.

[10] R.E.H. Sims, W. Mabee, J.N. Saddler, M. Taylor, An overview of second generation biofuel technologies, Bioresource Technology, 101 (2010) 1570-1580.

[11] A. Mohr, S. Raman, Lessons from first generation biofuels and implications for the sustainability appraisal of second generation biofuels, Energy Policy, 63 (2013) 114-122.

[12] Advanced Biofuels – GHG Emissions and Energy Balances (a report to IEA Bioenergy Task 39), Don O’Connor / (S&T)2 Consultants Inc, 2013.

[13] M. Mohammadi, G.D. Najafpour, H. Younesi, P. Lahijani, M.H. Uzir, A.R. Mohamed, Bioconversion of synthesis gas to second generation biofuels: A review, Renewable and Sustainable Energy Reviews, 15 (2011) 4255-4273.

[14] N. Doassans-Carrère, J.-H. Ferrasse, O. Boutin, G. Mauviel, J. Lédé, Comparative Study of Biomass Fast Pyrolysis and Direct Liquefaction for Bio-Oils Production: Products Yield and Characterizations, Energy & Fuels, 28 (2014) 5103-5111.

[15] J. Akhtar, N. Saidina Amin, A review on operating parameters for optimum liquid oil yield in biomass pyrolysis, Renewable and Sustainable Energy Reviews, 16 (2012) 5101-5109.

[16] D.C. Elliott, P. Biller, A.B. Ross, A.J. Schmidt, S.B. Jones, Hydrothermal liquefaction of biomass: Developments from batch to continuous process, Bioresource Technology, 178 (2015) 147-156.

http://www.nasdaq.com/markets/

https://ec.europa.eu/energy/node/73

References 78

[17] D.J. Roddy, C. Manson-Whitton, 5.10 – Biomass Gasification and Pyrolysis. In: Comprehensive Renewable Energy [A. Sayigh (ed.)], Elsevier, Oxford, 2012, pp. 133-153.

[18] C. Higman, M. van der Burgt, Gasification Processes. In: Gasification Gulf Professional Publishing, Burlington, 2008, pp. 91-191.

[19] J. Rostrup-Nielsen, L.J. Christiansen, Routes to Syngas. In: Concepts in Syngas Manufacture World Scientific, 2011.

[20] P. Basu, Gasification Theory. In: Biomass Gasification, Pyrolysis and Torrefaction Academic Press, Boston, 2013, pp. 199-248.

[21] P. Basu, Design of Biomass Gasifiers. In: Biomass Gasification, Pyrolysis and Torrefaction Academic Press, Boston, 2013, pp. 249-313.

[22] E4tech, Review of Technologies for Gasification of Biomass and Wastes, 2009.

[23] M. Siedlecki, W. De Jong, A.H.M. Verkooijen, Fluidized Bed Gasification as a Mature And Reliable Technology for the Production of Bio-Syngas and Applied in the Production of Liquid Transportation Fuels—A Review, Energies, 4 (2011) 389.

[24] P.K. Swain, L.M. Das, S.N. Naik, Biomass to liquid: A prospective challenge to research and development in 21st century, Renewable and Sustainable Energy Reviews, 15 (2011) 4917-4933.

[25] R.W.R. Zwart, Gas cleaning downstream biomass gasification – status report 2009, ECN, Petten, 2009.

[26] M.J.A. Tijmensen, A.P.C. Faaij, C.N. Hamelinck, M.R.M. van Hardeveld, Exploration of the possibilities for production of Fischer Tropsch liquids and power via biomass gasification, Biomass and Bioenergy, 23 (2002) 129-152.

[27] A.M. Ribeiro, J.C. Santos, A.E. Rodrigues, PSA design for stoichiometric adjustment of bio-syngas for methanol production and co-capture of carbon dioxide, Chemical Engineering Journal, 163 (2010) 355-363.

[28] N. Abdoulmoumine, S. Adhikari, A. Kulkarni, S. Chattanathan, A review on biomass gasification syngas cleanup, Applied Energy, 155 (2015) 294-307.

[29] P.J. Woolcock, R.C. Brown, A review of cleaning technologies for biomass-derived syngas, Biomass and Bioenergy, 52 (2013) 54-84.

[30] P. Sabatier, J.B. Senderens, Comptes Rendus de l'Académie des Sciences, 134 (1902) 514-516.

[31] BASF, German Patent, 293, 787, 1913.

[32] F. Fischer, H. Tropsch, Brennstoff-Chemie, 4 (1923) 276-285.

[33] J.E. Lesch, The German Chemical Industry in the Twentieth Century, Springer, 2000.

[34] A. Mittasch, M. Pier, US patent 1569775 A, 1926.

[35] S. Lee, Methanol Synthesis Technology, CRC Press,1989.

[36] http://demoplants.bioenergy2020.eu/projects/mapindex, accessed 2015-10-16.

http://demoplants.bioenergy2020.eu/projects/mapindex

References 79

[37] http://enerkem.com/, accessed 2015-10-16.

[38] G. Natta, U. Colombo, I. Pasquon, Direct Catalytic Synthesis of Higher Alcohols from Carbon Monoxide and Hydrogen. In: Catalysis [P.H. Emmett (ed.)], Reinhold, New York, 1957, pp. 131-174.

[39] A. Paggini, D. Sanfilippo, G. Pecci, I. Dybkjaer, Implementation of the methanol plus higher alcohols process by Snamprogetti, Enichem, Haldor Topsoe A/S "MAS technology". In: VII International Symposium on Alcohol Fuels Editions Technip, Paris, 1986, pp. 62-67.

[40] P. Forzatti, E. Tronconi, I. Pasquon, Higher Alcohol Synthesis, Catalysis Reviews, 33 (1991) 109-168.

[41] V. Subramani, S.K. Gangwal, A Review of Recent Literature to Search for an Efficient Catalytic Process for the Conversion of Syngas to Ethanol, Energy & Fuels, 22 (2008) 814-839.

[42] Energy Independence and Security Act of 2007. Public Law 110–140, 110th Congress of the United States.

[43] K.G. Fang, D.B. Li, M.G. Lin, M.L. Xiang, W. Wei, Y.H. Sun, A short review of heterogeneous catalytic process for mixed alcohols synthesis via syngas, Catalysis Today, 147 (2009) 133-138.

[44] G. Evans, C. Smith, 5.11 - Biomass to Liquids Technology. In: Comprehensive Renewable Energy [A. Sayigh (ed.)], Elsevier, Oxford, 2012, pp. 155-204.

[45] http://www.bloomberg.com/news/articles/2011-12-02/range-fuels-cellulosic-ethanol-plant-fails-as-u-s-pulls-plug, accessed 2015-11-19.

[46] B. Kovarik, Henry Ford, Charles Kettering and the Fuel of the Future, Automotive History Review, 32 (1998) 7-27.

[47] J. DiPardo, Outlook for Biomass Ethanol Production and Demand, DOE, EIA, 2002.

[48] R.J. Pearson, J.W.G. Turner, 5.16 - Renewable Fuels: An Automotive Perspective. In: Comprehensive Renewable Energy [A. Sayigh (ed.)], Elsevier, Oxford, 2012, pp. 305-342.

[49] S. Kumar, J.H. Cho, J. Park, I. Moon, Advances in diesel–alcohol blends and their effects on the performance and emissions of diesel engines, Renewable and Sustainable Energy Reviews, 22 (2013) 46-72.

[50] J. Campos-Fernández, J.M. Arnal, J. Gómez, M.P. Dorado, A comparison of performance of higher alcohols/diesel fuel blends in a diesel engine, Applied Energy, 95 (2012) 267-275.

[51] V.F. Andersen, J.E. Anderson, T.J. Wallington, S.A. Mueller, O.J. Nielsen, Vapor Pressures of Alcohol−Gasoline Blends, Energy & Fuels, 24 (2010) 3647-3654.

[52] R.J. Pearson, J.W.G. Turner, 3 - Using alternative and renewable liquid fuels to improve the environmental performance of internal combustion engines: key challenges and blending technologies. In: Alternative Fuels and Advanced Vehicle

http://enerkem.com/

http://www.bloomberg.com/news/articles/2011-12-02/range-fuels-cellulosic-ethanol-plant-fails-as-u-s-pulls-plug

http://www.bloomberg.com/news/articles/2011-12-02/range-fuels-cellulosic-ethanol-plant-fails-as-u-s-pulls-plug

References 80

Technologies for Improved Environmental Performance [R. Folkson (ed.)], Woodhead Publishing, 2014, pp. 52-89.

[53] D. Vidler, Classroom Manual for Automotive Engine Performance, Cengage Learning, 2003.

[54] J.E. Anderson, D.M. DiCicco, J.M. Ginder, U. Kramer, T.G. Leone, H.E. Raney-Pablo, T.J. Wallington, High octane number ethanol–gasoline blends: Quantifying the potential benefits in the United States, Fuel, 97 (2012) 585-594.

[55] B.M. Masum, H.H. Masjuki, M.A. Kalam, S.M. Palash, M. Habibullah, Effect of alcohol–gasoline blends optimization on fuel properties, performance and emissions of a SI engine, Journal of Cleaner Production, 86 (2015) 230-237.

[56] I. Gravalos, D. Moshou, T. Gialamas, P. Xyradakis, D. Kateris, Z. Tsiropoulos, Emissions characteristics of spark ignition engine operating on lower–higher molecular mass alcohol blended gasoline fuels, Renewable Energy, 50 (2013) 27-32.

[57] M.A. Costagliola, L. De Simio, S. Iannaccone, M.V. Prati, Combustion efficiency and engine out emissions of a S.I. engine fueled with alcohol/gasoline blends, Applied Energy, 111 (2013) 1162-1171.

[58] A. Elfasakhany, Experimental study of dual n-butanol and iso-butanol additives on spark-ignition engine performance and emissions, Fuel, 163 (2016) 166-174.

[59] B.M. Masum, H.H. Masjuki, M.A. Kalam, I.M. Rizwanul Fattah, S.M. Palash, M.J. Abedin, Effect of ethanol–gasoline blend on NOx emission in SI engine, Renewable and Sustainable Energy Reviews, 24 (2013) 209-222.

[60] V.R. Surisetty, A.K. Dalai, J. Kozinski, Alcohols as alternative fuels: An overview, Applied Catalysis A: General, 404 (2011) 1-11.

[61] J. Erjavec, R. Thompson, Automotive Technology: A Systems Approach, Cengage Learning, 2014.

[62] J.J. Spivey, A. Egbebi, Heterogeneous catalytic synthesis of ethanol from biomass-derived syngas, Chemical Society Reviews, 36 (2007) 1514-1528.

[63] S. Zaman, K.J. Smith, A review of molybdenum catalysts for synthesis gas conversion to alcohols: Catalysts, mechanisms and kinetics, Catalysis Reviews - Science and Engineering, 54 (2012) 41-132.

[64] G.J. Quarderer, G.A. Cochran, Catalytic process for producing mixed alcohols from hydrogen and carbon monoxide, EP 0119609 A1, The Dow Chemical Company, 1984.

[65] N.E. Kinkade, Process for producing alcohols from carbon monoxide and hydrogen using an alkali-molybdenum sulfide catalyst, EP 0149255 A2, Union Carbide Corporation, 1985.

[66] J. Iranmahboob, H. Toghiani, D.O. Hill, Dispersion of alkali on the surface of Co-MoS2/clay catalyst: a comparison of K and Cs as a promoter for synthesis of alcohol, Applied Catalysis A: General, 247 (2003) 207-218.

References 81

[67] H.C. Woo, T.Y. Park, Y.G. Kim, I.-S. Nam, J.S. Lee, J.S. Chung, Alkali-Promoted MoS2 Catalysts for Alcohol Synthesis: the Effect of Alkali Promotiom and Preparation Condition on Activity and Selectivity. In: Studies in Surface Science and Catalysis [L. Guczi, F. Solymosi, P.Tétényi (eds.)], Elsevier, 1993, pp. 2749-2752.

[68] R.R. Stevens, M.M. Conway, Mixed alcohols production from syngas, US patent 4831060 A, The Dow Chemical Company, 1989.

[69] N. Koizumi, K. Murai, T. Ozaki, M. Yamada, Development of sulfur tolerant catalysts for the synthesis of high quality transportation fuels, Catalysis Today, 89 (2004) 465-478.

[70] R. Andersson, M. Boutonnet, S. Järås, Correlation patterns and effect of syngas conversion level for product selectivity to alcohols and hydrocarbons over molybdenum sulfide based catalysts, Applied Catalysis A: General, 417–418 (2012) 119-128.

[71] D. Li, C. Yang, N. Zhao, H. Qi, W. Li, Y. Sun, B. Zhong, The performances of higher alcohol synthesis over nickel modified K2CO3/MoS2 catalyst, Fuel Processing Technology, 88 (2007) 125-127.

[72] D. Li, N. Zhao, H. Qi, W. Li, Y.H. Sun, B. Zhong, Ultrasonic preparation of Ni modified K2CO3/MoS2 catalyst for higher alcohols synthesis, Catalysis Communications, 6 (2005) 674-678.

[73] D.B. Li, C. Yang, H.R. Zhang, W.H. Li, Y.H. Sun, B. Zhong, A catalyst with high activity and selectivity for the synthesis of C2+-OH: ADM catalyst modified by nickel. In: Studies in Surface Science and Catalysis [B. Xinhe, X. Yide (eds.)], Elsevier, 2004, pp. 391-396.

[74] H. Qi, D. Li, C. Yang, Y. Ma, W. Li, Y. Sun, B. Zhong, Nickel and manganese co-modified K/MoS2 catalyst: high performance for higher alcohols synthesis from CO hydrogenation, Catalysis Communications, 4 (2003) 339-342.

[75] B.E. Concha, G.L. Bartholomew, C.H. Bartholomew, CO hydrogenation on supported molybdenum catalysts: Effects of support on specific activities of reduced and sulfided catalysts, Journal of Catalysis, 89 (1984) 536-541.

[76] J. Iranmahboob, D.O. Hill, Alcohol synthesis from syngas over K2CO3/CoS/MoS2 on activated carbon, Catal. Lett., 78 (2002) 49-55.

[77] Z.-r. Li, Y.-l. Fu, M. Jiang, T.-d. Hu, T. Liu, Y.-n. Xie, Active Carbon Supported Mo–K Catalysts Used for Alcohol Synthesis, Journal of Catalysis, 199 (2001) 155-161.

[78] M. Taborga Claure, S.-H. Chai, S. Dai, K.A. Unocic, F.M. Alamgir, P.K. Agrawal, C.W. Jones, Tuning of higher alcohol selectivity and productivity in CO hydrogenation reactions over K/MoS2 domains supported on mesoporous activated carbon and mixed MgAl oxide, Journal of Catalysis, 324 (2015) 88-97.

[79] V.R. Surisetty, Y. Hu, A.K. Dalai, J. Kozinski, Structural characterization and catalytic performance of alkali (K) and metal (Co and Rh)-promoted MoS2 catalysts for higher alcohols synthesis, Applied Catalysis A: General, 392 (2011) 166-172.

References 82

[80] V.R. Surisetty, I. Eswaramoorthi, A.K. Dalai, Comparative study of higher alcohols synthesis over alumina and activated carbon-supported alkali-modified MoS2 catalysts promoted with group VIII metals, Fuel, 96 (2012) 77-84.

[81] M.R. Morrill, N.T. Thao, H. Shou, R.J. Davis, D.G. Barton, D. Ferrari, P.K. Agrawal, C.W. Jones, Origins of unusual alcohol selectivities over mixed MgAl oxide-supported K/MoS2 catalysts for higher alcohol synthesis from syngas, ACS Catalysis, 3 (2013) 1665-1675.

[82] V. Surisetty, Research and development of Co and Rh-promoted alkali-modified molybdenum sulfide catalysts for higher alcohols synthesis from synthesis gas, PhD Thesis, University of Saskatchewan, 2010.

[83] D. Ferrari, G. Budroni, L. Bisson, N.J. Rane, B.D. Dickie, J.H. Kang, S.J. Rozeveld, Effect of potassium addition method on MoS2 performance for the syngas to alcohol reaction, Applied Catalysis A: General, 462–463 (2013) 302-309.

[84] A.A. Tsyganenko, F. Can, A. Travert, F. Maugé, FTIR study of unsupported molybdenum sulfide - in situ synthesis and surface properties characterization, Applied Catalysis A: General, 268 (2004) 189-197.

[85] S. Eriksson, U. Nylén, S. Rojas, M. Boutonnet, Preparation of catalysts from microemulsions and their applications in heterogeneous catalysis, Applied Catalysis A: General, 265 (2004) 207-219.

[86] M. Boutonnet, S. Lögdberg, E. Elm Svensson, Recent developments in the application of nanoparticles prepared from w/o microemulsions in heterogeneous catalysis, Current Opinion in Colloid & Interface Science, 13 (2008) 270-286.

[87] M. Sanchez-Dominguez, K. Pemartin, M. Boutonnet, Preparation of inorganic nanoparticles in oil-in-water microemulsions: A soft and versatile approach, Current Opinion in Colloid & Interface Science, 17 (2012) 297-305.

[88] M. Boutonnet, J. Kizling, P. Stenius, G. Maire, The preparation of monodisperse colloidal metal particles from microemulsions, Colloids and Surfaces, 5 (1982) 209-225.

[89] B.J. Palla, D.O. Shah, P. Garcia-Casillas, J. Matutes-Aquino, Preparation of nanoparticles of barium ferrite from precipitation in microemulsions, Journal of Nanoparticle Research, 1 (1999) 215-221.

[90] E.E. Svensson, S. Nassos, M. Boutonnet, S.G. Järås, Microemulsion synthesis of MgO-supported LaMnO3 for catalytic combustion of methane, Catalysis Today, 117 (2006) 484-490.

[91] S. Nassos, E.E. Svensson, M. Nilsson, M. Boutonnet, S. Järås, Microemulsion-prepared Ni catalysts supported on cerium-lanthanum oxide for the selective catalytic oxidation of ammonia in gasified biomass, Applied Catalysis B: Environmental, 64 (2006) 96-102.

[92] B. Kronberg, K. Holmberg, B. Lindman, Surface Chemistry of Surfactants and Polymers, Wiley, 2014.

[93] C.H. Bartholomew, R.J. Farrauto, Catalyst Characterization and Selection. In: Fundamentals of Industrial Catalytic Processes John Wiley & Sons, Inc., 2005, pp. 118-196.

References 83

[94] P. Gabbott, Principles and Applications of Thermal Analysis, Wiley, 2008.

[95] W.E.S. Unger, T. Gross, Catalyst Characterization. In: Handbook of Surface and Interface Analysis CRC Press, 2009, pp. 501-528.

[96] A.M. Venezia, X-ray photoelectron spectroscopy (XPS) for catalysts characterization, Catalysis Today, 77 (2003) 359-370.

[97] G. Leofanti, M. Padovan, G. Tozzola, B. Venturelli, Surface area and pore texture of catalysts, Catalysis Today, 41 (1998) 207-219.

[98] S. Brunauer, P.H. Emmett, E. Teller, Adsorption of Gases in Multimolecular Layers, Journal of the American Chemical Society, 60 (1938) 309-319.

[99] K. Sing, The use of nitrogen adsorption for the characterisation of porous materials, Colloids and Surfaces A: Physicochemical and Engineering Aspects, 187–188 (2001) 3-9.

[100] Z. Ma, F. Zaera, Characterization of Heterogeneous Catalysts. In: Surface and Nanomolecular Catalysis [R. Richards (ed.)], CRC Press, 2006.

[101] J.G. Santiesteban, C.E. Bogdan, R.G. Herman, K. Klier, Mechanism of C1-C4

Alcohol Synthesis over Alkali/MoS2 and Alkali/Co/MoS2 Catalysts, 9th International Congress on Catalysis, 1988, pp. 561-568.

[102] J.M. Christensen, P.M. Mortensen, R. Trane, P.A. Jensen, A.D. Jensen, Effects of H2S and process conditions in the synthesis of mixed alcohols from syngas over alkali promoted cobalt-molybdenum sulfide, Applied Catalysis A: General, 366 (2009) 29-43.

[103] J.M. Christensen, P.A. Jensen, N.C. Schioedt, A.D. Jensen, Coupling of Alcohols over Alkali-Promoted Cobalt-Molybdenum Sulfide, Chemcatchem, 2 (2010) 523-526.

[104] V.P. Santos, B. van der Linden, A. Chojecki, G. Budroni, S. Corthals, H. Shibata, G.R. Meima, F. Kapteijn, M. Makkee, J. Gascon, Mechanistic Insight into the Synthesis of Higher Alcohols from Syngas: The Role of K Promotion on MoS2 Catalysts, ACS Catal., 3 (2013) 1634-1637.

[105] R.G. Herman, Chapter 7: Classical and Non-Classical Routes for Alcohol Synthesis. In: Studies in Surface Science and Catalysis [L. Guczi (ed.)], Elsevier, 1991, pp. 265-349.

[106] R. Andersson, M. Boutonnet, S. Järås, On-line gas chromatographic analysis of higher alcohol synthesis products from syngas, Journal of Chromatography A, 1247 (2012) 134-145.

[107] R. Andersson, M. Boutonnet, S. Järås, Effect of CO2 in the synthesis of mixed alcohols from syngas over a K/Ni/MoS2 catalyst, Fuel, 107 (2013) 715-723.

[108] R. Andersson, M. Boutonnet, S. Järås, Higher alcohols from syngas using a K/Ni/MoS2 catalyst: Trace sulfur in the product and effect of H2S-containing feed, Fuel, 115 (2014) 544-550.

[109] W.M. Shen, J.A. Dumesic, C.G. Hill Jr, Criteria for stable Ni particle size under methanation reaction conditions: Nickel transport and particle size growth via nickel carbonyl, Journal of Catalysis, 68 (1981) 152-165.

References 84

[110] M. Lachowska, J. Skrzypek, Synthesis of higher alcohols. Enhancement by the addition of methanol or ethanol to the syngas. In: Studies in Surface Science and Catalysis [A. Parmaliana, D. Sanfilippo, F. Frusteri, A. Vaccari, F. Arena (eds.)], Elsevier, 1998, pp. 473-478.

[111] A. Kiennemann, H. Idriss, R. Kieffer, P. Chaumette, D. Durand, Study of the mechanism of higher alcohol synthesis on copper-zinc oxide-aluminum oxide catalysts by catalytic tests, probe molecules, and temperature programmed desorption studies, Industrial & Engineering Chemistry Research, 30 (1991) 1130-1138.

[112] J.M. Campos-Martín, J.L.G. Fierro, A. Guerrero-Ruiz, R.G. Herman, K. Klier, Promoter Effect of Cesium on C–C Bond Formation during Alcohol Synthesis from CO/H2 over Cu/ZnO/Cr2O3 Catalysts, Journal of Catalysis, 163 (1996) 418-428.

[113] G.J. Quarderer, R.R. Stevens, G.A. Cochran, C.B. Murchison, Preparation of ethanol and higher alcohols from lower carbon number alcohols, US patent 4825013 A, The Dow Chemical Company, 1989.

[114] A.K. Gunturu, E.L. Kugler, J.B. Cropley, D.B. Dadyburjor, A kinetic model for the synthesis of high-molecular-weight alcohols over a sulfided Co-K-Mo/C catalyst, Ind. Eng. Chem. Res., 37 (1998) 2107-2115.

[115] K.A.N. Verkerk, B. Jaeger, C.-H. Finkeldei, W. Keim, Recent developments in isobutanol synthesis from synthesis gas, Applied Catalysis A: General, 186 (1999) 407-431.

[116] M.A. Portillo Crespo, A.L. Villanueva Perales, F. Vidal-Barrero, M. Campoy, Effects of methanol co-feeding in ethanol synthesis from syngas using alkali-doped MoS2 catalysts, Fuel Processing Technology, 134 (2015) 270-274.

[117] G. Prieto González, J.M. Serra Alfaro, A. Martinez Feliu, J.L. Sanz Yagüe, J. Caraballo Bello, R. Arjona Antolín, Method for obtaining a multimetallic sulfureous catalyst and use thereof in a method for producing higher alcohols by catalytic conversion of synthesis gas, US patent 20120202898 A1, 2012.

[118] M. Taborga Claure, M.R. Morrill, J.W. Goh, S.-H. Chai, S. Dai, P.K. Agrawal, C.W. Jones, Insight into reaction pathways in CO hydrogenation reactions over K/MoS2 supported catalysts via alcohol/olefin co-feed experiments, Catalysis Science & Technology, (2016).

[119] M. Claeys, E. van Steen, Chapter 8 - Basic studies. In: Studies in Surface Science and Catalysis [S. André, D. Mark (eds.)], Elsevier, 2004, pp. 601-680.

[120] W. Böhringer, A. Kotsiopoulos, M. de Boer, C. Knottenbelt, J.C.Q. Fletcher, Selective Fischer-Tropsch wax hydrocracking — opportunity for improvement of overall gas-to-liquids processing. In: Studies in Surface Science and Catalysis [B.H. Davis, M.L. Occelli (eds.)], Elsevier, 2007, pp. 345-365.

[121] J. Eilers, S.A. Posthuma, S.T. Sie, The shell middle distillate synthesis process (SMDS), Catalysis Letters, 7 (1990) 253-269.

[122] L.P. Dancuart, R. de Haan, A. de Klerk, Processing of primary Fischer-Tropsch products. In: Studies in Surface Science and Catalysis [M.E. Dry, A.P. Steynberg (eds.)], Elsevier, 2004, pp. 482-532.

References 85

[123] D. Leckel, Hydrocracking of Iron-Catalyzed Fischer−Tropsch Waxes, Energy & Fuels, 19 (2005) 1795-1803.

[124] S.T. Sie, M.M.G. Senden, H.M.H. Van Wechem, Conversion of natural gas to transportation fuels via the shell middle distillate synthesis process (SMDS), Catalysis Today, 8 (1991) 371-394.

[125] R.F. Sullivan, J.W. Scott, The Development of Hydrocracking. In: Heterogeneous Catalysis American Chemical Society, 1983, pp. 293-313.

[126] V. Calemma, C. Gambaro, W.O. Parker Jr, R. Carbone, R. Giardino, P. Scorletti, Middle distillates from hydrocracking of FT waxes: Composition, characteristics and emission properties, Catalysis Today, 149 (2010) 40-46.

[127] P.P. Shah, G.C. Sturtevant, J.H. Gregor, M.J. Humbach, F.G. Padrta, K.Z. Steigleder, Fischer-Tropsch Wax Characterization and Upgrading: Final Report, UOP Inc., 1988.

[128] M.E. Dry, High quality diesel via the Fischer–Tropsch process – a review, Journal of Chemical Technology & Biotechnology, 77 (2002) 43-50.

[129] A. de Klerk, Shell Middle Distillate Synthesis (SMDS) Facilities. In: Fischer-Tropsch Refining Wiley-VCH Verlag GmbH & Co. KGaA, 2011, pp. 231-240.

[130] F.J.M. Schrauwen, Shell Middle Distillate Synthesis (SMDS) process. In: Handbook of petroleum refining processes [R.A. Meyers (ed.)], McGraw-Hill, New York, 2004, pp. 15.25–15.40.

[131] A. de Klerk, Oryx and Escravos Gas-to-Liquids Facilities. In: Fischer-Tropsch Refining Wiley-VCH Verlag GmbH & Co. KGaA, 2011, pp. 241-248.

[132] F.G. Botes, L.P. Dancuart, H.G. Nel, A.P. Steynberg, A.P. Vogel, B.B. Breman, J.H.M. Font Freide, 11 - Middle distillate fuel production from synthesis gas via the Fischer–Tropsch process. In: Advances in Clean Hydrocarbon Fuel Processing [M.R. Khan (ed.)], Woodhead Publishing, 2011, pp. 329-362.

[133] D. Lamprecht, L.P. Dancuart, K. Harrilall, Performance Synergies between Low-Temperature and High-Temperature Fischer−Tropsch Diesel Blends, Energy & Fuels, 21 (2007) 2846-2852.

[134] A. de Klerk, Fischer-Tropsch fuels refinery design, Energy & Environmental Science, 4 (2011) 1177-1205.

[135] A. de Klerk, Can Fischer−Tropsch Syncrude Be Refined to On-Specification Diesel Fuel?, Energy & Fuels, 23 (2009) 4593-4604.

[136] C. Bouchy, G. Hastoy, E. Guillon, J.A. Martens, Valorisation des cires de Fischer-Tropsch par hydrocraquage et hydroisomérisation sélective, Oil & Gas Science and Technology - Rev. IFP, 64 (2009) 91-112.

[137] H. Deldari, Suitable catalysts for hydroisomerization of long-chain normal paraffins, Applied Catalysis A: General, 293 (2005) 1-10.

[138] V. Calemma, S. Peratello, C. Perego, Hydroisomerization and hydrocracking of long chain n-alkanes on Pt/amorphous SiO2–Al2O3 catalyst, Applied Catalysis A: General, 190 (2000) 207-218.

References 86

[139] V. Calemma, S. Correra, C. Perego, P. Pollesel, L. Pellegrini, Hydroconversion of Fischer–Tropsch waxes: Assessment of the operating conditions effect by factorial design experiments, Catalysis Today, 106 (2005) 282-287.

[140] F. Alvarez, F.R. Ribeiro, G. Perot, C. Thomazeau, M. Guisnet, Hydroisomerization and Hydrocracking of Alkanes: 7. Influence of the Balance between Acid and Hydrogenating Functions on the Transformation of n-Decane on PtHY Catalysts, Journal of Catalysis, 162 (1996) 179-189.

[141] V. Calemma, C. Gambaro, Effect of Feed Distribution on Hydrocracking of Fischer-Tropsch Wax. In: Synthetic Liquids Production and Refining American Chemical Society, 2011, pp. 239-253.

[142] V. Calemma, S. Peratello, S. Pavoni, G. Clerici, C. Perego, Hydroconversion of a mixture of long chain n-paraffins to middle distillate: effect of the operating parameters and products properties. In: Studies in Surface Science and Catalysis [E. Iglesia, J.J. Spivey, T.H. Fleisch (eds.)], Elsevier, 2001, pp. 307-312.

[143] C. Gambaro, G. Pederzani, V. Calemma, Hydroconversion of Fischer-Tropsch waxes over platinum loaded catalyst: Assessment of product distribution, 2008 AIChE Spring National Meeting, Conference Proceedings, 2008.

[144] I. Rossetti, C. Gambaro, V. Calemma, Hydrocracking of long chain linear paraffins, Chemical Engineering Journal, 154 (2009) 295-301.

[145] D. Leckel, Low-Pressure Hydrocracking of Coal-Derived Fischer−Tropsch Waxes to Diesel, Energy & Fuels, 21 (2007) 1425-1431.

[146] D. Leckel, Noble Metal Wax Hydrocracking Catalysts Supported on High-Siliceous Alumina, Industrial & Engineering Chemistry Research, 46 (2007) 3505-3512.

[147] D. Leckel, Selectivity Effect of Oxygenates in Hydrocracking of Fischer−Tropsch Waxes, Energy & Fuels, 21 (2007) 662-667.

[148] D. Leckel, M. Liwanga-Ehumbu, Diesel-Selective Hydrocracking of an Iron-Based Fischer−Tropsch Wax Fraction (C15−C45) Using a MoO3-Modified Noble Metal Catalyst, Energy & Fuels, 20 (2006) 2330-2336.

[149] F. Pinna, Supported metal catalysts preparation, Catalysis Today, 41 (1998) 129-137.

[150] C.H. Bartholomew, R.J. Farrauto, Catalyst Materials, Properties and Preparation. In: Fundamentals of Industrial Catalytic Processes John Wiley & Sons, Inc., 2005, pp. 60-117.

[151] http://www.sasolgermany.de/silica_aluminas.html, accessed 2015-11-12.

[152] W. Daniell, U. Schubert, R. Glöckler, A. Meyer, K. Noweck, H. Knözinger, Enhanced surface acidity in mixed alumina–silicas: a low-temperature FTIR study, Applied Catalysis A: General, 196 (2000) 247-260.

[153] G. Bergeret, P. Gallezot, Particle Size and Dispersion Measurements. In: Handbook of Heterogeneous Catalysis Wiley-VCH Verlag GmbH & Co. KGaA, 2008.

http://www.sasolgermany.de/silica_aluminas.html

References 87

[154] S. Lowell, J. Shields, M. Thomas, M. Thommes, Volumetric Chemisorption: Catalyst Characterization by Static Methods. In: Characterization of Porous Solids and Powders: Surface Area, Pore Size and Density Springer Netherlands, 2004, pp. 283-296.

[155] C.H. Bartholomew, H2 adsorption on supported noble metals and its use in determining metal dispersion. In: Catalysis: Volume 11 [J.J. Spivey, S.K. Agarwal (eds.)], The Royal Society of Chemistry, 1994, pp. 93-126.

[156] A. Niquille-Röthlisberger, R. Prins, Hydrodesulfurization of 4,6-dimethyldibenzothiophene and dibenzothiophene over alumina-supported Pt, Pd, and Pt-Pd catalysts, Journal of Catalysis, 242 (2006) 207-216.

[157] V. Ragaini, R. Giannantonio, P. Magni, L. Lucarelli, G. Leofanti, Dispersion measurement by the single introduction method coupled with the back-sorption procedure: A chemisorption and TPD study of the different chemisorbed hydrogen species: II. Pd on Alumina, Journal of Catalysis, 146 (1994) 116-125.

[158] C. Karakaya, O. Deutschmann, A simple method for CO chemisorption studies under continuous flow: Adsorption and desorption behavior of Pt/Al2O3

catalysts, Applied Catalysis A: General, 445–446 (2012) 221-230.

[159] F. Pinna, F. Menegazzo, M. Signoretto, P. Canton, G. Fagherazzi, N. Pernicone, Consecutive hydrogenation of benzaldehyde over Pd catalysts: Influence of supports and sulfur poisoning, Applied Catalysis A: General, 219 (2001) 195-200.

[160] P. Canton, G. Fagherazzi, M. Battagliarin, F. Menegazzo, F. Pinna, N. Pernicone, Pd/CO Average Chemisorption Stoichiometry in Highly Dispersed Supported Pd/γ-Al2O3 Catalysts, Langmuir, 18 (2002) 6530-6535.

[161] H. Karge, E. Geidel, Vibrational Spectroscopy. In: Characterization I [H. Karge, J. Weitkamp (eds.)], Springer Berlin Heidelberg, 2004, pp. 1-200.

[162] C.A. Emeis, Determination of Integrated Molar Extinction Coefficients for Infrared Absorption Bands of Pyridine Adsorbed on Solid Acid Catalysts, Journal of Catalysis, 141 (1993) 347-354.

[163] T. Barzetti, E. Selli, D. Moscotti, L. Forni, Pyridine and ammonia as probes for FTIR analysis of solid acid catalysts, Journal of the Chemical Society, Faraday Transactions, 92 (1996) 1401-1407.

[164] T.K. Varadarajan, Catalysis by Solid Acids. In: Catalysis: principles and applications [B. Viswanathan, S. Sivasanker (eds.)], Narosa, 2002.

[165] L.-E. Sandoval-Díaz, J.-A. González-Amaya, C.-A. Trujillo, General aspects of zeolite acidity characterization, Microporous and Mesoporous Materials, 215 (2015) 229-243.

[166] H.P. Boehm, Acidic and basic properties of hydroxylated metal oxide surfaces, Discussions of the Faraday Society, 52 (1971) 264-275.

[167] M. Digne, P. Sautet, P. Raybaud, P. Euzen, H. Toulhoat, Use of DFT to achieve a rational understanding of acid–basic properties of γ-alumina surfaces, Journal of Catalysis, 226 (2004) 54-68.

References 88

[168] S.K. Das, S. Kapoor, H. Yamada, A.J. Bhattacharyya, Effects of surface acidity and pore size of mesoporous alumina on degree of loading and controlled release of ibuprofen, Microporous and Mesoporous Materials, 118 (2009) 267-272.

[169] R. Mueller, H.K. Kammler, K. Wegner, S.E. Pratsinis, OH Surface Density of SiO2 and TiO2 by Thermogravimetric Analysis, Langmuir, 19 (2003) 160-165.

[170] F. Regali, Hydroconversion of Model Fischer Tropsch Wax over Noble Metal / Silica-Alumina Catalysts, PhD Thesis, KTH, 2013.

[171] A. de Klerk, Hydroprocessing peculiarities of Fischer–Tropsch syncrude, Catalysis Today, 130 (2008) 439-445.

[172] M. Guisnet, P. Magnoux, Organic chemistry of coke formation, Applied Catalysis A: General, 212 (2001) 83-96.

[173] C.H. Bartholomew, R.J. Farrauto, Catalyst Deactivation: Causes, Mechanisms, and Treatment. In: Fundamentals of Industrial Catalytic Processes John Wiley & Sons, Inc., 2005, pp. 260-336.

[174] I. Nam, J. Seo, S. Hwang, I. Song, Deactivation behaviors of hybrid Fischer–Tropsch catalysts in the production of middle distillate from synthesis gas in a dual-bed reactor, Research on Chemical Intermediates, 36 (2010) 685-692.

[175] H.L. Coonradt, W.E. Garwood, Mechanism of Hydrocracking. Reactions of Paraffins and Olefins, Industrial & Engineering Chemistry Process Design and Development, 3 (1964) 38-45.

[176] M.J. Girgis, Y.P. Tsao, Impact of Catalyst Metal−Acid Balance in n-Hexadecane Hydroisomerization and Hydrocracking, Industrial & Engineering Chemistry Research, 35 (1996) 386-396.

[177] J.W. Thybaut, C.S. Laxmi Narasimhan, J.F. Denayer, G.V. Baron, P.A. Jacobs, J.A. Martens, G.B. Marin, Acid−Metal Balance of a Hydrocracking Catalyst: Ideal versus Nonideal Behavior, Industrial & Engineering Chemistry Research, 44 (2004) 5159-5169.

[178] J. Weitkamp, The Influence of Chain Length in Hydrocracking and Hydroisomerization of n-Alkanes. In: Hydrocracking and Hydrotreating [J.W. Ward, S.A. Qader (eds.)], American Chemical Society, 1975, pp. 1-27.

[179] J. Weitkamp, Catalytic Hydrocracking-Mechanisms and Versatility of the Process, ChemCatChem, 4 (2012) 292-306.

[180] V. Ponec, G.C. Bond, Reactions of alkanes and reforming of naphtha. In: Catalysis by Metals and Alloys Elsevier, 1995, pp. 583-677.

[181] J.H. Sinfelt, Specificity in Catalytic Hydrogenolysis by Metals. In: Advances in Catalysis [D.D. Eley, H. Pines, P.B. Weisz (eds.)], Academic Press, 1973, pp. 91-119.

[182] W.O. Haag, R.M. Dessau, Duality of Mechanism in Acid Catalyzed Paraffin Cracking, 8th International Congress on Catalysis, Berlin, Germany, 1984, pp. 305-315.

References 89

[183] S. Kotrel, H. Knözinger, B.C. Gates, The Haag–Dessau mechanism of protolytic cracking of alkanes, Microporous and Mesoporous Materials, 35–36 (2000) 11-20.

[184] F.C. Jentoft, B.C. Gates, Solid-acid-catalyzed alkane cracking mechanisms: evidence from reactions of small probe molecules, Top Catal, 4 (1997) 1-13.

[185] A. de Klerk, Refining Technology Selection. In: Fischer-Tropsch Refining Wiley-VCH Verlag GmbH & Co. KGaA, 2011, pp. 301-334.

[186] A. Kossiakoff, F.O. Rice, Thermal Decomposition of Hydrocarbons, Resonance Stabilization and Isomerization of Free Radicals, Journal of the American Chemical Society, 65 (1943) 590-595.

[187] L.B. Bragg, Rate of Cracking of Paraffin Wax, Industrial & Engineering Chemistry, 33 (1941) 376-380.

[188] H.H. Voge, G.M. Good, Thermal Cracking of Higher Paraffins, Journal of the American Chemical Society, 71 (1949) 593-597.

[189] A. de Klerk, Thermal Cracking of Fischer−Tropsch Waxes, Industrial & Engineering Chemistry Research, 46 (2007) 5516-5521.

[190] I. Gutman, The chemical formula CnH2n+2 and its mathematical background, The Teaching of Mathematics, XI_2 (2008) 53-61.

[191] L. Pellegrini, S. Locatelli, S. Rasella, S. Bonomi, V. Calemma, Modeling of Fischer–Tropsch products hydrocracking, Chemical Engineering Science, 59 (2004) 4781-4787.

[192] F. Regali, M. Boutonnet, S. Järås, Hydrocracking of n-hexadecane on noble metal/silica–alumina catalysts, Catalysis Today, 214 (2013) 12-18.

[193] D. Kubička, N. Kumar, T. Venäläinen, H. Karhu, I. Kubičková, H. Österholm, D.Y. Murzin, Metal−Support Interactions in Zeolite-Supported Noble Metals: Influence of Metal Crystallites on the Support Acidity, The Journal of Physical Chemistry B, 110 (2006) 4937-4946.

[194] J. Ancheyta, S. Sánchez, M.A. Rodríguez, Kinetic modeling of hydrocracking of heavy oil fractions: A review, Catalysis Today, 109 (2005) 76-92.

[195] B.D. Vandegehuchte, J.W. Thybaut, A. Martínez, M.A. Arribas, G.B. Marin, n-Hexadecane hydrocracking Single-Event MicroKinetics on Pt/H-beta, Applied Catalysis A: General, 441–442 (2012) 10-20.

[196] J.M. Schweitzer, P. Galtier, D. Schweich, A single events kinetic model for the hydrocracking of paraffins in a three-phase reactor, Chemical Engineering Science, 54 (1999) 2441-2452.

[197] J.C. Chavarría, J. Ramírez, H. González, M.A. Baltanas, Modelling of n-hexadecane hydroisomerization and hydrocracking reactions on a Mo/Hβ-alumina Bi-functional catalyst, using the single event concept, Catalysis Today, 98 (2004) 235-242.

[198] G. Martens, G.F. Froment, Kinetic modeling of paraffins hydrocracking based upon elementary steps and the single event concept. In: Studies in Surface Science and Catalysis [G.F. Froment, K.C. Waugh (eds.)], Elsevier, 1999, pp. 333-340.

References 90

[199] G.F. Froment, Single Event Kinetic Modeling of Complex Catalytic Processes, Catalysis Reviews, 47 (2005) 83-124.

[200] H. Kumar, G.F. Froment, A Generalized Mechanistic Kinetic Model for the Hydroisomerization and Hydrocracking of Long-Chain Paraffins, Industrial & Engineering Chemistry Research, 46 (2007) 4075-4090.

[201] M.A. Baltanas, K.K. Van Raemdonck, G.F. Froment, S.R. Mohedas, Fundamental kinetic modeling of hydroisomerization and hydrocracking on noble metal-loaded faujasites. 1. Rate parameters for hydroisomerization, Industrial & Engineering Chemistry Research, 28 (1989) 899-910.

[202] M. Adam, V. Calemma, F. Galimberti, C. Gambaro, J. Heiszwolf, R. Ocone, Continuum lumping kinetics of complex reactive systems, Chemical Engineering Science, 76 (2012) 154-164.

[203] I. Elizalde, M.A. Rodríguez, J. Ancheyta, Application of continuous kinetic lumping modeling to moderate hydrocracking of heavy oil, Applied Catalysis A: General, 365 (2009) 237-242.

[204] C.S. Laxminarasimhan, R.P. Verma, P.A. Ramachandran, Continuous lumping model for simulation of hydrocracking, AIChE Journal, 42 (1996) 2645-2653.

[205] I. Elizalde, M.A. Rodríguez, J. Ancheyta, Modeling the effect of pressure and temperature on the hydrocracking of heavy crude oil by the continuous kinetic lumping approach, Applied Catalysis A: General, 382 (2010) 205-212.

[206] H.M.S. Lababidi, F.S. AlHumaidan, Modeling the Hydrocracking Kinetics of Atmospheric Residue in Hydrotreating Processes by the Continuous Lumping Approach, Energy & Fuels, 25 (2011) 1939-1949.

[207] M.A. Baltanas, H. Vansina, G.F. Froment, Hydroisomerization and hydrocracking. 5. Kinetic analysis of rate data for n-octane, Industrial & Engineering Chemistry Product Research and Development, 22 (1983) 531-539.

[208] M. Steijns, G.F. Froment, Hydroisomerization and hydrocracking. 3. Kinetic analysis of rate data for n-decane and n-dodecane, Industrial & Engineering Chemistry Product Research and Development, 20 (1981) 660-668.

[209] G.F. Froment, Kinetics of the hydroisomerization and hydrocracking of paraffins on a platinum containing bifunctional Y-zeolite, Catalysis Today, 1 (1987) 455-473.

[210] L. Pellegrini, S. Bonomi, S. Gamba, V. Calemma, D. Molinari, The “all components hydrocracking model”, Chemical Engineering Science, 62 (2007) 5013-5020.

[211] L.A. Pellegrini, S. Gamba, V. Calemma, S. Bonomi, Modelling of hydrocracking with vapour–liquid equilibrium, Chemical Engineering Science, 63 (2008) 4285-4291.

[212] S. Gamba, L.A. Pellegrini, V. Calemma, C. Gambaro, Introduction of a Breakage Probability Function in the Hydrocracking Reactor Model, Industrial & Engineering Chemistry Research, 48 (2009) 5656-5665.

References 91

[213] C. Gambaro, V. Calemma, D. Molinari, J. Denayer, Hydrocracking of Fischer-Tropsch waxes: Kinetic modeling via LHHW approach, AIChE Journal, 57 (2011) 711-723.

[214] J.A. Martens, P.A. Jacobs, J. Weitkamp, Attempts to rationalize the distribution of hydrocracked products. I qualitative description of the primary hydrocracking modes of long chain paraffins in open zeolites, Applied Catalysis, 20 (1986) 239-281.

[215] J. Weitkamp, P.A. Jacobs, J.A. Martens, Isomerization and hydrocracking of C9 through C16 n-alkanes on Pt/HZSM-5 zeolite, Applied Catalysis, 8 (1983) 123-141.

[216] L.A. Pellegrini, S. Gamba, S. Bonomi, V. Calemma, Equilibrium Constants for Isomerization of n-Paraffins, Industrial & Engineering Chemistry Research, 46 (2007) 5446-5452.

[217] C. Gambaro, V. Calemma, Hydrocracking Different Fischer-Tropsch Waxes: Model Simulations. In: Synthetic Liquids Production and Refining American Chemical Society, 2011, pp. 255-277.

[218] I.R. Choudhury, K. Hayasaka, J.W. Thybaut, C.S. Laxmi Narasimhan, J.F. Denayer, J.A. Martens, G.B. Marin, Pt/H-ZSM-22 hydroisomerization catalysts optimization guided by Single-Event MicroKinetic modeling, Journal of Catalysis, 290 (2012) 165-176.

[219] J.A. Martens, P.A. Jacobs, Evidence for hypocage catalysis in FAU/ EMT zeolites derived from isomerisation and hydrocracking of heptadecane, Journal of Molecular Catalysis, 78 (1993) L47-L52.

[220] E.J.P. Feijen, J.A. Martens, P.A. Jacobs, Isomerization and hydrocracking of decane and heptadecane on cubic and hexagonal faujasite zeolites and their intergrowth structures. In: Studies in Surface Science and Catalysis [J.W. Hightower, W.N. Delgass, E. Iglesia, A.T. Bell (eds.)], Elsevier, 1996, pp. 721-729.

catalytic conversion of biomass-derived synthesis...

Documents