Post on 25-Jun-2020
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
R. Suárez París, V. Montes, M. Boutonnet, S. Järås
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
R. Suárez París, V. Montes, M. Boutonnet, S. Järås
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].
2.SECOND-GENERATION BIOFUELS 7
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].
2.SECOND-GENERATION BIOFUELS 8
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
2.SECOND-GENERATION BIOFUELS 9
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
2.SECOND-GENERATION BIOFUELS 10
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.
2.SECOND-GENERATION BIOFUELS 11
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
2.SECOND-GENERATION BIOFUELS 12
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.
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
3.HIGHER ALCOHOL SYNTHESIS 15
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.
3.HIGHER ALCOHOL SYNTHESIS 16
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].
3.HIGHER ALCOHOL SYNTHESIS 17
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
3.HIGHER ALCOHOL SYNTHESIS 18
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
3.HIGHER ALCOHOL SYNTHESIS 19
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].
3.HIGHER ALCOHOL SYNTHESIS 20
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.HIGHER ALCOHOL SYNTHESIS 21
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
3.HIGHER ALCOHOL SYNTHESIS 22
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.HIGHER ALCOHOL SYNTHESIS 23
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
3.HIGHER ALCOHOL SYNTHESIS 24
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.
3.HIGHER ALCOHOL SYNTHESIS 25
Figure 3-3. P&ID of the HAS reactor setup. Author's own elaboration.
3.HIGHER ALCOHOL SYNTHESIS 26
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
3.HIGHER ALCOHOL SYNTHESIS 27
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
3.HIGHER ALCOHOL SYNTHESIS 28
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.
3.HIGHER ALCOHOL SYNTHESIS 29
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.
3.HIGHER ALCOHOL SYNTHESIS 30
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.
3.HIGHER ALCOHOL SYNTHESIS 31
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
3.HIGHER ALCOHOL SYNTHESIS 32
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.
3.HIGHER ALCOHOL SYNTHESIS 33
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.
3.HIGHER ALCOHOL SYNTHESIS 34
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.HIGHER ALCOHOL SYNTHESIS 35
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
4. HYDROCONVERSION OF FISCHER-TROPSCH WAX 39
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
4. HYDROCONVERSION OF FISCHER-TROPSCH WAX 40
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
4. HYDROCONVERSION OF FISCHER-TROPSCH WAX 41
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
4. HYDROCONVERSION OF FISCHER-TROPSCH WAX 42
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.
4. HYDROCONVERSION OF FISCHER-TROPSCH WAX 43
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
4. HYDROCONVERSION OF FISCHER-TROPSCH WAX 44
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
4. HYDROCONVERSION OF FISCHER-TROPSCH WAX 45
°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
4. HYDROCONVERSION OF FISCHER-TROPSCH WAX 46
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].
4. HYDROCONVERSION OF FISCHER-TROPSCH WAX 47
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
4. HYDROCONVERSION OF FISCHER-TROPSCH WAX 48
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
4. HYDROCONVERSION OF FISCHER-TROPSCH WAX 49
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.
4. HYDROCONVERSION OF FISCHER-TROPSCH WAX 50
Figure 4-4. P&ID of the hydroconversion reactor setup. Author's own elaboration.
4. HYDROCONVERSION OF FISCHER-TROPSCH WAX 51
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.
4. HYDROCONVERSION OF FISCHER-TROPSCH WAX 52
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
4. HYDROCONVERSION OF FISCHER-TROPSCH WAX 53
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).
4. HYDROCONVERSION OF FISCHER-TROPSCH WAX 54
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.
4. HYDROCONVERSION OF FISCHER-TROPSCH WAX 55
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.
4. HYDROCONVERSION OF FISCHER-TROPSCH WAX 56
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.
4. HYDROCONVERSION OF FISCHER-TROPSCH WAX 57
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.
4. HYDROCONVERSION OF FISCHER-TROPSCH WAX 58
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.
4. HYDROCONVERSION OF FISCHER-TROPSCH WAX 59
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.
4. HYDROCONVERSION OF FISCHER-TROPSCH WAX 60
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).
4. HYDROCONVERSION OF FISCHER-TROPSCH WAX 61
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.
4. HYDROCONVERSION OF FISCHER-TROPSCH WAX 62
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].
4. HYDROCONVERSION OF FISCHER-TROPSCH WAX 63
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
4. HYDROCONVERSION OF FISCHER-TROPSCH WAX 64
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
4. HYDROCONVERSION OF FISCHER-TROPSCH WAX 65
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.
4. HYDROCONVERSION OF FISCHER-TROPSCH WAX 66
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.
4. HYDROCONVERSION OF FISCHER-TROPSCH WAX 67
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.
4. HYDROCONVERSION OF FISCHER-TROPSCH WAX 68
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.
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.
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
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.
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.