The effect of H2 S on oxidation properties of ZrO2

Department of Biotechnology and Chemical Technology

SCIENCE + TECHNOLOGY CROSSOVER DOCTORAL DISSERTATIONS

Aalto University

ART + DESIGN + ARCHITECTURE

The effect of H2S on oxidation properties of ZrO2 -based biomass gasification gas clean-up catalysts

Aalto University School of Chemical Technology Department of Biotechnology and Chemical Technology www.aalto.fi

BUSINESS + ECONOMY

Inkeri Kauppi

Aalto-DD 214/2015

9HSTFMG*agfida+

ISBN 978-952-60-6583-0 (printed) ISBN 978-952-60-6584-7 (pdf) ISSN-L 1799-4934 ISSN 1799-4934 (printed) ISSN 1799-4942 (pdf)

2015

The effect of H2S on oxidation properties of ZrO2-based biomass gasification gas clean-up catalysts Inkeri Kauppi

DOCTORAL DISSERTATIONS

Aalto University publication series DOCTORAL DISSERTATIONS 214/2015


A doctoral dissertation completed for the degree of Doctor of Science (Technology) to be defended, with the permission of the Aalto University School of Chemical Technology, at a public examination held at the lecture hall Ke2 of the school on 15th of January 2016.

Aalto University School of Chemical Technology Department of Biotechnology and Chemical Technology Industrial Chemistry

Supervising professor Prof. Juha Lehtonen Thesis advisor Dr. Sc. Jaana Kanervo Preliminary examiners Prof. Dr. Hab. Maria Ziolek Adam Mickiewicz University, Poland Prof. Dr. JH (Harry) Bitter Wageningen University, The Netherlands Opponents Prof. Lars J. Pettersson Kungliga tekniska högskolan (KTH, Royal Institute of Technology), Sweden

Aalto University publication series DOCTORAL DISSERTATIONS 214/2015 © Eeva Inkeri Kauppi ISBN 978-952-60-6583-0 (printed) ISBN 978-952-60-6584-7 (pdf) ISSN-L 1799-4934 ISSN 1799-4934 (printed) ISSN 1799-4942 (pdf) http://urn.fi/URN:ISBN:978-952-60-6584-7 Unigrafia Oy Helsinki 2015 Finland

Abstract Aalto University, P.O. Box 11000, FI-00076 Aalto www.aalto.fi

Author Eeva Inkeri Kauppi Name of the doctoral dissertation The effect of H2S on oxidation properties of ZrO2 -based biomass gasification gas clean-up catalysts Publisher School of Chemical Technology Unit Department of Biotechnology and Chemical Technology Series Aalto University publication series DOCTORAL DISSERTATIONS 214/2015 Field of research Industrial Chemistry Manuscript submitted 14 August 2015

Date of the defence 15 January 2016 Permission to publish granted (date) 28 October 2015 Language English Monograph

Article dissertation (summary + original articles)

Abstract Biomass gasification gas contains impurities which have to be removed before the gas can be utilized in e.g. energy or liquid fuels production. ZrO2-based catalysts can be used to oxidize tar impurities when oxygen is added to the gas. These catalysts have proven activity even when H2S is present. Moreover, an improving effect of H2S on ZrO2 oxidation activity has been observed at temperatures of 600 and 700 °C. Therefore, the reactivities of unsulfided and sulfided ZrO 2-based catalysts were studied in order to understand the enhancing effect of sulfur on oxidation activity during gasification gas clean-up. Different adsorption modes of H2S on the ZrO2-based catalysts were established. Molecular adsorption occurred on all the studied catalysts at room temperature. The molecularly adsorbed H2S species are weakly bound and thus are not likely to be present on the surface at the high temperatures of gasification gas clean-up. Therefore, they cannot be the source of the enhanced reactivity, either. Dissociative adsorption of H2S was suggested on cation-anion pairs or by titration of terminal hydroxyl groups on ZrO2 and doped ZrO2. However, the terminal SH groups formed in these surface processes cannot contribute to the observed oxidation activity improvement on ZrO2, as revealed by density functional theory (DFT) calculations. Stable sulfur species formed on the surface of ZrO2 during adsorption of H2S at elevated temperatures (at 100 °C and above). H2S reacts with the surface probably via replacement of surface lattice oxygen at specific defect sites. The amount of sulfur deposited on the surface in these high temperature processes was found to correlate with enhanced oxidation activity compared to unsulfided ZrO 2. This indicates that H2S at elevated temperatures produces surface sulfur which improves the redox properties of ZrO2. Based on DFT calculations, sulfur species in the lattice (multicoordinated SH or S) at specific sites can cause enhanced reactivity of lattice oxygen, but so far the exact structure of this site remains unknown. The interaction of H2S with ZrO2 is limited to specific surface sites, comprising not more than approximately 11 % of the surface. Studies on toluene oxidation (a model compound for tar) also indicated that tar oxidation occurs on specific sites where the intermediate species form. Based on observations, it is proposed that the oxidation improvement by sulfur occurs on these specific sites where sulfur improves the reactivity of surface lattice oxygen. The sulfurtolerance of ZrO2 is thus originated by the limited number of sites capable of binding sulfur.

Keywords zirconia, hydrogen sulfide, tar oxidation, gasification gas clean-up ISBN (printed) 978-952-60-6583-0 ISBN (pdf) 978-952-60-6584-7 ISSN-L 1799-4934 Location of publisher Helsinki Pages 57

ISSN (printed) 1799-4934 ISSN (pdf) 1799-4942 Location of printing Helsinki Year 2015 urn http://urn.fi/URN:ISBN:978-952-60-6584-7

Tiivistelmä Aalto-yliopisto, PL 11000, 00076 Aalto www.aalto.fi

Tekijä Eeva Inkeri Kauppi Väitöskirjan nimi Rikkivedyn vaikutus biomassan kaasutuskaasun puhdistuksessa käytettävien ZrO2 katalyyttien hapetusominaisuuksiin Julkaisija Kemian tekniikan korkeakoulu Yksikkö Biotekniikan ja kemian tekniikan laitos Sarja Aalto University publication series DOCTORAL DISSERTATIONS 214/2015 Tutkimusala Teknillinen kemia Käsikirjoituksen pvm 14.08.2015 Julkaisuluvan myöntämispäivä 28.10.2015 Monografia

Väitöspäivä 15.01.2016 Kieli Englanti

Yhdistelmäväitöskirja (yhteenveto-osa + erillisartikkelit)

Tiivistelmä Biomassan kaasutuksessa syntyvän tuotekaasun sisältämät epäpuhtaudet täytyy poistaa ennen kuin kaasua voidaan käyttää esimerkiksi energiantuotannossa tai nestemäisten polttoaineiden valmistukseen. Epäpuhtautena olevat tervat voidaan hapettaa ZrO2katalyyteillä hapen läsnä ollessa, jopa kaasun sisältäessä rikkivetyä (H2S). Aikaisemmin on todettu, että H2S:llä on ZrO2:n hapetusaktiivisuutta parantava vaikutus lämpötiloissa 600 ja 700 °C. Tämän vuoksi tässä työssä tutkittiin rikittämättömien ja rikitettyjen ZrO2-katalyyttien reaktiivisuutta. Tavoitteena oli löytää syyt H2S:n hapetusaktiivisuutta parantavaan vaikutukseen biomassan kaasutuskaasun puhdistuksessa. H2S adsorboitui molekuläärisesti kaikille katalyyteille huoneen lämpötilassa. Molekuläärisesti adsorboituneet muodot ovat sitoutuneet heikosti eivätkä ole läsnä katalyytillä edellä mainituissa korkeissa lämpötiloissa. Siksi molekuläärisesti adsorboitunut H2S ei ole syy havaittuun hapetusaktiivisuuden paranemiseen. Dissosiatiivinen H2S:n adsorptio tapahtui joko ZrO2:n kationi-anioni pareille tai titraten terminaalisia OH-ryhmiä muodostaen terminaalisia SH-ryhmiä. Terminaaliset SH-ryhmätkään eivät vaikuttaneet ZrO2-pinnan hapen reaktiivisuuteen, mikä havaittiin laskennallisesti. Yli 100 °C:ssa muodostui H2S:n adsorboituessa ZrO2:n pinnalle vahvasti sitoutuneita rikin muotoja, jotka ovat luultavimmin sitoutuneet hilaan syrjäyttäen pinnan happea erityispaikoilla. Tällä tavalla adsorboituneen rikin määrä korreloi lisääntyneen reaktiivisen pintahapen määrään verrattuna puhtaan ZrO2:n pinnan hapen reaktiivisuuteen viitaten siihen, että pinnan hapetus-pelkistys ominaisuudet paranivat. Laskennallisten tulosten perusteella ZrO2:n hilaan liittyneet rikin muodot (multikoordinoitunut SH tai S) voivat parantaa hilahapen reaktiivisuutta erityispaikoilla, kuitenkin näiden paikkojen tarkka luonne on vielä epäselvä. H2S:n vuorovaikutus ZrO2:n kanssa on rajoittunut tietyille erityispaikoille, jotka peittävät enintään 11 % pinnasta. Tutkimukset tervan malliaineena käytetyn tolueenin adsorptiosta ja hapetuksesta osoittivat myös, että tervojen hapetus tapahtuu erityispaikoilla, joille muodostuu reaktiivisia hapetuksen välituotteita. Oletettavasti rikin vaikutuksesta tapahtuva hapetusaktiivisuuden paraneminen tapahtuu näillä erityispaikoilla, joilla rikin läsnäolo parantaa hapetusreaktioon osallistuvan pintahapen reaktiivisuutta. ZrO2-katalyyttien hyvä rikin sietokyky perustuu osin rikkiyhdisteiden rajalliseen reaktiivisuuteen pinnan kanssa.

Avainsanat zirkonia, rikkivety, tervojen hapetus, kaasutuskaasun puhdistus ISBN (painettu) 978-952-60-6583-0 ISBN (pdf) 978-952-60-6584-7 ISSN-L 1799-4934 Julkaisupaikka Helsinki

ISSN (painettu) 1799-4934 Painopaikka Helsinki

ISSN (pdf) 1799-4942 Vuosi 2015

Sivumäärä 57

urn http://urn.fi/URN:ISBN:978-952-60-6584-7

Acknowledgements

The work for this thesis was conducted at Aalto University School of Chemical Technology, former Helsinki University of Technology, within the Research Group Industrial Chemistry. The work was made possible financially by Graduate School of Chemical Engineering (GSCE) funded by Academy of Finland. I would also like to thank Finnish Foundation for Technology Promotion and Alfred Kordelin foundation for financial support. I am indebted to my supervisors, Professors Outi Krause and Juha Lehtonen, for their support and for giving me the possibility to carry out this thesis work. I also want to thank the people who acted as my supervisors, Dr. Sanna Airaksinen, Dr. Juha Linnekoski, and Dr. Jaana Kanervo for their guidance and help. I specifically want to thank Dr. Jaana Kanervo for insightful discussions and support during the work, as well as Professor Outi Krause for her guidance. I am most grateful for Professor Leon Lefferts’ contribution to the progress of the thesis and the related publications, as well as for many hours of discussions that pushed the work forward. I thank Dr. Ella Rönkkönen for support and inspiration and most of all for her kind friendship throughout the years. I would also like to sincerely thank my co-author Ms. Tiia Viinikainen for her energy and positivity and help with research. Moreover, I would like to express my gratitude to my co-authors Dr. Sanna Airaksinen, Dr. Satu Korhonen, and Dr. Karoliina Honkala for their contributions. I thank Dr. Maija Honkela for her help. My warm thanks also to the laboratory personnel in the Research Group Industrial Chemistry for support and creating a relaxed working atmosphere. During the thesis work I have had the opportunity to pursue research abroad and I am grateful to my home university for making that possible. During 6 months in 2010 the work for this thesis was carried out in the Instituto de Catálisis y Petroleoquimica, CSIC, Madrid, Spain. I warmly thank Professor Dr. Miguel Bañares for his support, guidance and teaching. As co-authors I thank him and Dr. Søren Rasmussen for co-operation and discussions. Professor Dr. Pedro Ávila at CSIC is also thanked for co-operation. I sincerely thank Dr. Ricardo López Medina for his kind and professional help during experiments as well as everyone working in the Catalytic Spectroscopy Laboratory for their help and support, and most of all their friendship. The FTIR experiments were performed during summer 2013 in the University of Caen, in the Catalysis and Spectrochemistry Laboratory (LCS) under supervision of Dr. Laetitia Oliviero, whose expert professional guidance is gratefully acknowledged. Director of Research, Dr. Francoise Maugé and Professor Dr. Marco Daturi are also thanked 1

for facilitating the visit to LCS. Moreover, I thank Dr. Jianjun Chen for assistance in experimental work. I thank Ms. Nathalie Perrier for help in practicalities in Caen. I sincerely acknowledge both laboratories for providing me a space to carry out the research, and for the warm atmosphere which made me feel welcome. Finally, I kindly thank all my friends and family members for all the love and joy they bring into my life. Thank you for support and your patience throughout the years.

Helsinki, 26 November 2015 Eeva Inkeri Kauppi

2

Contents

Acknowledgements...................................................................................1 List of abbreviations and symbols........................................................... 5 List of publications .................................................................................. 7 Author’s contribution .............................................................................. 8 1.

Introduction.................................................................................. 9 1.1

Biomass gasification and gas clean-up..................................... 9

1.2

Oxidation of tars over ZrO 2-based catalysts ........................... 10

1.3

Properties and surface sites of ZrO 2 ........................................11

1.4

Effect of H2S on catalysis .........................................................12

1.5

Scope ........................................................................................13

2.

Experimental ...............................................................................15 2.1

Temperature-programmed techniques ...................................15

2.1.1 Temperature-programmed sulfidation....................................15 2.1.2

Temperature-programmed desorption ................................16

2.1.3

Temperature-programmed surface reaction of methanol ...16

2.1.4

Temperature-programmed reduction with H2 and CO........16

2.1.5

Temperature-programmed experiments with toluene ........16

2.2

Infrared spectroscopy of solid catalysts...................................17

2.2.1

Characterization of the surface sites by MeOH-DRIFTS.....17

2.2.2

Toluene adsorption and oxidation under DRIFTS...............17

2.2.3

Transmission IR experiments on sulfided ZrO 2...................18

2.3 3.

Computational work ................................................................19 Results and discussion.................................................................21

3.1

Surface properties of ZrO 2-based catalysts and effect of H2S..21

3.1.1 Methanol adsorption on ZrO 2-based catalysts ........................21 3.1.2

H2S adsorption modes and catalysts’ reactivity with MeOH22

3.1.3

MeOH surface species on ZrO 2 after H2S adsorption ......... 26

3.2 3.2.1

The effect of H2S on reactivity of surface oxygen on ZrO 2...... 29 Interaction of H2S with ZrO 2 ............................................... 29 3

3.2.2

Stability of adsorbed H2S .....................................................30

3.2.3

Reduction of unsulfided and sulfided ZrO 2 by CO............... 31

3.3 Redox properties of ZrO 2 and the effect of H2S studied on a molecular level .................................................................................................34 3.3.1 Calculations performed to study the improving effect of adsorbed H2S on reactivity of surface lattice oxygen.........................................34 3.3.2

ZrO 2 as a redox catalyst and defect sites acting in relevant catalysis ..............................................................................................36

3.4 Toluene adsorption and oxidation studies on ZrO 2-based biomass gasification gas clean-up catalysts......................................................37

4.

3.4.1

Surface species formed upon toluene adsorption and oxidation ..............................................................................................37

3.4.2

Gas-phase products from toluene adsorption and oxidation38

3.4.3

Possible toluene oxidation mechanism and sites involved on ZrO 2 ..............................................................................................41

Conclusions .................................................................................43

References..............................................................................................47

4

List of abbreviations and symbols

a.u.

arbitrary unit

DRIFTS

Diffuse Reflectance Infrared Fourier Transform Spectrum

FT

Fischer-Tropsch synthesis

FTIR

Fourier Transform Infrared

CO-TPR

Temperature-programmed reduction with CO

CPOM

catalytic partial oxidation of methane

c.u.s.

coordinatively unsaturated site

DFT

density functional theory

DME

dimethyl ether

DMS

dimethyl sulfide

H2-TPR

Temperature-programmed reduction with H2

TPA

Temperature-programmed adsorption

TPD

Temperature-programmed desorption

TPO

Temperature-programmed oxidation

TPS

Temperature-programmed sulfidation

t-OH

terminal hydroxyl group

m-OH

multicoordinated hydroxyl group

t-SH

terminal sulfide group

m-SH

multicoordinated sulfide groups

MeOH-TPSRTemperature-programmed surface reaction with methanol ML

monolayer (coverage)

MS

mass spectrometer/spectrometry

SNG

synthetic natural gas

į

bending mode (in spectroscopy) 5

ǌ as

asymmetric stretching mode (in spectroscopy)

ǌs

symmetric stretching mode (in spectroscopy)

WGS

water-gas-shift

6

List of publications

This doctoral dissertation consists of a summary and of the following publications which are referred to in the text by their numerals

I. Kauppi, Eeva Inkeri; Rönkkönen, Ella Hanne; Airaksinen Sanna; Rasmussen, Søren; Bañares, Miguel; Krause, Outi. 2012. Influence of H2S on ZrO 2based gasification gas clean-up catalysts: MeOH temperature-programmed reaction study. Elsevier. Applied Catalysis B: Environmental, 111 - 112, pages 605 - 613. 0926-3373. 10.1016/j.apcatb.2011.11.013. II. Kauppi, Eeva Inkeri; Kanervo, Jaana; Lehtonen, Juha; Lefferts, Leon. 2015. Interaction of H2S with ZrO 2 and its influence on reactivity of surface oxygen. Elsevier. Applied Catalysis B: Environmental, 164, pages 360 – 370. 10.1016/j.apcatb.2014.09.042. III. Kauppi, Eeva Inkeri; Honkala, Karoliina; Krause, Outi; Kanervo, Jaana; Lefferts, Leon. 2015. ZrO 2 acting as a redox catalyst. Accepted for publication in Topics in Catalysis 09/2015. IV. Viinikainen, Tiia; Kauppi, Eeva Inkeri; Korhonen, Satu; Lefferts, Leon; Kanervo, Jaana; Lehtonen, Juha. Molecular level insights to the interaction of toluene with ZrO2-based biomass gasification gas clean-up catalysts. Elsevier. Applied Catalysis B: Environmental, 142 – 143, pages 769 – 779. 0926-3373. 10.1016/j.apcatb.2013.06.008.

7

Author’s contribution

Publication I: Influence of H2S on ZrO 2-based gasification gas clean-up catalysts: MeOH temperature-programmed reaction study The author planned the research with the help of the co-authors, carried out the experiments except the MeOH-DRIFTS experiments, and interpreted the results and wrote the manuscript with the help of the co-authors. Publication II: Interaction of H2S with ZrO 2 and its influence on reactivity of surface oxygen The author planned, carried out the experiments and interpreted the results together with the co-authors and wrote the first as well as the final version of the manuscript. Publication III: ZrO 2 acting as a redox catalyst This review article consists of material that was made available to the author, including unpublished DFT results provided by Karoliina Honkala, and partly the author’s own previous work. The author contributed considerably to the selection of the data and wrote the first version of the manuscript as well as participated in finalizing the text and conclusions. Publication IV: Molecular level insights to the interaction of toluene with ZrO 2-based biomass gasification gas clean-up catalysts The author planned the experiments together with the other main author, carried out temperature-programmed experiments, carried out partly the DRIFTS-experiments, interpreted results and wrote the manuscript together with the other main author. This work has two main authors.

8

1. Introduction

1.1

Biomass gasification and gas clean-up

Among the thermochemical processes (combustion, pyrolysis, gasification) which convert biomass into value-added products, gasification has the highest efficiency. Gasification converts biomass through partial oxidation into synthesis gas which is a mixture of mainly CO and H2, but CO 2, H2O and CH4 and other light hydrocarbons are also present (Torres et al., 2007). The purified product can be targeted at liquid fuel production by Fischer-Tropsch (FT) synthesis, methanol synthesis, dimethyl ether production, synthetic natural gas (SNG) or H2 production (Kurkela et al., 2008). The gas can also be fed to gas turbines or fuel cells to generate electricity. The quality of the producer gas has to be adjusted to meet the end-use specifications. Generally, heavier purification measures have to be taken for synthesis purposes. The impurity tars, i.e. polyaromatic hydrocarbons having a molecular mass greater than benzene, present the main challenge for the utilization of the produced gas due to their tendency to plug downstream processing equipment. Therefore, their content has to be reduced. Nitrogen and sulfur compounds are also present in small quantities, depending on the fuel used (Juutilainen et al., 2006). The H2/CO-ratio of the gas is determined by the gasifying agent used (air, oxygen or steam) and is adjusted by the water-gas-shift reaction (WGS, CO + H2O ļ CO 2 +H2). A schematic representation of a biomass gasification process is shown in Figure 1.

Figure 1. Biomass gasification process (adapted from Kurkela et al., 2008).

An effective gas-cleaning step is key for a feasible biomass gasification process and must be added if the formation of tars cannot be avoided in the gasifier. Particulate dust and tar removal technologies can be divided into two categories, mainly referred to as primary and secondary methods (Göransson et al., 2011, Wang et al., 2008). Primary methods comprise treatments during gasification and secondary methods gas-cleaning after gasification, such as hot-gas cleanup. Studies and reviews are widely available on different hot-gas clean-up tech-

9

Introduction

nologies, including mechanical methods, tar reforming, and catalytic conversion of tars to less harmful compounds. The choice of technology depends on many factors, from the quality of the raw materials to the end-use of the gas. Catalytic hot-gas cleaning with a catalyst-coated monolith is a competitive way to convert tars into less harmful compounds (Torres et al., 2007, Simell et al., 1996). A gas clean-up catalyst must be resistant to high temperatures and sulfur tolerant, and still have high selectivity in tar oxidation, so that oxidations of CO and H2 are avoided. The studied catalysts are, for example, metal catalysts such as nickel, precious metal catalysts such as Rh and Pt, alkali metals, and dolomites (Sutton et al., 2001, Rönkkönen et al., 2011, Simell et al., 1996). ZrO 2 catalysts are active and selective in tar and ammonia oxidation at 600-900 °C when oxygen is added to the gas (Juutilainen et al., 2006), even when H2S is present (Rönkkönen et al., 2009). Oxidation over a ZrO 2-based catalyst is often used together with tar reforming over metal catalysts.

1.2

Oxidation of tars over ZrO2-based catalysts

ZrO 2 is a versatile material with high thermal stability, which has proven activity towards tar oxidation when oxygen is present (Juutilainen et al., 2006, Simell and Kurkela, 2004, Rönkkönen, 2014). It can be used as a catalyst or a catalyst support. The activities of ZrO 2 and doped ZrO 2 catalysts (Y 2O 3-ZrO2, SiO 2-ZrO 2 and La2O3-ZrO2) have been investigated previously, and their surface properties were related to their tar oxidation activity during gas clean-up (Rönkkönen, 2014, Viinikainen et al., 2009). The main tar decomposition reaction over ZrO 2 catalysts has been proposed (Juutilainen et al., 2006) to be a two-step oxidation, presented for naphthalene (a model compound for tar) in Equations 1 and 2 (Juutilainen et al., 2006, Rönkkönen et al., 2009b): C H +7O 10 CO+ 4 H O CO CO + O

(1) (2)

Since gasification gas is a complex mixture of gases, the reaction network contains several competitive and consecutive reactions (Rönkkönen et al., 2009b). The individual gas components affect tar decomposition activity, e.g. water has an inhibiting effect on the conversion of naphthalene (Rönkkönen et al., 2009b). H2S, which will always be present when real gasification gas feeds are processed, has a specific effect on the performance of ZrO 2 catalysts which will be discussed later. In the gas clean-up application, the oxygen is typically nearly completely consumed towards the end of the monolith, which results in changes in the reaction conditions along the monolith. At the inlet, where oxygen is still available, tar is oxidized to CO and CO 2. On the other hand, near the outlet of the monolith, steam and dry reforming reactions also contribute, as do hydrocracking and carbon formation (Rönkkönen et al., 2009b, Viinikainen et al., 2009). Understand-

10

Introduction

ing of e.g. WGS, reforming, and oxidation catalysis in general is required to obtain a complete picture of ZrO 2 performance during gas clean-up. Moreover, studies using individual impurity components are needed in order to determine the tar decomposition mechanism (relevant surface species formed and their reactions) and the sites involved.

1.3

Properties and surface sites of ZrO2

The oxidizing-reducing properties of ZrO 2 are established (Nakano et al., 1979), however, the oxidation ability of pure ZrO 2 is limited. It is a stable oxide which has been claimed to lose surface oxygen only after thermal treatment under a high vacuum at above 700 °C (Daturi et al., 1998). Nevertheless, ZrO2 can be used as an oxidation catalyst in some high-temperature applications (Zhu, 2005). ZrO 2 also has proven activity in e.g. reforming (Kaila, 2008), partial oxidation of methane (Zhu, 2005), and WGS (Graf et al., 2009, Kouva et al., 2014). It has been proposed that tar oxidation on ZrO 2-based catalysts during gasification gas clean-up proceeds via the Mars-van Krevelen mechanism (Rönkkönen, 2014). According to this mechanism, organic molecules are oxidized in redox cycles by surface lattice oxygen (Mars and van Krevelen, 1954). Oxidation via reaction with surface lattice oxygen creates vacancies, which are replenished by gas-phase oxygen (Mars and van Krevelen, 1954, Busca et al., 1996a). Surface lattice oxygen is present on ZrO 2 as coordinatively unsaturated (c.u.s.) oxygen and multicoordinated hydroxyl groups (m-OH), which exhibit different reactivity depending on the site/coordination. Surfaces of crystals are created when they are cut along a plane, generating atoms with different chemical environments. The surfaces are either flat or more open with steps and kinks, i.e. the minority sites. (Niemantsverdriet, 2007). Hydroxyl groups are generated on ZrO 2 upon adsorption of water and are present as terminal OH (t-OH) or m-OH. It is possible that OH groups also participate in redox cycles, however, their role in oxidation is not completely clear. The properties of ZrO 2 can be changed by adding dopants, such as Y 2O3 or CeO 2, to improve e.g. oxygen mobility on the surface and induce vacancy generation (Zhu et al., 2005a, Dutta et al., 2006). Improved ability to generate vacancies assists oxidation reactions on the oxide catalyst. Dopants also affect the surface area, which is low ion ZrO2. In general, the surface area of ZrO2 is increased by dopant addition. A change in the crystal structure from monoclinic to cubic or tetragonal or a mixture also occurs when adding dopants (Viinikainen et al., 2009). It has been observed that doping with La 2O 3-ZrO2 generates a catalyst with the highest activity towards tar decomposition in gasification gas clean-up application when H2S is present in the gas (Rönkkönen, 2014).

11

Introduction

1.4

Effect of H2S on catalysis

Sulfur compounds, most importantly H2S and COS, are present in the producer gas. In fact, the amount of H2S may even be as high as 500 ppm, depending on the feed stock used (Torres et al., 2007). H2S is known to poison especially metal catalysts even at extremely low gas-phase concentrations. Sulfur binds strongly on metal surfaces and the effects are often irreversible, leaving no chance of regeneration (Bartholomew et al., 1982). Thus, sulfur tolerance of the catalyst is essential during gasification gas clean-up. Interestingly, on ZrO2based catalysts, enhancement of naphthalene and ammonia oxidation in the presence of H2S has been noted during gas-cleaning experiments on synthetic biomass gasification gas. The effect was not observed on SiO2-ZrO2 (Rönkkönen et al., 2009a). Beneficial effects of H2S on catalyst activity have been reported at least by (Jackson et al., 1991, Ziolek et al., 1995, Erdöhelyi et al., 2004, Laosiripojana et al., 2010, Roushanafshar et al., 2012, Sugioka et al., 1989, Vincent et al., 2011, Stenberg et al., 1982) on various catalysts, including mainly metal oxides. The proposed reasons for the observed effects vary and may be linked with phenomena occurring on the surface of the catalyst or in the gas phase. For example, H2S has been reported to act as a gaseous promotor in the water-gas shift reaction, where H2S offers a pathway for CO to react via the formation of COS further to the products (Stenberg et al., 1982). On the other hand, adsorbed sulfur may cause surface modifications which change catalyst activity or selectivity. Different mechanisms have been suggested for the promoting effect of adsorbed H2S on the catalyst. For example, the reaction between CO 2 and CH4 over Rh/TiO 2 and Rh/SiO 2 catalysts was reported to be enhanced by selective poisoning of active sites (Erdöhelyi et al., 2004). Jackson et al. (Jackson et al., 1991) suggested that sulfur may modify the catalyst surface, changing the reaction mechanism. Sulfur-containing surface complexes, such as sulfate (SOx ) species, have known effects in promoting the acidity of the catalyst and enhancing mainly isomerization and condensation reactions (Sohn and Kim, 1989, Vera et al., 2002). Moreover, even weak interaction (physical adsorption) with H2S and the catalyst surface has been found to affect catalyst properties (Maugé et al., 2002). It is not commonly known what kinds of interactions between H2S and ZrO2 benefit oxidation reactions. The study by Rönkkönen et al. (Rönkkönen et al., 2009a) suggested that a strongly adsorbed form of H2S affects the performance of ZrO 2-based catalysts, since the effect was also seen after H2S was removed from the stream. This may also indicate a surface reaction of H2S with the catalyst. In general, H2S can interact with ZrO2 via three pathways: i) molecular adsorption on Zr4+ cations or interacting with hydroxyl groups, ii) dissociative adsorption on Zr4+-O 2- anion-cation pairs or replacing a t-OH group, or iii) exchange of sulfur with lattice oxygen (Ziolek et al. 1995, Travert et al., 2002).

12

Introduction

1.5

Scope

The main aim of this work was to clarify the reasons for the observed enhancement effect of H2S on tar oxidation activity on ZrO 2-based gasification gas cleanup catalysts. With regard to this aim, the target was to elucidate the effect of H2S on catalysts’ surface properties and generally the oxidizing/reducing properties on three ZrO 2-based catalysts. Studies using methanol as a probe also looked into why H2S improves the activities of ZrO 2 and Y 2O 3-ZrO2, whereas no effect can be found on SiO 2-ZrO 2. Since tar oxidation has been suggested to proceed via reaction with surface lattice oxygen, it was thought that adsorbed H2S improves the reactivity of surface lattice oxygen or induces the generation of new sites able to participate in redox cycles. Therefore, sulfur interactions with ZrO 2 surface were thoroughly examined. Temperature-programmed (TP) methods were utilized to obtain information on H2S adsorption pathways and the stability of adsorbed H2S on ZrO 2. The aim of the TP studies was primarily to obtain knowledge about how adsorbed H2S changes the reactivity of ZrO 2 towards oxidation reactions. With regard to this, a density functional theory (DFT) study was also conducted to find out the most stable adsorbed H2S species on the ZrO 2 surface and to determine how they modify the reactivity of surface oxygen on specific sites. Gaining more insightful knowledge about the effect of H2S requires a better understanding of the tar oxidation mechanism on ZrO 2. Studies on toluene (used as a tar model compound) adsorption and oxidation were targeted to give an estimation of the tar decomposition mechanism and site. Moreover, studies were aimed at discovering the relevant surface species formed from toluene. The role of defect sites and OH chemistry is fundamental in redox catalysis on ZrO2. Relevant to this work, the redox behavior of ZrO 2 and the sites involved were thus briefly reviewed, which served the aim of understanding the behavior of ZrO 2 in gasification gas clean-up application and the effect of H2S on oxidation catalysis.

13

2. Experimental

Full experimental details are given in Publications I-IV. Brief descriptions of the experimental techniques used are given here.

2.1

Temperature-programmed techniques

Temperature-programmed (TP) methods are techniques monitoring a chemical reaction while the temperature of the solid catalyst sample is increased linearly in time (Niemantsverdriet, 2007). A schematic presentation of the experimental setup used in the studies is given in Figure 2. The reactor, which is usually a Utube, is placed in a controllable oven and packed with the catalyst. The feed gases are specified by the experimental procedure used.

Figure 2. A schematic presentation of the TP technique used in the studies (adapted from Niemantsverdriet, 2007).

Relevant to this thesis, the specific qualitative TP techniques described in the following sections were used. A mass spectrometer was always applied for the analysis of gas-phase products. 2.1.1

T emperature-programmed sulfidation

Temperature-programmed sulfidation (TPS), where the sample is heated under an H2S atmosphere, is often used to describe the activation treatment in which oxidic catalysts that are active in their sulfided state are transformed into metal sulfides (e.g. alumina-supported molybdenum)(Niemantsverdriet, 2007). In this work (Publication II), the technique was used to study transformations occurring in the ZrO 2 structure under mild sulfiding conditions. It is known that ZrO 2 does not transform into ZrS2 (Clearfield, 1958); however, H2S adsorption occurs. Therefore, even though the bulk sulfide does not form, the term sulfided 15

Experimental

ZrO 2 is used in this work when referring to ZrO 2 treated with H2S. In the study (Publication II), H2S (in N 2, 500 ppm) was adsorbed on ZrO 2 isothermally at 30 °C (1 h) and via TPS with four different end temperatures of 100, 200, 300 and 400 °C, aiming to affect the amount (and nature) of the H2S adsorbed species on the surface. 2.1.2

T emperature-programmed desorption

Temperature-programmed desorption (TPD) provides information on the binding states of adsorbed elements (intermediates and products) on the catalyst surface. As the sample with adsorbate is heated, the energy transferred to the adsorbed species will cause it to desorb, giving an observable peak maximum (Tmax ) of the desorption temperature. In the publications regarding this thesis, TPD was used to study the thermal stability of adsorbed species and to probe the adsorption modes of H2S and toluene on the studied catalysts. TPD was performed after isothermal H2S adsorption at 40 °C and after H2S adsorption at 30-400 °C (respectively in Publications I and II) to study the interaction of H2S with the catalyst surface and the stability of adsorbed species. Toluene-TPD was performed on ZrO 2, Y 2O3-ZrO2, and SiO2ZrO 2 to study the decomposition of toluene adsorbed species (Publication IV). Accurate initial temperatures as well as the ramp rates are given in the relevant publications. 2.1.3

T emperature-programmed surface reaction of m ethanol

Methanol is a well-established surface probe (Badlani and Wachs, 2001, Tatibouët, 1997). The selectivity pattern of gas-phase products is used to characterize catalyst properties: dimethyl ether (DME) for acid sites, formaldehyde for redox, and CO 2 for basic sites (Badlani and Wachs, 2001, Tatibouët, 1997). The temperature-programmed surface reaction of methanol (MeOH-TPSR) from 100 to 700 °C under Ar flow was done for the unsulfided and sulfided (at 100 °C, 30 vol-% H2S in H2) ZrO 2, Y 2O3-ZrO2, and SiO 2-ZrO2 samples to study the effect of H2S addition on catalyst properties. H2S was adsorbed isothermally at 100 °C on the samples. The study is presented in Publication I. 2.1.4

T emperature-programmed reduction with H2 and CO

Temperature-programmed reductions with H2 and CO (H2- and CO-TPR) were applied in studies on sulfided ZrO 2 in order to reveal how sulfur modifies the reducibility of the ZrO 2 surface (Publication II). Both experiments were started at 30 °C and carried out until an end temperature of 600 °C with a ramp rate of 10 °C/min. H2 (4 vol-%) and (CO 5 vol-%) were fed during the ramp and at 600 °C for o.5 h. 2.1.5

T emperature-programmed ex periments with toluene

TP gas-phase analysis was applied to study the gas-phase products during adsorption and oxidation of toluene adsorbed species in detail (Publication IV). 16

Experimental

Toluene temperature-programmed adsorption (TPA) and temperature-programmed oxidation (TPO) from 30 to 600 °C (10 °C/min) was done for ZrO2, Y 2O 3-ZrO2, and SiO 2-ZrO 2 in the same test cycle. In this way, the adsorption could be monitored as well as the oxidation of all the deposited surface species during the ramp up to 600 °C.

2.2

Infrared spectroscopy of solid catalysts

In studies relevant to this thesis, in situ diffuse reflectance infrared Fourier transform spectroscopy (DRIFTS) combined with mass spectrometry was used (Publications I and IV). Sulfided ZrO 2 was also characterized in transmission mode. As a technique, DRIFTS is only qualitative, whereas the transmission technique can also yield the quantification of surface species. However, sample preparation for DRIFTS is easier than for transmission-type experiments (Armaroli et al., 2004). Catalyst surface sites can be characterized by in situ DRIFTS using specific probe molecules. Furthermore, the technique allows the formation of surface species (may be reaction intermediates) under relevant reaction conditions to be monitored. However, the most reactive species are usually either detected in very small amounts or not detected at all, and thus spectators rather than intermediates are detected under reaction conditions (Busca, 1996b). The DRIFTS equipment consisted of a Nicolet Nexus FTIR spectrometer and a Spectra-Tech high-temperature reaction chamber with ZnSe windows. Gasphase products were monitored using an on-line mass spectrometer (Omnistar, Pfeiffer Vacuum). The spectrum of an aluminum mirror measured under nitrogen flow (50 cm 3/min) was used as the background. 2.2.1

Characterization of the surface sites by MeOH-DRIFTS

When using MeOH as a probe, the catalyst surface can be characterized based on characteristic vibrations of the adsorbed methoxy type species on the catalyst surface derived from MeOH which show up in the IR spectra. The technique was applied when characterizing the surface of ZrO 2 and the differently doped oxides Y 2O 3-ZrO2 and SiO 2-ZrO2 at 100-500 °C (Publication I). 2.2.2

T oluene adsorption and ox idation under DRIFTS

Toluene adsorption and transformations with temperature as well as with oxygen were measured on ZrO 2, Y 2O 3-ZrO2, and SiO 2-ZrO 2 in four different kinds of DRIFTS experiments. TPD, TPA, TPO and TPSR-type experiments were conducted with the aim to find the relevant surface species formed from toluene in the absence and presence of gas-phase oxygen, and the results were linked with the results from the corresponding TP experiments. The details can be found in Publication IV.

17

Experimental

2.2.3

T ransmission IR ex periments on sulfided ZrO2

The IR setup (Figure 3) used in this work has three parts: the evacuation system, IR cell, and spectrometer. The IR cell pressure can be evacuated down to 10-3 kPa. A controlled amount of probe molecules can then be directed to the cell. The spectrometer is a Fourier Transform IR spectrometer (Nexus) from Nicolet with a MCT (Mercury Cadmium Telluride) detector. The spectra were collected using 256 scans with a resolution of 4 cm -1 .

Figure 3. Scheme of FTIR setup.

The IR cell used was a low-temperature glass IR cell (Figure 4). In this cell, the samples can be treated at high temperature (atmospheric pressure) with e.g. flowing H2S. A small volume (2.23 cm 3) of gas can be introduced in calibrated doses into the IR cell. The outlet at the upper part and the inlet at the basis of the cell allow the pellet to be under a gas flow.

18

Experimental

Figure 4. Low-temperature IR cell with flow system.

The powder ZrO 2 sample was pressed (3 tons) into the form of a pellet (area of 2 cm 2, mass in the range of 35-45 mg). Afterwards, the activation of the sample was carried out in situ in the IR cell. The pellet was activated under approx. 4 kPa of O 2 from room temperature (RT) to approx. 500 °C (heating 10 °C/min). The high temperature was maintained for two hours. Evacuation was performed for 1 h and continued while the sample was cooled down to room temperature. H2S was adsorbed on ZrO 2 at temperatures RT -> 400 °C -> RT. A mixture containing 500 ppm H2S in nitrogen was used.

2.3

Computational work

Adsorption of H2S was studied on a molecular level by applying density functional theory (DFT) calculations. Calculations were performed on edge and corner sites, which were modelled as representative minority sites on monoclinic ZrO 2. Computational details can be found in Publication III.

19

3. Results and discussion

3.1 3.1.1

Surface properties of ZrO2-based catalysts and effect of H2S Methanol adsorption on ZrO2-based catalysts

Surface species formed during methanol dissociative adsorption and decomposition as a function of temperature on the ZrO 2, Y 2O 3-ZrO2, and SiO 2-ZrO2 were studied with in situ DRIFTS in Publication I. Methanol dissociation on the ZrO 2 surface may occur by 1) adsorption on surface hydroxyl groups with formation of methoxy groups and water, or 2) on c.u.s. Zr4+-O 2- pairs to form methoxy-and hydroxyl groups (Fisher and Bell, 1999). Figure 5 shows spectra measured for the ZrO 2 catalyst during the MeOHDRIFTS experiment until 500 °C. The presented spectra were measured under nitrogen flow to minimize the effect of gas-phase methanol (the peak positions were similar under methanol). The corresponding spectra are not shown for the doped oxides, for brevity, but can be seen in Publication I.

Figure 5. In situ DRIFT spectra on pure ZrO2 during temperature-programmed experiment with methanol (Publication I).

Characterization with MeOH showed differences between the three studied catalysts, which were mainly seen in the transformations of methoxy groups with temperature. Hydroxyl groups characteristic to ZrO 2 and the mixed oxides 21

Results and discussion

(at 3800-3600 cm -1 ) were titrated during MeOH adsorption at 100 °C and restored with temperature, which can be seen on pure ZrO 2 in Figure 5. This suggests that dissociative adsorption took place, where hydroxyl groups were substituted by methoxy groups from methanol. In addition, hydrogen-bonded hydroxyls appeared during adsorption (broad band centered at ca. 3400 cm –1 ). Adsorption of methanol resulted in the appearance of peaks at 2920 and 2815 cm -1 (Fig. 5), which are attributed to methoxy species on ZrO 2 (Zr–OCH3) (Korhonen et al., 2007). The peak at 1165 cm -1 is also due to methoxy species, PRUHVSHFLILFDOO\WRǌ2&+3) of on-top methoxy species (Korhonen et al., 2007). Similar formation of methoxies was observed on Y2O 3-ZrO2. Two types of methoxies were seen on the SiO 2-ZrO 2 sample at 2927, 2830, 1443, and 1150 cm–1, (Zr–OCH3) and at 2953, 2850, and 1461 cm –1 (Si–OCH3) (Fisher and Bell, 1999). The new peaks appearing above 250 °C at 1573, 1380 and 1370 cm –1 (Fig. 5) DUHGXHWRǌ as&22 į&+ DQGǌ s(COO) vibrations of formates (Korhonen et al., 2007)7KHIRUPDWHǌ&H) peak at 2870 cm–1 appeared at 275 °C. Methoxies reacted to formates in a similar manner on ZrO 2 and Y 2O 3-ZrO2. On the SiO 2ZrO 2 formates appeared at above 250 °C, but the peak intensities were lower than on the other samples. MoreoverQRIRUPDWHǌ&+ VWUHWFKZDVGHWHFWHGDW approximately 2880-2870 cm –1 . This indicates that the formation of formates was not as significant on SiO 2-ZrO 2 as on ZrO 2 and Y 2O 3-ZrO2. It also seems that the methoxy species on Si-O phase do not evolve to form formates. Formates are formed on the surface via reaction with surface lattice oxygen (Busca et al., 1987). Based on methanol reactivity to formate on the different catalysts, it is suggested that the number of reacting sites of surface lattice oxygen is increased on ZrO 2 and Y 2O 3-ZrO2 compared to that on SiO 2-ZrO 2. Moreover, the oxygen contained on the SiO 2-ZrO2 phase does not seem to be reactive and it is suggested that oxygen vacancies cannot form on the silica phase. This may be due to the strong covalent nature of oxygen bonded to silicon atoms, which was noted in a study by del Monte et al. (del Monte et al., 2000). MeOH-DRIFTS characterization suggested different kind of distribution of dopants on the mixed oxides. Two phases exist on SiO 2-ZrO 2, whereas Y 2O3 seems more homogeneously distributed in the ZrO 2 matrix. 3.1.2

H2S adsorption modes and catalysts’ reactivity with MeOH

H2S retention on ZrO 2 and doped ZrO 2 H2S-TPD was conductedfrom 40 to 750 °C on ZrO 2, Y 2O 3-ZrO2, and SiO 2-ZrO2 (Publication I) to probe the adsorption modes of H2S with catalyst surfaces. Figure 6 presents desorbed H2S from the studied catalysts after adsorption at 40 °C. The high-temperature region is not shown, since no desorption of H2S was detected after ~300 °C.

22


Figure 6. H2S-TPD results (m/z=34) of ZrO2, Y2O3-ZrO2, and SiO2-ZrO2 catalysts. Heating rate 10 °C/min (Publication I).

H2S-TPD profiles (Figure 6) show one maximum for ZrO 2 and Y 2O 3-ZrO2 (at 100 and 115 °C, respectively), whereas the SiO 2-ZrO2 clearly shows two maxima (at 115 and 210 °C). This suggests that additional H2S adsorption sites exist on SiO 2-ZrO 2 compared to ZrO 2 and Y 2O 3-ZrO2. Furthermore, the intensity of the low-temperature peak decreased in the order Y 2O 3-ZrO2 > ZrO 2 > SiO 2-ZrO2. The latter peak on the SiO 2-ZrO 2 may be connected to H2S adsorbed on an interphase (between the SiO 2 and ZrO 2 phases on the catalyst), since the MeOHDRIFTS characterization revealed separate phases on this catalyst. Adsorption on the SiO 2 phase is unlikely, since H2S-TPD tests on pure SiO 2 showed that it does not adsorb H2S. It is realized that the observed phenomena occur on the surface and bulk sulfides do not form; however, the term sulfided ZrO 2 catalyst will be used for brevity. MeOH product distribution on the calcined and sulfided ZrO 2 catalysts The changes in the MeOH-TPSR product distribution on ZrO 2, Y 2O 3-ZrO2, and SiO 2-ZrO 2 were compared in order to determine the effect of H2S adsorption on catalyst acidic and basic properties, i.e. c.u.s. cationic Zr4+ as well as anionic O2sites, and different kinds of hydroxyl groups possessing Brønstedt acidity or basicity. Figure 7 shows the DME generation on the calcined and sulfided samples.

23


Figure 7. DME generation (MS signal, m/z= 45) in the MeOH-TPSR experiment on calcined and sulfided ZrO2, Y2O3-ZrO2, and SiO2-ZrO2 (Publication I).

24


Figure 7 shows that the desorption temperature of DME is lower for ZrO 2 than for the doped oxides (350 °C on ZrO 2, 480 and 370 °C on Y 2O 3-ZrO2 and SiO2ZrO 2 respectively). Moreover, the amount of DME generated was higher on the doped oxides than on the pure ZrO 2. DME is generated by the condensation of two methoxy groups on the catalyst surface (Bianchi et al., 1995), and selectivity to DME generally describes the pure dehydration ability of the catalyst, i.e. its Lewis acidity (Tatibouët, 1997). Therefore, the result suggests the following order in Lewis acidity: ZrO 2 Y 2O3ZrO 2 > ZrO 2. TPA gas-phase experiments from 30 to 600 °C revealed the formation of gasphase products during adsorption of toluene. The formation of benzene, CO2, H2 and H2O is presented in Figure 17.

39


Figure 17. MS responses for benzene (m/z= 78), CO2 (m/z=44), H2 (m/z=2) and H2O (m/z=18) during toluene-TPA experiments (Publication IV).

It can be seen that benzene and CO 2 were formed concurrently at high temperature (at 525 °C on ZrO 2 and Y 2O 3-ZrO2, and at 530 on SiO 2-ZrO 2, Figure 17). The molar amounts for CO 2 and benzene at high temperature were equivalent by estimation (approx. 0.1 ʅmol/m 2 for Y 2O 3-ZrO2). It is suggested that these products were formed into the gas phase from surface benzoate species detected during toluene-TPA DRIFTS experiments, which decomposes to benzene and CO 2 in the gas phase and reduces the surface. The fraction of surface oxygen taking part in this process is ~0.8 %, suggesting minority sites. 40


TPO experiments carried out after TPA showed mainly the formation of CO2, CO, H2O, H2 and consumption of O 2. The oxidation of carbonaceous surface species took place between 200 and 400 °C. The amount of carbonaceous deposits formed during adsorption at high temperatures was notable, as suggested by the intense evolution of CO 2 and CO. Benzoate species were suggested to be very stable with oxygen, based on the combined analysis of TPA under DRIFTS and gas-phase experiments. Therefore, it can be interpreted that benzoate is not the active intermediate in the toluene oxidation reaction, but is rather a spectator species. The TPO analyses also revealed that the nature of the carbonaceous deposits was different on the pure and doped ZrO 2. The deposits were more heterogeneous on the doped oxides. 3.4.3

Possible toluene oxidation mechanism and sites involved on ZrO2

Surface benzoate species could decompose to yield CO 2 and benzene, as suggested by their appearance in DRIFTS at temperatures below the temperature at which CO 2 and benzene were detected in the gas phase during the TP experiments. The amount of oxygen atoms in the CO 2 formed supposedly from surface benzoate corresponds to approx. 0.8 % of a ML of surface oxygen. So, a minor fraction of the surface is likely to participate, indicating that minority sites like edges and corners are involved (similar to the findings with other redox reactions and H2S adsorption in Publication II). The high stability of surface benzoates indicates that they are more likely spectator species rather than reaction intermediates in toluene oxidation. When toluene was fed together with oxygen to the catalyst and a temperature ramp was applied, surface benzyl species appeared simultaneously when the conversion of toluene and oxygen started. Therefore, the benzyl species was concluded to be the active intermediate in toluene oxidation reaction over ZrO2 and the doped ZrO 2 catalysts. Benzyl species form by the abstraction of one hydrogen atom from the methyl group of toluene and adsorb on the surface, possibly via the methylene (–CH2) group. The suggested different pathways for toluene interacting with ZrO 2-based gasification gas clean-up catalysts are shown in Figure 18. The mechanism is not conclusive, especially on how the benzene ring is oxidized; however, new knowledge was gained about the adsorption mechanisms of toluene and possible intermediate species during toluene oxidation on ZrO 2.

41


Figure 18. Adsorption and oxidation of toluene over ZrO2-based gasification gas clean-up catalysts; a schematic presentation (Publication IV).

Defect sites are responsible for the adsorption and oxidation of tar compounds, as suggested by the calculated values of adsorbed toluene (molecular) and gasphase products (fractions of a ML). This is in line with studies by Zhu et al. (Zhu et al., 2005a), who revealed that oxidation reactions occur at defect sites on ZrO2, as reviewed earlier. The participation of OH groups during oxidation is possible and could also play a role in toluene decomposition. The studies reviewing WGS on ZrO2 prove that surface OH groups at defect sites are active in converting CO to CO 2 and hydrogen. It is plausible that they could also be involved in oxidizing intermediates from adsorbed tar molecules. The effect of sulfur on toluene adsorption and oxidation can be estimated based on the studies conducted. If surface oxygen is exchanged for surface sulfur on the minority sites, where species derived from toluene supposedly also adsorb, the formation of benzoate could be hindered, since it requires the presence of two adjacent surface oxygen atoms. On the other hand, it is more difficult to estimate the effect of surface sulfur species on the formation and reactions of benzyl species. It is still under question as to how the tar molecules become oxidized by the surface lattice oxygen (the breaking of the benzene ring and further oxidation). It seems that the improvement mechanism with sulfur is related to these steps rather than the adsorption step, since the studies indicated that the initial adsorption of toluene on surface oxygen ions results in strongly adsorbed species not participating in oxidation. Moreover, the c.u.s. O 2- sites are more electronegative than the suggested S2- and therefore it is not expected that tar molecules would interact more favorably with sulfur-incorporated sites. It is suggested that the adsorbed form of sulfur increases the reactivity of surface oxygen at specific sites, which results in enhanced reaction rates of adsorbed intermediates from tar molecules with surface lattice oxygen.

42

4. Conclusions

The effect of H2S on ZrO 2-based catalysts was investigated in tar oxidation as a part of biomass gasification gas clean-up. The positive effect of H2S on tar oxidation activity had been observed earlier on the ZrO 2 and Y 2O 3-ZrO2 catalysts; however, no effect was observed on SiO 2-ZrO 2. The reactivities of unsulfided and sulfided ZrO 2-based catalysts were examined to understand the improvement in oxidation activity by sulfur during gasification gas clean-up. Moreover, H2S adsorption modes on ZrO 2 were thoroughly examined. Molecular and dissociative adsorption modes of H2S were established on ZrO2 and doped ZrO 2 based on temperature-programmed studies. On ZrO 2, the dissociative adsorption was suggested to yield t-SH species on the surface, eliminate t-OH species and possibly form new m-OH sites. The dissociatively adsorbed species were found to be strongly adsorbed even at low temperature (30 °C), whereas molecularly adsorbed H2S was weakly held. H2S was found to react with ZrO 2 to form H2O at elevated temperatures, producing activatedly adsorbed sulfur species. Those species are likely incorporated in the surface via the replacement of surface lattice oxygen (m-OH and c.u.s. O2-) at specific sites, i.e. minority sites such as edges, steps, and corners. They were also concluded to be strongly held. The amount of sulfur deposited on the surface in the hightemperature processes was found to correlate with the amount of surface oxygen removed, which was revealed by CO-TPR on differently sulfided ZrO 2. This indicates that at elevated temperatures H2S produces surface sulfur which improves the oxidation properties of ZrO 2 on minority sites. The reactivity of surface lattice oxygen was thus concluded to be enhanced by adsorbed H2S. In addition, CO 2 adsorption sites were blocked on the sulfided surface. Studies employing DFT calculations ruled out molecularly adsorbed H2S and t-SH species formed by dissociative adsorption as possible species to enhance the reactivity of surface oxygen on ZrO 2. The molecular adsorption of H2S was confirmed to be weak by DFT calculations, as already suggested by TP experiments. In fact, the molecularly adsorbed species are not even likely present on the surface at the high temperatures of gasification gas clean-up. The effect of adsorbed H2S on the reactivity of surface oxygen was tested using edge and corner models as representative minority sites. An increase in the reactivity of mOH and t-OH was found when m-SH was present on the corner model. Based on the calculations, it is concluded that the reactivity of surface lattice oxygen or t-OH is enhanced at very specific sites when m-SH is present, i.e. sulfur is 43

Conclusions

incorporated into the lattice on a minority site. The exact nature of the site is probably different from the model used, since quantitatively the calculation results do not resolve the observations. Stable sulfur surface species are likewise formed on doped oxides, since adsorbed H2S was not reversed by adsorbing methanol at 100 °C as indicated by MeOH-TPSR studies. Moreover, those stable species were likely incorporated in the lattice based on dimethyl sulfide generation from methanol on the sulfided catalysts at high temperature (500 °C). The reactivity of the deposited sulfur with methanol differed between ZrO2, Y 2O3- ZrO2, and SiO 2-ZrO2, and it was concluded that the reactivity of sulfur is reduced on the SiO 2-ZrO 2. Differences in the surface properties of the catalysts were also identified. Doping with SiO2 was concluded to produce a catalyst with two distinct phases on ZrO 2, with SiO2 forming islands on the ZrO 2 phase. On the other hand, doping with Y 2O 3 produced a well-distributed matrix. The reactivity of surface lattice oxygen was suggested to be lower on the SiO 2-ZrO 2 catalyst compared to the ZrO 2 and Y 2O3ZrO 2 catalysts. Vacancy formation is possibly more limited and cannot be enhanced by sulfur because of the covalent bonding to Si-atoms. SiO 2-ZrO 2 was also concluded to be the most acidic catalyst. H2S adsorbed on the Lewis acidic sites, i.e. the Zr4+ cationic sites, on all the catalysts. Thus, dissociative adsorption of H2S on Zr4+-O 2- pairs was suggested to also occur on the doped oxides. On ZrO 2, CO 2 was produced from methanol more favorably on the sulfided catalyst, whereas the effect was opposite on SiO 2-ZrO 2. No clear effect was found on Y 2O 3-ZrO2. Based on the TP studies, the adsorption of H2S occurs on similar sites on Y 2O 3-ZrO2 as on pure ZrO 2, whereas there are additional sites on SiO 2ZrO 2 to adsorb H2S. Therefore, the activity improvement effect during gasification gas clean-up is suggested to be originated by sulfur also interacting with the lattice on Y 2O 3-ZrO2. The oxidation properties of SiO 2-ZrO 2 cannot be improved by sulfur, probably because of the covalent nature of the bonds formed. The interaction of H2S with ZrO 2 is concluded to be limited to specific sites, comprising not more than approximately 11 % of a ML, based on the amount of adsorbed H2S. Strong adsorption found on a very limited amount of surface sites suggests that those surface sites must possess a significant difference in their nature. It seems that H2S does not reduce the catalyst but forms sulfur species on a limited amount of sites. The limited reactivity could be an advantage regarding applications with H2S content in streams, since sulfur poisoning does not occur at least with the relatively low concentrations considered here. The sulfur tolerance of ZrO 2 is originated by the low degree of sulfidation. It is concluded that ZrO 2-based catalysts are the optimal choice for applications with some H2S content, offering possibilities for the simplification of related processes when H2S does not necessarily have to be removed from the stream. The surface of ZrO 2 exhibits limited reducibility; however, part of the surface oxygen is able to participate in the redox cycles. Redox reactions occur at defect sites and hydroxyl groups could also be involved in oxidations. Therefore, hydroxyl groups probably also play a role during gasification gas clean-up, where water is always present. The reactivities of both the m-OH and t-OH groups

44

Conclusions

were suggested to be enhanced by sulfur present on the surface (as m-SH species) and hence the observed apparent enhancement in tar oxidation activity could also be connected to the increased reactivity of OH groups on ZrO 2. Tar decomposition could involve reactions where hydrocarbons react with hydroxyl groups on the surface generating CO and H2 to the gas phase. However, studies on the oxidation of tar molecules (transient and/or steady-state) in the presence of H2S but without the presence of other gases in gasification gas mixtures are needed. Toluene was used as a model compound for tar when studying its interaction with ZrO 2, Y 2O 3-ZrO2, and SiO 2-ZrO 2. The studies indicated that the active intermediate in the toluene oxidation is not benzoate, the formation of which requires surface oxygen, but the benzyl species which are formed by abstraction of the hydrogen atom from the methyl group of toluene. The molecular adsorption of toluene was found to be weak. Toluene adsorption and oxidation sites were concluded to be minority sites on ZrO 2, similarly to H2S adsorption sites. This is also in line with the observations on the redox ability of ZrO 2, which has been resolved to be limited to minority sites. It is unclear how the stable bonds in the benzene ring of tar molecules are broken when the molecule is oxidized on the surface. However, it is claimed that the reaction intermediates formed during tar oxidation are oxidized to CO or CO 2 via a reaction with the surface lattice oxygen in a Mars-van Krevelen type reaction. Interaction with surface oxygen atoms during initial adsorption of toluene on ZrO 2 seems not to be the preferred pathway in toluene oxidation, since the formed species are strongly adsorbed. The oxidation activity improvement mechanism with sulfur is therefore suggested to be related to the oxidation of the intermediate species, rather than the adsorption of tar molecules. This work contributes to understanding how adsorbed H2S modifies the oxidative properties of ZrO 2-based catalysts towards enhanced performance in tar oxidation during gasification gas clean-up. It is concluded that the improving effect of H2S must be originated by strongly adsorbed sulfur species present in the lattice at minority sites, where the intermediate species from tar molecules also form. Oxidation improvement by sulfur occurs on these specific sites where sulfur improves the reactivity of surface lattice oxygen.

45

References

(Azzam et al., 2007 ) Azzam, K. G., Babich, I. V ., Seshan, K., Lefferts, L., Bifunctional catalysts for single-stage water-gas shift reaction in fuel cell applications. Part 1. Effect of the support on reaction sequence J. Catal., 251 (2007) 1 53 – 1 62. (Anderson et al., 1 997 ) Anderson, J. R., Kleinke, H, Franzen, H. F., ZrO2-x Sx : Stabilization of cubic ZrO2 by partial exchange of oxygen by sulfur, J. Alloys Compd., 259 (1 997 ) L1 4 – L1 8. (Armaroli et al., 2004) Armaroli, T., Bécue, T., Gautier, S., Diffuse reflection infrared spectroscopy (DRIFTS): Application to the in situ analy sis of catalysts, Oil Gas Sci Technol, 59 (2004) 215 – 237 . (Badlani and Wachs, 2001 ) Badlani, M., Wachs, I. E., Methanol: a “smart” chemical probe molecule, Catal. Lett. 75 (2001) 1 37 – 1 49. (Bampenrat et al., 2008) Bampenrat, A., Meeyoo, V ., Kitiy anan, B., Rangsunv igit, P., Rirksomboon, T., Catalytic oxidation of naphthalene over CeO2-ZrO2 mixed oxide catalysts, Catal. Commun., 9 (2008) 2349 – 2352. (Bartholomew et al., 1 982) Bartholomew, C. H., Agrawal, P. K., Katzer, J. R., Sulfur poisoning of metals, Adv Catal., 31 (1 982) 1 35 – 242. (Busca et al., 1 987 ) Busca, G., Elmi, A. S., Forzatti, P., Mechanism of selective methanol ox idation over v anadium oxide –titanium oxide catalysts: A FT-IR and flow reactor study , J. Phys. Chem., 91 (1 987 ) 5263 – 5269. (Busca et al., 1 996a) Busca, G., Finocchio, E., Ramis, G., Ricchiardi, G., On the role of acidity in catalytic oxidation, Catal. Today, 32 (1996) 1 33 – 1 43. (Busca, 1 996b) Busca, G., Infrared studies of the reactive adsorption of organic molecules over metal oxides and of the mechanisms of their heterogeneously catalyzed ox idation, Catal. Today, 27 (1996) 457 – 496. (Clearfield, 1 958) A. Clearfield, The synthesis and properties of zirconium disulfide, J. Am. Chem. Soc., 80 (1 958) 6511 – 651 3. (Daturi et al., 1 998) Daturi, M., Binet, C., Bernal, S., Pérez Omil, J. A., Lav alley, J. C., FTIR study of defects produced in ZrO2 samples by thermal treatment, J. Chem. Soc. Faraday Trans., 94(8) (1 998) 1 143 – 1 1 47 . (del Monte et al., 2000) del Monte, F., Larsen, W., Mackenzie, J. D., Stabilization of tetragonal ZrO2 in ZrO2-SiO2 binary oxides, J. Am. Ceram. Soc., 83(3) (2000) 628 – 634. (Dutta et al., 2006) Dutta, G., Waghmare, U. V ., Baidy a, T., Hegde, M. S., Priolkar, K. R., Sarode, P. R., Reducibility of Ce 1 -x Zrx O2: origin of enhanced oxygen storage capacity, Catal. Lett., 108 (2006) 1 65 – 1 7 2. (Erdöhely i et al., 2004) Erdöhelyi, A., Fodor, K., Szailer, T., Effect of H2S on the reaction of methane with carbon dioxide over supported Rh catalysts, Appl. Catal., B, 53 (2004) 1 53 – 1 60. (Finocchio et al., 1 999) Finocchio, E., Daturi, M., Binet, C., Lav alley, J. C., Blanchard, G., Thermal ev olution of the adsorbed methoxy species on Ce x Zr 1-x O2 solid solution samples: a FT-IR study , Catal. Today, 52 (1 999) 53 – 63. (Fisher and Bell, 1 999) Fisher, I. A., Bell, A. T., A mechanistic study of methanol decomposition over Cu/SiO2, ZrO2/SiO2, and Cu/ZrO2/SiO2, J. Catal., 184 (1 999) 357 – 37 6.

47

References

(Graf et al., 2009) Graf, P. O., de V lieger, D. J. M., Mojet, B. L., Lefferts, New insights in reactivity of hy droxyl groups in water gas shift reaction on Pt/ZrO2, L., J. Catal., 262 (2009) 1 81 – 1 87 . (Göransson et al., 2011) Göransson, K., Söderlind, U., He, J., Zhang, W., Rev iew of syngas production v ia biomass DFBGs, Renew Sust Energ Rev, 16 (201 1) 482 – 492. (Hertl, 1 989) Hertl, W., Surface chemistry of zirconia polymorphs, Langmuir, 5 (1989) 96 – 1 00. (Jackson et al., 1 991) Jackson, S. D., Brandreth, B. J., Winstanley , D., The effect of preadsorbed sulfur on carbon monoxide hy drogenation ov er rhodium/silica, J. Catal., 129 (1 991 ) 540 – 543. (Juutilainen et al., 2006) Juutilainen, S., Simell, P., Krause, A. O. I., Zirconia: Selective ox idation catalyst for removal of tar and ammonia from biomass gasification gas, Appl. Catal., B, 62 (2006) 86 – 92. (Kaila, 2009) Kaila, R., Autothermal reforming of simulated and commercial fuels on zirconia supported mono- and bimetallic catalysts, Doctoral Thesis, Helsinki Univ ersity of Technology, Espoo 2008. (Kondo et al., 1 988) Kondo, J., Abe, H., Sakata, Y ., Maruya, K., Domen, K., Onishi, T., Infrared studies of adsorbed species of H2, CO and CO2 ov er ZrO2, J. Chem. Soc., Faraday Trans, 84(2) (1 988) 51 1 – 51 9. (Korhonen et al., 2007 ) Korhonen, S. T., Bañares, M. A., Fierro, J. L. G., Krause, A. O. I., Adsorption of methanol as a probe for surface characteristics of zirconia,- alumina,- and zirconia/alumina-supported chromia catalysts, Catal. Today, 126 (2007 ) 235 – 247 . (Kouv a et al., 201 4) S. Kouv a, J. Andersin, K. Honkala, J. Lehtonen, L. Lefferts, J. Kanerv o, Water and carbon ox ides on monoclinic zirconia: Ex perimental and computational insights, Phys. Chem. Chem. Phys., 16 (38) (201 4) 20650 – 20664. (Kurkela et al., 2008) Kurkela, E., Simell., P., McKeough, P., Kurkela, M., V TT publications 682, ”Sy nteesikaasun ja puhtaan polttokaasun v almistus”, Espoo 2008. (Laosiripojana et al., 201 0) Laosiripojana, N., Charojrochkul, S., Kim-Lohsoontorn, P., Assabumrungrat, Role and advantages of H2S in catalytic steam reforming over nanoscale CeO2-based catalysts, J. Catal., 276 (2010) 6 – 1 5. (Mars and v an Krev elen, 1 954) Mars, P., v an Krevelen, D. W., Ox idations carried out by means of v anadium oxide catalysts, Chem. Eng. Sci., (Spec. Suppl.), 3 (1954) 41 – 59. (Maugé et al., 2002) Maugé, F., Sahibed-Dine, A., Gaillard, M., Ziolek, M., Modification of the acidic properties of NaY zeolite by H2S adsorption – An infrared study, J. Catal., 207 (2002) 353 – 360. (Nakano et al., 1 97 9) Nakano, Y ., Iizuka, T., Hattori, H., Tanabe, K., Surface properties of zirconium oxide and its catalytic activity for isomerization of 1-butene, J. Catal., 57 (1 97 9) 1 – 1 0. (Niemantsv erdriet, 2007) Niemantsverdriet, J. W., Spectroscopy in Catalysis: An Introduction, Third Edition, Wiley -V CH V erlag GmbH & Co. KGaA, Weinheim, 2007 , pp. 1 1 – 38. (Roine, 2007 ) Roine, A., HSC-Chemistry for Windows 6.12, Outotec Research Oy, 2007. (Roushanafshar et al., 2012) Roushanafshar, M., Luo, J.-L., Vincent, A. L., Chuang, K. T., Sanger, A. R., Effect of hy drogen sulfide inclusion in sy ngas feed on the electrocatalytic activity of LST-Y DC composite anodes for high temperature SOFC applications, Int. J. Hydrogen Energy, 37 (2012) 7 762 – 7 7 70.

48

References

(Rönkkönen et al., 2009a) Rönkkönen, H., Simell, P. Reinikainen, M. Krause, O., The effect of sulfur on ZrO2-based gasification gas clean-up catalysts, Top. Catal., 52 (2009) 1 070 - 1 078. (Rönkkönen et al., 2009b) Rönkkönen, H., Rikkinen, E., Linnekoski, J., Simell, P., Reinikainen, M., Krause, O., Effect of gasification gas components on naphthalene decomposition over ZrO2, Catal. Today, 147 S (2009) S230 – S236. (Rönkkönen et al., 2011) Rönkkönen, H., Simell, P., Niemelä, M., Krause, O., Precious metal catalysts in the clean-up of biomass gasification gas part 2: Performance and sulfur tolerance of rhodium based catalyst, Fuel Process. Technol., 92 (2011) 1 881 – 1 889. (Rönkkönen, 2014) Rönkkönen, E., H., Cataly tic clean-up of biomass derived gasification gas with zirconia based catalysts, Doctoral Thesis, Aalto Univ ersity, 2014. (Simell et al., 1 996) Simell, P. Kurkela, E. Ståhlberg, P., Hepola, J., Cataly tic hot gas cleaning of gasification gas, Catal. Today, 27 (1996) 55-62. (Simell and Kurkela, 2004) EP 1 404785B1, 2004. Method for purification of gasification gas, V altion teknillinen tutkimuskeskus, Espoo, Finland, (Simell, P., Kurkela, E.). (Sohn and Kim, 1 989) Sohn, J. R., Kim, H. W., Cataly tic and surface properties of ZrO2 modified with sulfur compounds, J. Mol. Catal., 52 (1 989) 361 – 37 4. (Sutton et al., 2001 ) Sutton, D., Kelleher, B., Ross, J. R. H., Rev iew of literature on cataly sts for biomass gasification, Fuel Process. Technol., 73 (2001) 1 55 – 1 7 3. (Tatibouët, 1 997) Tatibouët, J. M., Methanol oxidation as a catalytic surface probe, Appl. Catal., A, 148 (1 997 ) 21 3 – 252. (Torres et al., 2007 ) Torres, W., Pansare, S. S., Goodwin, J. G., Hot gas remov al of tars, ammonia, and hy drogen sulfide from biomass gasification gas, Catal. Rev. Sci. Eng., 49 (2007 ) 407 – 456. (Trav ert et al., 2002) Travert, A., Manoilova, O. V ., Tsy ganenko, A. A., Maugé, F., Lav alley, J. C., Effect of hy drogen sulfide and methanethiol adsorption on acidic properties of metal oxides: An infrared study, J. Phys. Chem. B, 106 (2002) 1350 – 1 362. (V era et al., 2002) V era, C. R., Pieck, C. L., Shimizu, K., Parera, J. M., Tetragonal structure, anionic v acancies and catalytic activity of SO42--ZrO2 catalysts for n-butane isomerization, Appl. Catal., A, 230 (2002) 1 37 – 1 51. (V iinikainen et al., 2009) V iinikainen, T., Rönkkönen, H., Bradshaw, H., Stephenson, H., Airaksinen, S., Reinikainen, M., Simell, P., Krause, O., Acidic and basic surface sites of zirconia-based biomass gasification gas clean-up catalysts, Appl. Catal., A, 362 (2009) 1 69 – 1 7 7. (V incent et al., 2011) V incent, A. L., Luo, J.-L., Chuang, K. T., Sanger, A. R., Promotion of activ ation of CH4 by H2S in ox idation of sour gas ov er sulfur tolerant SOFC anode catalysts, Appl. Catal., B, 106 (201 1) 114 – 1 22. (Wang and Wachs, 2004) Wang, X., Wachs, I. E., Designing the activ ity/selectivity of surface acidic, basic, and redox active sites in the supported K2O-V 2O5/Al2O3 cataly tic system, Catal. Today, 96 (2004) 211 – 222. (Wang et al., 2008) Wang, L., Weller, C. L., Jones, D. D., Hanna, M. A., Contemporary issues in thermal gasification of biomass and its application to electricity and fuel production, Biomass Bioenerg, 32 (2008) 57 3 – 581 . (Zhu, 2005) J. Zhu, Catalytic partial oxidation of methane to synthesis gas over ZrO2based defective oxides, Doctoral Dissertation, Univ ersity of Twente, The Netherlands, 2005. (Zhu et al., 2005a) Zhu, J., v an Ommen, J. G., Bouwmeester, H. J. M., Lefferts, L., Activ ation of O2 and CH4 on y ttrium-stabilized zirconia for the partial ox idation of methane to sy nthesis gas, J. Catal., 233 (2005) 434 – 441 .

49

References

(Zhu et al., 2005c) Zhu, J., Albertsma, S., v an Ommen, J. G., Lefferts, L., Role of surface defects in activation of O2 and N2O on ZrO2 and y ttrium-stabilized ZrO2, J. Phys. Chem. B, 109 (2005) 9550 – 9555. (Ziolek et al., 1 994) Ziolek, M., Saur, O., Lamotte, J., Lav alley, J.-L., FTIR study of adsorption and transformation of methanethiol and dimethyl sulfide on zirconia, J. Chem. Soc. Faraday Trans., 90(7) (1 994) 1 029 – 1 032. (Ziolek et al., 1 995) Ziolek, M., Kujawa, J., Saur, O., Lav alley, J. C., Influence of hydrogen sulfide adsorption on the catalytic properties of metal oxides, J. Mol. Catal. A, 97 (1 995) 49 – 55.

50

I Kauppi, E. I., Rönkkönen, E. H., Airaksinen S. M. K., Rasmussen, S. B., Bañares, M. A., Krause, A. O. I., (2012) Influence of H2S on ZrO2-based gasification gas clean-up catalysts: MeOH temperature-programmed reaction study. Appl. Catal. B 111-112, 605613. Reproduced with permission from Elsevier B.V.

Applied Catalysis B: Environmental 111–112 (2012) 605–613

Contents lists available at SciVerse ScienceDirect

Applied Catalysis B: Environmental journal homepage: www.elsevier.com/locate/apcatb

Influence of H2 S on ZrO2 -based gasification gas clean-up catalysts: MeOH temperature-programmed reaction study E. Inkeri Kauppi a,∗ , E. Hanne Rönkkönen a , Sanna M.K. Airaksinen a , Søren B. Rasmussen b , b ˜ Miguel A. Banares , A. Outi I. Krause a a Aalto University, School of Chemical Technology, Department of Biotechnology and Chemical Technology, Research Group Industrial Chemistry, P.O. Box 16100, FI-00076 Aalto, Finland b Instituto de Catálisis y Petroleoquímica, ICP-CSIC, Calle Marie Curie 2, Cantoblanco E-28049, Madrid, Spain

a r t i c l e

i n f o

Article history: Received 26 August 2011 Received in revised form 3 November 2011 Accepted 6 November 2011 Available online 15 November 2011 Keywords: Zirconia catalyst Effect of H2 S MeOH adsorption MeOH-DRIFTS

a b s t r a c t Addition of H2 S in the gasification gas stream has been found to improve naphthalene and ammonia conversion over ZrO2 and Y2 O3 -ZrO2 catalysts, whereas over SiO2 -ZrO2 such effect has not been observed. The differences in the properties of the catalysts were studied by spectroscopic (MeOH-DRIFTS) and temperature-programmed methods (H2 S-TPD and MeOH-TPSR). Methanol was also used to probe the changes after sulfidation of the catalysts. H2 S adsorption sites were found to be different on the ZrO2 and Y2 O3 -ZrO2 catalysts compared with the ones on the SiO2 -ZrO2 catalyst. Sulfur was also found to be more reactive on the ZrO2 and Y2 O3 -ZrO2 catalysts than on SiO2 -ZrO2 . It was suggested, that the positive effect of H2 S on naphthalene conversion during gasification gas clean-up is connected with the reactions of adsorbed sulfur on their surface. In contrast, on SiO2 -ZrO2 the lack of such an effect may be related to limited reactivity of sulfur on the surface. © 2011 Elsevier B.V. All rights reserved.

1. Introduction Gasification is the first processing step required in synthesis gas production from biomass, which provides a prominent alternative route to environmentally friendly liquid fuels. The produced H2 rich gas from e.g. wood residues, straw or peat has to be cleaned from impurities such as tar and ammonia to avoid plugging in the downstream equipment [1,2]. Zirconia-based catalysts are known to be active in gasification gas clean-up processes in the presence of oxygen [3]. Since the biomass gasification gas contains considerable amounts of H2 S, even up to 500 ppm, sulfur tolerance of the catalyst is crucial [4]. The effect of H2 S on the gasification gas clean-up activities of ZrO2 , Y2 O3 -ZrO2 , and SiO2 -ZrO2 catalysts at 600–900 ◦ C was investigated by Rönkkönen et al. [4]. At 600 and 700 ◦ C addition of H2 S enhanced the activity of Y2 O3 -ZrO2 and ZrO2 , whereas mainly sulfur poisoning was observed for SiO2 -ZrO2 . The intensity of the enhancement effect was related to the catalyst’s Lewis basicity, and the amount of reactive oxygen on the catalyst. It was suggested that sulfur adsorption contributes to the generation of a new type of active sites, possibly also affecting the surface properties of the catalysts [4].

∗ Corresponding author. Tel.: +358 9 451 22666; fax: +358 9 470 22622. E-mail address: inkeri.kauppi@aalto.fi (E.I. Kauppi). 0926-3373/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.apcatb.2011.11.013

Generally, sulfur has been found to affect the behavior of catalysts in several applications. Addition of H2 S enhances the activity of ZrO2 in isopropanol decomposition, since when H2 S is adsorbed on the ZrO2 surface, oxygen is replaced by sulfur ions causing an increase in the basicity and/or redox properties of ZrO2 [5]. In an IR spectroscopic study performed by Travert et al. [6], adsorption of H2 S was shown to induce strong modifications in the surface acidic properties of SiO2 , Al2 O3 , TiO2 and ZrO2 . The concentration of Lewis acid sites, i.e. coordinatively unsaturated (c.u.s.) metal cationic Mn+ sites, on these catalysts was found to decrease by titration with adsorbed sulfur species. Simultaneously, an increase in the number of Brønsted acid sites was observed by the formation of OH− groups as a result of dissociative adsorption of H2 S [6]. The Lewis basicity of ZrO2 is claimed to originate from its c.u.s. O2− centers [7], whereas the Zr4+ sites are known to be centers for Lewis acidity [8]. Thus, the dry surface of ZrO2 has Zr4+ –O2− Lewis acid–base pairs, which may be saturated by dissociated water species forming hydroxyl groups. The hydroxyl groups are amphoteric in nature, thus acting as a source for either Brønsted acidity or basicity [8]. In the gasification gas clean-up, the catalyst performance is largely affected by the mobility of lattice oxygen and the formation of oxygen vacancies, since the lattice oxygen is believed to take part in the tar (toluene, naphthalene, etc.) oxidation reactions [2,4,9]. In general, (intrinsic) oxygen vacancies are formed on ZrO2 via reduction of the surface. Extrinsic oxygen vacancies are induced by other metals in the lattice, which enhance the

606

E.I. Kauppi et al. / Applied Catalysis B: Environmental 111–112 (2012) 605–613

performance of e.g. Y2 O3 -ZrO2 compared to ZrO2 [10]. It has been observed for the partial oxidation of methane that the Y2 O3 -ZrO2 system possesses superior properties compared to ZrO2 due to the increased lattice oxygen diffusion and fast oxygen exchange of the doped oxide [11]. On the contrary, SiO2 is not expected to create oxygen vacancies on ZrO2 [12]. Studies on the influence of H2 S suggest that the modification of the catalyst properties are caused by changes in the acidic, basic, or redox properties of the catalyst. These properties can be studied by the methanol probe reaction. Vibrational spectroscopic techniques characterize the catalyst surface based on the adsorbed methoxy type species derived from methanol [13]. On the other hand, the gaseous desorption products generally indicate the chemical nature of the surface sites; formaldehyde (CH2 O) indicates the presence of redox sites, dimethyl ether (CH3 OCH3 , DME) acidic sites, and carbon oxides basic sites. Several publications exist on MeOH adsorption and its surface reaction on metal oxides surfaces [14–19]. Less is known about MeOH adsorption on the sulfided metal oxide surfaces. The aim of this study was to elucidate the observed H2 S-induced changes in the properties of the zirconia-based catalysts, especially regarding the sulfided catalysts’ different behavior in naphthalene conversion during gasification gas clean-up (improvement for ZrO2 and Y2 O3 -ZrO2 compared to no effect for SiO2 -ZrO2 ). First, methanol was used as a surface probe molecule to characterize the untreated catalysts. The catalysts were also probed with H2 S. Finally, MeOH adsorption was performed on the sulfided catalysts to study the effect of H2 S on the catalyst properties. 2. Experimental ZrO2 , Y2 O3 -ZrO2 with 5 mol-% Y2 O3 , and SiO2 -ZrO2 with 8 mol% SiO2 from MEL Chemicals were used as catalysts. The catalysts were calcined at 800 ◦ C for 1 h. These catalysts have been previously characterized by Viinikainen et al. [2]. The surface characterization indicated that the surface sites of these catalysts varied. Raman and XRD characterizations showed that ZrO2 was monoclinic, but Y2 O3 -ZrO2 and SiO2 -ZrO2 were mixtures of tetragonal and cubic structures. The specific surface area was found to differ (for ZrO2 , Y2 O3 -ZrO2 , and SiO2 -ZrO2 the surface areas were 24, 53, 92 m2 /g, respectively) [2]. 2.1. H2 S retention on the catalyst surface H2 S temperature-programmed desorption (TPD) experiments were carried out in an Altamira AMI-100 characterization system to study the sulfur retention and probe the adsorption modes of H2 S on the studied catalyst. Prior to each experiment the samples were dried in N2 (30 cm3 /min) for 2 h at 150 ◦ C. A gas mixture containing 15 cm3 /min N2 (AGA, 99.999%) and 15 cm3 /min H2 S in H2 (AGA, 0.1% H2 S in H2 ) was passed through powder form catalyst samples at 40 ◦ C for 2 h. Thereafter the sample was flushed for 1 h with N2 (30 cm3 /min) to remove weaker held species so that only the chemisorbed H2 S species remained on the surface. H2 S-TPD was performed under 30 cm3 /min N2 flow with a temperature ramp from 40 to 750 ◦ C (10 ◦ C/min). Desorbed H2 S was measured with a mass spectrometer (ThermoStarTM , Pfeiffer Vacuum). 2.2. Catalyst characterization by probing with MeOH The surface properties of the catalysts were probed with methanol. MeOH-DRIFTS experiments were performed to yield information about the surface based on the specific vibrations of the formed methoxy species. MeOH-TPSR was carried out on the

calcined and sulfided catalysts to study the chemical nature of the surface sites and the effect of H2 S. 2.2.1. In situ DRIFTS–MS measurements with MeOH surface probing The surface species formed during methanol decomposition on the calcined ZrO2 , Y2 O3 -ZrO2 and SiO2 -ZrO2 samples were studied by in situ diffuse reflectance infrared Fourier transform (DRIFT) spectroscopy combined with mass spectrometry (MS). The DRIFTS measurements were performed using a Nicolet Nexus FTIR spectrometer and a Spectra-Tech high temperature/high pressure reaction chamber with ZnSe windows. Gaseous products were monitored on-line by a mass spectrometer (Omnistar, Pfeiffer Vacuum). The spectrum of an aluminum mirror measured under nitrogen flow was used as the background, and the outlet was continuously analyzed by MS. The total gas flow rate was kept at 50 cm3 /min. Methanol decomposition was studied as a function of temperature from 100 to 500 ◦ C. A fresh sample was used in each experiment and pre-treated by in situ calcination with 10% O2 /N2 (AGA, air 99.999%, N2 99.999%) at 600 ◦ C for 2 h. The methanol feed (2000 ppm MeOH in N2 , AGA) was started at 100 ◦ C and continued at this temperature for 30 min. Spectra were collected once every minute (4 cm−1 resolution, 30 scans) for the first 10 min, and thereafter once every 5 min (100 scans). After 30 min the sample cell was flushed with nitrogen to obtain a spectrum without the contribution by gas-phase methanol. Then methanol was redirected to the cell and the temperature was increased stepwise to 200 ◦ C. Spectra were recorded every 25 ◦ C (100 scans). At 200 ◦ C, the sample cell was flushed with nitrogen and a spectrum was recorded. Thereafter, the methanol flow was resumed and the temperature increased. The experiment was continued by feeding methanol vapor to the chamber during heating with spectra collected every 25 ◦ C (100 scans). Nitrogen flushes were done at 300, 400, and 500 ◦ C. At 500 ◦ C the sample was flushed with nitrogen and oxidized with diluted air. 2.2.2. Temperature-programmed surface reaction of MeOH The temperature-programmed surface reaction of MeOH (MeOH-TPSR) of the ZrO2 , Y2 O3 -ZrO2 and SiO2 -ZrO2 samples were carried out using 0.2 g of particles with a size distribution between 0.425 and 0.85 mm, made from crushed pellets. The ZrO2 powders used to obtain those pellets had all been calcined at 800 ◦ C. The ZrO2 , Y2 O3 -ZrO2 , and SiO2 -ZrO2 samples were first calcined in O2 /He gas mixture (20 vol-% O2 in He) from room temperature to 550 ◦ C (ramp 10 ◦ C/min) to remove adsorbed moisture and possible carbonaceous residues. The sample was transferred to a stainless steel reactor of 1 cm diameter, which was kept at 100 ◦ C. Methanol adsorption for the calcined samples was performed under a total flow of 100 cm3 /min, P = 110 kPa, of a 2000 ppm methanol in Ar. After complete saturation of the sample, the sample was purged with Ar at a flow rate of 50 cm3 /min. Thereafter the temperature was increased to 700 ◦ C (temperature ramp 10 ◦ C/min) under Ar flow. The detection of the desorbing gases at the microreactor outlet was performed by mass spectrometry (Omnistar, Balzers), using a Channeltron detector. To study the effect of H2 S on the catalysts, the samples were sulfided prior to MeOH-TPSR experiments. After calcination, the catalyst sample (0.2 g) was transferred to a Pyrex microreactor of 5 mm diameter, operating with a total flow of 50 cm3 /min, P = 110 kPa. After a pre-treatment at 100 ◦ C with H2 S (30 vol-% in H2 ), the system was cooled to room temperature in a nitrogen atmosphere. Subsequently, the sample was transferred to stainless steel microreactor and MeOH-TPSR was carried out following the same procedure as for the calcined samples.


607

Fig. 1. In situ DRIFT spectra measured for pure ZrO2 during temperature-programmed experiment with methanol.

3. Results and discussion 3.1. MeOH probe reaction on the calcined samples 3.1.1. MeOH DRIFTS results Surface species formed during methanol adsorption and decomposition on the ZrO2 , Y2 O3 -ZrO2 , and SiO2 -ZrO2 samples were

studied as a function of temperature with in situ DRIFTS; online MS analyzed the gaseous products formed. Figs. 1–3 show selected spectra measured for the three catalysts during MeOH-DRIFTS experiments. The presented spectra were measured under nitrogen flow to minimize the effect of gas phase methanol. The peak positions were similar under methanol, which increased the intensities of the absorption bands.

Fig. 2. In situ DRIFT spectra measured for Y2 O3 -ZrO2 during temperature-programmed experiment with methanol.

608


Fig. 3. In situ DRIFT spectra measured for SiO2 -ZrO2 during temperature-programmed experiment with methanol.

The calcined ZrO2 sample (Fig. 1) showed peaks due to hydroxyl groups at 3773, 3724, 3674 and 3650 cm−1 . The first three are assigned to terminal Zr–OH, bibridged (Zr)2 –OH and tribridged (Zr)3 –OH groups, respectively, on monoclinic zirconia [2,7]. Adsorption of methanol on the calcined zirconia at 100 ◦ C resulted in consumption of hydroxyl groups and appearance of hydrogenbonded hydroxyls (broad band centered at ca. 3400 cm−1 ). New peaks appeared at 2920 and 2815 cm−1 attributed to the as (CH3 ) and s (CH3 ) vibrations of methoxy species on zirconia (Zr–OCH3 ), and at 1165 cm−1 due to (OCH3 ) vibration of on-top methoxy species [13,20]. Molecularly adsorbed methanol may have been present, as suggested by a shoulder at 2950 cm−1 [13]. The methyl deformation band was seen at 1460 cm−1 [13]. Heating ZrO2 above 250 ◦ C led to new peaks at 1573, 1380 and 1370 cm−1 due to as (COO), ı(CH), and s (COO) vibrations of formates [13]. The formate (CH) peak at 2870 cm−1 was apparent from 275 ◦ C. The formates disappeared above 450 ◦ C whereas methoxy species were still apparent at 500 ◦ C. In addition, peaks of carboxylate or carbonate [13] species were present at approximately 1540 and ∼1430 cm−1 at 500 ◦ C, and a gradual reappearance of hydroxyl groups with temperature was observed during the entire experiment. The calcined Y2 O3 -ZrO2 sample (Fig. 2) exhibited peaks at 3757 and 3677 cm−1 , which are, in accordance with earlier results obtained for this sample [2], attributed to terminal and tribridged hydroxyl groups, respectively, on tetragonal Y2 O3 -doped zirconia [21]. Adsorption of methanol resulted in consumption of hydroxyl groups and in appearance of methoxy peaks at 2920, 2810, and 1160 cm−1 similar to the calcined ZrO2 sample. Additional peaks were seen at 1444, 1417, 1075, and 1025 cm−1 during heating at 200 ◦ C and above. Formates at 2877, 1580, 1382 and 1356 cm−1 became increasingly apparent at 300 and above, along with carboxylates/carbonates (1446 and 1335 cm−1 ). Hydroxyl groups were partly restored with temperature. The calcined SiO2 -ZrO2 sample (Fig. 3) showed peaks at 3732 and 3671 cm−1 due to silanol Si–OH species, and tribridging hydroxyl groups on zirconia, respectively [2]. With adsorption of methanol the intensities of these hydroxyl groups decreased, whereas a new hydroxyl band appeared at 3572 cm−1 possibly

due to molecularly adsorbed methanol and/or hydrogen-bonded hydroxyls. Two types of methoxy species were seen: Zr–OCH3 at 2927, 2830, 1443, and 1150 cm−1 , and Si–OCH3 at 2953, 2850, and 1461 cm−1 [19]. With increasing temperature, formates appeared above 250 ◦ C, but the peak intensities were lower than for the other samples. Furthermore, no formate (CH) peak was observed at approximately 2880–2870 cm−1 . This may indicate that the formation of the formate species was not as significant on this sample as on the other ones. The methoxy groups were present still at 500 ◦ C, and carbonate/carboxylate-type species were also seen (1560, 1455, and 1352 cm−1 ). Simultaneous mass spectrometry results (not shown) indicated formation of H2 with all three samples above 250 ◦ C with the amount continuously increasing with temperature. DME formation was seen only on the SiO2 -ZrO2 sample between 300 and 400 ◦ C. A small amount of CO2 was produced with the calcined ZrO2 and the SiO2 -ZrO2 samples at above 400 ◦ C. In addition, methyl fragments were detected throughout the whole temperature range. Other products that may form from methanol include formaldehyde and carbon monoxide but their formation could not be detected due to overlapping of their mass numbers with other components. 3.1.2. Temperature-programmed surface reaction of MeOH on the calcined samples The MeOH-TPSR experiments on the ZrO2 , Y2 O3 -ZrO2 and SiO2 ZrO2 were carried out to determine the effect of H2 S on the catalyst properties. For this the changes in the MeOH-TPSR product distribution were compared for the calcined and sulfided samples. The formation of DME is presented in Fig. 4 for the calcined and sulfided samples. For CO2 , formaldehyde, CO, and H2 the Tmax values and the corresponding peak areas are presented in Tables 1 and 2, respectively. The changes due to pre-adsorbed H2 S are discussed in the next section. Fig. 4 shows that the amount of DME generated decreased in the order SiO2 -ZrO2 > Y2 O3 -ZrO2 ZrO2 . Desorption temperature is lower for ZrO2 (350 ◦ C) than for Y2 O3 -ZrO2 and SiO2 -ZrO2 (480 and 370 ◦ C, respectively). CO2 is produced on a wide temperature range on all the catalysts (Tmax of the most intense peak shown in Table 1). Two CO2


609

Fig. 4. DME MS-signal (m/z = 45) in the MeOH-TPD experiment on calcined and sulfided ZrO2 , Y2 O3 -ZrO2 , and SiO2 -ZrO2 . Table 1 Tmax (◦ C) for formaldehyde, CO2 , CO and H2 during MeOH-TPSR experiment on calcined and sulfided ZrO2 , Y2 O3 -ZrO2 , and SiO2 -ZrO2 (the most intense maximum is reported, shoulders in the parentheses). Sample

Formaldehyde

CO2

CO

H2

ZrO2 ZrO2 (sulf.) Y2 O3 -ZrO2 Y2 O3 -ZrO2 (sulf.) SiO2 -ZrO2 SiO2 -ZrO2 (sulf.)

170 200 220 220 250 220

580 (170, 380) 580 (150, 420) 350 290 560 (360) 530 (270)

400 (300) 510 (400) 580 560 (420) 510 (370) 550

400 510 580 560 510 (370) 550

Table 2 Peak areas (a.u. s × 10−8 ) for formaldehyde, CO2 , CO and H2 during MeOH-TPSR experiment on calcined and sulfided ZrO2 , Y2 O3 -ZrO2 , and SiO2 -ZrO2 . Sample

Formaldehyde

CO2

CO

H2

ZrO2 ZrO2 (sulf.) Y2 O3 -ZrO2 Y2 O3 -ZrO2 (sulf.) SiO2 -ZrO2 SiO2 -ZrO2 (sulf.)

0.077 0.109 0.474 0.236 0.338 0.306

0.047 0.052 0.115 0.070 0.155 0.019

2.46 5.51 4.49 4.12 4.29 4.41

1.13 0.826 3.46 1.69 3.54 3.30

610


maxima are detected on the SiO2 -ZrO2 catalyst, whereas Y2 O3 -ZrO2 shows only a broad maximum. ZrO2 shows only minor formation of CO2 with multiple maxima. CO2 production on the studied catalysts decreased in the SiO2 -ZrO2 Y2 O3 -ZrO2 > ZrO2 order (Table 2). Table 2 shows that the amount of formaldehyde formed on the catalysts decreases in the order Y2 O3 -ZrO2 > SiO2 -ZrO2 > ZrO2 . Formaldehyde desorbs on a wide temperature range. The peak maxima appear, however, at low temperatures (Table 1). Also MeOH desorbs at approximately this temperature (not shown), which is attributed to desorption of molecular MeOH present on the catalysts or the reverse reaction of the methoxy species [13]. In Table 1, the CO and H2 Tmax values for the calcined Y2 O3 ZrO2 and SiO2 -ZrO2 are higher (∼580 and 510 ◦ C, respectively) than those on the calcined ZrO2 sample (∼400 ◦ C). The CO peaks for ZrO2 , Y2 O3 -ZrO2 , and SiO2 -ZrO2 appear concurrently with the H2 peaks (Table 1). The peaks on the Y2 O3 -ZrO2 , and SiO2 -ZrO2 are more intense than on the ZrO2 (Table 2). 3.1.2.1. Methanol adsorption and reaction pathways. Methanol may adsorb on oxide surfaces either molecularly or dissociatively. The dissociative adsorption has been suggested to take place via two ways: by adsorption on surface hydroxyl groups with formation of methoxy groups and water, or on c.u.s. Zr4+ –O2− pairs with formation of methoxy- and hydroxyl groups [18,19]. As noted during the MeOH-DRIFTS experiments, hydroxyl groups decreased in intensity during the adsorption of methanol on all three samples, suggesting that dissociative adsorption took place on the hydroxyl groups, which would be substituted by the methoxies. MeOH-DRIFTS results show the formation of only one type of methoxy species with the calcined ZrO2 and Y2 O3 -ZrO2 samples, and two types of species with the SiO2 -ZrO2 sample. This suggests that yttria is homogeneously distributed in the Y2 O3 -ZrO2 sample and it does not form a separate phase; whereas the silica and the zirconia phases are more separate and two distinct methoxy groups can be formed on the SiO2 -ZrO2 sample. Methoxy groups react on catalyst surface to produce DME via condensation of two methoxy groups [18]. DME selectivity is used to describe the pure dehydration ability of the catalyst, related generally to its Lewis acidic character [14,16]. Based on the intensities of the DME peaks during MeOH-TPSR (Fig. 4), it is concluded that catalyst acidity for the calcined samples decreases in the order SiO2 ZrO2 > Y2 O3 -ZrO2 ZrO2 in agreement with previous studies on these catalysts [2]. This is in accordance with the MeOH-DRIFTS results, which also indicated highest number of Lewis acidic sites on the SiO2 -ZrO2 sample, and also with Bosman et al. [22] who reported that doping ZrO2 with SiO2 increases the acidity [22]. The Lewis acid sites were comparatively weak on the ZrO2 sample as indicated by the relatively low temperature of desorption for DME (Fig. 4), also in agreement with previous studies on the same catalysts [2]. During the MeOH-DRIFTS experiments the peaks of the methoxy species decreased in intensity with increasing temperature on all three samples, whereas those of the formate species became apparent near 300 ◦ C, and thereafter decreased before 500 ◦ C. This suggests that the methoxy species formed in the dissociative adsorption of methanol reacted to formate (HCOO*) species. Similar trends have already been reported for methanol decomposition on zirconia-based catalysts [18]. The transformation of methoxy to formate has been suggested [23] to proceed through an intermediate dioxymethylene species, an adsorbed form of formaldehyde. Dioxymethylene may decompose to gas-phase formaldehyde on the redox sites [14]. Wang and Wachs associate formaldehyde formation with redox sites on the catalyst [14]. However, some studies also suggest that the formation of formaldehyde may be connected with basic O2− sites and the reducibility of the catalyst [24,25]. Roozeboom

et al. concluded that the selectivity to formaldehyde increases with decreasing reducibility of the catalyst [25]. Formaldehyde may also be formed via a reaction between methanol and hydroxyl group [19]. Based on the generation of formaldehyde (Tables 1 and 2) the redox sites on ZrO2 , Y2 O3 -ZrO2 , and SiO2 -ZrO2 catalysts are suggested to be varied, since formaldehyde production differed. A small part of the formaldehyde could come from reactions of methanol and hydroxyl groups, since methanol desorbed at the same time. Also, according to Viinikainen et al. the redox properties of these oxides differ, which affects their reactivity towards oxidation reactions during gasification gas clean-up [2]. CO2 may be originated from the decomposition of formate and/or carbonate species [18,26], which were present on all the catalysts during MeOH-DRIFTS experiments above 300 ◦ C and after (however, formates mostly disappeared before 500 ◦ C, whereas carbonates/carboxylates were still present). The formation of CO2 acts as an indicator of the MeOH oxidation processes [16]. In general it can be used to depict the Lewis basicity of the catalyst [14,16,18,27,28]. This was demonstrated by Garcia Cortez et al. who performed MeOH-TPSR experiments on VOx /Al2 O3 with different amounts of K doping, clearly showing the increase in CO2 production with increasing basicity of the catalyst [27]. In the MeOH-DRIFTS experiments (MS results, not shown) CO2 forms on ZrO2 and SiO2 -ZrO2 samples, suggesting that these had more basic sites than the Y2 O3 -ZrO2 sample. However, during MeOH-TPSR experiments CO2 formation decreased in the order SiO2 -ZrO2 Y2 O3 -ZrO2 > ZrO2 , illustrating the order of the Lewis basicity of the catalysts (Table 2). In the MeOH-DRIFTS experiments the methoxy species detected on SiO2 -ZrO2 were Zr–OCH3 and Si–OCH3 . Apparently separate formates appeared with increasing temperature (above 250 ◦ C). It is thus probable that the methoxy species on the SiO2 -ZrO2 sample evolve to formates yielding the two CO2 peaks reported in Table 1 (at ∼360 and 560 ◦ C, the latter being very intense). Also, only one DME peak is observed in the MeOH-TPSR experiments for SiO2 ZrO2 . The formation of CO2 was low on the ZrO2 catalyst (Table 2) suggesting that the strength of its basic sites is rather weak. This is consistent with Korhonen who reported that highly basic c.u.s. sites were not observed on ZrO2 [26]. Also Bianchi et al. observed the generation of CO2 to be minor in a MeOH-TPSR experiment on ZrO2 aerogel [18]. The higher formation of CO2 with Y2 O3 -ZrO2 compared to ZrO2 was expected since its surface was reported to consist mainly of strongly basic O2− centers [2]. The simultaneous production of CO and H2 means that they must originate from the decomposition of formate or methoxy species [18,19]. The CO and H2 Tmax on ZrO2 at lower temperatures (Table 1) suggest that the formate groups on the ZrO2 were more reactive than the ones on Y2 O3 -ZrO2 and SiO2 -ZrO2 . According to Badlani and Wachs the surface methoxy decomposition temperature for ZrO2 is 326 ◦ C and 560 ◦ C for SiO2 [17]. MeOH probing showed differences in the catalyst characteristics. SiO2 -ZrO2 possessed stronger Lewis acidic and basic properties compared to the pure ZrO2 and Y2 O3 -ZrO2 . The redox properties of Y2 O3 -ZrO2 are stronger than on the other catalysts studied. It was also suggested, that doping ZrO2 with yttria or silica results in different kind of distribution of the dopants in ZrO2 phase; silica exists as separate phases on SiO2 -ZrO2 , whereas the distribution of yttria in Y2 O3 -ZrO2 is more homogeneous. 3.2. H2 S retention on the surfaces and the effect of H2 S on the properties of ZrO2 -based catalysts H2 S adsorption on metal oxides has been suggested to proceed via the following pathways: (1) exchange of surface oxygen to sulfur whereupon water is desorbed, (2) dissociative adsorption of H2 S to HS− and H+ with formation of a surface hydroxyl group [5], or a


OH−

water molecule if the dissociation of H2 S occurs on an group [6], and (3) coordinatively bonded hydrogen sulfide [29]. Pathway 1 can also be described as an interaction of H2 S with a defective site, or oxygen vacancy, which involves incorporation of S2− in the lattice [30]. 3.2.1. Temperature-programmed desorption of H2 S H2 S adsorption and desorption was studied for the ZrO2 , Y2 O3 ZrO2 and SiO2 -ZrO2 samples to investigate the interaction of H2 S with the catalyst surface. Pure SiO2 was also tested; however, virtually no sulfur compounds were detected in the H2 S-TPD spectrum (results for the pure SiO2 sample not shown), showing the inert nature of SiO2 . Also Travert et al. [6] found that on SiO2 , only undissociative adsorption of H2 S occurs. Thus, it seems that undissociatively adsorbed H2 S is relatively easily desorbed, as also indicated earlier by Datta and Cavell [31]. Water was produced during H2 S adsorption on all the samples (not shown). This indicates the dissociative adsorption of H2 S via exchange of surface oxygen to sulfur, or via the replacement of the surface hydroxyl groups by SH− groups, both pathways resulting in water generation [5,6,30]. H2 S was found to dissociate on ZrO2 surfaces at higher temperatures (room temperature and above) also by Travert et al. [6]. Fig. 5 shows the H2 S desorption profiles. The high temperature region is not shown, since no H2 S desorbed after ∼300 ◦ C. A minor peak for SO2 was detected only on SiO2 -ZrO2 at 660 ◦ C (not shown), suggesting that the sulfur may also react with lattice oxygen. The H2 S-TPD profiles (Fig. 5) showed one maximum for ZrO2 at 100 ◦ C and for Y2 O3 -ZrO2 at 115 ◦ C. The SiO2 -ZrO2 sample had two distinct maxima; one at 115 ◦ C and another one at 210 ◦ C, the latter being very intense indicating strong H2 S adsorption. This suggests that H2 S occupies different sites on SiO2 -ZrO2 than it does on ZrO2 and Y2 O3 -ZrO2 . Sulfur retention on the surface after TPD cannot be ruled out. The surface characterization in the earlier studies [2] and MeOH probing of the three ZrO2 -based catalysts indicated that the surface sites of these catalysts were different. It has also been established, that oxygen mobility is enhanced on the Y2 O3 -ZrO2 catalyst compared to the pure ZrO2 [11], whereas on the SiO2 -ZrO2 oxygen mobility is more limited due to the strong covalent nature of the SiO bond [12]. The intensity order of the H2 S peak at approximately 100 ◦ C is Y2 O3 -ZrO2 > ZrO2 > SiO2 -ZrO2 , which corresponds to the order in which oxygen vacancies are generated [2,12], and also to the order in which H2 S affects the reactivity in the gasification gas application. It is suggested that this peak be due to dissociative adsorption of H2 S and H2 S adsorption via replacement of oxygen in the lattice. IR results indicated two phases on SiO2 -ZrO2 , which could explain the two H2 S desorption peaks detected in its H2 S desorption spectrum. However, pure SiO2 did not adsorb hydrogen sulfide. Therefore it is not expected that H2 S adsorbs on the SiO2 phase on SiO2 -ZrO2 but may however be adsorbed on an “interphase”. According to characterization studies, highly acidic Zr4+ centers were only present on SiO2 -ZrO2 [2]. Thus, the H2 S desorption peak at 210 ◦ C observed on SiO2 -ZrO2 may be connected to the dissociative adsorption of H2 S on the Zr4+ acid sites, or the adsorption of H2 S on some SiO2 –ZrO2 interphase. 3.2.2. Temperature-programmed surface reaction of MeOH on the sulfided samples MeOH-TPSR experiments on the sulfided ZrO2 , Y2 O3 -ZrO2 and SiO2 -ZrO2 samples were performed to study the effect of H2 S on catalysts characteristics. During the adsorption of MeOH on the sulfided catalyst surface at 100 ◦ C, no sulfur-containing species were detected in the mass spectra (adsorption profiles not shown). No H2 S was also detected

611

during MeOH-TPSR. Thus, H2 S cannot be replaced by MeOH during its adsorption, indicating strong adsorption of H2 S. More water desorbed during adsorption of MeOH from the sulfided samples than from the calcined ones, which may suggest increased number of surface hydroxyl groups after sulfidation. Fig. 4 shows that the amount of DME produced on the sulfided Y2 O3 -ZrO2 sample was lower than on the calcined Y2 O3 -ZrO2 sample, whereas a similar decrease was not as evident on the ZrO2 or SiO2 -ZrO2 samples. With ZrO2 and SiO2 -ZrO2 however, the temperature of the DME peak shifted upwards. This suggests that sulfur adsorption affected the nature of the Lewis acid sites on ZrO2 and SiO2 -ZrO2 (the adsorption of DME stronger on the sulfided ZrO2 and SiO2 -ZrO2 sample). It is suggested that at least part of the H2 S adsorbed on the Lewis acidic sites responsible for DME production on all the catalysts studied or that the adsorption of H2 S changed the nature of these sites. Also Travert et al. [6] found that on a ZrO2 catalyst, sulfur adsorbs partly on its Lewis acidic sites [6]. Sulfidation increased CO2 production for the ZrO2 catalyst (Table 2). On the contrary, for the SiO2 -ZrO2 catalyst, sulfidation significantly decreased CO2 production (Table 2). On Y2 O3 -ZrO2 the change was not as evident, however a slight decrease in the intensity of CO2 peak and temperature of desorption was seen. It is suggested that H2 S treatment enhanced the basic properties of the ZrO2 catalyst (or, moreover, the mobility of surface oxygen) responsible for CO2 production, whereas on the SiO2 -ZrO2 catalyst the basic character was markedly reduced. This is in accordance with the earlier suggestion by Rönkkönen et al. that sulfur influences the basic properties of these catalysts [4]. 3.2.2.1. DMS formation. Dimethyl sulfide (DMS) was identified during MeOH-TPSR as shown in Fig. 6, indicating that the pre-adsorbed H2 S reacted with MeOH to DMS. The Tmax for DMS on the studied samples is at approximately 490–530 ◦ C. A shoulder is seen on SiO2 ZrO2 at 390 ◦ C. No H2 S or other sulfur-containing molecules were detected during MeOH adsorption (not shown), indicating that the species from pre-adsorbed H2 S were not replaced by the adsorbing MeOH. This and the high DMS Tmax values indicate that sulfur is rather strong bound to the surface. The amount of DMS formed decreased in the order Y2 O3 ZrO2 ZrO2 > SiO2 -ZrO2 , which is the same order as that of H2 S desorbed at around 100 ◦ C in the H2 S-TPD experiments (Fig. 5). Y2 O3 -ZrO2 had the highest amount of adsorbed sulfur species or the adsorbed sulfur species were more reactive, as indicated by the amount of DMS formed. When comparing the results of SiO2 ZrO2 in the H2 S-TPD and MeOH-TPSR experiments, two maxima are detected for H2 S and DMS (Figs. 5 and 6) indicating at least two sites for H2 S adsorption and DMS generation on SiO2 -ZrO2 . The temperature difference between the H2 S Tmax and DMS Tmax does not allow one to conclude that the same sulfur species are present in the H2 S-TPD and MeOH-TPSR experiments. Nevertheless, the different behavior of SiO2 -ZrO2 in both experiments is noted. Very few studies on MeOH adsorption on sulfided catalysts have been reported, yet the reaction between MeOH and H2 S on metal oxides catalysts and DMS formation was reported by Ziolek et al. [32]. DMS formation pathways on ZrO2 catalysts are still unknown, since the active form of sulfur from adsorbed H2 S has not been established. However, literature suggests, that the reaction to DMS occurs between SH− and/or S2− (formed in the dissociative adsorption of H2 S [5,6]) and methoxy species [32]. According to Travert et al. [6] a H2 S-treated ZrO2 surface has a higher amount of hydroxyl groups, which are believed to have increased Brønsted acidic character. On the contrary, they reported that SH groups did not reveal any proton-donating ability [6]. The sulfided surface may also have increased redox or basic properties because of the adsorbed sulfur replaced an oxygen atom in the lattice, as suggested by Ziolek et al. [5]. This was related with the

612


Fig. 5. H2 S-TPD-profiles of the studied ZrO2 , Y2 O3 -ZrO2 and SiO2 -ZrO2 catalysts (m/z = 34).

Fig. 6. MS-signal (m/z = 47) illustrating the generation of DMS in the MeOH-TPD experiment on ZrO2 , Y2 O3 -ZrO2 , and SiO2 -ZrO2 .

lower electronegativity of S2− compared to O2− [5]. In view of this, Fig. 7 presents the sulfided ZrO2 surface with the “new” hydroxyl groups acting as acid and the adjacent sulfur atom as a base, following DMS formation as speculated. It is possible that upon the adsorption of MeOH on the sulfided surface water is released from MeOH during its adsorption (Fig. 7) and DMS is generated via the reaction of CH3 S and methoxy groups (Fig. 7). It is thus speculated that sulfur on the surfaces of the studied ZrO2 catalysts acted as a base and caused MeOH to react by donating its OH− group (breaking of C–O bond in MeOH). This may explain the greater amounts of desorbing water during MeOH adsorption on the sulfided ZrO2 surfaces compared to those amounts on the calcined ZrO2 surfaces in our experiments. It may be the combination of the two features (increased number of Brønsted acidic OH− groups and the more basic sulfur atoms compared with O2− ) that acts as a source for the observed reactivity towards the conversion of methanol to DMS.

It is also possible, that the reaction occurs between the methoxy and SH− species. However, during the MeOH-TPSR experiments it was noted that sulfur is very strongly bound to the surface, and the elimination of SH− groups would have been expected during MeOH adsorption. Moreover, the amount of the generated DMS on different catalysts decreases in correlation with the order in which oxygen vacancies are generated (Y2 O3 -ZrO2 ZrO2 > SiO2 ZrO2 ), which suggests the active form of sulfur in the lattice. The migration of S in to the lattice during TPSR can neither be ruled out. In summary, DMS formation possibly involves interaction with a sulfur atom incorporated in the lattice of the ZrO2 catalysts. DMS is formed most probably via the condensation of CH3 S and CH3 O species. Based on our study, there is a clear difference in the behavior of the studied ZrO2 -based catalysts with H2 S, SiO2 -ZrO2 possessing the most diverging characteristics. The adsorbed H2 S has different reactivities on the surfaces of the ZrO2 -based catalysts, as indicated

Fig. 7. Suggested pathway for MeOH adsorption and the generation of DMS on the sulfided ZrO2 catalysts.


by the DMS profiles, which can be correlated to the observed effect in naphthalene conversion during gasification gas clean-up. Based on the results from MeOH-TPSR on sulfided catalysts, it is speculated that the limited reactivity of H2 S on SiO2 -ZrO2 hinders its activity during gas cleaning (whether some sulfur is irreversibly adsorbed cannot be concluded here but is subject to further study). It is also suggested, that the interaction of sulfur with the lattice is seen in the gasification gas clean-up as an enhancement effect on Y2 O3 -ZrO2 and ZrO2 . 4. Conclusions We probed surfaces of ZrO2 , Y2 O3 -ZrO2 , and SiO2 -ZrO2 catalysts with the aim of explaining the differences between the samples regarding the H2 S-induced changes in naphthalene conversion during gasification gas clean-up; namely improvement on ZrO2 and Y2 O3 -ZrO2 compared to no effect or poisoning on SiO2 -ZrO2 . MeOH was used as a surface probe in addition to H2 S. The following conclusions can be drawn: (1) MeOH-DRIFTS experiments uncover the existence of two phases on SiO2 -ZrO2 , whereas yttria is more homogeneously distributed on/in Y2 O3 -ZrO2 . This probably explains the desorption profiles of H2 S, which differ for SiO2 -ZrO2 compared to ZrO2 and Y2 O3 -ZrO2 (two peaks for SiO2 -ZrO2 and one for the other catalysts). It is concluded that H2 S adsorption on ZrO2 and Y2 O3 -ZrO2 occurs mainly on one kind of sites and there are at least two H2 S adsorption sites on the SiO2 -ZrO2 . H2 S was found to adsorb dissociatively on ZrO2 surfaces. (2) MeOH-TPSR products on sulfided ZrO2 catalysts differed from those on the calcined samples showing that H2 S affected the reactivity of the surfaces. H2 S-treatment was also found to affect the nature of the acidic and basic sites on the ZrO2 catalysts; Lewis acidity decreased especially on Y2 O3 -ZrO2 , whereas Lewis basicity was differently affected; on the ZrO2 catalyst it increased and on the SiO2 -ZrO2 it markedly decreased. The effect on the Lewis basicity of Y2 O3 -ZrO2 was not as evident as on the other catalysts. (3) Sulfur reacted with methoxy species to DMS during heating. The amount of DMS formed was in the order Y2 O3 ZrO2 > ZrO2 > SiO2 -ZrO2 , showing that the sulfur species were least reactive on SiO2 -ZrO2 . Two DMS desorption maxima were found on SiO2 -ZrO2 , whereas on the ZrO2 and Y2 O3 -ZrO2 DMS showed only one intense maximum. (4) The order of the generated DMS corresponds to the intensity of the sulfur effect on the properties of the catalysts in gasification gas clean-up, as well as the order of desorbed H2 S at around 100 ◦ C in the H2 S-TPD experiments. This is also the order in which oxygen vacancies are expected to be generated on these catalysts. (5) H2 S adsorption on the sites present on ZrO2 and Y2 O3 -ZrO2 has a positive effect on naphthalene conversion during gasification gas clean-up which may be connected with the reactions of adsorbed sulfur on their surface. On the contrary, on SiO2 ZrO2 the observed poisoning effect may be due to reduced

613

reactivity of sulfur on the surface, which was illustrated by the lower amount of the sulfur-containing product from MeOH. Acknowledgments The authors would like to thank Miikka Tamminen, Ma del Puerto Martín and Juan Miguel Ramos for assistance during the experimental work. MEL Chemicals is thanked for donating the zirconia samples. The financing of the Finnish Funding Agency for Technology and Innovation (TEKES), Academy of Finland, the Technical Research Centre of Finland (VTT), the Spanish Ministry of Science and Innovation (CTQ2008-04361/PPQ, CTM2008-06876CO2-02/TECNO), the Comunidad de Madrid (CAM) (Program S0505/AMB/0406), JAE (Junta de Ampliación de Estudios) are gratefully acknowledged. Dr. Pedro Ávila, Dr. Juha Linnekoski, and Dr. Pekka Simell are thanked for co-operation and European Science Foundation COST Action D36 for support. References [1] J. Hepola, P.A. Simell, Appl. Catal., B 14 (3–4) (1997) 287–321. [2] T. Viinikainen, H. Rönkkönen, H. Bradshaw, H. Stephenson, S. Airaksinen, M. Reinikainen, P. Simell, O. Krause, Appl. Catal., A 362 (2009) 169–177. [3] S. Juutilainen, P. Simell, O. Krause, Appl. Catal., B 62 (2006) 86–92. [4] H. Rönkkönen, P. Simell, M. Reinikainen, O. Krause, Top. Catal. 52 (2009) 1070–1078. [5] M. Ziolek, J. Kujava, O. Saur, J.C. Lavalley, J. Mol. Catal. A: Chem. 97 (1995) 49–55. [6] A. Travert, O.V. Manoilova, A.A. Tsyganenko, F. Maugé, J.C. Lavalley, J. Phys. Chem. B 106 (2002) 1350–1362. [7] B. Bachiller-Baeza, I. Rodriquez-Ramos, A. Guerrero-Ruiz, Langmuir 14 (1998) 3556. [8] G. Cerrato, S. Bordiga, S. Barbera, C. Morterra, Surf. Sci. 377–379 (1997) 50–55. [9] M. Labaki, S. Siffert, J.-F. Lamonier, E.A. Zhilinskaya, A. Aboukaïs, Appl. Catal., B 43 (2003) 261–271. [10] J. Zhu, Catalytic partial oxidation of methane to synthesis gas over ZrO2 -based defective oxides, Doctoral dissertation, University of Twente, The Netherlands, 2005. [11] J. Zhu, J.G. van Ommen, H.J.M. Bouwmeester, L. Lefferts, J. Catal. 233 (2005) 434–441. [12] F. Del Monte, W. Larsen, J.D. Mackenzie, J. Am. Ceram. Soc. 83 (2000) 628–634. ˜ [13] S.T. Korhonen, M.A. Banares, J.L.G. Fierro, A.O.I. Krause, Catal. Today 126 (2007) 235–247. [14] X. Wang, I. Wachs, Catal. Today 96 (2004) 211–222. [15] I.E. Wachs, J.-M. Jehng, W.J. Ueda, Phys. Chem. B 109 (2005) 2275–2284. [16] J.M. Tatibouët, Appl. Catal., A 148 (1997) 213–252. [17] M. Badlani, I.E. Wachs, Catal. Lett. 75 (2001) 137–149. [18] D. Bianchi, T. Chafik, M. Khalfallah, S.J. Teichner, Appl. Catal., A 123 (1995) 89–110. [19] I.A. Fisher, A.T. Bell, J. Catal. 184 (1999) 357–376. [20] K.T. Jung, A.T. Bell, J. Catal. 204 (2001) 339–347. [21] J. Zhu, J.G. van Ommen, L. Lefferts, Catal. Today 117 (2006) 163–167. [22] H.J.M. Bosman, A.P. Pijpers, W.M.A. Jaspers, J. Catal. 161 (1996) 551–559. [23] G. Busca, T. Montanari, C. Resini, G. Ramis, U. Costantino, Catal. Today 143 (2009) 2–8. [24] G. Busca, Catal. Today 27 (1996) 457–496. [25] F. Roozeboom, P.D. Cordingley, P.J. Gellings, J. Catal. 68 (1981) 464–472. [26] S. Korhonen, Effect of support material on the performance of chromia dehydrogenation catalysts, Doctoral dissertation, Helsinki University of Technology, Finland, 2008. ˜ [27] G. Garcia Cortez, J.L.G. Fierro, M.A. Banares, Catal. Today 78 (2003) 219–228. [28] L.E. Briand, W.E. Farneth, I.E. Wachs, Catal. Today 62 (2000) 219–229. [29] M. Sugioka, T. Nakyama, Y. Uemichi, T. Kanazuka, React. Kinet. Catal. Lett. 41 (1990) 345–350. [30] T.J. Toops, M. Crocker, Appl. Catal., B 82 (2008) 199–207. [31] A. Datta, R.G. Cavell, J. Phys. Chem. 89 (1985) 450–454. [32] M. Ziolek, J. Kujawa, O. Saur, J.C. Lavalley, J. Phys. Chem. 97 (1993) 9761–9766.

II Kauppi E. I., Kanervo J. M., Lehtonen J., Lefferts L., (2015) Interaction of H2 S with ZrO2 and its influence on reactivity of surface oxygen. Appl. Catal. B 164, 360-370. Reproduced with permission from Elsevier B. V.

Applied Catalysis B: Environmental 164 (2015) 360–370

Contents lists available at ScienceDirect


Interaction of H2 S with ZrO2 and its influence on reactivity of surface oxygen E.I. Kauppi a,∗ , J.M. Kanervo a , J. Lehtonen a , L. Lefferts a,b a Aalto University, School of Chemical Technology, Department of Biotechnology and Chemical Technology, Research Group Industrial Chemistry, P.O. Box 16100, 00076 Aalto, Finland b University of Twente, Faculty of Science & Technology and MESA+ Institute for Nanotechnology, P.O. Box 217, 7500 AE Enschede, The Netherlands

a r t i c l e

i n f o

Article history: Received 1 July 2014 Received in revised form 15 September 2014 Accepted 20 September 2014 Keywords: ZrO2 Biomass gasification gas clean-up H2 S adsorption TPD H2 -TPR CO-TPR

a b s t r a c t ZrO2 catalysts can be efficiently applied to convert tars into less harmful compounds in biomass gasification gas clean-up, also in the presence of H2 S. In fact, H2 S has even been observed to enhance naphthalene (a model compound for tar) oxidation activity on ZrO2 . Sulfur binding and its effect on the reactivity of surface oxygen species, including OH groups as well surface lattice oxygen, on ZrO2 was studied using temperature-programmed methods. At 30 ◦ C approx. half of the adsorbed H2 S was physisorbed and the other part dissociated titrating terminal OH groups. This dissociated H2 S was strongly adsorbed. Moreover, it was found that after treatment with H2 S until 400 ◦ C half of the adsorbed species were irreversibly adsorbed. H2 S adsorbed inducing desorption of water at two distinct temperatures (∼170 and 280 ◦ C) during temperature ramp. Thus, H2 S adsorption at elevated temperatures produced surface sulfur by replacement of surface lattice oxygen at two types of minority sites (max. 5% of a monolayer). Therefore we suggest that multicoordinated OH groups and surface lattice oxygen at defective sites are involved. CO-TPR revealed increased reactivity of surface lattice oxygen on ZrO2 with increasing amount of sulfur on the surface. In addition, strong adsorption of carbonates was suppressed by the presence of sulfur. The observed H2 S-induced enhancement of naphthalene oxidation in gasification gas clean-up is suggested to be caused by the increased reactivity of surface oxygen species in the vicinity of sulfur species on the surface. © 2014 Elsevier B.V. All rights reserved.

1. Introduction Syngas can be produced from biomass by gasification. The product gas can be targeted to energy, H2 , or liquid fuels production, affording a CO2 neutral route to produce fuels. The main challenge in the utilization of product gas from biomass gasification arises because of tars (modeled as toluene and naphthalene) present in the produced syngas. Sulfur compounds, such as H2 S and COS, and ammonia are also present as impurities. The content of tar in the gas is very low, but due to its tendency to plug downstream processing equipment it has to be removed from the gas to ppb level. A competitive way to convert tars to less harmful compounds is catalytic hot-gas cleaning, which can take place inside the gasifier or in a specially designed tar removal unit [1,2]. The types of catalysts studied include metal catalysts such as nickel, alkali metals, and precious metals (Rh, Pt) [3], and e.g. dolomites [1]. ZrO2 -based catalysts are

∗ Corresponding author. Tel.: +358 50 5300129. E-mail address: inkeri.kauppi@aalto.fi (E.I. Kauppi). http://dx.doi.org/10.1016/j.apcatb.2014.09.042 0926-3373/© 2014 Elsevier B.V. All rights reserved.

known to be active in tar decomposition when oxygen is added to the gas [4]. The level of H2 S in biomass gasification gas may be as high as 500 ppm [5], thus sulfur tolerance of the catalyst is essential. Metal catalysts are known to be poisoned severely by H2 S. Apparently, sulfur bonds so strongly to metal surfaces that marked loss of activity may occur at extremely low gas-phase concentrations [6]. For example, even 0.5–15 ppm of sulfur is enough to cause severe deactivation of conventional nickel steam reforming catalysts [7]. Sulfur tolerance of precious metals (Rh, Pt) may be somewhat higher [3]. Interestingly, studies on ZrO2 -based gasification gas clean-up catalysts have shown that H2 S may even have an enhancing effect on naphthalene conversion in catalytic hot-gas cleaning. Enhancement of naphthalene oxidation activity was observed mainly at 600 and 700 ◦ C on pure ZrO2 and Y2 O3 –ZrO2 , but not on SiO2 –ZrO2 . [2,3] Other studies also report H2 S having a promoting effect in catalysis on metal oxides [8–11]. Laosiripojana et al. [8] investigated steam reforming of methane over CeO2 -based catalysts in the presence of H2 S. Their studies indicated that an appropriate amount of

E.I. Kauppi et al. / Applied Catalysis B: Environmental 164 (2015) 360–370

H2 S promotes the reforming activity and, moreover, the enhancement effect was connected with the formation of various Ce–O–S phases during the reaction [8]. H2 S has also been found to affect the reaction between CH4 and CO2 over TiO2 and SiO2 supported Rh-catalysts, supposedly by selective poisoning of active sites [10]. Regarding oxidation reactions, Vincent et al. [9] reported recently a strong promoting effect of H2 S on methane oxidation over sulfur resistant metal oxide catalysts (a mixture of La0.4 Sr0.6 TiO3 and Y2 O3 –ZrO2 ) [9]. Chemical species adsorbed on the catalyst surface affect the catalytic activity profoundly [12]. However, it is not commonly known what kind of interactions with H2 S benefit catalyst operation. While it is known that sulfate (SOX ) species affect superacidic properties of ZrO2 promoting mainly isomerization or condensation type of reactions [12–14], it remains unclear what kind of sulfur derived surface species affect promotion of oxidation reactions on ZrO2 . Possibly, adsorption of S on ZrO2 modifies the surface chemistry and thus the catalytic properties, or S is involved via reactions with the species to be oxidized, offering new reaction pathways. Some studies suggest that sulfur interaction with lattice is connected with the promoting effects of H2 S on ZrO2 [2,11]. Adsorption of H2 S induces structural changes on ZrO2 via O2− atoms being replaced by S2− in the lattice, which has been estimated to increase basicity and/or redox properties of ZrO2 [11]. It has also been reported that when O sites are replaced by S the average Zr–O distance increases resulting in increased ionicity of the Zr–O bond [15]. In addition, H2 S dissociated surface species are expected to alter the properties of ZrO2 . Travert et al. [16] observed creation of very acidic surface hydroxyl groups (and consumption of lower coordinated ones) upon H2 S dissociative adsorption on ZrO2 . H2 S dissociated species have been found to occupy Zr4+ sites, thus lowering the Lewis acidity of ZrO2 [16,17]. The above mentioned H2 S-induced changes on the ZrO2 surface affect the adsorption of reactant molecules and thus, effects on catalytic activity are observed. Among characterization techniques to study H2 S adsorption, IR is relatively easily applied in situ and has been used to study the adsorption of H2 S on metal oxides, mainly Al2 O3 (e.g. [18]), but also ZrO2 (e.g. [11,16]). However, challenges arise due to the weak response of S–H in infrared. The stretching vibrations of S–H give weak IR bands due to their low extinction coefficient, and the bending vibration has even lower intensity in the region where many metal oxides are even not transparent to IR beam [16]. Regarding characterization of S in the lattice, Datta and Cavell [18] noted that it is not possible to monitor the presence of an M–S bond by infrared spectroscopy and suggested that XPS technique could be more suitably used for this purpose [18]. Sohn and Kim [12] studied modification of ZrO2 by sulfur compounds via infrared and photoelectron spectroscopies. By XPS (ex-situ) they were only able to confirm the presence of oxidized sulfur species (S6+ ) and they also noted that the reduced species are fully oxidized in O2 -containing atmosphere [12]. Laosiripojana et al. [8] did postreaction characterizations (XRD, XPS) on CeO2 -based reforming catalysts to study the effect of H2 S and information on sulfur-containing phases was attained. In the study it was also noted, that in situ XPS studies are needed to determine the oxidation states during reaction [8]. Such studies demand dedicated equipment due to contamination with sulfur species. Bringing the sulfided sample to ambient is thus likely to affect the amount and nature of sulfur on the surface and therefore render ex situ characterization ineffective. Oxidation of naphthalene is thought to involve surface oxygen on ZrO2 [2]. Thus, the aim of this work is find out how sulfur manipulates the reactivity of ZrO2 towards oxidation reactions. Oxygen is present on the ZrO2 surface in terminal or multicoordinated hydroxyl (t-OH, m-OH) groups and surface lattice oxygen with different level of unsaturation, i.e. on terraces and steps, kinks and

361

defects. Temperature-programmed techniques were used to discover the effect of adsorbed H2 S on the reactivity of oxygen on ZrO2 . 2. Experimental Monoclinic ZrO2 from MEL Chemicals was used as catalyst. The specific surface area was 24 m2 /g (measured by BET-method), which was reported in [19] among other physical and chemical properties. The density of O-atoms on the surface of ZrO2 was estimated to be 1.5·1019 O/m2 , which was calculated based the density of O-atoms in the unit cell assuming that (1 0 0) faces are exposed [19,20]. The purity of ZrO2 was studied by XPS analysis and no impurity metals were detected (except for Hf). TP-experiments were performed using a sample of 0.1 g (particle size 0.25–0.42 mm). The catalyst powder had been pre-calcined at 800 ◦ C for 1 h and pressed into pellets, which were crushed and sieved to the desired particle size. TP-experiments were carried out in an Altamira AMI-100 characterization system. The gaseous products were analyzed with a mass spectrometer (MS, Omnistar GSD320, Pfeiffer Vacuum). The experiments were carried out using calcined and reduced samples. Pre-treatment of the samples was done in-situ under O2 /He (10 vol-% O2 in He, AGA) at 600 ◦ C for 2 h (ramp rate 10 ◦ C/min) following reduction in 10 vol-% H2 /He (H2 99.999%, AGA) for 15 min to mimic the state of the surface during gasification gas clean up (the concentration of reducing gases in gasification gas is high). All further treatments were done in-situ and the sample was never exposed to ambient during the TP-cycles. 2.1. Temperature-programmed sulfiding (TPS) TPS was performed flowing 50 cm3 /min of H2 S/N2 (500 ppm of H2 S, AGA) through the catalyst bed first at 30 ◦ C for 0.5 h. The gas was purposely free of H2 to prevent reduction of the surface by H2 . However, the gas mixture may have contained ppm levels of water as impurity. After initial adsorption at 30 ◦ C, temperature was increased from 30 ◦ C to the end temperature at a rate of 10 ◦ C/min under similar H2 S flow. Sulfiding was continued at the end temperature for 1 h, whereafter the sample was cooled down under H2 S-containing flow. The sample was flushed with inert for 1 h at 30 ◦ C to remove weaker held species before probing of the H2 S-treated surface. H2 S was adsorbed on ZrO2 isothermally at 30 ◦ C (1 h) and via TPS at four different final temperatures of 100, 200, 300 and 400 ◦ C aiming to affect the amount (and nature) of H2 S adsorbed species on the surface. The differently sulfided samples are designated H2 Sads30 ◦ C , TPS30–100–30 ◦ C , TPS30–200–30 ◦ C , TPS30–300–30 ◦ C and TPS30–400–30 ◦ C . The consumption of H2 S during H2 S adsorption was calculated by numerically integrating the area of the H2 S signal compared to 500 ppm in the feed gas. 2.2. Probing the sulfided surface The sulfided ZrO2 surface was tested via temperatureprogrammed methods; temperature-programmed desorption (TPD), temperature-programmed reduction with hydrogen (H2 TPR), and temperature-programmed reduction with CO (CO-TPR). The TPD, H2 -TPR, and CO-TPR results with sulfided ZrO2 were compared to results obtained on unsulfided (calcined and reduced) ZrO2 to determine the effect of adsorbed H2 S. TPD was performed after flushing the sample at 30 ◦ C with N2 (50 cm3 /min, AGA, 99.999%) after TPS30–400–30 ◦ C for 1.5 h. Thereafter a temperature ramp was applied from 30 to 600 ◦ C under similar N2 flow (10 ◦ C/min).

362


Fig. 1. TPS profile (MS signals for H2 S, H2 O, and H2 ) from 30 to 400 ◦ C on ZrO2 (500 ppm H2 S in N2 ). In the figure also the isothermal adsorption stage at 30 ◦ C (0.5 h) is shown.

H2 -TPR was performed after TPS30–400–30 ◦ C . A gas mixture of H2 /N2 (50 cm3 /min, 4 vol-% H2 ) was admitted to the reactor at 30 ◦ C and let to dwell for 0.5 h. Then, a temperature ramp was applied under H2 -containing flow from 30 to 600 ◦ C (10 ◦ C/min) and the end temperature was maintained for 0.5 h. CO-TPR was performed after 1 h He flush (50 cm3 /min, AGA, 99.996%) at 30 ◦ C on the five differently sulfided ZrO2 samples described before. A total flow of 50 cm3 /min CO in He (5 vol-% CO) was passed through the samples first at 30 ◦ C for 0.5 h and during ramp from 30 to 600 ◦ C (10 ◦ C/min). The sample was let to dwell at 600 ◦ C for 0.5 h. The gas phase products were analyzed by MS. Compounds analyzed and their respective m/z values monitored during TPD and H2 -TPR runs were H2 , (m/z = 2), He (m/z = 4), water (m/z = 18), H2 S (m/z = 34, 33, and 32), CO or N2 (m/z = 28), CO2 (m/z = 44) and SO2 (m/z = 64 and 48). During CO-TPR, the production of any possible sulfided products from CO were also monitored (methanethiol, COS and CS2 , m/z = 45, m/z = 60, and m/z = 76, respectively). For calibrating the MS, one point calibration with constant feed was used (for H2 S 500 ppm H2 S/N2 , and for CO2 1 vol-% CO2 /He). 3. Results 3.1. Sulfiding of ZrO2 Temperature-programmed sulfiding (TPS) of the catalysts was carried out to gain information about H2 S adsorption sites and sulfur binding on the ZrO2 surface. H2 S was adsorbed on pre-reduced ZrO2 isothermally at 30 ◦ C and via TPS from ambient temperature to 100, 200, 300 and 400 ◦ C. Fig. 1 shows the result of a TPS run with end temperature of 400 ◦ C, where the dotted line indicates the baseline of the H2 S signal (based on a blank experiment), used to calculate the amounts of H2 S that adsorb and desorb at different stages of the temperature program. The results of TPS runs with final temperatures 100, 200, and 300 ◦ C are presented in supplementary information in Figs. S1, S2, and S3, respectively. The signal m/z = 34 is shown for H2 S and also the signals m/z = 33 and m/z = 32 for SH− and S2− showed similar profiles. H2 S was consumed immediately at 30 ◦ C when it was introduced into the reactor, which can be seen in Fig. 1 between 240 and 1000 s. When a temperature ramp was applied at approx. 2050 s, H2 S desorbed with a maximum at 55 ◦ C with minor formation of water. This was similar in all the experiments (shown in S1, S2,

and S3). During temperature ramp from 30 to 400 ◦ C, H2 S was consumed at approx. 170 and 280 ◦ C with concurrent formation of water (temperatures corresponding roughly to H2 O production and H2 S consumption maxima, Fig. 1). The same processes took place during TPS30–300–30 ◦ C as can be seen in Fig. S3 during temperature ramp. Simultaneous H2 S consumption and H2 O production was also seen during TPS experiments until 100 and 200 ◦ C (Figs. S1 and S2). The peak areas for desorbing water during ramp were calculated and correlated with adsorbing H2 S amounts, to confirm the relevance of the results. Their linear correlation is shown in S4. H2 S was also consumed during cooling, similarly for all experiments (not shown). The multiple processes for H2 S consumption indicate that at least three types of sites exist to adsorb H2 S on the ZrO2 surface. Apparently two of these sites are able to release oxygen, since separate maxima for water were detected concurrently with H2 S consumption. Finally, H2 S adsorbed at approx. 400 ◦ C with simultaneous formation of H2 in Fig. 1 (the amount of H2 being 0.26 ␮mol) indicating that at this temperature H2 S started to decompose forming H2 (no significant H2 signal was detected in all other experiments in Figs. S1–S3). Experiments with an empty reactor showed that this does not occur without a catalyst. It was also noted, that H2 was produced (exponential increase with H2 S consumption) during TPS until 600 ◦ C, also suggesting decomposition of H2 S rather than surface reaction. It is thought that under the reducing conditions of gasification gas clean-up elemental sulfur is not formed (other components in the gas mixture are likely to stabilize H2 S [21]), and therefore the maximum temperature in TPS experiments was limited to 400 ◦ C. Table 1 shows the amounts of H2 S adsorbed during the TPS experiments and the estimations of respective monolayer coverage (%, calculated as amount of H2 S per number of O atoms on the surface of ZrO2 sample). The “sum” indicates the amount that was retained on the catalyst, based on the cumulative amounts that were adsorbed, subtracting the amount that desorbed. The amount of H2 S adsorbed at 30 ◦ C was reproducible 3.5 ␮mol, corresponding to approx. 6% of a monolayer, in all experiments (Table 1). Approximately half of this amount was desorbed during temperature ramp at 55 ◦ C (Table 1), which can be attributed to desorption of physisorbed H2 S. The amount of H2 S adsorbed during cooling was similar, so apparently the process is reversible as would be expected for physisorption. Minor amounts of water also desorbed right after heating was started. The amount of adsorbed


363

Table 1 Calculated amounts (␮mol) of adsorbed and desorbed H2 S during adsorption at 30 ◦ C and TPS runs at varied final temperatures on 0.1 g ZrO2 . Value in parentheses indicates the ML amount (% of surface coverage).

At 30 ◦ C Desorption during ramp Activated adsorption Adsorption during cooling Sum

H2 Sads30 ◦ C

TPS30–100–30 ◦ C

TPS30–200–30 ◦ C

TPS30–300–30 ◦ C

TPS30–400–30 ◦ C

3.5 (5.8) – – – 3.5 (5.8)

3.6 (6) 1.7 (2.9) 0.4 (0.7) 1.2 (2) 3.5 (5.9)

3.5 (6) 1.7 (2.8) 1.4 (2.3) 1.8 (3) 4.7 (8.6)

3.5 (6) 1.7 (2.9) 2.6 (4.3) 1.6 (2.8) 6.0 (10.1)

3.5 (5.9) 1.7 (2.8) 3.2 (5.4) 1.6 (2.6) 6.5 (11)

H2 S species increased from 0.4 to 3.2 ␮mol by increasing sulfidation end temperature from 100 to 400 ◦ C (Table 1), indicating that H2 S adsorption was activated with temperature (adsorption at elevated temperatures will therefore be denoted activated adsorption). However, the increase was less significant when the end temperature was increased from 300 to 400 ◦ C, at which temperature H2 S started to decompose to H2 and S. Therefore, sulfidation capacity is apparently saturating around 400 ◦ C. TPS with end temperature of 400 ◦ C was also performed on pre-treated and wetted catalyst (a flow of 50 cm3 /min H2 O/He, 1000 ppm, was passed through the catalyst at 30 ◦ C for 0.5 h and thereafter flushed for 0.5 h prior to TPS). Wetting the catalyst only suppressed H2 S adsorption at low temperatures (30 ◦ C), whereas at higher temperatures, after additional H2 O had desorbed, the adsorption pattern was similar compared to that on the reduced catalyst (not shown). The results indicated that the amount of physisorbed H2 S was decreased by molecularly adsorbed water on the catalyst. 3.2. Reactivity of sulfided ZrO2 3.2.1. TPD and H2 -TPR; thermal stability and reactivity towards H2 TPD and H2 -TPR were performed on unsulfided ZrO2 and after TPS30–400–30 ◦ C . The results from both experiments will be presented and discussed together in order to describe the effect of sulfidation on the properties of ZrO2 . Fig. 2A shows H2 S desorption during TPD on unsulfided ZrO2 and ZrO2 after TPS30–400–30 ◦ C . As can be seen, H2 S desorbs from the sulfided sample with a sharp maximum at 80 ◦ C, which can be connected with residual molecularly adsorbed H2 S. Desorption of weakly held species was also seen during TPS (Fig. 1) after heating was started. In the beginning of the temperature ramp during TPD (Fig. 2B) the signal for water rises indicating that some molecular water was also present. Fig. 3 shows the results of H2 -TPR experiments on unsulfided and sulfided ZrO2 . The sulfided sample shows a maximum for H2 S at approx. 75 ◦ C corresponding to low temperature H2 S desorption during TPD and can therefore be assigned to desorption of residual molecular H2 S. Also water desorbed at low temperature after heating was started (Fig. 3B) which is assigned to molecularly adsorbed water. A broad maximum for desorbing H2 S was observed in Fig. 2A also at approx. 300 ◦ C during TPD after TPS30–400–30 ◦ C . Based on earlier H2 S-TPD studies on ZrO2 catalysts, when H2 S was adsorbed isothermally at 40 ◦ C, only one peak for desorbing H2 S was detected (maximum at 100 ◦ C) [17]. Therefore it is proposed that, the H2 S species that desorb at approx. 300 ◦ C after TPS30–400–30 ◦ C result from surface species formed via an activated process. Water desorbed during TPD on the unsulfided ZrO2 with a maximum at 360 ◦ C (Fig. 2B), whereas the sulfided ZrO2 (via TPS30–400–30 ◦ C ) hardly showed any H2 O desorption. Particularly it can be seen that the peak at approx. 360 ◦ C in Fig. 2B, which is attributed to dehydroxylation of ZrO2 [22–24], was diminished on the sulfided sample. Apparently there are less hydroxyl groups that are able to react to water on the sulfided surface. It is thus suggested that H2 S adsorption titrated hydroxyl groups on the surface, which is in line with data from Travert et al. [16].

During H2 -TPR on the sulfided sample a sharp maximum for H2 S appeared at approx. 320 ◦ C with simultaneous formation of H2 O and consumption of H2 (Fig. 3). The amount of H2 S formed at 320 ◦ C was calculated to be 2.5 ␮mol and that of H2 O 6 ␮mol for the peak at 320 ◦ C, based on rough estimation. On the other hand, H2 consumption was calculated to be 11.4 ␮mol. The sulfided sample also showed a maximum for water at 375 ◦ C. It is probable that part of the H2 is consumed in this process (reduction of surface oxygen species), since the amount of H2 consumed is higher than the calculated molar amounts of H2 S and H2 O formed at 320 ◦ C. On the unsulfided sample water was formed during H2 -TPR at 350 ◦ C (Fig. 3B) which is suggested to originate from dehydroxylation, similarly as during the TPD experiments. Furthermore, the consumption of H2 on the unsulfided ZrO2 was zero, since no clear change was observed in the H2 curve (ZrO2 is not reducible below 600 ◦ C). Only the sulfided sample showed a distinct minimum in the H2 curve (320 ◦ C, Fig. 3C) indicating consumption of hydrogen. No SO2 formation was detected on the sulfided samples during TPD nor H2 -TPR suggesting that ZrO2 , unlike e.g. TiO2 [10,25], is not able to provide oxygen to form SO2 . The amount of desorbing H2 S during TPD was calculated to be 2.1 ␮mol, corresponding to 29% of the total amount of H2 S that was present on the surface after TPS30–400–30 ◦ C . The total amount of H2 S desorbed during H2 -TPR corresponds to 53% of the total amount of H2 S that was left on the surface after TPS. An additional temperature-programmed oxidation experiment was performed after TPD to check if any S species remained on the surface after TPD. The results showed SO2 formation at room temperature right after O2 /He (10 vol-%) feed was started, so evidently sulfur species remained which are easily oxidized at room temperature.

3.2.2. CO-TPR on sulfided vs. unsulfided ZrO2 The surfaces sulfided to different extent (H2 Sads30 ◦ C , TPS30–100–30 ◦ C , TPS30–200–30 ◦ C , TPS30–300–30 ◦ C and TPS30–400–30 ◦ C ) were studied in CO-TPR in order to determine the reactivity of the catalysts towards CO. Fig. 4A and B show CO2 , H2 S, H2 O and H2 generation during COTPR experiment on unsulfided and sulfided ZrO2 (TPS30–400–30 ◦ C ). CO consumption cannot be reported because of a too low signal to noise ratio. Fig. 4A and B show that H2 O desorbs from the unsulfided ZrO2 showing no significant maxima, whereas maxima are seen on the sulfided ZrO2 at approx. 190 and 380 ◦ C. CO was not found to react directly with adsorbed H2 S on the sulfided surface, i.e. no products containing C and S could be detected with MS. On the other hand, it is possible that if such products were formed they were retained on the surface even after 600 ◦ C. Remarkably, the sulfided ZrO2 (TPS30–400–30 ◦ C ) clearly shows an additional peak for CO2 at low temperature (190 ◦ C) in Fig. 4B as compared to unsulfided sample in Fig. 4A, for which only one CO2 peak is detected at high temperature, concurrently with a H2 peak. Instead the low-temperature generation of CO2 occurs with simultaneous formation of H2 S and H2 O. H2 was generated simultaneously with CO2 having maxima at 480 and 535 ◦ C on unsulfided and sulfided ZrO2 , respectively.

364


Fig. 2. MS signal for A) H2 S (m/z = 34) and B) H2 O (m/z = 18) during TPD from 30 to 600 ◦ C (ramp rate 10 ◦ C/min) on unsulfided and sulfided ZrO2 (via TPS30–400–30 ◦ C ).

Table 2 shows temperatures at which maxima were detected for CO2 , H2 O, H2 and H2 S on unsulfided and differently sulfided ZrO2 samples. CO2 and H2 are generated at high temperature (∼530 ◦ C) on all the sulfided ZrO2 samples, corresponding maxima are seen on the unsulfided ZrO2 at 480 ◦ C. The additional maximum for CO2 at 190 ◦ C is seen on all the sulfided catalysts. The intensity of the peak clearly increases with increasing sulfidation temperature and on the sample sulfided at 30 ◦ C this is barely noticeable (Fig. 5). Minor amounts of water desorbed during the CO-TPR experiments on all the catalysts. H2 S desorbed at 190 ◦ C after TPS30–200–30 ◦ C , TPS30–300–30 ◦ C , and TPS30–400–30 ◦ C . The amounts of CO2 generated on unsulfided and differently sulfided ZrO2 as shown in Fig. 5 are presented in Table 3. The total amount of oxygen atoms removed from the ZrO2 surface by CO

oxidation was no more than 5% of a ML (% of the total number of oxygen atoms in one monolayer on the ZrO2 surface). It can be seen that increasing the sulfiding temperature (and therefore the amount of activatedly adsorbed species) caused an increase in the amount of the CO2 produced at 190 ◦ C as already discussed above. On the other hand, H2 S adsorption isothermally at 30 ◦ C and via TPS at 100 ◦ C produced additional high temperature CO2 maxima at 500 and 440 ◦ C, respectively, during subsequent CO-TPR (Fig. 5). These appeared without simultaneous H2 production. H2 S treatment causes the amount of CO2 to increase, independent of the end temperature of TPS. The possible presence of sulfate (SOX ) species was considered to affect the differing properties of ZrO2 after sulfidation and thus the CO-TPR experiment was performed also on sulfated ZrO2 (ZrO2

Table 2 CO2 , H2 O, H2 and H2 S (m/z = 44, 18, 2 and 34, respectively) maxima (◦ C) at unsulfided and differently sulfided ZrO2 surfaces during CO-TPR (5 vol-% CO).

CO2 H2 O H2 H2 S

Unsulfided

H2 Sads30 ◦ C

TPS30–100–30 ◦ C

TPS30–200–30 ◦ C

TPS30–300–30 ◦ C

TPS30–400–30 ◦ C

480 170, 250, 360 480 –

(190), 500, 530 60, 350 530 –

190, 440, 530 70, 180, 290, 370 530 –

190, (315, 425), 535 75, 175, (300, 475) 535 190

190, (310, 390), 535 70, 170, 390 535 190

190, (260, 390), 535 (80), 180, (390) 535 (80), 190


365

Fig. 3. MS signal for (A) H2 S (m/z = 34), (B) H2 O (m/z = 18), and (C) H2 (m/z = 2) during H2 -TPR from 30 to 600 ◦ C (ramp rate 10 ◦ C/min) on unsulfided and sulfided ZrO2 (via TPS30–400–30 ◦ C ).

366


Fig. 4. Formation of CO2 , H2 S, H2 O and H2 (MS signals m/z = 44, 34, 18 and 2, respectively) during CO-TPR on (A) unsulfided and (B) sulfided via TPS30–400–30 ◦ C ZrO2 .

treated with H2 SO4 ). The results showed CO2 formation with simultaneous release of SO2 at 570 ◦ C, a temperature which is consistent with literature [26]. Therefore it is evident that the interaction of sulfided zirconia with CO is not related to the presence of any sulfates on the surface. 4. Discussion This discussion concentrates on how H2 S adsorption modifies the surface of monoclinic ZrO2 and how these modifications affect

the reactivity of surface oxygen. By surface oxygen we mean terminal or multi-coordinated hydroxyls on the surface (t-OH, m-OH) and surface lattice oxygen with different level of unsaturation, i.e. on terraces as well as defective sites. Products evolved during H2 TPR and CO-TPR on samples sulfided to different degree provide information on the ability of the surface to supply oxygen for oxidation reactions and indirectly on possible sulfur surface species that induce these changes. CO-TPR results showed that adsorbed sulfur species do not form any gas-phase products with CO in the studied temperature range. This confirms that the effects discussed

Table 3 Amount of CO2 generated (␮mol) during CO-TPR on unsulfided and differently sulfided catalysts. The value in parentheses indicates the quantity of oxygen removed from the ZrO2 surface (% of a ML). CO2 (␮mol)

Unsulfided

H2 Sads30 ◦ C

TPS30–100–30 ◦ C

TPS30–200–30 ◦ C

TPS30–300–30 ◦ C

TPS30–400–30 ◦ C

∼190 ◦ C ∼500 ◦ C Tot

0 (0) 2.7 (2.3) 2.7 (2.3)

0.3 (0.3) 4.0 (3.4) 4.3 (3.7)

0.9 (0.7) 3.7 (3.1) 4.9 (4.1)

1.6 (1.4) 2.2 (1.9) 5.5 (4.6)

2.6 (2.2) 2.1 (1.8) 6.1 (5.1)

2.9 (2.4) 2.1 (1.8) 6.2 (5.2)


367

Fig. 5. CO2 generation on unsulfided and differently sulfided ZrO2 during CO-TPR from 30 to 600 ◦ C (10 ◦ C/min, 5 vol-% CO in He).

below are due to incorporation of S in ZrO2 , instead of opening new reaction pathways via S-containing intermediates.

4.1. H2 S adsorption on ZrO2 H2 S was adsorbed isothermally on ZrO2 at 30 ◦ C and in a temperature-programmed mode from 30 to 100, 200, 300 and 400 ◦ C. Approximately half of the H2 S adsorbed isothermally at 30 ◦ C was physisorbed and relatively easily removed from the surface at 55 ◦ C (the instability of physisorbed H2 S has been indicated also earlier [17,18]). This process is reversible since the amount adsorbed during cooling corresponds to the physisorbed amount (Table 1). Desorption of the weakly held, physisorbed H2 S was seen during TP-treatments on the sulfided catalysts in the initial part of the temperature ramp (TPD, H2 -TPR and CO-TPR in Figs. 2A, 3A, and 4B, respectively). Molecularly adsorbed water on the surface seems to hinder the adsorption of physisorbed H2 S, based on the TPS result on wetted surface (not shown). In addition to physisorbed H2 S, half of the H2 S species adsorbed at 30 ◦ C were retained on the surface when heating was started during TPS (Table 1). Catalyst properties were affected even after isothermal adsorption at 30 ◦ C (as observed with CO-TPR, Fig. 5), confirming H2 S dissociation and formation of stable surface species, even at this low temperature. Travert et al. studied H2 S and CH3 SH adsorption on metal oxides, including ZrO2 , by FTIR [16]. They observed disappearance of original high frequency (OH) band and increase and shift of the low-frequency (OH) band. At the same time a broad band accompanied with a bending HOH vibration was observed. Weak bands for SH were also observed at low temperature. [16] Thus, dissociation of H2 S at low temperature occurs on terminal hydroxyl (t-OH) groups (Eq. 1) and coordinatively unsaturated surface oxygen anions (O2− ) generating new multicoordinated hydroxyl (m-OH) groups and SH groups (Eq. 2). t-OH(s) + H2 S(g) → H2 O(g) + SH(s)

(1)

O2− (s) + H2 S(g) → m-OH(s) + SH(s)

(2)

Water desorbed in the beginning of the temperature ramp during the TPS experiments (Fig. 1). TPD results after TPS30–400–30 ◦ C showed desorption of minor amounts of water, and the water peak corresponding to dehydroxylation at 340 ◦ C (Fig. 2B) was nonexistent. Dehydroxylation involves t-OH and m-OH groups [22–24], so it is suggested that at 30 ◦ C H2 S dissociated, titrating terminal hydroxyl groups on the surface via Reaction 1. H2 S may also dissociate via Reaction 2 forming new m-OH groups. Activated H2 S uptake took place during TPS, generating water at two different temperatures (∼170 and 280 ◦ C, Fig. 1), suggesting reaction with surface oxygen. It has been proposed that several metal oxides, including ZrO2 , are able to exchange surface lattice oxygen with S from adsorbing H2 S via following reaction H2 S + O2− surf. → H2 O + S2− surf. [11,25,27–29]. Ziolek et al. [11] suggested that this affects increase of basicity or redox properties on ZrO2 . For ZrO2 , the surface oxygen sites available for this type of exchange would be, in principle, m-OH groups, surface lattice oxygen at terraces, or surface lattice oxygen at edges or low coordination sites. The observation that the maximal amount of H2 S adsorbed is in the order of 5% of a ML (Table 1), is suggesting that minority sites, i.e. defective sites, are involved instead of oxygen at terraces. Therefore we suggest that TPS-induced incorporation of sulfur on the surface is replacing oxygen at m-OH groups at lower temperature (170 ◦ C) and surface lattice oxygen at low coordination sites at the higher end of the temperature ramp. Sulfur in the lattice might also result via an activated conversion of –SH groups formed via H2 S dissociated at lower temperatures. According to Beck et al. [25], H2 S exposure to TiO2 (rutile) above 350 ◦ C forms mostly strongly bound sulfide species which can remain on the surface or diffuse into the lattice [25]. Also in our study, diffusion of sulfide (S2− ) into the bulk lattice cannot be completely excluded. The H2 S uptake leveled off between 300 and 400 ◦ C. Also H2 S decomposition started at temperatures above 360 ◦ C resulting in deposition of elemental S, based on the occurrence of H2 generation in Fig. 1. It is known that bulk sulfide (ZrS2 ) does not form upon exposure to H2 S [30], which is also consistent with the small amounts of H2 S adsorbed during TPS experiments. In principle, XRD could provide information on structure modifications by sulfur.

368


Fig. 6. Relationship between surface sulfur content of ZrO2 and the amount of surface oxygen removed by CO oxidation at approx. 190 ◦ C during CO-TPR on differently sulfided catalysts.

However, the amount of sulfur deposited on the surface calculated by integration of the TPS profiles was showing sub-monolayer quantities of adsorbed H2 S predicting that only specific sites on ZrO2 surface are interacting. Since no bulk effects are expected, also XRD characterization of the sulfided ZrO2 material is unlikely to reveal any distortions in bond lengths and so to give any additional information. 4.2. Stability of adsorbed species According to mass balances during TPD and H2 -TPR experiments only approximately 30% of the retained H2 S desorbed and 50% could be removed by H2 . Thus, the adsorbed S species are highly stable under reducing conditions, suggesting very strong adsorption. In contrast, presence of oxygen would result in removal via SOX formation even at mild temperatures, confirming that any exsitu characterization including exposure to ambient cannot provide further insight. Water desorption maxima during H2 -TPR (Fig. 3) and TPD (Fig. 2) on unsulfided ZrO2 were very similar (350 and 340 ◦ C, respectively). In line with this, no H2 consumption was detected, indicating that unsulfided ZrO2 does not contain any reducible surface oxygen species. In contrast, H2 consumption was detected at 320 ◦ C on sulfided ZrO2 , along with H2 S and H2 O production. Additional H2 O was produced on the sulfided ZrO2 also at 375 ◦ C. Furthermore, a significantly smaller dehydroxylation peak is observed on the sulfided sample in TPD (Fig. 2). Thus, water formation during H2 -TPR is significantly enhanced on sulfided ZrO2 . Simultaneous production of H2 S and H2 O at 320 ◦ C can be attributed to reaction between dissociated H2 (via Reaction 3) and neighboring SH and OH groups (Reactions 4 and 5, respectively). H2 (g) + 2 ∗ (s) → 2H ∗ (s)

(3)

H ∗ (s) + SH(s) → H2 S(g) + ∗(s)

(4)

H ∗ (s) + OH(s) → H2 O(g) + ∗(s)

(5)

Since H2 S and H2 O are produced in the gas-phase simultaneously when H2 is adsorbed, its dissociation (Reaction 3) must be the rate-limiting step.

It should be noted that during (and after) the TPD and H2 -TPR experiments the sample still contain S, which probably is responsible for the enhanced reducibility of the sample. Anderson et al. [15], suggested that S incorporated in the ZrO2 lattice decreases the Zr–O bond strength since the volume of the unit cell increases. Therefore it is considered, that the reactive oxygen species are present in the lattice, probably at the surface. Alternatively, it may be argued that surface m-OH, which otherwise would not be reducible, are being destabilized by the presence of S. 4.3. Reactivity of oxygen on sulfided ZrO2 Unmodified ZrO2 is known to bind CO molecularly on surface Zr cationic sites [22,24] and as stable formates on terminal hydroxyls via an activated equilibrium process (Eq. (6) describes formate formation) [31–33]. CO + Zr–OH ↔ Zr–HCOO

(6) 200 ◦ C.

Molecular CO desorbs already below Decomposition of formate occurs under CO atmosphere by 500 ◦ C yielding CO2 and H2 in the gas phase [22,31,34,35], whereas in the absence of CO formate decomposes to CO and terminal hydroxyls as Reaction 6 is an equilibrium reaction. In this study CO2 and H2 were detected on unsulfided ZrO2 at 480 ◦ C during CO-TPR, whereas these maxima were shifted to approx. 530 ◦ C after all the sulfidation treatments (Table 3). Simultaneous generation of CO2 and H2 is attributed to decomposition of formate, reducing the ZrO2 surface (reductive decomposition). The formation of formates requires (terminal) hydroxyl groups [33,34] whereas their reductive decomposition requires additionally m-OH [36]. The higher temperature of formate decomposition on the sulfided surfaces indicates higher stability of formate on those catalysts or, alternatively shortage of t-OH and/or m-OH since they react with H2 S as discussed above. Additional CO2 formation, without concurrent formation of any other product, was seen at 500 and 440 ◦ C on ZrO2 after isothermal H2 S adsorption at 30 ◦ C and after TPS30–100–30 ◦ C , respectively (Fig. 5). CO2 can be formed via decomposition of surface carbonate (bidentate or monodentate) [32,34], which are formed on the surface via reaction of CO with surface lattice oxygen and decompose at temperatures above 300 ◦ C [31,32,34]. Thus, additional


generation of CO2 on the ZrO2 samples after H2 S adsorption at 30 ◦ C or TPS30–100–30 ◦ C may be caused by CO reacting with surface oxygen during CO-TPR. We propose that the reactivity of the zirconia is modified by the presence of sulfur species adsorbed at low temperature (possibly SH species), very similar to as observed with H2 -TPR, discussed above. The formed CO2 is retained on the surface as monodentate or polydentate carbonates, which are known to be the most stable form of adsorbed CO2 and are also adsorbed on c.u.s. O2− centers [19,37,38] (i.e. surface lattice oxygen at defects). The decrease of CO2 desorption temperature on increasing the sulfidation degree (TPS30–100–30 ◦ C compared to H2 Sads30 ◦ C ) is suggested to be due to the increasing concentration of sulfur species on the surface. Sulfiding at elevated temperatures caused a new lowtemperature (∼190 ◦ C) CO2 maximum during subsequent CO-TPR (Fig. 5). The amount of oxygen removed by CO oxidation at this temperature (Table 3) increased with the amount of activatedly adsorbed H2 S (Table 1) and a reasonable linear correlation is presented in Fig. 6. The slightly nonlinear end of the curve confirms that H2 S adsorption via this activated process (amount of surface sulfur) saturated after TPS at 300 ◦ C and H2 S decomposition to elemental S started when further increasing temperature. The activatedly adsorbed H2 S species are very strongly bound on the surface by replacing O2− by S2− at low coordination sites. Therefore, increasing sulfidation temperature creates more lattice sulfur that increasingly destabilizes oxygen in the surface, enhancing the reactivity with CO. The same effect was observed during H2 -TPR, as surface oxygen species became reactive towards H2 (see Fig. 3). Additionally, the low temperature of CO2 desorption indicates that the sites where CO2 adsorbs as mono/bidentate carbonate are blocked on the sulfided surface. Naphthalene oxidation is expected to involve surface oxygen (the mechanism has been suggested to follow the Mars–van Krevelen mechanism [39]), so it is probable that its enhancement in the presence of H2 S is connected with the pronounced reactivity of surface oxygen on the sulfur-containing surface. Thus, the observed effects of H2 S on ZrO2 can explain the reasons for naphthalene oxidation enhancement in the presence of H2 S during gasification gas clean-up. The study of Rönkkönen et al. [2] suggested that enhancement of naphthalene conversion was also connected to an adsorbed form of sulfur on the surface, since the effect remained after H2 S was removed from the atmosphere [2]. It is thought that under reducing conditions of the gasification gas-mixture the nature of sulfur compounds on the catalyst surface would be similar to the surface species in this study. Therefore, the effect is suggested to be caused by sulfur surface species that have replaced surface oxygen on ZrO2 and enhance reactivity of surface lattice oxygen.

5. Conclusions Temperature-programmed sulfiding study on ZrO2 showed that at low temperature half of the H2 S adsorbed can be attributed to physisorbed H2 S, which is weakly adsorbed and leaves the surface when heating. The other half dissociates titrating terminal OH groups on the surface. The dissociated H2 S at 30 ◦ C was also found to be strongly adsorbed. It is concluded that sulfur exchanges with surface oxygen in activated processes based on water formation at two distinct temperatures (∼170 and 280 ◦ C) during TPS until 400 ◦ C. It is suggested that sulfur replaces oxygen at lower temperature with m-OH sites and at higher temperature with surface lattice oxygen at defective sites (minority sites). Minority sites on ZrO2 surface are able to bind sulfur more preferably than oxygen, and moreover, sulfur adsorption on these sites enhances reactivity of surface lattice oxygen. Sulfur on the surface increased reactivity of surface lattice oxygen (reducibility of the surface) with

369

increasing amount of surface sulfur. In addition, H2 S was found to block CO2 adsorption sites, i.e. sites for carbonate formation, on ZrO2 . It is suggested that the observed enhancement of oxidation reactions during gasification gas clean-up is connected with increased reactivity of surface oxygen on sulfided ZrO2 surfaces. Acknowledgments The authors thank Prof. Outi Krause for critical reading of the manuscript and fruitful discussions. Ms. Sonja Kouva, Ms. Ella Rönkkönen, and Ms. Tiia Viinikainen are also warmly thanked for inspiring discussions. Ms. Heidi Meriö-Talvio is acknowledged for assistance during experimental work. MEL Chemicals is thanked for donating the zirconia samples. Financing from the Ministry of Education of Finland and the Academy of Finland are gratefully acknowledged. The Finland Distinguished Professor Programme (FiDiPro) funded by the Finnish Funding Agency for Technology and Innovation (TEKES) is acknowledged for financial support. Appendix A. Supplementary data Supplementary material related to this article can be found, in the online version, at http://dx.doi.org/10.1016/j.apcatb. 2014.09.042. References [1] P. Simell, E. Kurkela, P. Ståhlberg, J. Hepola, Catal. Today 27 (1996) 55–62. [2] H. Rönkkönen, P. Simell, M. Reinikainen, O. Krause, Top. Catal. 52 (2009) 1070–1078. [3] H. Rönkkönen, P. Simell, M. Niemelä, O. Krause, Fuel Process. Technol. 92 (2011) 1881–1889. [4] S. Juutilainen, P. Simell, O. Krause, Appl. Catal. A 62 (2006) 86. [5] W. Torres, S.S. Pansare, J.G. Goodwin, Catal. Rev. Sci. Eng. 49 (2007) 407–456. [6] C.H. Bartholomew, P.K. Agrawal, J.R. Katzer, Adv. Catal. 31 (1982) 135–242. [7] J.R. Rostrup-Nielsen, in: J.R. Anderson, M. Boudart (Eds.), Catalysis, Science and Technology, Springer-Verlag, Berlin, 1984, pp. 95–104. [8] N. Laosiripojana, S. Charojrochkul, P. Kim-Lohsoontorn, S. Assabumrungrat, J. Catal. 276 (2010) 6–15. [9] A.L. Vincent, J.-L. Luo, K.T. Chuang, A.R. Sanger, Appl. Catal. B 106 (2011) 114–122. [10] A. Erdöhelyi, K. Fodor, T. Szailer, Appl. Catal. B 53 (2004) 153–160. [11] M. Ziolek, J. Kujava, O. Saur, J.C. Lavalley, J. Mol. Catal. A 97 (1995) 49–55. [12] J.R. Sohn, H.W. Kim, J. Mol. Catal. 52 (1989) 361–374. [13] C.R. Vera, C.L. Pieck, K. Shimizu, J.M. Parera, Appl. Catal. A 230 (2002) 137–151. [14] G.D. Yadav, J.J. Nair, Microporous Mesoporous Mater. 33 (1999) 1–48. [15] J.R. Anderson, H. Kleinke, H.F. Franzen, J. Alloys Compd. 259 (1997) L14–L18. [16] A. Travert, O.V. Manoilova, A.A. Tsyganenko, F. Maugé, J.C. Lavalley, J. Phys. Chem. B 106 (2002) 1350–1362. ˜ [17] E.I. Kauppi, E.H. Rönkkönen, S.M.K. Airaksinen, S.B. Rasmussen, M.A. Banares, A.O.I. Krause, Appl. Catal. B. 111–112 (2012) 605–613. [18] A. Datta, R.G. Cavell, J. Phys. Chem. 89 (1985) 450–454. [19] T. Viinikainen, H. Rönkkönen, H. Bradshaw, H. Stephenson, S. Airaksinen, M. Reinikainen, P. Simell, O. Krause, Appl. Catal. A 362 (2009) 169. [20] T. Viinikainen, I. Kauppi, S. Korhonen, L. Lefferts, J. Kanervo, J. Lehtonen, Appl. Catal. B 142–143 (2013) 769–779. [21] T. Song, M. Zhang, 4th International Conference on Bioinformatics and Biomedical Engineering (iCBBE), Chengdu, China, 18–20 June, 2010, pp. 1–4. [22] V. Bolis, C. Morterra, M. Volante, L. Orio, B. Fubini, Langmuir 6 (1990) 695–701. [23] D. Bianchi, T. Chafik, M. Khalfallah, S.J. Teichner, Appl. Catal. A 123 (1995) 89–110. [24] C. Morterra, L. Orio, Mater. Chem. Phys. 24 (1990) 247–268. [25] D.D. Beck, J.M. White, C.T. Ratcliffe, J. Phys. Chem. 90 (1986) 3123–3131. [26] T. Zhu, L. Kundakovic, A. Dreher, M. Flytzani-Stephanopoulos, Catal. Today 50 (1999) 381–397. [27] C.L. Liu, T.T. Chuang, I.G. Dalla Lana, J. Catal. 26 (1972) 474–476. [28] A.M. Deane, D.L. Griffiths, I.A. Lewis, J.A. Winter, A.J. Tench, J. Chem. Soc. Faraday Trans. 1 (75) (1975) 1005. [29] T.J. Toops, M. Crocker, Appl. Catal. B 82 (2008) 199–207. [30] A. Clearfield, J. Am. Chem. Soc. 80 (1958) 6511–6513. [31] W. Hertl, Langmuir 5 (1989) 96–100. [32] J. Kondo, H. Abe, Y. Sakata, K. Maruya, K. Domen, T. Onishi, J. Chem. Soc. Faraday Trans. 1 (842) (1988) 511–519.

370


[33] P.O. Graf, D.J.M. de Vlieger, B.L. Mojet, L. Lefferts, J. Catal. 262 (2009) 181–187. [34] D.G. Rethwisch, J.A. Dumesic, Langmuir 2 (1986) 73–79. [35] X. Mugniery, T. Chafik, M. Primet, D. Bianchi, Catal. Today 52 (1999) 15–22. [36] S. Kouva, J. Andersin, K. Honkala, J. Lehtonen, L. Lefferts, J. Kanervo, Phys. Chem. Chem. Phys. 16 (38) (2014) 20650–20664.

[37] B. Bachiller-Baeza, I. Rodriguez-Ramos, A. Guerrero-Ruiz, Langmuir 14 (1998) 3556–3564. [38] P. Stelmachowski, S. Sirotin, P. Bazin, F. Maugé, A. Travert, Phys. Chem. Chem. Phys. 15 (2013) 9335–9342. [39] A. Bampenrat, V. Meeyoo, B. Kitiyan, P. Rangsungvigit, T. Rirksomboon, Catal. Commun. 9 (2008) 2349–2352.

Supplementary information for: Interaction of H2S with ZrO2 and its influence on reactivity of surface oxygen E. I. Kauppi1*, J. M. Kanervo1, J. Lehtonen1, L. Lefferts1,2 1

Aalto University, School of Chemical Technology, Department of Biotechnology and Chemical Technology, Research Group Industrial Chemistry, P.O. Box 16100, FI-00076 Aalto, Finland.

2

University of Twente, Faculty of Science & Technology and MESA+ Institute for Nanotechnology, P.O. Box 217, 7500 AE Enschede, The Netherlands Fax: +358 50 5300129; *e-mail: [email protected]

Figure S1. TPS profile (MS signals for H2S, H2O, and H2) from 30 to 100 °C on ZrO2 (500 ppm H2S in N2). In the figure also the isothermal adsorption stage at 30 °C (0.5 h) is shown.



Figure S4. The area of desorbing H2O as a function of adsorbed H2S amount during TPS at increasing end temperatures.

III Kauppi E. I., Honkala K., Krause A. O. I., Kanervo J. M., Lefferts L. ZrO2 acting as a redox catalyst. Accepted for publication in Topics in Catalysis.

ZrO2 acting as a redox catalyst E. I. Kauppi1, K. Honkala2, A. O. I. Krause1, J. M. Kanervo1, L. Lefferts1,3 1

Aalto University, School of Chemical Technology, Department of Biotechnology and Chemical Technology, Research Group Industrial Chemistry, P.O. Box 16100, FI-00076 Aalto, Finland 2

University of Jyväskylä, Department of Chemistry, Nanoscience Center, P. O. Box 35, 40014 Jyväskylä, Finland

3

University of Twente, Faculty of Science & Technology and MESA+ Institute for Nanotechnology, P.O. Box 217, 7500 AE Enschede, The Netherlands *Leon Lefferts [email protected]

Abstract Surface defects are discussed and reviewed with regards to the use of ZrO2 in applications involving interactions with CO, H2, CH4, CO2, water and hydrocarbons. Studies of catalytic partial oxidation of methane reveal that part of the surface lattice oxygen in terraces can be removed by methane at high temperatures (e.g. 900 °C). The reaction proceeds via a surface confined redox mechanism. The studies presented here also highlight that defects play a decisive role in the water-gas-shift reaction, since the reaction is likely carried out via OH groups present at defect sites, which are regenerated by dissociating water. Hydroxyl chemistry on ZrO2 is briefly reviewed related to the studies presented. Finally, new density functional theory calculations were conducted to find out how H2S interacts with ZrO2 surface (defect sites), in order to explain enhancement of activity in naphthalene and ammonia oxidation by H2S. Molecularly adsorbed H2S as well as terminal SH species (produced by dissociation of H2S) cannot be responsible for enhanced reactivity of surface oxygen. In contrast, multi-coordinated SH induced a relatively weak increase of the reactivity of neighboring OH groups according to thermodynamic calculations. Probably, the right active site responsible for the observed H2S-induced enhancement of oxidation activity on ZrO2 is yet to be discovered.

Keywords: ZrO2, CPOM, WGS, tar oxidation, H2S, hydroxyl groups, redox

1

1 Introduction This review presents an overview of research performed at Aalto University and the University of Twente on studies on ZrO2 catalyzed oxidation over the past 10 years. The reactions studied all targeted production of hydrogen and synthesis gas (syngas, a mixture of H2 and CO), via partial oxidation of methane (CPOM), water-gas-shift (WGS) and tar decomposition as part of clean-up of biomass gasification gas. Pure ZrO2 is a stable oxide that cannot be reduced under the conditions relevant for catalysis. However, it is claimed that surface oxygen can be removed in vacuum at temperatures above 700 °C [1, 2]. Reduction of ZrO2 has been suggested to follow a mechanism according to which surface OH groups formed by dissociatively activated hydrogen react to form water, and therefore the role of OH groups in directing the catalytic properties of the oxide should be considered [2]. So, reactivity of surface oxygen on ZrO2 is connected to the formation of OH groups, which will always be present on the surface under conditions relevant to catalysis. Part of the surface oxygen (coordinatively unsaturated O2-) on ZrO2 can also be reduced with H2 producing water and leaving an oxygen vacancy, which is again quenched in the presence of water [3]. Water can also dissociate to yield OH groups on the surface, so it seems that there is an equilibrium between H2, H2O, surface hydroxyl groups and oxygen vacancies depending on the reaction conditions. The data discussed in this review include ZrO2 with and without yttrium doping, as well as different crystal structures. Despite the fact that doping will have an important influence on the properties of ZrO2, we do not make systematic distinction because generally the properties discussed here are not influenced qualitatively. The goal of this short review is to combine information available now on ZrO2 catalyzed reactions (CPOM, WGS and tar decomposition) in which redox cycles at the surface of ZrO2 are involved, focusing on the role of specific surface sites including OH groups. In addition, and connected to this we will present new results on DFT calculations in order to explain the improved performance of ZrO2 in tar decomposition in the presence of H2S.

2 Partial oxidation of methane Syngas is traditionally produced via steam-reforming of methane (SR), which yields a H2-rich syngas. However, if the syngas is targeted to Fischer-Tropsch (FT) synthesis, a more favorable H2/CO-ratio is attained by CPOM [4]. FT technology has become very important since it provides an alternative route to liquid fuels based on e.g. natural gas [5]. Another advantage of CPOM is the fact that the reaction (CH4 +1/2 O2 Æ CO + 2 H2) is mildly exothermic, in contrast to the extremely endothermic SR reaction. 2

Two mechanisms for CPOM are considered, the indirect mechanism and the direct mechanism. In the indirect mechanism methane is first combusted to CO2 and H2O, and the following reforming reactions, i.e. SR and dry-reforming (DR) of unconverted methane, generate CO and H2. This, however, creates large temperature gradients, the demand for high operation temperatures and a risk for hotspots in the reactor since exothermic deep oxidation and endothermic reforming reactions proceed in different zones in the reactor. The direct mechanism yields CO and H2 directly from methane without deep oxidation, requiring high temperatures and usually a noble metal catalyst. Metallic catalysts suffer from sintering and metal loss via evaporation at high temperatures, which can be simply prevented by avoiding the presence of the metal. ZrO2 is a potential alternative catalyst, with the disadvantage of a significantly lower activity. [4]

Figure 1. Yields of CO, H2O, CO2 and H2 as a function of methane conversion at 600 °C (different conversion levels at steady state were reached by varying the amount of catalyst). Reprinted from ref. [4], Copyright (2004), with permission from Elsevier.

Formaldehyde (CH2O) had been considered a reaction intermediate in CPOM [6], however, it was questioned whether it was the only source of CO and H2. Zhu et al. conducted a study in order to complete the reaction scheme of CPOM over yttrium-stabilized ZrO2 (YSZ) catalyst. [4] Catalytic experiments with methane and formaldehyde in transient as well as in steady-state operation were conducted and CPOM was also studied with in-situ FTIR. The major products formed during CPOM over YSZ were CO, H2, CO2 and H2O. Fig. 1 shows that there is a linear relationship between their yields and methane conversion at 600 °C. The product selectivities are independent of methane conversion indicating that CO, H2, CO2 and H2O are all primary products of CPOM. This implies that,

3

surprisingly, methane is better activated on YSZ than CO or H2. [4] Also, ZrO2 apparently allows running CPOM partly via the direct mechanism. In the study YSZ was used, however, the general trends observed here are also valid for un-doped ZrO2. The proposed reaction scheme for methane oxidation includes dissociation of methane on O sites at the surface, finally forming surface formate, as observed by in-situ IR, which technique was unfortunately limited to relatively mild temperatures (400-470 °C). CH2O was indicated as an intermediate product, which is fast transformed to rather stable surface formate. Formate is decomposed to CO2 and H2, but also to CO and H2O. CO and H2 are formed via both decomposition of adsorbed formaldehyde and formate, whereas CO2 forms mainly via decomposition of formate. Some CO2 is also produced via oxidation of formate to carbonates, which decomposes to yield CO2. This is a minor pathway that may be relevant at high temperatures [4]. The influence of surface hydroxyl species on formaldehyde conversion to formate was studied by Zhu et al. [7]. It was concluded that water improves the selectivity to CO and H2 by increasing surface hydroxyl coverage, suppressing oxidation of formaldehyde to formate on the surface. Surface hydroxyls were suggested to partially block the lattice oxygen anions needed for oxidation of formaldehyde. [7] Methane activation at high temperature was studied further on ZrO 2 and YSZ with transient pulse experiments concentrating especially on the role of surface lattice anions [8]. CH4 was pulsed over preoxidized YSZ (and ZrO2) at 900 °C and a mixture of CO, H2, CO2 and H2O was produced during pulsing. This is in line with the product mixture in steady-state experiments and is the result of CH4 reacting with active oxygen species on the catalyst surface. [8] On continuing methane pulsing, it was noted that CO2 and H2O were formed only during the first four pulses, whereas formation of CO and H2 continued during methane pulses until the 22nd pulse. After that, CO formation ceased and CH4 decomposed to H2 and carbon deposits on the catalyst surface. Thus, part of the surface oxygen was indeed available to oxidize CH4 to CO2 and H2O at 900 °C. The amount of surface oxygen that can be extracted by methane on ZrO2 and YSZ corresponded to approx. 14 and 8.5 % of a monolayer of oxygen atoms on the surface, respectively. During subsequent O2 pulsing, after methane pulsing, CO was formed exclusively during the first 12 pulses. Thereafter also CO2 was detected in the products. This was due to oxidizing of carbon deposits on the surface. After pulse 17 constant oxygen signals were observed, indicating no further consumption of oxygen. Fig. 2 shows the amounts of oxygen atoms in the pulses leaving the reactor on YSZ; the dotted line indicates the amount of oxygen contained in each pulse. The difference between the amount of oxygen pulsed and the amount of oxygen detected in the product pulse allows calculation of the amount of oxygen consumed by re-oxidation of the ZrO2. The

4

amounts of oxygen restored in the materials during re-oxidation were similar to the amounts extracted by CH4, in the order of 10 % of the monolayer capacity. [8] It was ensured that these properties are not due to any redox-active contamination, based on careful analyses with LEIS. [9]

Figure 2. Oxygen pulses over 0.3 g YSZ; amount of oxygen contained in the products of each pulse. Reprinted from ref. [8], Copyright (2005), with permission from Elsevier.

To summarize, the study by Zhu et al. [8] revealed that CPOM over ZrO2 and YSZ proceeds via Marsvan Krevelen mechanism [10], be it that this is confined to a minor fraction of surface lattice oxygen. Originally, the reactive O species were assumed to be surface lattice oxygen species, however, it is conceivable that OH groups at the ZrO2 surface are involved. Increasing the crystal size by calcining at high temperature was found to increase the number of active sites per m2 [9]. In general, large crystals obtained by treatment at high temperature contain less low coordination sites and structural defects. Since large crystals are more active per m2 at high temperature [9, 11] the redox mechanism seems structure sensitive. The surface confined redox mechanism in CPOM apparently proceeds on sites on flat terrace on YSZ. Structural defects, such as Zr3+ cations on corners, edges, and kinks, seem not responsible for CPOM at high temperature. [11] ZrO2 and YSZ show relatively low but significant activity for CPOM. CPOM on ZrO2 proceeds via direct partial oxidation of methane, which results in milder temperature profiles in the reactor compared to deep oxidation. However, selectivity is far from complete and hence reforming reactions do contribute, like in the indirect mechanism.

3 WGS; formate chemistry and reactivity of hydroxyl groups on ZrO2 5

WGS (CO + H2O ↔ H2 + CO2) process is key technology for production of hydrogen. Complete conversion of CO is desired when pure H2 is targeted. High CO conversion can be achieved by combining high-temperature-shift (350-500 °C) and low-temperature-shift (200-250 °C) using Fe-based and Cu-based catalysts, respectively, in the well-established two-staged process. On the other hand, in gas to liquids (GTL) or biomass to liquids (BTL) applications the goal is to achieve a suitable H2/CO ratio for FT-synthesis, which can be adjusted by carrying out the WGS reaction to the required extent. The role of formates in the mechanism of WGS has been and is being discussed intensively in literature. This might be relevant also considering CPOM, since formate has been observed with IR on ZrO2 during CPOM at relatively low temperatures (approx. 400 °C) [4]. The study by Graf et al. showed that CO reacts with terminal hydroxyls (t-OH) to surface formate on pure ZrO2. [12] Those formates were found to be in equilibrium as CO was reversibly released into the gas phase on flushing with inert gas. The multi-coordinated hydroxyls (m-OH) present on ZrO2 surface were found not to interact with CO at temperatures studied, i.e. 240-400 °C. On Pt/ZrO2 catalyst, the formation of surface formate was found to occur similarly via interaction with t-OH. However, decomposition of formate at temperatures between 200 and 400 °C results in formation of gas-phase CO2 and H2, consuming m-OH and requiring the presence of Pt. Furthermore, Pt also provides a site to linearly bind CO, which is easily observed in FTIR. Thus, formate was found to be a reactive intermediate in WGS on the Pt/ZrO2 at 300-400 °C. The WGS reaction mechanism was presented on Pt/ZrO2 by Graf et al. (Fig. 3), where the decomposition of formate (step 2) is suggested to be ratedetermining. [12]

Figure 3. WGS mechanism on Pt/ZrO2. Reprinted from ref. [12], Copyright (2008), with permission from Elsevier.

6

Based on CO pulsing experiments performed earlier by Azzam et al. [13] and steady-state WGS experiments, Graf et al. noted that partial reduction of the support takes place during WGS. The role of water in WGS mechanism was concluded to be regeneration of the OH groups and re-oxidation of the surface (step 3 in Fig. 3) [12]. This is in line with the study by Azzam et al., establishing the role of the oxide support in activating water as well as its role in the WGS mechanism (redox step). [13] Azzam et al. demonstrated that CO could be transformed to formate and further to CO2 and H2 on a hydroxylated Pt/ZrO2 surface, whereas if the surface was regenerated by N2O and thus leaving virtually no reactive OH groups on the surface, only CO2 was formed. Therefore, WGS reaction can not be carried out on the non-hydroxylated surface, but CO can still be oxidized with lattice oxygen. Also, some lattice oxygen could be removed from Pt/ZrO2 either via reaction with hydroxyls (when ZrO2 was regenerated with H2O) or reaction with lattice oxygen (when ZrO2 is oxidized with N2O) even at 300 °C. [13] These results confirm that part of the surface lattice oxygen on ZrO2 is reactive, as was already observed by Zhu et al. [8] and that N2O gives rise to particular active surface oxygen species. [11] It was suggested by Graf et al. [12], based on data in [11, 13], that the redox step (step 3) is limited to structural defects on ZrO2. Furthermore, Graf et al. proposed that the regeneration of the surface with water occurs by dissociation of water on Zr3+ ions on structural defects producing two hydroxyl groups, t-OH and m-OH. Thus, the WGS reaction mechanism was proposed to involve defect sites (edges and kinks) on the support [12]. The apparent reactivity of formate was suggested to depend on the distance to Pt particles [12], implying that the reaction proceeds on, or close to, the metal-support interphase and the role of metal is to promote the reaction occurring on ZrO2. Thus, further studies on pure ZrO2 are needed for further insight. Kouva et al. performed a study utilizing a combination of surface characterization techniques, i.e. DRIFTS (diffuse reflectance infrared Fourier transform spectroscopy) and Temperature Programmed Reaction [14]. The type of the experiment, i.e. continuous CO feed with temperature rise, was of paramount importance in the recovery of the interactions between carbon oxides and surface OH groups, allowing observations of surface reactions of surface formate species at temperatures under which thermodynamics is favoring decomposition of these species. [14] In this study, the properties and quantity of hydroxyl groups on ZrO2 were changed by the pre-treatment method so that sample i) was calcined and reduced with 5 % H2 at 600 °C and cooled to 100 °C in He flow, sample ii) was calcined 600 °C and re-hydroxylated with water (0.1 vol-%) during cooling down to 100 °C, and iii) was calcined and reduced with 5% H2 at 600 °C followed by cooling down under H2O-contining flow. The obtained degrees of hydration were: i) 0.15, ii) 0.25, and iii) 0.50 fractions of a 7

ML. These surfaces were further characterized via CO-and CO2-temperature-programmed surface reaction (CO- and CO2-TPSR) which were also performed under DRIFTS. Fig. 4 presents the results of CO-TPSR experiments (gas-phase) on ZrO2 samples with varied degrees of hydration [14]. It can be noted how samples i) and ii) behave qualitatively similarly, only the adsorption of CO is somewhat higher on the more hydrated ii) sample at around 300 °C. On the sample iii), i.e. the sample with the highest degree of hydration, a new low-temperature release of water and CO is observed. Finally, all samples release concurrently CO and H2O at around 400 °C, as well as CO2 and H2 at above 500 °C. It can be noted that the effect of surface hydration is mainly seen in the low temperature window since the high-temperature processes do not differ greatly. [14]

Figure 4. CO-TPSR as a function of temperature on ZrO 2 samples with varied hydroxyl coverages, i) 0.15, ii) 0.25, iii) 0.50 of a ML. Level 0 refers to the feed level of CO. Experiments in [14]. Reproduced from ref. [14] with permission from the PCCP Owner Societies.

CO adsorption in the temperature window up to 300 °C (Fig. 4) is indicative of activated formate formation and was not found to be affected by surface hydration. The adsorption of CO during flushing the sample with CO at 100 °C (not shown in Fig. 4), before starting the temperature program, is thus kinetically limited. Full adsorption is not achieved, but adsorption is significant, explaining the

8

seemingly imbalance between the amounts of CO adsorption and desorption in Fig. 4. The effect is masked to some extent by the low temperature desorption that is observed exclusively on the most hydrated surface (iii). The release of CO and water at low temperature observed on sample iii) remarks a new weakly bound CO adsorption state. Surprisingly, this state cannot be detected with IR spectroscopy and more research would be required to understand this phenomenon. [14] It was observed that linear adsorption of CO on Zr4+ cations becomes possible when the OH-coverage on the oxide is decreased; it was suggested this may provide a favourable pathway for forming formates. Formation of formate via linear CO interacting with t-OH and its further decomposition on ZrO2 is presented in Fig. 5. On the other hand, high degree of hydration blocks the adsorption sites for linear CO, but creates hydroxyl sites allowing formate formation directly from gas phase CO. This, however, is a route with a higher activation barrier as compared to the route via CO linearly adsorbed on Zr4+ cations, based on DFT calculations. The formation of CO2 and H2 can be connected to reductive decomposition of surface formate via reducing the ZrO2 surface. This was proven to be a valid reaction pathway also on pure ZrO2 without the metal, at high temperatures (above 500 °C) however. [14]

Figure 5. Formation and decomposition of formates on pure ZrO 2. Reproduced from ref. [14] with permission from the PCCP Owner Societies.

Based on density functional theory (DFT) model calculations presented in the study by Kouva et al., it was indicated that both carbonates and hydroxyl groups preferably form on low-coordinated sites. This is in agreement with the observation that t-OH is required for the formation of formate. Carbonates could be a possible reaction intermediate in CO oxidation on ZrO2 even at high temperatures. [14] Also Zhu et al., concluded that carbonate might play a role in methane oxidation. [4] The effect of the surface area and thus the amount of structural defects on the surface chemistry described above is currently subject of further studies.

9

4 H2S adsorption on ZrO2 and the effect on its redox properties Many industrial gas-mixtures of CO, CH4, CO2 and H2 contain also varying amounts of H2S, coalderived syngas even 2000 ppm. H2S is well known to poison especially metal catalysts. However, also beneficial effects are known, which occur typically on metal oxide catalysts. For example, methane oxidation was found to be promoted by H2S on YSZ, which was studied as a sulfur-tolerant catalyst to be used in solid oxide fuel cell applications [15]. Also sour water gas shift, i.e. WGS in the presence of H2S tolerant catalyst, has been investigated recently with regards to converting H 2S-containing streams in integrated gasification combined cycle (IGCC) plants [16]. It can be noted that catalysts tolerating H2S offer possibilities to simplify upstream processes, reducing capital as well as operating cost. Biomass gasification yields a syngas mixture, which is complex and needs treatment before further use [17]. Tars are a major challenge in utilization of the gas. ZrO2-based catalysts have been proven competent in oxidizing tars during biomass gasification gas clean-up [18], where also H2S is present. Moreover, it has been shown that the redox/oxidative properties of ZrO 2 are enhanced by H2S [19]. H2S (100 ppm) improved naphthalene and ammonia oxidation activity in biomass gasification gas clean-up experiments on ZrO2 (and on YSZ), under conditions where carbon oxides, hydrogen, and water are also present at 600 and 700 °C [19]. Oxidation of organic compounds often involves redox cycles during which the organic molecule is oxidized by surface lattice oxygen leaving a reducing center (the Marsvan Krevelen mechanism), and its occurrence requires the catalysts’ ability to generate oxygen vacancies [10, 20]. Therefore, it is suggested that H2S improves the reactivity of surface lattice oxygen or increases the amount of sites able to participate in the redox cycles. Later studies targeted understanding how H2S influences the reactivity of surface oxygen on ZrO2 [21]. Temperature-programmed reduction with CO (CO-TPR) was performed on five samples sulfided to different extent via temperature-programmed sulfiding (TPS). The results were compared with those on pure ZrO2 to study how adsorbed H2S changes the CO reactivity of the samples (Fig. 6). H2S was adsorbed on the samples isothermally at 30 °C, and in a temperature-programmed manner between 30 to 100, 200, 300, or 400 °C (samples labeled as H2Sads 30 °C, TPS30-100-30 °C, TPS30-200-30 °C, TPS30-300-30 °C or TPS30-400-30 °C), respectively. H2S adsorption on ZrO2 can be considered to occur molecularly, dissociatively, or by exchange of sulfur with surface lattice oxygen [21-23]. During TPS experiments, H2S adsorption at 30 °C was observed to be partly reversible via molecular adsorption, and partly irreversible suggesting dissociative adsorption. At elevated temperatures H2S adsorption caused water desorption. It was suggested that exchange of

10

sulfur with lattice oxygen occurs at elevated temperatures, resulting in water generation as observed, indicating surface reaction of H2S with ZrO2 via an activated process. The amount of H2S adsorbed at 30 °C was 5.8 % of a ML and the additional amounts, i.e. activatedly adsorbed H2S at elevated temperatures, were 0.7, 2.3, 4.3, and 5.4 % of a ML on the samples TPS 30-100-30 °C, TPS30-200-30 °C, TPS30300-30 °C

or TPS30-400-30

°C,

respectively. The sub-ML amounts of adsorbed/reacted H2S indicated that

minority sites, i.e. defects such as edges, steps or kinks, are involved in interactions between H 2S and ZrO2. [21] After the sulfidation step the catalysts were treated with CO with increasing temperature from 30 to 600 °C and the outlet was analyzed with MS. During these CO-TPR experiments, CO2 was generated on the sulfided ZrO2 samples at lower temperatures than on the pure oxide (Fig. 6). Furthermore, it was noted that the amount of CO2 generated at approx. 200 °C increased with increasing amount of activatedly adsorbed sulfur species. The reaction with H2S at temperatures of 100 °C and higher apparently modified the ZrO2 surface so that CO could be oxidized to CO2 at remarkably lower temperatures than on the unsulfided ZrO2. This is in agreement with the results in [14]. It was further suggested that reaction of H2S with defective sites increases the reactivity of surface lattice oxygen on ZrO 2. [21] However, it could not be ruled out that molecularly and dissociatively adsorbed species, without reacting with the lattice, also might change the reactivity.

Figure 6. CO2 generation on unsulfided and differently sulfided ZrO2 during CO-TPR from 30 to 600 °C (10 °C/min, 5 vol-% CO in He). Experiments in [21]. Reprinted from ref. [21], Copyright (2014), with permission from Elsevier.

11

The CO2-peak evolving at higher temperatures (approx. 500 °C), which is also observed on pure ZrO2 but at somewhat lower temperature, is the only one accompanied by H2 generation (not shown in Fig. 6). Kouva et al. attributed this to decomposition of surface formates to yield CO2 and H2 with consumption of t-OH and m-OH on the surface [14]. It was suggested that the sites for formate generation and their reductive decomposition (t-OH and m-OH) were modified by sulfur. Moreover, the sites to adsorb carbonates were blocked on the sulfided surface, since CO2 could be formed and released to gas-phase at low temperature (~200 °C). [21] The obtained CO-TPR results call for better understanding of the interaction between H2S and ZrO2. Despite IR-spectroscopy can be and has been used to study H2S adsorption on oxides [22, 23], the information obtained is rather limited due to low extinction coefficients of S-H stretching vibrations. Unfortunately, surface lattice sulfur species could not be detected and it is not possible to distinguish between molecular and dissociative adsorption. Density functional theory (DFT) calculations could be employed to obtain atomic level information on the interaction between H2S and ZrO2. Similar calculations have been previously employed to study H2S on CeO2 [24] and on YSZ [25, 26] but, to the best of our knowledge, no computational study of H2S adsorption on ZrO2 has been published to date. We apply DFT to shed light on the nature of sulfur species present on ZrO2 and secondly to determine how different sulfur species modify the redox properties of ZrO2, as observed experimentally. We note that sulfur species must be relatively strongly bound to ZrO2 to be able to modify the reactivity of neighboring oxygen atoms. Therefore, the first aim of calculations is to identify the structure of stable sulfur species on ZrO2. Calculations are performed on minority sites on ZrO2, since the TP-experiments indicated that H2S adsorption is limited to such sites. These sites are modelled employing edge and corner site models. The edge sites are presented with a four-layer thick stepped (2 ī 2) ZrO2 surface with a (1x2) unit cell. The highly uncoordinated corner sites were modeled by doubling the unit cell of the (2 ī 2) surface in the x direction and removing atoms from the step to form a corner site. We also assume that low coordinated sites are partially hydroxylated under reaction conditions and thus one t-OH and one m-OH species are placed on the model surfaces. The models are shown in Fig. 7 and the details of calculations are presented in Supplementary information.

12

Figure 7. SH adsorption on the edge model (ZrO 2 (2 ī 2)) a front and side view (A and B) and on the corner model a front and side view (C and D). Zr atoms are grey, oxygen is red, sulfur is yellow and hydrogen is white. Here, the SH group binds to a Zr atom which is missing one oxygen.

To start with, the adsorption of molecular H2S at the hydroxylated edge of ZrO2 (2 ī 2) is addressed. According to calculations, H2S resides on a Zr ion at a distance of 4.3 Å. The adsorption is weakly exothermic with adsorption energy of -14.5 kJ/mol, confirming that molecularly adsorbed H2S cannot be held responsible for modifying redox properties of the surface. Moreover, due to the weak binding, molecular H2S cannot be significantly present on the surface at temperatures where increased reactivity has been observed. Weak interaction between H2S and ZrO2 is also in line with experimental findings [21, 23, 27] and calculations on H2S adsorption on CeO2 [28] and on YSZ (111) [25]. Dissociative adsorption of H2S has been suggested to take place on the strong Lewis acid-base pair site [22, 23]. Moreover, H2S can also undergo a dissociation process, where lattice oxygen is exchanged with sulfur and water is released. The other possible process is that H titrates a terminal OH (t-OH) group and forms a water molecule and a terminal SH (t-SH) group as follows t-OH (s) + H2S (g) ↔t-SH (s) +H2O (g), where (s) refers to surface species and (g) to gas-phase species. Computed reaction enthalpy for the above-described exchange process is +28 kJ/mol on a hydroxylated edge model. In the case that lattice oxygen is replaced with sulfur upon H2S dissociation, the computed reaction enthalpy is nearly thermo-

13

neutral being only 8 kJ/mol endothermic for the hydroxylated edge model. Considering the general accuracy of DFT calculations, this interaction may also be mildly exothermic. Next we addressed H2S dissociation on an oxygen vacancy at the hydroxylated edge, where no water is released but a new m-OH forms together with a t-SH group, filling the vacancy. This process is strongly exothermic having a reaction enthalpy of -69.5 kJ/mol. These results demonstrate that the thermodynamics of dissociation depends strongly on the nature of the site. The fact that the dissociation is energetically more favorable on oxygen deficient ZrO2 is in line with recent DFT calculations on ZnO [29] and CeO2 [111] [28] surfaces. If it is assumed that ZrO2 is partly reduced and a corner model is applied, the dissociative adsorption of H2O is computed to be exothermic by -216 kJ/mol on a vacancy site at the edge, while the dissociative adsorption of H2S to the same site is exothermic only by -140 kJ/mol. This indicates that from a thermodynamical point of view oxygen vacancies are easily mended under water atmosphere. However, temperature and pressure might have an effect which is not predicted with the calculated model, since experiments in [21] indicate that during H2S adsorption at elevated temperatures H2O is indeed released. We turn next to calculations on reactivity of ZrO2 in the presence of sulfur employing again an edge model. For comparison, we first consider the case, where hydrogen is used as reducing agent to form water on a hydroxylated ZrO2 surface without sulfur. For this process reaction enthalpy is +205 kJ/mol. Next, we replace one t-OH at the edge with a t-SH group (Fig. 7A and 7B). This leads to a configuration, where m-OH, which is to be reduced with H2, and t-SH share one Zr atom. The computed reaction enthalpy for m-OH reduction to H2O is slightly higher in the presence of -SH (+217 kJ/mol) as compared to +205 kJ/mol in the absence of -SH. This result suggests that t-SH groups on the edge of the ZrO2 (2 ī 2) are not responsible for enhanced reducibility. Increasing the coverage of SH groups at a step edge has no significant influence. Thus, the observed discrepancy between calculations and experimental results is not due to -SH coverage. As a next step, we considered how a multi-coordinated SH (m-SH) group, present on a corner model (Fig. 7C and D), would destabilize lattice oxygen (m-OH) and t-OH. m-SH might modify the electronic properties of ZrO2 as it binds to two Zr atoms and is thus more integrated into the oxide lattice. Moreover, some Zr atoms at the corner are less coordinated than Zr atoms at the edge of a (2 ī 2) surface, presenting higher structural flexibility. Again, the reactivity of oxide is studied employing H 2 as a reducing agent. We consider the removal of both t-OH and m-OH species but one species at the time. On a non-sulfided surface, the reduction of m-OH to H2O with gas-phase hydrogen costs +273 kJ/mol

14

while the reduction of t-OH is slightly more favorable with reaction enthalpy of +239 kJ/mol. The presence of m-SH decreases the reduction enthalpies with 36 kJ/mol and 31 kJ/mol for m-OH and t-OH species, respectively. These findings agree with experiments in a sense that the presence of m-SH slightly improves the reactivity. However, if we compare the absolute values, the edge model without SH still gives the highest reducibility among the computed systems. In general, the calculations suggest that minority sites are needed to adsorb H 2S dissociatively but we cannot solely determine the composition and geometry of the site from the present results. One of the main reasons why computational results only partially support experimental findings could be that the structure of the active site and the ZrO2 surface structure remain unresolved. Throughout this study, we employed edge and corner models based on a ZrO2 (2 ī 2) surface to mimic a minority site on a ZrO2 nanoparticle. However, these systems apparently do not represent the minority sites correctly. This, together with the fact that the exact concentration of OH groups and oxygen vacancies is unknown, as well as uncertainty about the bulk structure of ZrO2, might be responsible for the observed discrepancies. Without knowing the properties of active sites more precisely, one way to proceed with calculations would be to explore a variety of models for the surface of ZrO2 nanoparticles, varying concentrations of oxygen vacancies and OH/SH groups using first principles atomistic thermodynamics [30]. The calculations demonstrate that the interaction of molecular H2S or t-SH with ZrO2 is too weak to increase the reactivity of surface oxygen on ZrO2. Thermodynamics does not favour the formation of the bulk sulfide but since some sites are able to bind H2S the nature of these sites is different. Furthermore, calculations suggest that possibly m-SH species, but more likely a minority site with a so far unknown structure, is needed to explain enhanced oxidation activities observed during CO-TPR [21] and tar oxidation in the gasification gas media [19]. Limited interaction with H2S can also be seen as a benefit for oxidation catalysis, since the catalysts’ sulfur tolerance is improved.

5 Conclusions This review discusses the catalytic properties and surface chemistry of ZrO2 in catalytic partial oxidation of methane (CPOM), water-gas-shift reaction (WGS), and tar decomposition. The catalytic properties are discussed in terms of (sub-monolayer) redox capacity of the ZrO2 surface. A fraction of the surface lattice oxygen on ZrO2 was revealed to be involved in redox cycles, part of the cycle being formation of oxygen vacancies at terraces (comprising ~10 % of the monolayer capacity) during CPOM. Reduction of the surface via decomposition of formates to H2 and CO2 was presented to

15

proceed involving t-OH and m-OH surface species. DFT calculations revealed a relatively weak enhancing effect of m-SH on the reactivity of neighboring t-OH and m-OH sites. The effect is too small to account for the significant enhancement of the reducibility of the ZrO2 surface by the presence of sulfur. We therefore conclude that the structure of the actual active site is yet to be discovered. ZrO2 is clearly not just a catalyst support, but provides intrinsic catalytic activity. Application of ZrO2 as a catalyst, mitigating disadvantages of metallic catalysts i.e. activation by sulfur instead of deactivation as well as thermal stability, or as a reactive catalyst support is therefore promising for practical applications.

Acknowledgments The authors are grateful for all contributions of PhD students and postdocs that have contributed over the years, resulting in many references in this manuscript. Financial support of the Foundation of Applied Sciences (STW), Netherlands Organisation for Scientific Research (NWO), Advanced Chemical Technologies for Sustainability (ACTS) as well as the FiDiPro program of the Finnish Funding Agency for Innovation (Tekes) is gratefully acknowledged. The computer resources were provided by the Finnish IT Centre for Science (CSC) Finland.

References [1] Daturi M, Binet C, Bernal S, Pérez Omil JA, Lavalley JC (1998) FTIR study of defects produced in ZrO2 samples by thermal treatment. J Chem Soc Faraday Trans 94(8):1143-1147 [2] Kogler M, Köck E-M, Bielz T, Pfaller K, Klötzer B, Schmidmair D, Perfler L, Penner S (2014) Hydrogen Surface Reactions and Adsorption Studied on Y2O3, YSZ, and ZrO2. J Phys Chem C 118:8435-8444 [3] Eder D, Kramer R (2002) The stoichiometry of hydrogen reduced zirconia and its influence on catalytic activity Part 1: Volumetric and conductivity studies. Phys Chem Chem Phys 4:795-801 [4] Zhu J, van Ommen JG, Lefferts L (2004) Reaction scheme of partial oxidation of methane to synthesis gas over yttrium-stabilized zirconia. J Catal 225:388-397 [5] Holmen A (2009) Direct conversion of methane to fuels and chemicals. Catal Today 142:2-8 [6] Steghuis AG, van Ommen, JG, Lercher JA (1998) On the reaction mechanism for methane partial oxidation over yttria/zirconia. Catal Today 46:91-97 [7] Zhu J, van Ommen JG, Lefferts L (2006) Effect of surface OH groups on catalytic performance of yittriumstabilized ZrO2 in partial oxidation of CH4 to syngas. Catal Today 117:163-167 [8] Zhu J, van Ommen JG, Bouwmeester HJM, Lefferts L (2005) Activation of O2 and CH4 on yttrium-stabilized zirconia for the partial oxidation of methane to synthesis gas. J Catal 233:434-441 [9] Zhu J, van Ommen JG, Knoester A, Lefferts L (2005) Effect of surface composition of yttrium-stabilized zirconia on partial oxidation of methane to synthesis gas. J Catal 230:291-300 [10] Mars P, van Krevelen DW (1954) Oxidations carried out by means of vanadium oxide catalysts. Chem Eng Sci (Spec Suppl) 3:41-59 [11] Zhu J, Albertsma S, van Ommen JG, Lefferts L (2005) Role of Surface Defects in Activation of O2 and N2O on ZrO2 and Yttrium-Stabilized ZrO2. J Phys Chem B 109:9550-9555

16

[12] Graf PO, de Vlieger DJM, Mojet BL, Lefferts L (2009) New insights in reactivity of hydroxyl groups in water gas shift reaction on Pt/ZrO2. J Catal 262:181-187 [13] Azzam KG, Babich IV, Seshan K, Lefferts L (2007) Bifunctional catalysts for single-stage water–gas shift reaction in fuel cell applications.: Part 1. Effect of the support on the reaction sequence. J Catal 251:153-162 [14] Kouva S, Andersin J, Honkala K, Lehtonen J, Lefferts L, Kanervo J (2014) Water and carbon oxides on monoclinic zirconia: experimental and computational insights. Phys Chem Chem Phys 16(38):20650-20664 [15] Vincent AL, Luo J-L, Chuang KT, Sanger AR (2011) Promotion of activation of CH 4 by H2S in oxidation of sour gas over sulfur tolerant SOFC anode catalysts. Appl Catal B 106:114-122 [16] Liu B, Xu H, Zhang Z (2012) Platinum based core–shell catalysts for sour water–gas shift reaction. Catal Commun 26:159-163 [17] Rönkkönen EH (2014) Catalytic clean-up of biomass derived gasification gas with zirconia based catalysts, Doctoral Thesis, Aalto University [18] Juutilainen S, Simell P, Krause AOI (2006) Zirconia: Selective oxidation catalyst for removal of tar and ammonia from biomass gasification gas. Appl Catal B 62:86-92 [19] Rönkkönen H, Simell P, Reinikainen M, Krause O (2009) The Effect of Sulfur on ZrO 2-Based Biomass Gasification Gas Clean-Up Catalysts. Top Catal 52:1070-1078 [20] Busca G, Finocchio E, Ramis G, Ricchiardi G (1996) On the role of acidity in catalytic oxidation. Catal Today 32:133-143 [21] Kauppi EI, Kanervo JM, Lehtonen J, Lefferts L (2015) Interaction of H2S with ZrO2 and its influence on reactivity of surface oxygen. Appl Catal B 164:360-370 [22] Ziolek M, Kujava J, Saur O, Lavalley JC (1995) Influence of hydrogen sulfide adsorption on the catalytic properties of metal oxides. J Mol Catal A 97:49-55 [23] Travert A, Manoilova OV, Tsyganenko AA, Maugé F, Lavalley JC (2002) Effect of Hydrogen Sulfide and Methanethiol Adsorption on Acidic Properties of Metal Oxides: An Infrared Study. J Phys Chem B 106:1350-1362 [24] Chen H-T, Choi Y, Liu M, Lin MC (2007) A First-Principles Analysis for Sulfur Tolerance of CeO 2 in Solid Oxide Fuel Cells. J Phys Chem C 111:11117-11122 [25] Chu X, Li Z, Zhang Y, Yang Z, (2013) Can H 2S poison the surface of yttria-stabilized zirconia?. Int J Hydrogen Energy 38:8974-8979 [26] Chu X, Zhang Y, Li S, Yang Z (2014) Ab initio atomistic thermodynamics study on the sulfur tolerance mechanism of the oxygen-enriched yttria-stabilized zirconia surface. Surface Science 622:16-23 [27] Kauppi EI, Rönkkönen EH, Airaksinen SMK, Rasmussen SB, Bañares MA, Krause AOI (2012) Influence of H 2S on ZrO2-based gasification gas clean-up catalysts: MeOH temperature-programmed reaction study. Appl Catal B 111112:605-613 [28] Mayernick AD, Li R, Dooley,KM, Janik MJ (2011) Energetics and Mechanism for H 2S Adsorption by CeriaLanthanide Mixed Oxides: Implications for the Desulfurization of Biomass Gasifier Effluents. J Phys Chem C 115:24178-24188 [29] Ling L, Wu J, Song J, Han P, Wang B (2012) The adsorption and dissociation of H2S on the oxygen-deficient Zn(1010) surface: A density functional theory study. Comput Theor Chem 1000:26-32 [30] Reuter K, Scheffler M (2001) Composition, structure, and stability of RuO 2(110) as a function of oxygen pressure. Phys Rev B 65:035406(1)-035406(11)

17

Electronic supplementary information for: ZrO2 acting as a redox catalyst E. I. Kauppi1, K. Honkala2, A. O. I. Krause1, J. M. Kanervo1, L. Lefferts1,3 1

Aalto University, School of Chemical Technology, Department of Biotechnology and Chemical Technology, Research Group Industrial Chemistry, P.O. Box 16100, FI-00076 Aalto, Finland 2

University of Jyväskylä, Department of Chemistry, Nanoscience Center, P. O. Box 35, 40014 Jyväskylä, Finland

3

University of Twente, Faculty of Science & Technology and MESA+ Institute for Nanotechnology, P.O. Box 217, 7500 AE Enschede, The Netherlands *Leon Lefferts [email protected]

Computational methods Density functional theory calculations were performed in the real space grid implementation of the projector augmented wave (PAW) [1] method as implemented in the GPAW program package [2]. The Kohn-Sham equations were solved self-consistently using the PBE functional [3]. The grid spacing was set to 0.2 Å in all directions of the supercell and applied throughout the work. The (2x2x1) MonkhorstPack k-point sampling was applied to the studied slab models. The lattice parameters were optimized in the previous study [4].

References [1] Blöchl BE (1994) Projector augmented-wave method. Phys Rev B Condens Mater Phys 50:17953-17979 [2] Enkovaara J, Rostgaard C, Mortensen J J, Chen J, Dulak M, Ferrighi L, Gavnholt J, Glinsvad C, Haikola V, Hansen HA, Kristoffersen HH, Kuisma M, Larsen AH, Lehtovaara L, Ljungberg M, Lopez-Acevedo O, Moses PG, Ojanen J, Olsen T, Petzold V, Romero NA, Stausholm-Møller J, Strange M, Tritsaris GA, Vanin M, Walter M, Hammer B, Häkkinen H, Madsen GKH, Nieminen RM, Nørskov JK, Puska M, Rantala TT, Schiøtz J, Thygesen KS, Jacobsen KW (2010) Electronic structure calculations with GPAW: a real-space implementation of the projector augmented-wave method. J Phys Condens Matter 22:253202(1)-253202(24) [3] Perdew JP, Burke K, Ernzerhof M (1996) Generalized Gradient Approximation Made Simple. Phys Rev Lett 77:3865-3868 [4] Kouva S, Andersin J, Honkala K, Lehtonen J, Lefferts L, Kanervo J (2014) Water and carbon oxides on monoclinic zirconia: experimental and computational insights. Phys Chem Chem Phys 16(38):20650-20664

18

IV Viinikainen T., Kauppi E. I., Korhonen S., Lefferts L., Kanervo J. M., Lehtonen J. (2013) Molecular level insights to the interaction of toluene with ZrO2 -based biomass gasification gas clean-up catalysts. Appl. Catal. B 142-143, 769-779. Reproduced with permission from Elsevier B. V.

Applied Catalysis B: Environmental 142–143 (2013) 769–779

Contents lists available at SciVerse ScienceDirect


Molecular level insights to the interaction of toluene with ZrO2 -based biomass gasification gas clean-up catalysts Tiia Viinikainen a,∗ , Inkeri Kauppi a , Satu Korhonen a , Leon Lefferts a,b , Jaana Kanervo a , Juha Lehtonen a a b

Aalto University, School of Chemical Technology, Department of Biotechnology and Chemical Technology, P.O. Box 16100, 00076 Aalto, Finland University of Twente, Faculty of Science & Technology, P.O. Box 217, 7500 AE Enschede, The Netherlands

a r t i c l e

i n f o

Article history: Received 27 February 2013 Received in revised form 31 May 2013 Accepted 15 June 2013 Available online 22 June 2013 Keywords: ZrO2 Toluene adsorption Gasification gas cleaning In situ DRIFTS TP gas-phase analysis

a b s t r a c t Gasification of biomass, followed by ZrO2 -catalyzed hot gas clean-up at 600–900 ◦ C for the oxidation of impurities (such as tar), is an environmentally attractive way to produce heat and power or synthesis gas. The interaction of toluene (as a model compound for tar) with ZrO2 -based gasification gas clean-up catalysts was studied by in situ DRIFTS and temperature-programmed gas-phase analysis. Toluene was found to interact in four ways with ZrO2 surfaces: forming molecularly adsorbed toluene, surface benzoate species, carbonaceous deposits and surface benzyl species. The adsorption of toluene in the absence of gas-phase oxygen at ambient temperature on ZrO2 -based catalysts resulted in weakly adsorbed molecular toluene, while the adsorption of toluene at higher temperatures yielded carbonaceous deposits and surface benzoate species. Combined analysis of infrared and TP data showed that some of the benzoate species decomposed, producing benzene and carbon dioxide. Surface benzyl species, on the other hand, were detected on the surface of ZrO2 -based catalysts only in the presence of gas-phase oxygen at a temperature where toluene and oxygen started to convert. Therefore, it is suggested that benzyl species are the active intermediates from toluene to carbon oxides during the gasification gas cleaning over ZrO2 -based catalysts. © 2013 Elsevier B.V. All rights reserved.

1. Introduction Gasification of biomass is a thermochemical process, which converts biomass into syngas (mixture of CO and H2 ) [1]. The applications of the product gas include production of energy, H2 , second generation liquid biofuels via Fischer–Tropsch synthesis and several chemicals. However, the product gas also contains impurities such as tar (aromatic hydrocarbons heavier than benzene), and therefore it has to be cleaned before use [2]. Hot-gas cleaning with a catalyst-coated monolith after biomass gasification is a preferable choice for tar decomposition [3]. Since the gas from the gasifier is a complex mixture of compounds, several decomposition and equilibrium reactions take place simultaneously during catalytic gas cleaning. Thus, the determination of tar decomposition reaction rates in the presence of other gas components, such as CO, H2 , CO2 or H2 O, is challenging [4]. Zirconia-based catalysts have shown to selectively oxidize tar molecules into CO and CO2 during hot gas clean-up at 600–900 ◦ C when a small amount of oxygen is added to the gas [5]. However, this addition of oxygen changes the tar decomposition

∗ Corresponding author. Tel.: +358 50 5300 367. E-mail address: tiia.viinikainen@aalto.fi (T. Viinikainen). 0926-3373/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.apcatb.2013.06.008

mechanism and makes the determination of tar decomposition reaction rates even more complicated. Furthermore, changes in the composition of the gas along the catalyst-coated monolith have been measured [4,6,7]. At the oxygen-rich zone in the inlet of the monolith, tar molecules preferably use the oxygen available and are oxidized mainly to CO and CO2 . However, near the outlet of the monolith, oxygen is nearly or completely consumed, and therefore other reactions (such as steam and dry reforming, hydrocracking, carbon formation, etc.) might be preferred. The activity of selected ZrO2 -based catalysts in the decomposition of tar from synthetic gasification gas in the presence of oxygen (3 vol%) have been reported to decrease in the order of ZrO2 > Y2 O3 –ZrO2 > SiO2 –ZrO2 [7]. Certain physical and chemical properties have been related to the tar decomposition activity of ZrO2 -based gasification gas clean-up catalysts [5,7]. For example, Juutilainen et al. suggested that increased acidity is not favorable for ZrO2 -based gasification gas cleaning catalyst [5]. In contrast to the high total amount and strength of acidic surface sites, the high total amount and strength of basic surface sites seem to be essential for an active ZrO2 tar decomposition catalyst [7]. Furthermore, the combination of suitable redox properties of ZrO2 -based gasification gas cleanup catalysts as well as their tendency to generate oxygen vacancies has been related to high activity in tar decomposition [7].

T. Viinikainen et al. / Applied Catalysis B: Environmental 142–143 (2013) 769–779

770

Table 1 The specific surface areas as well as the amounts of basic and acidic surface sites of the ZrO2 -based catalysts [7]. Catalyst

Specific surface area (m2 /g)

ZrO2 Y2 O3 –ZrO2 SiO2 –ZrO2

24 53 92

Amount of basic sites (␮mol/g) 84 146 49

Amount of acidic sites (␮mol/g) 22 36 78

Molecular level insight into the ZrO2 -based catalysts under gasification gas conditions offers essential information needed, e.g. when optimizing the catalyst composition. Furthermore, such knowledge combined with the understanding of the influence of gas components on the catalyst performance allows the design and optimization of the cleaning process to ensure high-quality product gas. Since the gasification gas is a complex matrix, a novel approach is needed for overcoming the challenge of finding the tar decomposition mechanism. Our approach is to diminish the complexity by studying simpler subsystems with fewer components and fewer reactions at a time. The first step in this approach is to understand the interaction of a single tar component with ZrO2 -based catalysts. Later the complexity is increased by adding components one by one. By combining in situ spectroscopic methods with temperatureprogrammed gas-phase analysis, unique knowledge on the relevant surface species and reaction mechanisms is obtained. In this study the adsorption of toluene (as a model compound for tar) is investigated to obtain an insight into toluene decomposition during gasification gas cleaning. The adsorption of toluene has been studied widely over various reducing and non-reducing oxide catalysts [8–17]. However, the mechanism with which toluene interacts with ZrO2 -based gasification gas clean-up catalysts has not yet been established. Therefore, the adsorption of toluene over selected ZrO2 -based catalysts was studied at temperatures from 30 ◦ C to as high as 600 ◦ C, 600 ◦ C being the relevant temperature for hot-gas cleaning. A reduction pretreatment was applied to the catalysts in order to reduce the amount of surface hydroxyl groups on the catalyst to imitate the reducing conditions of the gasification gas (gasification gas always contains hydrogen and CO). The aim of this study was to discover the relevant surface species formed from toluene in the absence and presence of gas-phase oxygen as well as to estimate the decomposition mechanism of the adsorbed toluene-derived species over ZrO2 , Y2 O3 –ZrO2 and SiO2 –ZrO2 . 2. Experimental Pure ZrO2 , 5 mol% Y2 O3 –ZrO2 , and 8 mol% SiO2 –ZrO2 were provided by MEL Chemicals, where Y2 O3 –ZrO2 and SiO2 –ZrO2 were prepared by co-precipitation. The catalysts were calcined in static air at 800 ◦ C for 1 h. The specific surface areas as well as the amounts of basic and acidic surface sites of the catalysts are shown in Table 1. Other previously measured physical and chemical properties of the catalysts have been reported elsewhere [7,18]. Two complementary techniques were used to study the interaction of toluene with ZrO2 -based gasification gas clean-up catalysts: in situ DRIFTS and temperature-programmed gas-phase analysis. Three types of experiments were carried out with both techniques: (1) temperature-programmed desorption of toluene (TPD), (2) temperature-programmed adsorption of toluene (TPA), and (3) temperature-programmed oxidation of adsorbed surface species (TPO). The feed gas compositions and heating rates were matched as close as possible in the DRIFTS and TP gas-phase analysis in order to guarantee accurate comparison between surface species and gasphase products. This allows surface phenomena to be linked with gas-phase composition. In addition, the interaction of toluene and

oxygen over ZrO2 -based catalysts was studied via the temperatureprogrammed surface reaction between toluene and oxygen (TPSR) using only in situ DRIFTS. Toluene was chosen as a model compound for tar.

2.1. In situ DRIFTS experiments In situ diffuse reflectance infrared Fourier transform spectroscopy (DRIFTS) measurements were performed with a Nicolet Nexus FTIR spectrometer and a Spectra-Tech high temperature/high pressure chamber. The total gas flow through the reaction chamber (passing over the catalyst sample) was kept at 50 cm3 /min throughout the experiments. A catalyst sample powder (approx. 30 mg) was pretreated by in situ calcination with 10% O2 /N2 (synthetic air 99.99%, N2 99.999%, AGA) at 600 ◦ C for 2 h followed by N2 flush for 1 h. In all experiments, the spectrum of an aluminum mirror measured (4 cm−1 resolution, 200 scans) under nitrogen flow was used as the background. After the calcination, the catalyst sample was pre-reduced with 5% H2 /N2 (H2 99.999%, AGA) at 600 ◦ C for 15 min. Spectra were recorded every minute (4 cm−1 , 30 scans) for the first 8 min on stream and then after 9 and 12 min (4 cm−1 , 100 scans). After the reduction, the catalyst sample was flushed with nitrogen for 30 min while cooling down to the adsorption temperature of toluene. In the TPD experiments, toluene (750 ppm toluene in He with 1–2 ppm water and even smaller amount of methanol as impurities, AGA) was fed to the sample cell at 30 ◦ C and in the TPA experiments at 100 ◦ C for 30 min. Spectra were collected once every minute (4 cm−1 , 30 scans) for the first 5 min, and thereafter once every 5 min (4 cm−1 , 100 scans). After the adsorption of toluene, the samples were flushed with nitrogen for 30 min to obtain spectra without the contribution of gas-phase toluene. In the TPD experiments , desorption of the adsorbed species from the samples was followed with increasing temperature stepwise from 30 to 600 ◦ C. During the heating of the sample, spectra (4 cm−1 , 100 scans) were collected every 25 ◦ C, i.e. every 4 min. In the TPA experiments, toluene flow was directed back to the sample cell after the nitrogen flush at 100 ◦ C and temperature was increased stepwise from 100 to 200 ◦ C. At 200 ◦ C, the sample cell was flushed with nitrogen for 5 min and a spectrum (4 cm−1 , 100 scans) was recorded. In the TPA experiments, spectra (4 cm−1 , 100 scans) were collected every 25 ◦ C, i.e. every 5 min under toluene flow and additionally at 100, 200, 300, 400, 500 and 600 ◦ C under nitrogen flow. For comparison, an additional TPA experiment was performed using a calcined ZrO2 sample, i.e. without hydrogen pre-reduction. In the TPO experiments, toluene was adsorbed at 600 ◦ C for 30 min and then the sample was flushed with nitrogen while cooling down to 30 ◦ C. Next, the adsorbed species were oxidized (2% O2 /N2 ) with increasing temperature stepwise from 30 to 600 ◦ C while spectra (4 cm−1 , 100 scans) were collected every 25 ◦ C, i.e. every 5 min. In the TPSR experiments, 45 cm3 /min toluene (750 ppm in He) and 5 cm3 /min N2 (yielding a gas mixture of 675 ppm toluene, 90% He and 10% N2 ) was first fed to the sample at 100 ◦ C for 1 h. Thereafter N2 was replaced by synthetic air, resulting in a gas mixture of 675 ppm toluene, 90% He, 2% oxygen and 8% N2 . The oxygen concentration in the TPSR experiments was kept as low as possible but still resulting in an oxygen surplus, thus matching the oxygen feed composition in the gasification gas cleaning. After a 30-min stabilizing period, the sample cell was heated in the toluene-helium-air mixture from 100 to 600 ◦ C. The spectra (4 cm−1 , 100 scans) were collected every 25 ◦ C, i.e. every 5 min. The gas flow from the reactor in the TPSR experiments was followed with a Pfeiffer Vacuum Omnistar mass spectrometer. It should be noted that accurate kinetic data cannot be obtained because there


is not a well-defined contact between the gas flow and the catalyst bed. 2.2. Temperature-programmed gas-phase analysis Temperature-programmed (TP) gas-phase analysis was applied to study in detail the gas-phase products during adsorption of toluene. TP gas-phase analyses were designed to mimic the DRIFTS experiments. However, when preparing the samples, the powder samples were first pressed into pellets. Furthermore, to obtain the desired particle size of 0.25–0.42 mm, the pellets were crushed and sieved. The sample amount used in the experiments was 0.1 g. TP experiments were carried out in an Altamira AMI-100 characterization system. Prior to the experiments, a similar pre-treatment method was applied as in the DRIFTS experiments: the samples were calcined in situ in O2 /He flow of 50 cm3 /min (5 vol% O2 in He, AGA) at 600 ◦ C for 2 h, and reduced in 5 vol% H2 /He for 15 min (H2 99.999%, AGA). Gaseous products were analyzed with a mass spectrometer (MS, OmniStar GSD320, Pfeiffer Vacuum). The m/z values monitored in each experiment were 2 (H2 ), 4 (He), 15 (CH4 ), 16 (O2 ), 18 (H2 O), 28 (CO), 32 (O2 ), 44 (CO2 ), 65 (toluene), 78 (benzene), 91 (toluene), 92 (toluene), 106 (benzaldehyde), 108 (benzyl alcohol) and 122 (benzoic acid). The TPD experiments were similar to the corresponding DRIFTS experiments: first toluene (750 ppm in He with 1–2 ppm water and even smaller amount of methanol as impurities, AGA) with a flow of 49 cm3 /min was fed to the samples at ambient temperature for 30 min, then the reactor was flushed with a flow of 50 cm3 /min of He (99.996%, AGA) for 30 min and a TPD experiment was performed under the same He flow with a temperature ramp from ambient temperature to 600 ◦ C (at the rate of 10 ◦ C/min). The desorbed toluene amounts were quantified. Toluene-derived species possibly remaining on the ZrO2 surface after the TPD heat ramp was investigated by TPO. The results showed no significant amounts of CO2 , and therefore it could be concluded that no toluene remained on the surface after desorption at 600 ◦ C. A blank TPD experiment was performed on Y2 O3 –ZrO2 to follow the formation of gas-phase products from the catalyst surface without toluene adsorption. The TPA and TPO experiments were carried out in the same test cycle: first TPA followed by TPO, thus slightly differing from the corresponding DRIFTS experiments. During the TPA, a total flow of 49 cm3 /min of a gas mixture containing toluene (750 ppm in He, AGA) was passed through the sample first at ambient temperature for 30 min. Thereafter, a temperature ramp was applied under the same toluene flow (ambient to 600 ◦ C, 10 ◦ C/min). The toluene and possible products from its reactions (CO2 , CO, H2 , H2 O, benzene) during adsorption with increasing temperature were recorded with MS. Toluene flow was kept constant at 600 ◦ C for 15 min, after which the reactor was flushed with a He flow of 50 cm3 /min for 30 min and cooled down to 30 ◦ C to start the TPO experiment. The TPO experiment was carried out from ambient temperature to 600 ◦ C (ramp rate 10 ◦ C/min) under a gas mixture of 25 cm3 /min O2 /He (5 vol%) and 25 cm3 /min He, resulting in a total flow of 50 cm3 /min and an oxygen content of 2.5 vol%. The sample was let to dwell at 600 ◦ C for 30 min. The consumption of oxygen and the formation of oxidation products (CO2 , CO, H2 , and H2 O) were monitored with MS. The possible formation of oxygen containing benzene derivatives was monitored (m/z values of 106, 108, and 122), but these species were not observed. 3. Results The nature of adsorbed toluene species in the absence and presence of gas-phase oxygen was studied in order to evaluate their interaction over ZrO2 -based gasification gas clean-up catalysts.

771

Toluene TPD was studied to discover the surface species formed upon the adsorption of toluene at low temperature and to evaluate their thermal stability. Toluene TPA was studied to follow the evolution of surface species and gas-phase products as a function of temperature in the absence of gas-phase oxygen. TPO of the toluene-derived surface species was studied to determine their reactivity toward oxygen. TPSR between toluene and oxygen was studied to discover additional surface species during toluene oxidation. 3.1. Temperature-programmed desorption of toluene (TPD) 3.1.1. Surface species during toluene TPD Toluene adsorption at 30 ◦ C produced several peaks in the spectra of ZrO2 , Y2 O3 –ZrO2 , and SiO2 –ZrO2 (all the vibrations and their assignments are collected in Table 2). The spectra of pure ZrO2 in the region of 3200–2600 cm−1 recorded during the toluene TPD experiment are shown in Fig. 1a. Small peaks at 3030, 2926, 2818, 1601, and 1492 cm−1 remained in the spectrum after toluene adsorption and nitrogen flush at 30 ◦ C. The peak at 3030 cm−1 can be assigned to the C H stretching modes of the aromatic ring of toluene and the peaks at 1601 and 1492 cm−1 (Fig. S1a in Supplementary data) to the skeleton vibrations of the aromatic ring of toluene [19]. The peak at 2926 cm−1 could be assigned to (CH) modes of the methyl group of toluene [19] but the lower intensity compared to that of peak at 3030 cm−1 suggests that this assignment is unlikely. Therefore, the small peaks at 2926 and 2818 cm−1 are assigned to impurity-originated (trace amount of CH3 OH was analyzed from the feed) methoxy species together with the small peaks at 1153 and 1057 cm−1 (not shown) [18,20]. Supplementary data related to this article found, in the online version, at http://dx.doi.org/10.1016/j.apcatb.2013.06.008. The spectra of Y2 O3 –ZrO2 and SiO2 –ZrO2 in the region of 3200–2600 cm−1 collected during the toluene TPD experiments are shown in Fig. 1b and c, respectively. The number of peaks in the spectra of both Y2 O3 –ZrO2 and SiO2 –ZrO2 was higher compared to that of pure ZrO2 after the nitrogen flush. Furthermore, the intensity of the peaks in the spectra of SiO2 –ZrO2 was greater than in the spectra of the other catalysts. The peaks in the spectra of Y2 O3 –ZrO2 and SiO2 –ZrO2 at ca. 3080, 3060 and 3030 cm−1 can be assigned to the C H stretching modes of the aromatic ring of toluene, and the bands at ca. 1600 and 1495 cm−1 (Fig. S1b and c in Supplementary data) to the skeleton vibrations of the aromatic ring of toluene [19]. The peak at ca. 1600 cm−1 in the spectrum of SiO2 –ZrO2 was evidently much stronger and broader than the other peaks in the spectra. However, the assignment of this peak is unclear. In addition, the peaks at ca. 1450 and 1380 cm−1 over Y2 O3 –ZrO2 and SiO2 –ZrO2 (Fig. S1b and c in Supplementary data) can be assigned to the asymmetric and symmetric bending modes of the methyl group [19], while the bands at ca. 2925 and 2880 cm−1 can be assigned to (CH) modes of the methyl group of toluene [19]. Furthermore, a small peak at 1177 cm−1 in the spectra of Y2 O3 –ZrO2 suggests that methoxy species might also be present [18]. In addition, a peak at ca. 1420 cm−1 in the spectra of the doped zirconias (Fig. S1b and c in Supplementary data) could refer to surface carbonates together with the peak at ca. 1450 cm−1 . In addition, the peak at ca. 3770 cm−1 (not shown) in the spectrum of ZrO2 and Y2 O3 –ZrO2 , previously assigned to terminal OH groups of ZrO2 [7], disappeared during the adsorption of toluene, whereas a relatively broad peak appeared at ca. 3575 cm−1 . Similar results were obtained by Hernández-Alonso et al. [19], who suggested that the peak at ca. 3575 cm−1 indicates that the adsorption of toluene occurs via hydrogen bonding on the isolated (terminal) OH groups of the catalyst surface. Furthermore, some water entered as a minor impurity in the toluene/He mixture and accumulated on the surface of ZrO2 -based catalysts, thereby produced an extremely


772

Table 2 Vibrations and assignments of IR peaks appearing in the spectra of ZrO2 -based catalysts in toluene adsorption experiments. Vibrations

Assignment

References

3770 3575 3085, 3055, 3025 3070 2960, 2880 2925, 2880 2925, 2820, 1153, 1057 2880, 1563, 1443, 1365 1600, 1580, 1495 1510, 1410 1450, 1370 1440, 1430 1440, 1430

Terminal OH-groups Hydrogen-bonded OH stretching vibration Molecularly adsorbed toluene (C H stretching vibrations of the aromatic ring) C H stretching vibrations of the aromatic ring Benzyl species (C H stretching vibrations of CH2 group) Molecularly adsorbed toluene (C H stretching vibrations of the methyl group) Methoxy species (CH3 and O CH3 vibrations) Formate species (C H and COO vibrations) Molecularly adsorbed toluene (skeletal vibrations of the aromatic ring) Benzoate species (COO vibrations) Molecularly adsorbed toluene (bending vibrations of the methyl group) Monodentate carbonatesa on Y2 O3 -doped ZrO2 Polydentate carbonatesa on SiO2 -doped ZrO2

[7,17] [19] [19] [15] [21] [19] [18,20] [18,20] [19] [8,15] [19] [7,22] [7,17]

a

The assignments of monodentate and polydentate carbonates on doped zirconias are based on the thermal stability of these species. Further details can be found in [7].

Fig. 1. DRIFTS spectra from the TPD experiments in the region of 3200–2600 cm−1 of calcined and reduced (a) ZrO2 , (b) Y2 O3 –ZrO2 and (c) SiO2 –ZrO2 after toluene adsorption at 30 ◦ C followed by heating in nitrogen from 30 to 500 ◦ C.

broad peak in the spectra of all the catalysts across the range of interest. Upon heating the samples after the adsorption of toluene and the nitrogen flush (Fig. 1), the peaks assigned to molecular toluene vanished from the spectra of all of the ZrO2 -based catalysts at temperatures below 200 ◦ C, indicating desorption of adsorbed molecular toluene. Small new peaks at 2880, 1563, 1443 and 1365 cm−1 appeared at 225 ◦ C to the spectrum of pure ZrO2 , suggesting that small amounts of formate species might be present. Methoxy species observed over pure ZrO2 desorbed at 400 ◦ C, while the formate species desorbed at 425 ◦ C, in accordance with methanol adsorption experiments in the literature [18,20]. During the heat ramp, the terminal OH groups of the surface of the ZrO2 -based samples were restored (not shown). 3.1.2. Gas-phase products during toluene TPD The gas-phase products during adsorption of toluene (m/z = 91) at ambient temperature followed by desorption with heating up to 600 ◦ C were studied over the ZrO2 -based catalysts. Fig. 2 shows the adsorption of toluene and the following TPD. All the ZrO2 based catalysts were able to adsorb toluene. The lowest amount of toluene was adsorbed on pure ZrO2 and the highest amount on SiO2 -doped ZrO2 . This is shown in Fig. 2, where the signal of toluene on ZrO2 levels off first, indicating that this sample is saturated with toluene sooner than the doped zirconias. It appears that desorption of toluene was taking place already during the isothermal inert flush, but it accelerated with heating. The inset

in Fig. 2 shows the temperature dependence of toluene desorption peaks. The temperatures of the desorption rate maxima (Tmax , ◦ C) of toluene over ZrO , Y O –ZrO and SiO –ZrO were 70, 80, 2 2 3 2 2 2 and 88 ◦ C, respectively. The amount of desorbed toluene per area of catalyst on ZrO2 , Y2 O3 –ZrO2 , and SiO2 –ZrO2 was calculated to be approximately 0.13, 0.23, and 0.23 ␮mol/m2cat , corresponding to less than 4% of a monolayer. In addition to toluene, trace amounts of CO2 and CO were detected on all the catalysts during TPD (not shown). A blank TPD experiment, i.e. without toluene adsorption, showed no CO2 evolution from Y2 O3 –ZrO2 . The formation of carbon oxides in toluene TPD experiments has also been observed by Saqer et al. [23] over Al2 O3 -supported metal oxide catalysts. They related the formation of CO2 to the reduction of the catalyst surface and to the activity of surface oxygen [23]. ZrO2 -based catalysts in this work, on the other hand, were pretreated by hydrogen reduction and therefore were expected to possess only a minor redox capacity. Thus it is expected that trace amount of feed-originated methanol (or CO2 ) was retained on the surface as carbonates (in accordance with ref [18]) that were detected in the DRIFTS spectra (Section 3.1.1) and subsequently released at characteristic temperatures. Furthermore, significant amounts of water desorbed during TPD from all the catalysts. The Tmax values (◦ C) of the H2 O desorption curves were 290 ◦ C (ZrO2 ), 340 ◦ C (Y2 O3 –ZrO2 ) and 385 ◦ C (SiO2 –ZrO2 ). A blank TPD experiment (without toluene adsorption) on Y2 O3 –ZrO2 indicated a minor desorption process of water with similar qualitative characteristics in similar temperature range


773

Fig. 2. Toluene adsorption at ambient temperature followed by desorption to 600 ◦ C over ZrO2 , Y2 O3 –ZrO2 and SiO2 –ZrO2 . The inset shows the toluene desorption peaks as a function of temperature.

(∼250–450 ◦ C). Therefore, the desorption of water during TPD is likely related to dehydroxylation of the ZrO2 surfaces at these temperatures. For example, Bianchi et al. [24] connected water formation during methanol TPD over ZrO2 to dehydroxylation of the surface, where two hydroxyl groups are combined to yield water in the gas phase. The water evolution in the toluene TPD was tenfold in amount compared to the blank TPD. This is explained by more extensive hydroxylation of zirconia caused by water that entered as a minor impurity in the toluene/He mixture and its accumulation on the catalyst during the preceding adsorption stage. 3.2. Temperature-programmed adsorption of toluene (TPA) 3.2.1. Surface species during toluene TPA In the TPA experiments, toluene adsorption on the calcined and hydrogen reduced ZrO2 , Y2 O3 –ZrO2 and SiO2 –ZrO2 samples was studied in the temperature range of 100 to 600 ◦ C. The spectra in the range of 1700–1300 cm−1 are shown in Fig. 3 for all of the catalysts. Toluene adsorption at 100 ◦ C was similar to the adsorption of toluene at 30 ◦ C. Several peaks appeared in the spectra of all the ZrO2 -based catalysts, the peak assigned to the terminal OH groups (at ca. 3770 cm−1 ) disappeared, and a broad peak at 3575 cm−1 appeared (not shown). Small peaks at 2928 (Fig. S2a in Supplementary data) and 1489 cm−1 (Fig. 3a) remained in the spectrum of pure ZrO2 during nitrogen flush after the adsorption of toluene at 100 ◦ C. The peak at 2928 cm−1 can be assigned to (CH) modes of the methyl group [19], and the peak at 1490 cm−1 can be assigned to the skeleton vibrations of the aromatic ring of toluene [19]. Supplementary data related to this article found, in the online version, at http://dx.doi.org/10.1016/j.apcatb.2013.06.008. The number of peaks after adsorption of toluene at 100 ◦ C was again higher in the spectra of both Y2 O3 –ZrO2 (Fig. 3b) and SiO2 –ZrO2 (Fig. 3c) compared to that of pure ZrO2 (Fig. 3a). The peaks in the spectra of Y2 O3 –ZrO2 and SiO2 –ZrO2 at ca. 3080, 3060, 3030, 2925 (Fig. S2b and c in Supplementary data), 1600, 1495, and 1450 cm−1 can be assigned to the adsorption of molecular toluene similar to the adsorption at 30 ◦ C (Fig. 1). Similarly to the TPD experiments, the peak at ca. 1600 cm−1 in the spectrum of SiO2 –ZrO2 was much stronger and broader than the other peaks in the spectra. The assignment of this peak remains unclear. In addition, a peak

at ca. 1420 cm−1 appeared in the spectra of both Y2 O3 –ZrO2 and SiO2 –ZrO2 . The strong bands at ca. 1450 and 1420 cm−1 have been previously assigned to the adsorbed CO2 species [7]. The existence of these carbonate species is related to the feed-originated minor CH3 OH impurity in accordance with the results of the gas-phase analysis during the toluene TPD experiments (Section 3.1.2). Heating under toluene flow from 100 to 600 ◦ C resulted in the appearance of new peaks at temperatures above 300 ◦ C in the spectra of pure ZrO2 (Fig. 3a) at ca. 3070 (Fig. S2a in Supplementary data), 1580, 1510, and 1410 cm−1 . The intensity of these new peaks increased with increasing temperature. The peaks at ca. 1510 and 1410 cm−1 can be assigned to asymmetric and symmetric stretching modes of adsorbed benzoate species, whose presence indicate that the C-H bonds in the methyl group of toluene can dissociate [8]. First hydrogen is abstracted from the methyl group of toluene and a surface benzoate is formed with two surface oxygen atoms of the catalyst [8]. In addition, the bands appearing at 3070 and 1585 cm−1 can be assigned to the skeleton vibrations of the aromatic ring of benzoate species [8,19]. The peaks at 1540 and 1340 cm−1 are original peaks of the ZrO2 sample. The reference TPA experiment on calcined ZrO2 showed the same surface species as on the calcined and reduced ZrO2 : molecularly adsorbed toluene at low temperatures and benzoate species at higher temperatures. However, the amount of benzoate species was higher on the calcined than on the calcined and reduced ZrO2 surface, which seems logical because the calcined sample has more oxygen available on the surface of the catalyst. In the spectra of Y2 O3 –ZrO2 (Fig. 3b), peaks at 3080, 3060, 3030 (Fig. S2b in Supplementary data), 1600, and 1494 cm−1 vanished below 300 ◦ C, the peak at 1447 cm−1 remained in the spectra throughout the temperature range, and the intensity of the peak at 1420 cm−1 grew with increasing temperature. In addition, a new band at ca. 3070 cm−1 (Fig. S2b in Supplementary data), which can be assigned to C-H stretching of the aromatic ring [15], appeared to the spectrum of Y2 O3 –ZrO2 at temperatures above 400 ◦ C. Previously, bands at ca. 1450 and 1420 cm−1 (which vanish below 500 ◦ C) have been assigned to monodentate carbonates formed in the adsorption of CO2 to the surface of Y2 O3 –ZrO2 [7]. Therefore, the thermal stability of the peaks suggests that at lower temperatures, the peaks at 1447 and 1418 cm−1 can be assigned to monodentate

774


Fig. 3. DRIFTS spectra from the TPA experiments in the region of 1700–1300 cm−1 of calcined and reduced (a) ZrO2 , (b) Y2 O3 –ZrO2 and (c) SiO2 –ZrO2 during toluene adsorption at 100–600 ◦ C. Spectra recorded after a 5-min nitrogen flush at each temperature.

carbonates [7], and at higher temperatures the peak at 1418 cm−1 can be assigned to the benzoate species together with the increasing peak at 1512 cm−1 , which is first detected at 400 ◦ C [8]. The peak at 1339 cm−1 is an original peak of the Y2 O3 –ZrO2 sample. In the spectra of SiO2 –ZrO2 (Fig. 3c), the peak at 1495 cm−1 vanished below 200 ◦ C and the peaks at 3080, 3060, and 3030 cm−1 below 300 ◦ C (Fig. S2c in Supplementary data), while the peaks at 1598, 1448, and 1419 cm−1 vanished above 300 ◦ C. New peaks appeared in the spectra of SiO2 –ZrO2 at 3070, 1504, and 1406 cm−1 above 300 ◦ C. The peaks appearing at lower temperatures (1448 and 1419 cm−1 ) can be assigned to polydentate carbonates formed by the adsorption of CO2 to the surface of SiO2 –ZrO2 in accordance with our previous findings (monodentate carbonates on Y2 O3 –ZrO2 and polydentate carbonates on SiO2 –ZrO2 ) [7] and the increasing peaks appearing at higher temperatures (1504 and 1406 cm−1 ) to benzoate species [8]. In addition, the peak at 3070 cm−1 can be assigned to C-H stretching of the aromatic ring of benzoate [15]. The peak 1359 cm−1 is an original peak of the SiO2 –ZrO2 sample. 3.2.2. Gas-phase products during toluene TPA The formation of gas-phase products from toluene over zirconiabased catalysts was followed from ambient temperature to 600 ◦ C and thereafter during isothermal hold under toluene flow at the end temperature for 30 min. In the beginning of the experiment, the toluene feed was switched on and toluene adsorption on the catalysts continued until adsorption equilibrium was reached. Immediately when the heating was started, the net desorption of toluene began as the toluene outlet flow level rose above the toluene feed level. As an example, Fig. 4 shows the TPA response of toluene as a function of time with ZrO2 and illustrates the net desorption peaks for all the catalysts as a function of temperature. The toluene desorption peaks in the TPA experiments were more intense than in the TPD experiments, suggesting that during the inert flush prior to the TPD a fraction of toluene was lost. The temperature range of the net desorption of toluene in the TPA was lower with all of the catalysts (Tmax values 30, 55, and 40 ◦ C for ZrO2 , Y2 O3 –ZrO2 , and SiO2 –ZrO2 , respectively) compared to the TPD experiments as well. The toluene net desorption amounts were 1.42, 1.23, and 0.82 ␮mol/m2cat over ZrO2 , Y2 O3 –ZrO2 , and SiO2 –ZrO2 , respectively. Interestingly, the toluene net desorption amount was highest over ZrO2 in contrast to the toluene desorption amounts in the TPD experiments. All of the toluene net desorption

amounts were higher than in the TPD experiments because the lack of inert flush before the heating was started in the TPA experiments. After the net desorption peaks, toluene response returned to the feed level and remained there until almost the end of the temperature ramp. During the isothermal hold at 600 ◦ C, toluene was moderately consumed over pure ZrO2 and Y2 O3 –ZrO2 and less over SiO2 –ZrO2 . The responses of components other than toluene were similar to those observed during toluene TPD. The most striking difference between the TPD and the TPA of toluene was the formation of benzene accompanied by CO2 formation at 520–530 ◦ C in the TPA. Fig. 5 shows the responses of CO2 and benzene in detail. CO2 formation was observed around two temperatures over all the studied catalysts. Small peaks for CO2 were first detected at temperatures of 300–350 ◦ C as in the TPD experiments (Section 3.1.2) and can be related to a feed-originated minor (CH3 OH) impurity. A maximum for CO2 at high temperature (525 ◦ C for ZrO2 and Y2 O3 –ZrO2 , or 530 ◦ C for SiO2 –ZrO2 ) was common for all catalysts and was accompanied by the formation of benzene (C6 H6 ) at the same temperatures. Furthermore, the estimated molar amounts of CO2 and benzene desorbing at around 530 ◦ C were approximately equivalent (∼0.1 ␮mol/m2 for Y2 O3 –ZrO2 ). The intensities of both CO2 and benzene peaks decreased in the order of Y2 O3 –ZrO2 > ZrO2 > SiO2 –ZrO2 , which is interestingly the same order as the order of basicity on these catalysts [7]. Over ZrO2 and Y2 O3 –ZrO2 , the formation of CO2 , benzene (Fig. 5), and CO (not shown) increased after 550 ◦ C. These signals continued to increase even after 600 ◦ C was reached (not shown) simultaneously with toluene consumption. Moreover, some H2 was produced (Fig. 5) at high temperature (approx. 500 ◦ C) over pure ZrO2 and Y2 O3 –ZrO2 . However, the formation of H2 increased (Fig. 5) considerably when toluene feed was continued at 600 ◦ C for 30 min over all the studied catalysts. The increasing H2 amount in the gas phase might be related to dehydrogenation of toluene forming carbonaceous deposits on the catalyst surface and, moreover, the exponential shape suggests that this is autocatalytic. Carbon formation has been suggested to occur during gasification gas cleaning via hydrocarbon decomposition reactions producing H2 and carbon on the catalyst by Simell et al. [3]. Water was produced during the TPA experiments over ZrO2 , Y2 O3 –ZrO2 , and SiO2 –ZrO2 around 310, 375, and 400 ◦ C, respectively (Fig. 5). The H2 O Tmax values (◦ C) were similar to the TPD experiments


775

Fig. 4. Toluene response for ZrO2 in TPA: toluene/He flow switched on at 20 ◦ C, followed by constant heating and isothermal hold at 600 ◦ C. The inset shows the temperature dependence of toluene desorption peaks over all the ZrO2 -based catalysts.

(Section 3.1.2) and the origin of desorbing water is related here as well to the water that entered as a minor impurity in the toluene/He mixture and causing further hydration of the ZrO2 surfaces.

3.3. Temperature-programmed oxidation of adsorbed toluene species (TPO) 3.3.1. Surface species during toluene TPO In the TPO experiments, toluene was adsorbed at 600 ◦ C for 30 min and ZrO2 -based catalyst samples were flushed with nitrogen while cooling down to 30 ◦ C. Next, the oxidation of the adsorbed species was followed from 30 to 600 ◦ C. Peaks at ca. 3070 (not shown), 1600, 1530 (1504 for SiO2 –ZrO2 ), 1490, 1440, and 1410 cm−1 appeared during the adsorption of toluene at 600 ◦ C and remained in the spectrum of all the ZrO2 -based catalysts during cooling in nitrogen to 30 ◦ C (Fig. 6). The assignment of these peaks equals the assignments of the corresponding peaks in the higher temperatures of the TPA experiments. However, the peak at ca. 1600 cm−1 , although with very low intensity, might suggest formation of carbonaceous deposits on the catalysts. The significantly higher intensity of the peaks assigned to surface benzoates compared to the peak assigned to carbonaceous deposits is due to the higher extinction coefficient of C O bonds in benzoates than that of C C bonds in carbonaceous deposits. During oxidation, the peak assigned to carbonaceous deposits disappeared above 300 ◦ C from the spectra of the catalysts. The adsorbed benzoate species, on the other hand, vanished at temperatures above 400 ◦ C from the spectrum of Y2 O3 –ZrO2 and above 500 ◦ C from the spectra of ZrO2 and SiO2 –ZrO2 , indicating differing reactivity of benzoates toward oxygen.between the catalyst samples. In addition, small new peaks in the range of 3000–2800 cm−1 (not shown) in the spectra of all ZrO2 -based catalysts were observed during the heating of the samples in oxygen, which can be related to the rearranging of the adsorbed species before oxidation.

3.3.2. Gas-phase products during toluene TPO The surface species that were irreversibly adsorbed on the surface of the ZrO2 -based catalysts during the TPA experiments were oxidized in the TPO experiments by heating from ambient temperature to 600 ◦ C under O2 /He flow. The responses of O2 , CO2 , CO, H2 O, and H2 were of interest here. Fig. 7 shows the evolution of CO2 (m/z = 44) and CO (m/z = 28) and the consumption of oxygen (m/z = 32) during the temperature-programmed oxidation over the studied catalysts. As indicated by the consumption of oxygen and the evolution of carbon oxides, the oxidation of carbonaceous surface species took place between 200 ◦ C and 400 ◦ C. The evolution of carbon oxides was intense during TPO, suggesting that the amount of retained carbonaceous deposits was notable, thus confirming the assumed carbon laydown at the high temperature part of the TPA run. With ZrO2 , one clear maximum was detected at approx. 330 ◦ C for CO2 and CO. Over Y2 O3 –ZrO2 the formation of CO2 and CO was detected at 325 ◦ C and 335 ◦ C, respectively. Another evolution of both carbon oxides occurred with Y2 O3 –ZrO2 catalyst at approx. 400 ◦ C. The formation of CO2 over SiO2 –ZrO2 was clearly lower in amount with higher Tmax (maximum at 350 ◦ C) than over the other catalysts (Fig. 7). Moreover, the total ratio of formed CO to CO2 was highest over SiO2 –ZrO2 and O2 consumption was the lowest. Tmax values (◦ C) for water during TPO were 340, 400, and 435 ◦ C respectively for ZrO2 , Y2 O3 –ZrO2 and SiO2 –ZrO2 . The water curves again exhibited a similar shape and order of Tmax values (◦ C) as in the TPA experiments (Fig. 5) with different catalysts. Hydrogen formation (not shown) was detected in trace amounts over all the catalysts during the temperature range of the main oxidation events, Tmax values (◦ C) for H2 on ZrO2 , Y2 O3 –ZrO2 and SiO2 –ZrO2 being 355, 400, and 450 ◦ C, respectively. On ZrO2 , the formation of all the oxidation products (CO2 , CO, H2 and H2 O) occurred at 330–355 ◦ C (CO2 and CO shown in Fig. 7). Over Y2 O3 –ZrO2 and SiO2 –ZrO2 , on the other hand, the formation of H2 and H2 O can be correlated with the additional formation of CO2 and CO at higher temperature (a shoulder can be seen in

776


Fig. 7. CO2 and CO formation and O2 consumption (m/z values of 44, 28, and 32 respectively) from ambient to 600 ◦ C during the TPO-experiments. Fig. 5. MS signals for benzene, CO2 , H2 , and H2 O (m/z = 78, 44, 2, and 18 respectively) over ZrO2 -based catalysts from ambient to 600 ◦ C during the toluene TPA experiments. CO2 formation over all catalysts is also shown during the TPD experiments for comparison.

Fig. 6. DRIFTS spectra from the TPO experiments in the region of 1700–1300 cm−1 of calcined and reduced (a) ZrO2 , (b) Y2 O3 –ZrO2 and (c) SiO2 –ZrO2 after toluene adsorption at 600 ◦ C and subsequent cooling to 30 ◦ C during oxidation from 30 to 600 ◦ C. All spectra were recorded under 2% O2 /N2 .


777

Fig. 8. DRIFTS spectra from the TPSR experiments in the region of 3200–2800 cm−1 of calcined and reduced (a) ZrO2 , (b) Y2 O3 –ZrO2 and (c) SiO2 –ZrO2 during co-adsorption of toluene and oxygen from 100 to 600 ◦ C.

Fig. 7). Therefore, it is suggested that the carbonaceous deposits are more heterogeneous on the doped zirconias than on pure ZrO2 . 3.4. Temperature-programmed surface reaction between toluene and oxygen (TPSR) The co-adsorption of toluene and oxygen and their reactions were studied from 100 to 600 ◦ C by in situ DRIFTS. Analogously to the TPD and the TPA experiments, molecularly adsorbed toluene species (peaks at ca. 3080, 3060, 3030, 2925, 2820, 1600, 1545 and 1490 cm−1 ) were observed at low temperatures and adsorbed benzoate species (peaks at ca. 3070 (Fig. 8), 1510 and 1410 cm−1 (Fig. S3 in Supplementary data) at higher temperatures in the spectra of all the ZrO2 -based catalysts. However, at temperatures above 300 ◦ C, new peaks compared to the toluene TPA experiments could be seen in the spectra of all the catalysts. The peaks at ca. 2960 and 2880 cm−1 in the spectra of all the catalysts above 300 ◦ C can be assigned to the symmetric and asymmetric C H stretching vibrations of a CH2 group characteristic of benzyl species [21]. When toluene is adsorbed dissociatively giving one hydrogen atom from its methyl group, benzyl species are formed. Furthermore, the peaks at ca. 3070 and ca. 1580 cm−1 can be assigned to the skeleton vibrations of the aromatic ring of both benzyl and benzoate species [8,19]. Interestingly, only small, barely detectable benzyl peaks were detected in the reference TPA experiment where toluene was adsorbed on the calcined ZrO2 , therefore suggesting that gasphase oxygen increased the surface concentration of the benzyl species. Supplementary data related to this article found, in the online version, at http://dx.doi.org/10.1016/j.apcatb.2013.06.008. In the TPSR experiments, the gas flow from the DRIFTS cell was followed with an on-line mass spectrometer. As an example, the MS signals for the consumption of toluene and oxygen in the TPSR experiment over Y2 O3 –ZrO2 are shown in Fig. 9. Over all the ZrO2 -based catalysts, both toluene and oxygen started to convert at temperatures above 300 ◦ C (shown for Y2 O3 –ZrO2 in Fig. 9). The product formation was similar for all the ZrO2 -based catalysts: water was formed below 300 ◦ C due to desorption of physisorbed water and the oxidation products (CO2 , CO, H2 O, and H2 ) started to form above 300 ◦ C.

4. Discussion 4.1. Adsorption of toluene in the absence of gas-phase oxygen over ZrO2 -based catalysts Toluene adsorption in the absence of gas-phase oxygen was studied over ZrO2 , Y2 O3 –ZrO2 , and SiO2 –ZrO2 . The lowtemperature-adsorption of toluene resulted in molecularly adsorbed toluene on the surface of all the ZrO2 -based catalysts (Fig. 1). Molecularly adsorbed toluene desorbed from the catalysts below 200 ◦ C (Figs. 1 and 2). A similar relatively weak toluene adsorption mode that becomes unstable below 200 ◦ C has been observed for a number of oxides [23,25]. The toluene curves in the TP experiments showed that the amount of toluene desorbing from the catalyst surface decreased in the order of SiO2 –ZrO2 > Y2 O3 –ZrO2 > ZrO2 (Fig. 2). When the specific surface areas are taken into account, the amount of desorbed toluene in the TPD experiments was approximately 0.13, 0.23, and 0.23 ␮mol/m2cat on ZrO2 , Y2 O3 –ZrO2 , and SiO2 –ZrO2 , respectively. The total acidity and total basicity of these ZrO2 -based catalysts (Table 1) have been measured in our previous work [7], but there is no clear correlation between toluene adsorption capacity and

Fig. 9. Consumption of toluene and oxygen in the TPSR experiment over Y2 O3 –ZrO2 .

778


the acidity or the basicity of these catalysts. The differences in capacities per surface area for pure and doped zirconias obtained in this work are attributed to different chemical compositions that induce different crystalline phases of samples: pure ZrO2 being monoclinic and doped samples containing tetragonal and cubic phases [7]. The adsorption of toluene in the absence of gas-phase oxygen (TPA experiments) revealed benzoate species on ZrO2 as well as on SiO2 –ZrO2 at temperatures above 300 ◦ C and on Y2 O3 –ZrO2 above 400 ◦ C (Fig. 3). Benzoate species are formed when hydrogen is first abstracted from the methyl group of toluene and then a benzoate ion is formed using two surface oxygen atoms of the catalyst [8]. The analyses of gas-phase products during toluene adsorption in the absence of gas-phase oxygen showed that benzene and CO2 were formed over all the catalysts at high temperatures of 525 ◦ C (530 ◦ C for SiO2 –ZrO2 ) (Fig. 5). The temperature at which benzene and CO2 were formed (Fig. 5) is higher than the temperature at which surface benzoate species started to appear in the spectra of the catalysts (Fig. 3). This suggests that some of the surface benzoate species undergo scission to benzene and CO2 . This is also supported by the estimated equimolar amounts of CO2 and benzene (approx. 0.1 ␮mol/m2 for Y2 O3 –ZrO2 ) produced, thus suggesting decarboxylation of surface benzoate species. The amount of O contained in the CO2 detected corresponds roughly to 0.8% of surface oxygen atoms in the Y2 O3 –ZrO2 surface. The small fraction of the surface oxygen species involved clearly indicated that minority sites like edge or corner sites are responsible, similar to the results of Zhu et al. [26], who reported somewhat higher amount of O removed from yttria-doped zirconia with methane at much higher temperature (900 ◦ C). Furthermore, it may well be that surface OH groups are involved as a minority site, similar to the results of Jonson et al. [27], who reported that decarboxylation of benzoate species on V2 O5 /Al2 O3 -C catalysts benzene (and CO2 ) requires OH groups of the catalyst surface. Although the gasification gas mixture contains several compounds other than applied in this work, the role of surface benzoates species can be speculated. If toluene adsorbs as benzoates on the catalyst surface during gasification gas cleaning, the formation of unwanted side products is possible. Benzene formation via surface benzoates was detected during the TPA experiments (Figs. 3 and 5). Benzene has also been previously observed during gasification gas cleaning. Even though benzene is not classified as tar, its presence is not desired if the product gas is targeted, for example, at Fischer–Tropsch synthesis [6]. Furthermore, according to the literature, the ammoxidation of toluene

goes through adsorbed surface benzoates that react with ammonia (ammonia is always present in the gasification gas) to benzonitrile [28]. Benzonitrile has previously been detected during gasification gas cleaning over ZrO2 -based catalysts [28]. Benzonitrile is a harmful compound and is also classified as tar, thus an undesirable component in the product gas. 4.2. Oxidation of toluene-derived surface species Essential information on the oxidation of toluene-derived species was gathered in the TPO experiments, where residual carbon-containing species that originate from toluene adsorption were removed from the surface of the catalysts by oxidation. The major formation of oxidation products in the gas phase was observed at approx. 330 ◦ C (Fig. 5), whereas the surface benzoate species vanished from the spectrum of Y2 O3 –ZrO2 above 400 ◦ C and from the spectra of ZrO2 and SiO2 –ZrO2 above 500 ◦ C (Fig. 3). Therefore, it seems that benzoate species are not responsible for the major formation of oxidation products. However, the small peak at ca. 1600 cm−1 tentatively assigned to carbonaceous deposits disappeared from the spectra of all the catalysts above 300 ◦ C (Fig. 3), thus suggesting that it could be related to the major formation of oxidation products detected in the gas phase (Fig. 5) at the temperature in question. The fact that surface benzoate species are more stable than the carbonaceous deposits toward oxygen.seems to be surprising at first sight but is actually expected, since adsorbed benzoate species have been reported to be highly stable according to Hernández-Alonso et al. [19]. In the TPO experiments, the formation of CO2 and CO started at a lower temperature over ZrO2 and Y2 O3 –ZrO2 than over SiO2 –ZrO2 (Fig. 7). Furthermore, it is evident that the carbonaceous deposits on the doped zirconias are more heterogeneous than on pure ZrO2 based on the TPO results; over pure ZrO2 the carbon on the surface was oxidized in one step, whereas on the doped zirconias two events could be distinguished (Fig. 7). The simultaneous adsorption of toluene and oxygen showed yet another surface species on ZrO2 -based catalysts that was not observed in any other experiments; at temperatures above 300 ◦ C benzyl species appeared in the spectra (Fig. 8). Benzyl species are formed from toluene when one hydrogen atom from the methyl group is abstracted and toluene is adsorbed via the methylene group ( CH2 ) on the surface of the catalyst [21]. At temperatures above 300 ◦ C, the consumption of both toluene and oxygen was initiated and increased with increasing temperature (Fig. 9). The temperature where consumption of toluene and oxygen started

Fig. 10. Schematic presentation of the mechanistic steps involved in the adsorption and oxidation of toluene over ZrO2 -based gasification gas clean-up catalysts.


(Fig. 9) correlates well with the formation temperature of the surface benzyl species (Fig. 8). Furthermore, at higher temperatures (above 500 ◦ C) where the consumption of toluene and oxygen started to stabilize (Fig. 9), the surface benzyl species vanished from the spectrum (Fig. 8). The surface benzoate species, on the other hand, showed evidently strong peaks in the spectra of ZrO2 -based catalysts still at 600 ◦ C (Fig. 8). Of all the surface species that were derived from toluene, benzoates were also found to be least reactive toward oxygen. Therefore, it can be suggested that the surface benzoate species are spectator species in toluene oxidation, similar to the finding of Paulis et al. [16], who observed that benzoate species became spectators in the total oxidation of toluene over Pd/Al2 O3 catalyst as well. Furthermore, the surface benzyl species seem to be the intermediate of the toluene oxidation reaction over ZrO2 -based catalysts, similar to the postulation of Busca [29] for several vanadia catalysts (partial oxidation) as well as for ␣-Fe2 O3 (total oxidation). 5. Conclusions The interaction of toluene with selected ZrO2 -based gasification gas clean-up catalysts in the absence and presence of gas-phase oxygen was studied with two complementary techniques: in situ DRIFTS and temperature-programmed gas-phase analysis. The combination of these two techniques offered unique knowledge on the adsorbed surface species from toluene as well as information on their decomposition and/or oxidation mechanisms over ZrO2 , Y2 O3 –ZrO2 , and SiO2 –ZrO2 . Using this knowledge, the nature of the adsorbed toluene-derived surface species (the mechanistic steps are shown schematically in Fig. 10) over ZrO2 -based gasification gas clean-up catalysts could thus be evaluated. The adsorption of toluene in the absence of gas-phase oxygen leads to three types of surface species over ZrO2 -based gasification gas clean-up catalysts: molecularly adsorbed toluene, surface benzoate species, and carbonaceous deposits. Molecularly adsorbed toluene desorbs at low temperatures without reacting. Surface benzoate species are formed when toluene adsorbs on the ZrO2 -based catalysts at temperatures above 400 ◦ C. In the absence of gas-phase oxygen, some of the benzoate species decompose into benzene and CO2 corresponding roughly to a removal of 0.8% ML of surface oxygen atoms from the surface of Y2 O3 –ZrO2 . When toluene and oxygen are co-fed to ZrO2 -based catalysts, the presence of benzyl species can be detected. The benzyl species appear to the spectra of all ZrO2 -based catalysts at the same temperature where toluene and oxygen start to convert and oxidation products (CO2 , CO, H2 O and H2 ) start to form. However, at higher temperatures (above 500 ◦ C) where the consumption of toluene and oxygen started to stabilize, the surface benzyl species vanished from the spectrum. The surface benzoate species, on the other hand, were still present at 600 ◦ C, indicating strong adsorption of these species. Therefore, it is suggested that the highly stable surface benzoate species are spectators, while the surface benzyl species seem to be a key intermediate in the toluene oxidation reaction.

779

Acknowledgements The authors would like to thank Ms. Kaisa Vikla and Ms. Heidi Meriö-Talvio for their assistance with the experiments. The authors are grateful for the funding from the Ministry of Education of Finland and the Academy of Finland. Also the Finland Distinguished Professor Programme (FiDiPro) funded by the Finnish Funding Agency for Technology and Innovation (TEKES) is acknowledged for the financial support. The zirconia samples were kindly provided by MEL Chemicals. References [1] P. Gallezot, A. Kiennemann, in: G. Ertl, H. Knözinger, F. Schüth, J. Weitkamp (Eds.), Handbook of Heterogeneous Catalysis, 2nd ed., Wiley-VCH, Weinhem, 2008, pp. 2447–2476. [2] D. Sutton, B. Kelleher, J.R.H. Ross, Fuel Processing Technology 73 (2001) 155–173. [3] P. Simell, E. Kurkela, P. Ståhlberg, J. Hepola, Catalysis Today 27 (1996) 55–62. [4] H. Rönkkönen, E. Rikkinen, J. Linnekoski, P. Simell, M. Reinikainen, O. Krause, Catalysis Today 147S (2009) S230–S236. [5] S.J. Juutilainen, P.A. Simell, A.O.I. Krause, Applied Catalysis B 62 (2006) 86–92. [6] H. Rönkkönen, P. Simell, M. Reinikainen, O. Krause, Topics in Catalysis 52 (2009) 1070–1078. [7] T. Viinikainen, H. Rönkkönen, H. Bradshaw, H. Stephenson, S. Airaksinen, M. Reinikainen, P. Simell, O. Krause, Applied Catalysis A 362 (2009) 169–177. [8] A.A. Davydov, Materials Chemistry and Physics 19 (1988) 97–112. [9] M. Niwa, M. Sago, H. Ando, Y. Murakami, Journal of Catalysis 69 (1981) 69–76. [10] S.L.T. Andersson, Journal of Catalysis 98 (1986) 138–149. [11] G. Busca, F. Cavani, F. Trifirò, Journal of Catalysis 106 (1987) 471–482. [12] N.R. Avery, Journal of the Chemical Society: Chemical Communications 3 (1988) 153–154. [13] H. Miyata, T. Ohno, F. Hatayama, Journal of the Chemical Society: Faraday Transactions 91 (1995) 3505–3510. [14] R. Méndez-Román, N. Cardona-Martı´ınez, Catalysis Today 40 (1998) 353–365. [15] S. Besselmann, E. Löffler, M. Muhler, Journal of Molecular Catalysis A 162 (2000) 401–411. [16] M. Paulis, L.M. Gandı´ıa, A. Gil, J. Sambeth, J.A. Odriozola, M. Montes, Applied Catalysis B 26 (2000) 37–46. [17] B. Bachiller-Baeza, J.A. Anderson, Journal of Catalysis 212 (2002) 231–239. ˜ [18] E.I. Kauppi, E.H. Rönkkönen, S.M.K. Airaksinen, S.B. Rasmussen, M.A. Banares, A.O.I. Krause, Applied Catalysis B 111–112 (2012) 605–613. [19] M.D. Hernández-Alonso, I. Tejedor-Tejedor, J.M. Coronado, M.A. Anderson, Applied Catalysis B 101 (2011) 283–293. ˜ [20] S.T. Korhonen, M.A. Banares, J.L.G. Fierro, A.O.I. Krause, Catalysis Today 126 (2007) 235–247. [21] G. Busca, T. Zerlia, V. Lorenzelli, A. Girelli, Reaction Kinetics and Catalysis Letters 27 (1985) 429–432. [22] C. Morterra, G. Cerrato, L. Ferroni, Journal of the Chemical Society: Faraday Transactions 91 (1995) 125–132. [23] S.M. Saqer, D.I. Kondarides, X.E. Verykios, Applied Catalysis B 103 (2011) 275–286. [24] D. Bianchi, T. Chafik, M. Khalfallah, S.J. Teichner, Applied Catalysis A 123 (1995) 89–110. [25] R.I. Slioor, J.M. Kanervo, T.J. Keskitalo, A.O.I. Krause, Applied Catalysis A 344 (2008) 183–190. [26] J. Zhu, J.G. van Ommen, H.J.M. Bouwmeester, L. Lefferts, Journal of Catalysis 233 (2005) 434–441. [27] B. Jonson, B. Rebenstorf, R. Larsson, S.L. Andersson, S.T. Lundin, Journal of the Chemical Society: Faraday Transactions 82 (1986) 767–783. [28] S.J. Juutilainen, P.A. Simell, A.O.I. Krause, in: A.V. Bridgwater, D.G.B. Boocock (Eds.), Science in Thermal and Chemical Biomass Conversion, CPL Press, Speen, Berkshire, England, 2006, pp. 821–831. [29] G. Busca, E. Finocchio, G. Ramis, G. Ricchiardi, Catalysis Today 32 (1996) 133–143.

Molecular level insights to the interaction of toluene with ZrO2-based biomass gasification gas clean-up catalysts

Tiia Viinikainen a,*, Inkeri Kauppi a, Satu Korhonen a, Leon Lefferts a,b, Jaana Kanervo a and Juha Lehtonen a a

Aalto University, School of Chemical Technology, Department of Biotechnology and Chemical Technology, P.O. Box 16100, 00076 Aalto, Finland.

b

University of Twente, Faculty of Science & Technology, P.O. Box 217, 7500 AE Enschede, The Netherlands. * Corresponding author: [email protected] +358 50 5300 367

Figure S1. DRIFTS spectra from the TPD experiments in the region of 1700-1300 cm-1 of calcined and reduced a) ZrO2, b) Y2O3-ZrO2 and c) SiO2-ZrO2 after toluene adsorption at 30 °C followed by heating in nitrogen from 30 to 500 °C.




b


Figure S2. DRIFTS spectra from the TPA experiments in the region of 3200-2800 cm-1 of calcined and reduced a) ZrO2, b) Y2O3-ZrO2 and c) SiO2-ZrO2 during toluene adsorption at 100-600 °C. Spectra recorded after a 5-min nitrogen flush at each temperature.




b


Figure S3. DRIFTS spectra from the TPSR experiments in the region of 1700-1300 cm-1 of calcined and reduced a) ZrO2, b) Y2O3-ZrO2 and c) SiO2-ZrO2 during co-adsorption of toluene and oxygen from 100 to 600 °C.

Department of Biotechnology and Chemical Technology

SCIENCE + TECHNOLOGY CROSSOVER DOCTORAL DISSERTATIONS

Aalto University

ART + DESIGN + ARCHITECTURE

The effect of H2S on oxidation properties of ZrO2 -based biomass gasification gas clean-up catalysts

Aalto University School of Chemical Technology Department of Biotechnology and Chemical Technology www.aalto.fi

BUSINESS + ECONOMY

Inkeri Kauppi

Aalto-DD 214/2015

9HSTFMG*agfida+

ISBN 978-952-60-6583-0 (printed) ISBN 978-952-60-6584-7 (pdf) ISSN-L 1799-4934 ISSN 1799-4934 (printed) ISSN 1799-4942 (pdf)

2015


DOCTORAL DISSERTATIONS