Synthesis and characterization of nanoporous materials

SYNTHESIS AND CHARACTERIZATION OF NANOPOROUS MATERIALS:

NANOZEOLITES AND METAL-ORGANIC FRAMEWORKS

Thèse

Thanh Vuong Gia

Doctorat en Génie Chimique

Philosophiae Doctor (Ph.D)

Québec, Canada

© Thanh Vuong Gia, 2013

iii

Résumé

Dans ce travail, deux types de nanomatériaux poreux ont été obtenues: des

nanozéolithes et des matériaux à réseau organométallique (MOF). Pour les nanozéolithes,

deux nouvelles méthodes de synthèses ont été développé: une méthode à phase unique et

une méthode biphasique. Dans la méthode à phase unique, une quantité du gel-zéolithique

est ajoutée à une solution de toluène/n-butanol contenant l‟agent silylant organosilane.

Après 12 heures à 60oC, une phase homogène est obtenue. Ce mélange est traité

hydrothermalement pour produire une nanozéolithe fonctionalisée. En revanche, la

méthode de synthèse à deux phases, implique l‟introduction de l‟organosilane mélangé à un

solvant organique dans le gel de zéolithe aqueux conduisant ainsi à un mélange biphasique.

Après mélange et traitement hydrothermal, des nanozéolithes fonctionalisées par silylation

sont obtenu dans la phase organique et de larges cristaux de zéolithes sont obtenus dans la

phase aqueuse. En principe, les deux méthodes utilisent l‟organosilane pour empêcher la

croissance des cristaux. Le solvant organique joue le rôle de dispersant des nanozéolithes

fonctionalisées avec l‟organosilane à partir de la phase aqueuse, et contrôle le processus de

croissance des nanozéolithes. Ces deux méthodes de synthèse sont applicables autant aux

zéolithes MFI que FAU, telles que silicatite-1, ZSM-5 et NaY. Elles peuvent être étendues

à la synthèse d‟autres types de zéolithes. L‟activité catalytique de ces nanozéolithes a été

évaluée pour le craquage de FCC. Les résultats indiquent que la nanozéolithe de type FAU

montre une bonne activité dans cette réaction.

Pour l‟étude des matériaux à réseau organométallique (MOF), une nouvelle

approche a été développé pour la synthèse de MIL-88B en utilisant un cluster neutre de

métaux mixtes bimétalliques Fe2Ni(µ3-O). Les clusters occupent les nœuds du réseau MIL-

88B à la place du mono-métal Fe3 (µ3-O) avec un anion compensateur. Ce dernier est le

cluster formant le réseau du Fe3MIL-88B non-poreux qui est obtenu par la méthode

conventionnelle. De ce fait, en absence des anions compensateurs dans la structure, Fe2Ni

MIL-88B devient un matériau poreux. De plus, avec la combinaison de la flexibilité de

MIL-88B et des métaux mixtes comme nœuds dans le réseau, la porosité peut être contrôlée

par échange avec des ligands terminaux du réseau. Ceci nous a permis de moduler d‟une

iv

manière réversible la porosité de MIL-88B à différents niveaux, ainsi que la surface

spécifique et le volume de pores dépendant de taille de ligands échangés. Le mécanisme de

synthèse a été aussi étudié pour les matétiaux Fe3-MIL88B et Fe2Ni-MIL88B. Les résultats

montrent que pour la synthèse de Fe3-MIL88B, le mono-métal Fe3-MOF-235 est comme le

précurseur pour la formation de MIL-88B. Dans le cas d‟utilisation de métaux mixtes

Fe2Ni(µ3-O), les mono-métal Fe3-MOF-235 est formés en premier lors de la synthèse du

métal mixte Fe2Ni-MIL88B. Il est montré que la présence de l‟anion FeCl4- est déteminante

dans la formation de la phase initiale MOF-235 et dans le succès de la synthèse du MIL-

88B mono- ou bimétallique

L‟anion FeCl4- est très important pour le succès de la formation de MOF-235. Un

mécanisme d‟anion médiateur dans la formation de MOF-235 a été suggéré.

v

Abstract

In this thesis, two types of nanoporous materials: nanozeolites and metal-organic

frameworks were studies. For nanozeolites, two novel methods e.g. single-phase and two-

phases were reported for the synthesis of nanozeolites. In the single-phase synthesis

method, a proper amount of zeolite gel solution was added to a toluene/n-butanol solution

containing an organosilane. After 12 hours at 60oC, a single phase mixture was obtained.

This mixture was then subjected to hydrothermal crystallization to produce uniform

functionalized nanozeolites. In contrast, the two-phase synthesis method involved the

introduction of an organic solvent containing organosilane to the aqueous zeolite gel

solution, resulting in a two-phase mixture. Upon mixing and hydrothermal treatment of this

mixture, organosilane-functionalized nanozeolites were obtained in the organic phase

whereas, large zeolite crystals were found in the aqueous phase. In principle, both methods

employed the use of organosilane to inhibit the crystal growth. The organic solvent acted as

the medium for the dispersion of nanozeolites functionalized with organosilane from the

aqueous phase, which led to the complete halt of the growth process. These two methods

were demonstrated to be applicable to the synthesis of MFI and FAU nanozeolites such as

silicalite-1 and NaY, and could be applied to the synthesis of other types of zeolites.

Catalytic activity of the synthesized nanozeolites was evaluated by the cracking reaction of

FCC feed. The result showed that FAU nanozeolites can be good catalysts for the cracking

reaction.

For the study of the metal-organic frameworks (MOF), a new rational approach was

developed for the synthesis of mixed metal MIL-88B metal organic framework based on

the use of neutral bimetallic cluster, such as Fe2Ni(µ3-O) cluster. Unlike the conventional

negative charged single metal cluster, the use of neutral bimetallic cluster as a framework

node avoids the need of compensating anion inside porous MIL-88B system; thus such a

bimetallic MIL-88B becomes porous. The flexibility of the mixed metal MIL-88B can be

controlled by terminal ligands with different steric hindrance. This allows us to reversibly

customize the porosity of MIL-88B structure at three levels of specific surface area as well

as the pore volume. Synthesis mechanism was also studied. It was found that the

vi

monometallic Fe3-MOF-235 is the precursors to the formation of MIL-88B. MOF-235

comes first then later transforms to Fe3-MIL-88B or acts as seeds for the formation of

mixed Fe2Ni-MIL88B. FeCl4- anion is very important to the successful formation of MOF-

235. An anion mediated mechanism of the formation of MOF-235 is suggested

vii

List of Contents

Résumé .................................................................................................................................. iii Abstract ................................................................................................................................... v List of Contents .................................................................................................................... vii List of Tables .......................................................................................................................... x List of Figures ........................................................................................................................ xi

Acknowledgement ................................................................................................................ xv Preface ................................................................................................................................. xvi Chapter 1. Introduction ........................................................................................................... 1

1.1. Zeolite .......................................................................................................................... 1 1.1.1. Background ........................................................................................................... 1

1.1.2. Structure ................................................................................................................ 2 1.1.3. General Properties of Zeolites .............................................................................. 5

1.1.4. Applications .......................................................................................................... 6 1.2. Nanozeolites ................................................................................................................. 7

1.2.1. Background ........................................................................................................... 7 1.2.2. Synthesis ............................................................................................................... 8

1.2.3. Recent advances in application of nanozeolites ................................................. 26 1.3. Metal-organic frameworks ......................................................................................... 29

1.3.1. Background ......................................................................................................... 29 1.3.2. Design principles of MOFs ................................................................................. 33 1.3.3. Synthesis ............................................................................................................. 36

1.4. Some applications of MOFs ...................................................................................... 38

1.4.1. Adsorption .......................................................................................................... 38 1.4.2. MOF as catalysts ................................................................................................. 45

References ......................................................................................................................... 53

Chapter 2. Experimental ....................................................................................................... 61 2.1. Synthesis .................................................................................................................... 61

2.1.1. Preparation of clear zeolite gel solution ............................................................. 61 2.1.2. Synthesis of nanozeolites using clear gel solution in aqueous medium

(conventional method) .................................................................................................. 62 2.1.3. Synthesis of nanozeolites in organic medium..................................................... 63 2.1.4. Preparation of silica containing nanozeolites ..................................................... 65 2.1.5. Synthesis of MIL-88B metal-organic framework ............................................... 65

2.2. Characterization ......................................................................................................... 66

2.2.1. FTIR Spectroscopy ............................................................................................. 66 2.2.2. Raman spectroscopy ........................................................................................... 68

2.2.3. UV-Vis spectroscopy .......................................................................................... 68 2.2.4. Energy-dispersive X-ray spectroscopy ............................................................... 71 2.2.5. X-ray Diffraction (XRD) .................................................................................... 71 2.2.6.

29Si Magic Angle Spinning Nuclear Magnetic Resonance Spectroscopy (MAS

NMR) ............................................................................................................................ 73

2.2.7. Scanning electron microscope (SEM) ................................................................ 74 2.2.8. Transmission Electron Microscope (TEM) ........................................................ 75 2.2.9. Nitrogen Adsorption/Desorption Isotherms........................................................ 75

viii

2.2.10. Cracking reaction ............................................................................................. 78

References ........................................................................................................................ 79

Chapter 3. A New Route for the Synthesis of Uniform Nanozeolites with Hydrophobic

External Surface in Organic Solvent Medium ..................................................................... 81 Résumé ............................................................................................................................. 81 Abstract ............................................................................................................................ 81 References ........................................................................................................................ 88

Supporting information .................................................................................................... 89 Chapter 4. Synthesis of Silylated Nanozeolites in the Presence of Organic Phase: Two-

Phase and Single Phase Methods ......................................................................................... 95 Résumé ............................................................................................................................. 95 Abstract ............................................................................................................................ 95

4.1. Introduction ............................................................................................................... 96 4.2. Experimental ............................................................................................................. 99

4.2.1. Synthesis of silylated silicalite-1 using the two phase and single-phase methods

...................................................................................................................................... 99

4.2.2. Two-phase method ............................................................................................. 99 4.2.3. Single-phase method ........................................................................................ 100

4.2.4. Conventional method (synthesis of nanozeolites in aqueous medium) ........... 101 4.2.5. Characterization ............................................................................................... 101

4.3. Results and discussion ............................................................................................. 102

4.4. Conclusion ............................................................................................................... 113 Acknowledgments .......................................................................................................... 114

References ...................................................................................................................... 115

Chapter 5. Synthesis of Nanozeolite-Based FCC Catalysts and their Catalytic Activity in

Gasoil Cracking Reaction ................................................................................................... 117 Résumé ........................................................................................................................... 117

Abstract .......................................................................................................................... 118 5.1. Introduction ............................................................................................................. 119 5.2. Materials and methods ............................................................................................ 121

5.2.1. Synthesis of nanofaujasite ................................................................................ 121 5.2.2. Synthesis of nanofaujasite-based FCC catalysts .............................................. 121

5.2.3. Characterization ............................................................................................... 121 5.2.4. MAT cracking evaluation ................................................................................. 122

5.3. Results and discussion ............................................................................................. 123 5.3.1. Synthesis of nanozeolites ................................................................................. 123

5.3.2. Synthesis of FCC .............................................................................................. 133 5.3.3. Catalytic test ..................................................................................................... 136

5.4. Conclusion ............................................................................................................... 144

References ...................................................................................................................... 146 Chapter 6. Synthesis and Engineering Porosity of Mixed Metal Fe2Ni-MIL-88B Metal-

Organic Framework ............................................................................................................ 149 Résumé ........................................................................................................................... 149 Abstract .......................................................................................................................... 149

6.1. Introduction ............................................................................................................. 150 6.2. Experiments ............................................................................................................. 154 6.3. Results ..................................................................................................................... 155

ix

6.3.1. Synthesis of Mixed Metal Fe2Ni-MIL-88B with Different Terminal Ligands . 155

6.3.2. Reversible Breathing Control Using Terminal Ligand ..................................... 159

6.3.3. Adsorption Analysis ......................................................................................... 160 6.4. Discussion ................................................................................................................ 163 6.5. Conclusion ............................................................................................................... 167 Refrences ........................................................................................................................ 168 Supporting information ................................................................................................... 170

Chapter 7. Direct Synthesis and Mechanism for the Formation of Mixed Metal Fe2Ni-MIL-

88B ...................................................................................................................................... 185 Résumé ............................................................................................................................ 185 Abstract ........................................................................................................................... 185 7.1. Introduction .............................................................................................................. 186

7.2. Experimental Section ............................................................................................... 187 7.3. Results ...................................................................................................................... 188 7.4. Discussion ................................................................................................................ 194

7.5. Conclusion ............................................................................................................... 198

References ....................................................................................................................... 199 Supporting Information ................................................................................................... 219

Chapter 8. Conclusion ......................................................................................................... 233 List of Pulications ............................................................................................................... 239

x

List of Tables

Table 1.1. Molecular sieve types synthesized in nanosized form, synthesis conditions, and

crystal size[9] ............................................................................................................... 16 Table 2.1. Structure insensitive and sensitive framework vibrations of zeolites [4] ........... 67

Table 2.2. FTIR band assignment in the wavenumber 400 – 800 cm-1

............................... 67 Table 4.1. Physico-chemical properties of the calcined silylated nanozeolite and zeolite

samples prepared from the same clear zeolite gel, using different methods: the two-

phase, single-phase and conventional methods. ......................................................... 102 Table 5.1. Physicochemical properties of nanofaujasite samples. .................................... 133

Table 5.2. BET analysis of nanozeolite-based FFC catalyst samples. .............................. 134 Table 6.1. IR analysis of the MIL-88B samples ................................................................ 156

Table 6.2. Crystal parameters of the Fe2Ni-MIL-88B samples ......................................... 158 Table 6.3. Porosity of Fe2Ni-MIL-88B ............................................................................. 163 Table 7.1. FTIR band assignment in the wavenumber 400 – 800 cm

-1 ............................. 217

Table 7.2. Raman band assignments ................................................................................. 217

Table 7.3. Fe and Ni atomic percentages calculated from EDS spectra ............................ 217 Table 7.4. Comparison of the crystal parameters of MIL-88B and MOF-235 .................. 218

xi

List of Figures

Figure 1.1. Secondary building units and their symbols. Number in parenthesis indicates

frequency of occurrence .................................................................................................. 3 Figure 1.2. (a) MFI and (b) FAU structures ........................................................................... 4

Figure 1.3. Inter-conversion of Brönsted and Lewis acid sites .............................................. 5 Figure 1.4. Calculated surface to bulk atom ratios for spherical nanocrystals[13] ................ 8 Figure 1.5. Synthesis of nanozeolites from clear solutions [18] ............................................ 9 Figure 1.6. Two pathways of zeolite formation ................................................................... 10 Figure 1.7. A scheme for the crystallization mechanism of silicalite-1[28] ........................ 12

Figure 1.8. The "nanoslab" hypothesis: (a) the precursor unit containing one TPA cation

and (b) schematic representation of nanoslab formation by aggregation of precursor

units, as determined by XRS and GPC [32] ................................................................. 12

Figure 1.9. Nucleation and growth model of zeolite A and zeolite Y as represented by

TEM [34, 35] ................................................................................................................ 13 Figure 1.10. Schematic illustration of the mechanism proposed for nanoparticle evolution

and crystal growth by aggregation. ............................................................................... 14 Figure 1.11. Schematic illustration of confined space synthesis.[101]................................ 21 Figure 1.12. Schematic representation of synthesis of template-free zeolite nanocrystals by

using in situ thermoreversible polymer hydrogels.[104] .............................................. 23 Figure 1.13. Schematic representation of revered microemulsion ...................................... 24

Figure 1.14. Schematic representation of microemulsion-microwave synthetic method[115]

...................................................................................................................................... 26

Figure 1.15. Rational design of zeotype and MOF from the basis of zeolite ...................... 29 Figure 1.16. Crystal structure of HKUST-1 (left) and MOF-5 (right) ................................. 31

Figure 1.17. Number of metal–organic framework (MOF) structures reported in the

Cambridge Structural Database (CSD) from 1978 through 2006. The bar graph

illustrates the recent dramatic increase in the number of reports, while the inset shows

the natural log of the number of structures as a function of time, indicating the

extraordinarily short doubling time for MOF structures compared to the total number

of structures archived in the database.[129] ................................................................. 32

Figure 1.18. Two geometrically different but topologically identical nets. ......................... 34 Figure 1.19. MOF-5 structure from linking octahedral SBUs ............................................. 35

Figure 1.20. Design of MOF-5 from simple fcc structure ................................................... 36 Figure 1.21. Structure of NU-100 and its linker .................................................................. 40

Figure 1.22. The two different linkers used in MOF-210 (left) and the crystal structure of

MOF-210 displaying its unique topology with two different types of pores (right). ... 40 Figure 1.23. Excess hydrogen uptake at 77 K versus BET specific surface area (BET ssa)

for various high-porosity MOFs. The symbols denote measurements conducted by

different research groups (circles: at 20 bar by Hirscher et al.[193] triangles: at 60 bar

by Kaskel et al.[194]; squares: saturation values by Yaghi et al.).[195] ...................... 41 Figure 1.24. MOF-117 structure and comparison of the volumetric CO2 capacity of

crystalline MOF-177 relative to zeolite 13X pellets, MAXSORB carbon powder, and

pressurized CO2.[200] ................................................................................................... 42 Figure 1.25. ZIF-69 structure ............................................................................................... 43

xii

Figure 1.26. A portion of the structure of the sodalite-type framework of Cu-BTTri (1)

showing surface functionalization of a coordinatively unsaturated Cu(II) site with

ethylenediamine, followed by attack of an amino group on CO2 . .............................. 44 Figure 1.27. Detail of Pd-MOF, showing the 4-membered and the two 6-membered rings.

3D arrangement of the sodalite cages in sodalite-type frameworks. ............................ 46 Figure 1.28. (a) Basic building block of Cu(2-pymo)2 and (b) Diagram of the asymmetric

unit of the Co(PhIM)2 framework ................................................................................ 47

Figure 1.29. Transformation of ZIF-90 (A) by Reduction with NaBH4, and reaction with

ethanolamine to give ZIF-91 (B) and ZIF-92 (C) [213] ............................................... 49 Figure 1.30. (a) Eight-coordinate molecular building block that could be represented as a

tetrahedral building unit, (b) [H2TMPyP]4+

porphyrin, (c) crystal structure of rho-

ZMOF (left), hydrogen atoms omitted for clarity, and schematic presentation of

[H2TMPyP]4+

porphyrin ring enclosed in rho-ZMOF R-cage (right, drawn to scale)

[217] ............................................................................................................................. 51 Figure 2.1. Methods studied for the synthesis of nanozeolites ........................................... 61

Figure 2.2. Scheme of the autoclave: (1) a cylindrical stainless steel vessel, (2) a Teflon

cylindrical beaker, (3) a flat Teflon cover for closing the Teflon beaker, (4) a flat

stainless steel cover which was tightened up to part (1) by six screws.[3] .................. 63

Figure 2.3. The excitation process ....................................................................................... 69 Figure 2.4. Electronic energy levels and transitions. .......................................................... 69 Figure 2.5. Principle of EDX spectroscopy ......................................................................... 71

Figure 2.6. Diffraction of X-ray beams on a crystal lattice ................................................. 72 Figure 2.7. Types of sorption isotherms.[18] ...................................................................... 75

Figure 2.8. t-plot method. .................................................................................................... 78

Figure 3.1. XRD patterns of the as-made silylated zeolite and zeolite samples prepared

from the same zeolite gel in solvent medium in the presence of organosilane and in

aqueous medium in the absence of organosilane, respectively: (A) silicalite-1; (B)

faujasite. ....................................................................................................................... 85 Figure 3.2. TEM images of the as-made samples: (A) silylated nanosilicalite-1, (B)

silylated nanofaujasite. ................................................................................................. 86

Figure 4.1. Schematic representation of the two-phase synthesis method. ....................... 100 Figure 4.2. Schematic representation of the single-phase synthesis method. ................... 103

Figure 4.3. XRD patterns of the as-made silicalite-1 samples, (a) sample prepared using the

conventional method in aqueous medium, (b) AP silicalite-1, (c) OP silicalite-1 using

the two-phase method and (d) SOP silicalite-1 using single-phase method. ............. 104

Figure 4.4. SEM micrographs of the as-made samples, (A) silylated OP silicalite-1 and (B)

silylated AP silicalite-1. ............................................................................................. 106 Figure 4.5. TEM micrograph of the as-made SOP nanosilicalite-1 sample prepared using

the single-phase method. ............................................................................................ 107

Figure 4.6. FTIR spectra of the silicalite-1 samples prepared using the single-phase method

(A) and the conventional method (B). ........................................................................ 108

Figure 4.7 29

Si MAS NMR spectra of the silicalite-1 samples prepared from the same

zeolite gel solution using (a) the conventional method in aqueous medium without

organosilane, (b) as-made SOP nanosilicalite-1 using single-phase method in organic

solvent and (c) calcined SOP nanosilicalite-1. ........................................................... 109

xiii

Figure 4.8 29

Si MAS NMR spectra of the as-made silicalite-1 samples prepared from the

same zeolite gel solution using the two-phase method: (a) as-made AP nanosilicalite-1

and (b) as-made OP nanosilicalite-1. .......................................................................... 110 Figure 4.9. Nitrogen adsorption/desorption isotherms of the calcined samples: (A) SOP

silicalite-1, (B) OP silicalite-1 and (C) AP silicalite-1 (inset: t-plot curve). .............. 112 Figure 5.1. FT-IR spectra of the prepared nanofaujasite samples: (A) FAU–TOL2D

prepared using toluene and pre-heated zeolite gel for 2 days at 90 °C, (B) FAU–

FOR2D prepared using formamide and pre-heated zeolite gel for 2 days at 90 °C, (C)

FAU–FOR4D prepared using formamide and pre-heated zeolite gel for 4 days at

90 °C, and (D) zeolite Y reference. ............................................................................ 126 Figure 5.2. XRD patterns of nanofaujasite samples prepared: (A) FAU–TOL2D in toluene,

(B) FAU–FORM2D in formamide from the zeolite gel pre-heated at 90 °C for 2 days,

(C) FAU–FORM4D in formamide from the zeolite gel pre-heated at 90 °C for 4 days,

and (D) zeolite Y standard. ......................................................................................... 127 Figure 5.3. TEM images of (A) the sample FAU–TOL2D prepared in toluene from the

zeolite gel pre-heated at 90 °C for 2 days, (B) sample FAU–FOR2D prepared in

formamide from the zeolite gel pre-heated at 90 °C for 2 days, and (C) the sample

FAU–FOR4D prepared in formamide from the zeolite gel pre-heated at 90 °C for 4

days. ............................................................................................................................ 128 Figure 5.4.

29Si MAS NMR spectra of the as-made faujasite prepared in aqueous medium

in the absence of organosilane (conventional method) and silylated faujasite samples:

(A) FAU–TOL4D using toluene pre-heated for 4 days, (B) FAU–FOR2D using

formamide pre-heated for 2 days, (C) FAU–TOL2D using toluene pre-heated for 2

days, and (D) FAU-Standard using conventional method. ......................................... 130

Figure 5.5. N2 adsorption desorption isotherms of (A) FAU–TOL2D prepared in toluene

from the zeolite gel pre-heated at 90 °C for 2 days, (B) FAU–FOR2D prepared in

formamide from the zeolite gel pre-heated at 90 °C for 2 days, and (C) FAU-FOR4D

prepared in formamide from the zeolite gel pre-heated at 90 °C for 4 days. .............. 132 Figure 5.6. XRD patterns of the nanozeolite-based FCC catalyst samples prepared from the

corresponding 40, 24 and 100 nm nanozeolites: (A) FCC–TOL2D, (B) FCC–FOR2D

and (C) FCC–FOR4D. ................................................................................................ 133 Figure 5.7. SEM image of (A) FCC–FAU–TOL2D, (B) FCC–FAU–FOR2D and (C) FCC–

FAU-FOR4D. ............................................................................................................. 134 Figure 5.8. N2 adsorption desorption isotherms of (A) FCC–FAU–TOL2D, (B) FCC–

FAU–FOR2D and (C) FCC–FAU–FOR4D. .............................................................. 135 Figure 5.9. Relationship between conversion and catalyst-to-oil ratio of different prepared

FCC-samples. .............................................................................................................. 137 Figure 5.10. Correlation of dry gas yield with conversion of different prepared FCC-

samples. ....................................................................................................................... 138

Figure 5.11. Correlation of LPG yield with conversion of different prepared FCC-samples.

.................................................................................................................................... 139 Figure 5.12. Correlation between gasoline yield and conversion of different prepared FCC-

samples. ....................................................................................................................... 140 Figure 5.13. Relationship between gasoline selectivity and conversion of different prepared

FCC-samples. .............................................................................................................. 141 Figure 5.14. Relationship between LCO yield and conversion. ........................................ 142

xiv

Figure 5.15. Relation between HCO yield and conversion of different prepared FCC-

samples. ...................................................................................................................... 143

Figure 5.16. Relationship between coke yield and conversion of different prepared FCC-

samples. ...................................................................................................................... 144

Figure 6.1. XRD patterns of Fe2Ni-MIL-88B.H2O (a) and XRD simulation of the Fe3-MIL-

88B (b). ....................................................................................................................... 155 Figure 6.2. UV-Vis spectra of Fe2Ni-MIL-88B.Bp (a), Fe2Ni-MIL-88B.Pz (b), Fe2Ni-

MIL-88B.Py (c), Fe2Ni-MIL-88B.DMF (d) Fe2Ni-MIL-88B.H2O (e) and Fe3-MIL-

88B (f) ........................................................................................................................ 157

Figure 6.3. XRD patterns of Fe2Ni-MIL-88B samples, the planes of open phase are in

black, the planes of dense phase are in red and placed in boxes. Fe2Ni-MIL-88B.Bp

(a), Fe2Ni-MIL-88B.Pz (b), Fe2Ni-MIL-88B.Py (c), Fe2Ni-MIL-88B.DMF (d) and

Fe2Ni-MIL-88B.H2O (e) ............................................................................................ 159 Figure 6.4. N2 adsorption isotherms at 77 K (A) and pore size distributions (B) of Fe2Ni-

MIL-88B.Bp (a), Fe2Ni-MIL-88B.Pz (b), Fe2Ni-MIL-88B.Py (c), Fe2Ni-MIL-

88B.DMF (d) and Fe2Ni-MIL-88B.H2O (e). .............................................................. 162 Figure 6.5. CO2 adsorption isotherms at 273 K of Fe2Ni-MIL-88B.Bp (a), Fe2Ni-MIL-

88B.Pz (b), Fe2Ni-MIL-88B.Py (c), Fe2Ni-MIL-88B.DMF (d) and Fe2Ni-MIL-

88B.H2O (e) ................................................................................................................ 163 Figure 7.1. UV-Vis spectra of the samples Fe3-NO3-x (A) and Fe3-Cl-x (B) prepared using

Fe(NO3)3.9H2O and FeCl3.6H2O, respectively, at different synthesis times. ............ 202 Figure 7.2. UV-Vis spectra of Fe2Ni-NO3-x (A) and Fe2Ni-Cl-x (B) samples prepared

using Fe(NO3)3.9H2O and FeCl3.6H2O, respectively at different synthesis times ..... 204 Figure 7.3. Transmittance FTIR spectra of the samples of Fe3-NO3-x (A) and Fe3-Cl-x (B)

at different synthesis times ......................................................................................... 205 Figure 7.4. Transmittance FTIR spectra of Fe2Ni-NO3-x (A) and Fe2Ni-Cl-x (B) at

different synthesis times ............................................................................................. 207 Figure 7.5. Raman spectra of the samples Fe3-NO3-x (A) and Fe3-Cl-x (B) at different

synthesis times ............................................................................................................ 209

Figure 7.6. Raman spectra of Fe2Ni-NO3-x (A) and Fe2Ni-Cl-x (B) at different synthesis

times. .......................................................................................................................... 211

Figure 7.7. XRD patterns of Fe3-NO3-x (A) and Fe3-Cl-x (B) at different synthesis times.

(*) MOF-235 phase, (#): MIL-88B phase. ................................................................. 212

Figure 7.8. XRD patterns of Fe2Ni-NO3 (A) and Fe2Ni-Cl (B) at different synthesis time.

(*) MOF-235 phase, (#): MIL-88B phase .................................................................. 213 Figure 7.9. Representative HRTEM and EDS acquiring positions of Fe2Ni-Cl-12h crystal

.................................................................................................................................... 214

xv

Acknowledgement

I would like to thank the Laval Foundation for granting me the scholarship for this

study.

I would like to my deep and sincere gratitude to my supervisor, Professor Trong-On

Do. His wide knowledge and creative thoughts have been of great value for me. His

understanding, encouraging and patience guidance have provided a good basis for the

present thesis. It was a great pleasure to me to conduct this thesis under his supervision.

But it would have never come to such a happy ending if I hadn‟t met the love of my

life Phuong Trinh Nguyen midway of this study. It was only with her that I found the

courgage to go on with the rest of the study. She even gave up on her own academic carrier

to support me. I am indebted to her love forever.

I would like to thank Professor Serge Kaliaguine for giving me the access to his

laboratory where some of my experiments were carried out.

I wish to express my warm and sincere thanks to Dr. Hoang-Vinh Thang, Dr.

Bousselham Echchahed and my colleagues Minh-Hao Pham and Dinh-Cao Thang who

gave me invaluable thoughtful insights, advice, support, discussions and encouragements.

Special gratitude is also given to all the professors, staff and graduate students of the

Department of Chemical Engineering from Laval University for their great assistance and

cooperation.

xvi

Preface

This thesis documents our study in two most interesting nanoporous materials:

zeolites and metal-organic framework (MOF).

As one of the most important nanoporous materials, zeolites have been widely used

in catalysis, adsorption and ion exchange. Since their industrial introduction in 1954, the

annual market for zeolites has grown to 1.7 million tons. The reason zeolites can “enjoy”

such a great success is largely due to their interesting features: (i) high surface area, (ii)

uniform pore size structure, (iii) controllable acidity/basicity, and (iv) high

hydrothermostability. Inspired by zeolites, there are continuous attempts to search for other

materials which can duplicate zeolites‟ features, hence their name zeotypes. These attempts

have found some interesting zeotypes, however even AlPO4, which is regarded as one of

the best zeotypes, yet falls short on catalytic activity due to its neutral nature. That is why

when it comes to nanoporous materials, zeolites are still the first choice for researchers.

Nevertheless it is not to say that zeolites have no drawbacks. In fact due to the pore size

constraints, the use of zeolites are less effective when large molecules are involved.

Mesoporous materials can cope with large molecules but their amorphous nature leads to

their low hydrothermostablity and low catalytic activity. So the challenge for zeolite

science nowadays is to expand the application of zeolites to include large and bulky

molecules.

Most of new materials can be found with inspiration from nature as in case of

zeolite: with the knowledge of natural zeolites new successful methods to synthesize

zeolites were proposed. Rarely a new material comes out of solely intellectual vigor

without any precedent clues in nature. And the discovery of MOF was done in that unusual

fashion. While zeolites are solely inorganic materials, MOFs are “ambitious” ones

combining the whole two main categories of chemistry into crystalline structures: organic

entities connecting to each other via inorganic metal clusters. The resulting products are

huge and growing collections of MOFs with uniform nanoporous structure, extremely high

surface area. With MOFs, the nanoporous material researchers can have both organic part

and inorganic part to tinker with. Versatility and flexibility are the key and attractive

features of MOFs, as the choice of metal cluster and organic entities is almost infinite.

xvii

Researches on MOF are explosively growing, but there are still many problems in the

synthesis of MOF needed to be addressed: how the MOF forms, how to control porosity

etc.

In this thesis, we look at both zeolites and MOFs and choose to solve some of the

most pressing issues in these great materials. The thesis is built largely on our published or

submitted papers in recent years. The first author of these papers is also the author of this

thesis.

The thesis starts with an introduction in chapter 1 and then chapter 2 is an overview

of the techniques used in our study. In chapter 3, pulished in Journal of the American

Chemical Society 2007, 129 (13), 3810-3811, we suggest an approach to overcome the

pore size limit of zeolites: reducing zeolite crystal size from microns down to tens of

nanometers thus gaining more exposure of active sites to large molecules. Synthesis of

nanozeolites is not an easy task, without protection nanozeolites tend to aggregate or

dissolve to form large and stable crystals. We devise a new method to obtain highly

uniform nanozeolites with the hydrophobic surface and controlled crystal size. The main

distinction in our approaches is that an organic solvent is used as a medium for the

crystallization instead of water. The zeolite precursors (nanoslabs) are functionalized with

organic silane groups thus become hydrophobic and able to be well dispersed in the organic

solvent. Because the crystallization occurs in the organic phase and the zeolite precursors

are protected by functional groups. The aggregation can be avoided, hence, resulting in

small and uniform nanozeolites with the hydrophobic external surface. Chapter 4, published

in Microporous and Mesoporous Materials 2009, 120 (3), 310-316, pushes forward our new

method of synthesis of nanozeolites. Attention was put onto the organic solvent. We found

that depending on the amount of solvent one can have a single phase or two phase system.

The results revealed that the single-phase method allows producing uniform/small

nanosilicalite-1, whereas the two-phase one can bring two separate products: nanosized and

microsized zeolite crystals in organic phase and in aqueous phase, respectively. Chapter 5,

published in Applied Catalysis A: General 2010, 382 (2), 231-239, is our ultimate catalytic

test of our nanozeolite. A series of FCC catalyst containing our nanozeolites with different

sizes were prepared and tested on the gasoil cracking. The catalytic test results confirmed

xviii

our anticipation: the nanozeolites exhibit higher activity, this is due to higher external

surface area and higher number of active sites available.

Chapter 6 and 7 are dedicated to MOF. In chapter 6, Published in Dalton

Transactions 2013, 42 (2), 550-557, we take on one of the well-known MOFs: MIL-88B.

MIL-88B is best known for its structural breathing effect upon adsorption. That is, upon

interaction with various solvent molecules, the framework of MIL-88B can “swell” up and

down greatly and reversibly without breaking structure. However, MIL-88B is not actually

porous since solvent molecules are needed to pack its pores to sustain its swelling, when

solvent is removed, the structure shrinks and its pores are blocked by the charge balancing

anions. To overcome these drawbacks, the original MIL-88B was modified rationally.

Neutral mixed metal cluster was used instead of the single metal cluster which requires

compensating anion. Thus the anion blockage issue is avoided, the new MIL-88B is

genuinely porous upon removal of solvent. Next, taking advantage of the breathing effect,

we use rigid terminal ligands as “pillars” to sustain the structure. Now there is no need of

solvent molecules and the swelling degree of the structure will be determined by the size of

the terminal ligand. This rational approach has allowed us to control the porosity of MIL-

88B at three distinct levels in terms of surface area and pore size. Chapter 7, which was

submmited to CrystEngComm, continues our adventure with MIL-88B structure, we will

delve into very essential problems in the synthesis of MOFs: why just a seemingly trivial

substitution of iron chloride with iron nitrate results in a complete failure in the synthesis;

and how the phase competition wears on to finalize the desired products. It was amazing to

us that we could answer these problems with the knowledge coming from zeolite science. I

propose the established mechanisms of the synthesis of zeolites are very helpful to

understand the surprising mechanism by which chroride ion promotes the formation of

MIL-88B structure. Alike zeolites, the synthesis of MIL-88B starts with the formation of a

kinetically favored phase, and then if the synthesis time suffices, the thermokinetically

stable phase will come out.

Finally chapter 8 will concluded our thesis with some additional recommendations.

Bellow is the list of publications of which the contents are used in this thesis.

xix

1. G.T. Vuong and T.O. Do, A new route for the synthesis of uniform nanozeolites

with hydrophobic external surface in organic solvent medium. Journal of the

American Chemical Society, 2007. 129(13): p. 3810-3811.

2. G.T. Vuong and T.O. Do, Synthesis of silylated nanozeolites in the presence of

organic phase: Two-phase and single-phase methods. Microporous and

Mesoporous Materials, 2009. 120(3): p. 310-316.

3. G.T. Vuong., V.T. Hoang, D.T. Nguyen, and T.O. Do, Synthesis of nanozeolites

and nanozeolite-based FCC catalysts, and their catalytic activity in gas oil

cracking reaction. Applied Catalysis A: General, 2010. 382(2): p. 231-239.

4. G.T. Vuong, M.H. Pham, and T.O. Do, Synthesis and Engineering Porosity of

mixed metal Fe2Ni- MIL-88B Metal-Organic Framework. Dalton Transactions,

2013, 42, 550-557.

5. G.T. Vuong, M.H. Pham, and T.O. Do, Direct Synthesis and Mechanism for the

Formation of Mixed Metal Fe2Ni-MIL-88B. CrystatEngComm, submitted, 2013.

1

Chapter 1. Introduction

1.1. Zeolite

1.1.1. Background

Zeolites (Greek, zein,"to boil", lithos, "a stone") are aluminosilicates that have well-

defined porous structures. The term was originally coined in the 18th

century by a Swedish

mineralogist named Axel Fredrik Cronstedt who observed, upon rapidly heating a natural

mineral that the stones began to dance about as the water evaporated. Using the Greek

words which mean "stone that boils", he called this material as zeolite.[1]

Strictly speaking, zeolites are defined as crystalline microporous aluminosilicates

with pore structures consisting of sharing TO4 tetrahedra, where T is Si or Al. Zeolites can

be described with the following empirical formula:[2]

Mn+

1/n . AlO2- . x SiO2 . yH2O

Where M – counter ion

n – counter ion valence

x – silicon/aluminum ratio

y – content of hydrate water

Owing to the well-defined pore structure, zeolites are also known as "molecular

sieves". The term molecular sieve refers to a particular property of these materials, i.e., the

ability to selectively adsorb molecules based primarily on a size exclusion process. This is

due to a very regular pore structure of molecular dimensions. The maximum size of the

molecular or ionic species that can enter the pores of a zeolite is controlled by the diameter

of the pore channels.

2

1.1.2. Structure

The flexibility of the zeolite Si-O-Si bond explains the fact that about 200 structures

have been determined. Indeed, there is little energetic difference (10-12 kJ/mol) between

these remarkable porous silicates and higher density phases such as quartz. More than 150

zeolite types have been synthesized and 48 naturally occurring zeolites are known.[3]

The structure commission of the International Zeolite Association (IZA) provides

up to date classification by framework type. Each framework is assigned a three-letter code,

recognized by the IUPAC Commission on Zeolite Nomenclature.[3] According to the IZA

structure commission, zeolite frameworks can be thought to consist of finite or infinite (i.e.,

chain- or layer-like) component units. The primary building units are single TO4 tetrahedra.

The finite units which have been found to occur in tetrahedral frameworks are shown in

Figure 1.1. These secondary building units (SBU), which contain up to 16 T-atoms, are

derived assumption that the entire framework is made up of one type of SBU only. A unit

cell always contains an integral number of SBUs. In some instances, combinations of SBUs

have been encountered. However, it should be noted that the SBUs are only theoretical

topological building units and should not be considered to be or equated with species that

may be in the solution/gel during the crystallization of a zeolitic material.[3]

Zeolites can be also classified on grounds of their pore openings and the

dimensionality of their channels. Thus, one distinguishes small pore zeolites (eight-

membered-ring pores), medium pore zeolites formed by ten-membered rings, large pore

zeolites with twelve-membered-ring pores and extra-large pore zeolite category. This

classification simplifies comparisons in terms of adsorptive, molecular sieving and catalytic

properties.

3

Figure 1.1. Secondary building units and their symbols. Number in parenthesis indicates

frequency of occurrence

Two important and industrially relevant structures, MFI and FAU are depicted in

Figure 1.2. The channels in the MFI structure are formed by 5-1 building units linked

4

together. These building units render a framework of zigzag 10-membered ring channel

(5.1 x 5.5 Å) and intersecting straight 10-membered ring channels (5.3 x 5.6 Å).

(a)

(b)

Figure 1.2. (a) MFI and (b) FAU structures

The FAU structure (structure of zeolite Y) is made up of 6-6 SBUs. In addition, it is

possible to consider the sodalite cage, a truncated octahedron that has eight hexagonal and

six square faces, as basic structure of zeolite Y. The FAU structure is formed when half of

the octahedral faces are joined together to form hexagonal prisms. The spherical internal

5

cavity generated when eight sodalite cages are joined is called the -cage (or supercage)

and is about 13 Å in diameter. Entry into the spherical -cage can occur through four

identical openings that are 7.4 Å wide. The dimensions of zeolite Y allow reasonably large

molecules to penetrate the internal pores, since compounds may extend through a prism

into two connecting -cages.

1.1.3. General Properties of Zeolites

Figure 1.3. Inter-conversion of Brönsted and Lewis acid sites

As mentioned above, the presence of Al in the structure of zeolites results in the

formation of anion sites within the framework. Charge neutralization may occur by either

protonation or by interaction with a metal cation or a hydronium ion. Thus, both Brönsted

and Lewis acidities may be present within the zeolite framework. The protonation of the

Al-O-Si oxygen center can result in Brönsted acidity in the zeolites structure. Lewis acidity

is typically related to the compensating metal ions and defects in the aluminosilicate

framework. Brönsted acid sites in zeolites can change into Lewis acid sites through

dehydroxylation on heating.[4]

Although zeolites are usually considered acid catalysts, cation substitution with Rb

and Cs, as well as metal doping, creates a basic zeolite.[5] The presence of heavy metal

cations is believed to increase the negative charge on the aluminum center, which is

transferred to the adjacent oxygen atom, creating a basic site.[6]

6

The hydrophobicity is an important characteristic of zeolites since it can have a

profound influence on their chemical reactivity. Zeolites containing charges are normally

hydrophilic materials that, depending on the framework Si/Al ratio, can be more or less

selective adsorbents for polar or nonpolar molecules. However, silicalite-1 which is a pure

silica zeolite is a highly hydrophobic material. In contrast, FAU zeolite with the Si/A1 ratio

between 2 and 5 is a highly hydrophilic absorbent. It is then clear that the polarity of a

given zeolite could be controlled by controlling the Si/Al ratio by direct synthesis or by

postsynthesis treatments, and this, together with appropriate control of the number of

silanol groups by synthesis or postsynthesis treatments, should make it possible to prepare

zeolite catalysts within a wide range of surface polarities.[7]

1.1.4. Applications

Since their successful introduction as commercial molecular sieves in 1954,

synthetic zeolites have grown to an estimate $1.6-1.7 billion industry of which detergents

represent the largest volume.[8] LTA-type zeolites have been used to substitute phosphate

compounds in the water softening process in laundry. However, the largest market value for

zeolites is in refinery catalysis. FCC (Fluid Catalytic Cracking) catalysts account for more

than 95% of zeolite catalyst consumption and consist of various forms of zeolite Y. MFI-

type zeolites are the second most used catalyst, primarily because they are added to FCC

catalysts for octane number enhancement. Zeolites are also employed in the drying and

purification of natural gas, separation of paraffins and desulfurization processes. Despite

being in a relatively early stage of development, zeolites are also used in fine chemicals

production such as oxidation and acylation.[8]

Zeolite science appears to be a mature science and is still a very dynamic field.

Discoveries of new zeolites continuously open new areas of development. New trends at

the beginning of this century include environmental applications such as De-NOx catalysis

and hydrocarbon storage in vehicles powered with diesel or gasoline engines, and

biopharmaceutical applications. Zeolites can also be used in the nuclear industry for

radioactive waste storage. Applications of zeolite material science still play an important

role in many areas of technology.

7

1.2. Nanozeolites

1.2.1. Background

Nanozeolites are a type of zeolites which have the particle distribution and sizes of

less than 200 nm.[9] Compared to “ordinary” zeolites of which the particle diameters are of

micrometer order, nanozeolites represent very small particle size, the narrowness of their

particle size distributions (often monodisperse) and especially, the fact that they are

composed of discrete particles (single crystal) rather than aggregates.

One of the advantages of nanozeolites is their higher external surface area. The

external surface is of vital important in numerous processes, including adsorption and

catalysis. For example, in the fluidized catalytic cracking (FCC) process, the commercial

catalysts are manufactured by dispersion of 1 micron FAU and MFI zeolites in an

amorphous alumina-silica matrix. For cracking to occur, gasoil molecules must pass

through the matrix and reach the surface of the zeolite crystals. The molecules then diffuse

through the micropores of zeolites until they reach an active site. Due to the zeolite

structure, molecules larger than 7.4 Å cannot reach active sites located inside the zeolites.

This problem can be solved by replacing the micrometer-sized zeolites with the

corresponding nanozeolites. The substitution could lead to the decrease in the diffusional

resistance and the increase in the external surface area, hence raising the number of active

sites available for large molecules.[10-12] Zeolite particles in the 10-100 nm range can

bring in new applications of zeolites. The huge surface areas of the nanosized materials

dictate that many of the atoms are on the surface, thus allowing good “atom economy” in

surface-gas and surface-liquid reactions. Figure 1.4 illustrates the calculated numbers of

atoms on spherical solid nanoparticles (iron) that are surface or bulk (interior) atoms. The

ratio of atoms available on the surface increases as the crystal size decreases. A 20 nm

particle has about 10% atoms present on the surface. This feature demonstrates that it is

necessary to be very small in order to benefit from the atom economy desired.[13]

8

Figure 1.4. Calculated surface to bulk atom ratios for spherical nanocrystals[13]

Besides the improvement on the external surface areas, nanozeolites have been

found to be excellent “building blocks” for constructing structured materials.[14]

Hierarchical porous materials with controlled porosity microstructure are of great interest

for catalysis and separation applications.[15-17] These porous structures can be fabricated

by templated self-assembly of silicalite nanocrystals (nanosilicalite-1) and the structures so

obtained include long zeolite fibers, micro-patterned zeolite films and micro-macroporous

zeolite structures. The use of preformed zeolite nanocrystals for preparation of supported

zeolite films and membranes is one of major applications of nanozeolites. The small size of

nanozeolites offers high homogeneity and intactness of the zeolite layer and reduces the

number of defects in the film, such as crack and pinholes.

1.2.2. Synthesis

1.2.2.1. Principles

A typical synthesis of nanozeolites using this method can be described as follows

(Figure 1.5):

9

- Amorphous reactants containing silica and alumina are mixed together with a

structure directing agent (SDA) source, usually in a basic (high pH) medium, resulting in a

clear solution.

- The aqueous reaction mixture is heated, often (for reaction temperatures around

100 °C) in a sealed autoclave.

- For some time after raising to synthesis temperature, the reactants remain

amorphous.

- After the above “induction period”, crystalline zeolite product can be detected.

- Gradually, essentially all amorphous material is replaced by an approximately

equal mass of zeolite crystals (which are recovered by filtration, washing and drying).

Figure 1.5. Synthesis of nanozeolites from clear solutions [18]

It is clear that a substantial understanding of the mechanism of the formation of

zeolite is necessary for the synthesis of nanozeolite. However, the study on the formation of

zeolites, although dates back to its earliest days,[2] has not reached its conclusion yet.[18-

20] The formation of zeolite is still a developing subject which receives a lot of interest and

10

its study would not only benefit the zeolite science but also contribute to the understanding

of the crystallization and growth of materials. The difficulties in the study of zeolite

synthesis originate from its inherent synthesis conditions which involve high temperature

and closed system. Hence in situ studies often require sophisticated equipment and setup.

However, thanks to the application of new characterization techniques and the rapid

development in computation chemistry in the last two decades, some significant discoveries

have been found, giving us a better picture of zeolite formation mechanism.[21, 22]

Figure 1.6. Two pathways of zeolite formation

Unlike the crystallization of common solid materials which often involves only two

distinct phases, the liquid phase and the solid phase,[23] the crystallization of zeolite

implicates three phases: liquid phase, amorphous phase (or gel) and crystalline phase of

Aluminosilica gel

Zeolite

Solution pathway Solid pathway

11

zeolite.[2] And it is the determination of the role of the amorphous phase that is the main

issue in the study of the crystallization of zeolite.[18, 19] Based on the role of the

amorphous phase there are two possible pathways in zeolite formation (Figure 1.6): (i) the

solid pathway in which the amorphous phase is a precursor to zeolite, the transformation to

zeolite taking place inside the amorphous phase via structure rearrangement; and (ii) the

solution pathway in which the amorphous phase is merely a nutrient source for crystal

growth by its dissolution to release active aluminosilicate monomer to the solution.

The solid pathway was suggested by Flanigen and Breck as early as 1960[2, 24, 25].

However subsequent studies on synthesis of zeolite A by Kerr and Zhdanov [19, 26, 27] in

the next two decades favored the solution pathway. By the end of 1980s the solution

pathway had been accepted, the nature of gel had very little impact on the final zeolite

structure. The new clear solution synthesis of zeolite and the application of advanced

characterization methods such as NMR and TEM in 1990s saw a numerous reports

emphasizing the role of the intermediate amorphous phase. The clear solution synthesis

method utilizes a clear solution of starting materials, hence facilitating the feasibility of in

situ techniques, allowing better imaging of the nucleation.

For example, in the crystallization of silicalite-1, an all-silica and hydrophobic

zeolite, it was revealed by de Moor et al[28] that hydrophobic silicates and structure

directing agents (SDA) are assembled by hydrophobic interaction, which results in the

formation of primary units (ca. 2.8 nm) in the solution prior to nucleation. Subsequently,

the nucleation occurs via aggregation of primary units. The primary units were also

incorporated directly into the crystalline phase during crystal growth. As the result,

nucleation and subsequent crystal growth mechanisms are described by a cluster

aggregation scheme (Figure 1.7). Further study was carried out by Jacobs et al.[29-33]. The

silica species in an aged clear sol (which crystallizes silicalite upon heating) were extracted.

The resulting powder was characterized by a wide variety of methods leading to the

identification of constituent "nanoslabs" having dimensions 1.3 × 4.0 × 4.0 nm and having

the MFI structure with nine intersections per particle, each constituent unit containing a

TPA cation (Figure 1.8a). Aggregation of precursor units leads to larger particles

measuring up to 15.6 × 8 × 8 nm and eventually to the crystalline colloidal MFI-type

12

material which forms the final product of the synthesis (Figure 1.8b). Hence, the authors

proposed that the formation of the final silicalite-1 crystals was resulted from stacking the

nanoslabs.

Figure 1.7. A scheme for the crystallization mechanism of silicalite-1[28]

Figure 1.8. The "nanoslab" hypothesis: (a) the precursor unit containing one TPA cation

and (b) schematic representation of nanoslab formation by aggregation of precursor units,

as determined by XRS and GPC [32]

13

Figure 1.9. Nucleation and growth model of zeolite A and zeolite Y as represented by

TEM [34, 35]

14

In the crystallization of hydrophilic zeolites (low Si/Al zeolites) using clear solution

method, the presence of nanosized amorphous gel particles in the pre-crystallized solution

was also found.[34, 35] From the TEM observations, Mintova et al[34, 35] revealed that,

these amorphous gel particles have different sizes, depending on the starting materials and

the zeolite structure. The particles in the synthesis of nanozeolite A was about 5 nm

whereas, those in the synthesis of nanozeolite Y was 25-35 nm. The authors proposed that

the mechanism involved the aggregation of these particles (Figure 1.9).

The nucleation and growth of zeolite can be slowed down by lowering synthesis

temperature thus allowing detailed observation. In a synthesis of silicalite-1 at room

temperature that lasted over a year by Tsapatsis et al and other groups, the evolution of the

intermediate gel from the starting clear solution was well documented.[36] The authors

found that 5 – 10 nm nanoparticle gels, although appearing at early stages[37], are subject

to slow structure rearrangement steps before reaching a critical precursor state that allows

their aggregation to form zeolite nanoparticle (Figure 1.10). Model calculations based on

the proposed mechanism are in good agreement with observation data.[38, 39]

Figure 1.10. Schematic illustration of the mechanism proposed for nanoparticle evolution

and crystal growth by aggregation.

These studies using clear solution synthesis hence emphasize the important role of

the intermediate amorphous phase, implying the preference of the solid pathway. However,

even in the clear solution synthesis, evidence for the solution pathway was also found.[40-

42]. In the synthesis of zeolite A from clear solution, Bronić et al[42] found that the zeolite

15

formation is similar to the synthesis involving heterogeneous aluminosilicate gel, namely

by (i) precipitation of an amorphous aluminosilicate gel precursor, (ii) formation of the

particles of quasi-crystalline phase (nuclei) inside the gel matrix, (iii) „„releasing‟‟ of the

nuclei from the gel matrix during its dissolution (autocatalytic nucleation) and (iv) growth

of the nuclei (crystals) from the liquid phase. The remark for the clear solution synthesis is

the need of high temperature (60 oC) to induce the formation of the gel while in the

conventional synthesis, the gel comes readily at ambient temperature.

In summary, the two pathways: solution and solid ones are likely the two extremes

of the formation of zeolite. Indeed, the formation of zeolite would always involve the

cooperation of these both pathways, however, in certain conditions that one pathway will

dominate over the other one. The conventional synthesis of micron-sized zeolite in which

the amorphous phase forms fast at large quantity would favor the solution pathway while

the clear solution synthesis would prefer the solid pathway.

Generally in the synthesis of nanoparticle materials, the crucial step is to intercept

the crystal growth of the particle right after the end of the nucleation stage, thus allowing

the control of the desired particle size and shape. For nanozeolite synthesis, the dominate

formation pathways are very important because it is the factor to determine whether it is

possible to effectively intercept the growth process. By its nature due to the presence of

large gel phase in the conventional synthesis, there is no clear separation between the

nucleation and the growth, in fact both processes wear on in parallel. It is possible to

control the growth process in this case but the undesired impact is that the nucleation is

affected also. Keep in mind that the formation of zeolite is very sensitive to the nucleation

step, the interception of zeolite growth would even prevent the formation of zeolite,

yielding amorphous products. In contrast the clear solution synthesis allows separated

nucleation and growth, hence the control of the growth process would be possible without

disturbing the nucleation process. A number of nanozeolites with different structures have

been synthesized such as FAU, MFI, LTA, MOR… Most of them were prepared using

clear solutions or gels, however, other methods such as confined space synthesis and

synthesis using growth inhibitor have been found to be useful to synthesize these materials.

16

1.2.2.2. Synthesis from clear solutions

The synthesis of nanozeolites from clear solutions was first discovered by Shoeman

et al[43] and Verduijn.[44] Since then this method has been widely used in the synthesis of

nanozeolites (Table 1.1).

Table 1.1. Molecular sieve types synthesized in nanosized form, synthesis conditions, and

crystal size[9]

Type Molar composition of the clear synthesis solution Temp, oC

Crystal

size range,

nm

Ref

AEI 1Al2O3:3.16P2O5:3.16(TEA)2O:186H2O 100,150,1

70 120-240 [45]

AEL 1.6i-Pr2NH:1.3-1.73P2O5:1.1Al2O3:35/70H2O:0/0.8-1.2HF 160,200 100-800 [46]

AFI 0.7-1.1(TEA)2O:0.6-1.0Al2O3:1.1P2O5:50H2O 160, 150-

160(mw) 50-300 [47, 48]

AFI 1(TEA)2O:1Al2O3:1.32P2O5:110H2O 90,110,16

0(mw) 80-600 [49]

BEA Al2O3:16-400SiO2:5.16-105(TEA)2O:240-6400H2O 140 10-200 [50]

BEA 0.48Na2O:9TEAOH:0.25Al2O3:25SiO2:295H2O 100 60 [51, 52]

BEA SiO2:0.2(TEA)2O:11.8H2O 100 100 [53]

FAU 5.5Na2O:1.0Al2O3:4.0SiO2:190H2O 60 20-100 [54]

FAU

2.46(TMA)2O:0.04Na2O:Al2O3:3.4SiO2:370H2O


100

100

[55]

FAU 3.4SiO2:0.83-1.7Al2O3:2.3(TMA)2O:0.1NaCl:300H2O 100 80 [56]

FAU 4Na2O:0.2Al2O3:1.0SiO2:200H2O 25 100-300 [56]

FAU 1.00Al2O3:4.35SiO2:1.40-3.13(TMA)2O(2OH

-):0-

2.40(TMA)2O(2Br-): 0.048Na2O:249.00H2O

100 32-120 [57]

FAU 2.46(TMA)2O:0.032-

0.43Na2O:1.0Al2O3:3.40SiO2:370H2O:13.6EtOH

100,130,

100+130 75-137 [58, 59]

FAU 5.5 Na2O:Al2O3:10 SiO2:180 H2O:x NaCl 100 80 - 200 [60]

GIS 1Al2O3:4.17SiO2:2.39(TMA)2O:253H2O 100 30-50 [61]

LTA

2.0-2.3(TMA)2O:0.2-0.5Na2O:Al2O3:3.4SiO2:370H2O


100

230-240

[43, 55]

LTA 1.12-3.6SiO2:1.0Al2O3:1.5-7(TMA)2O:0.007-0.28NaCl:276-

500H2O 100

50+130-

900 [34, 62]

LTA 6.1-15.8SiO2:Al2O3:17Na2O:0.9-6.5(TMA)2O:389H2O:3iPrO2 80 50-100 [63]

LTA 0.3Na2O:11.25SiO2:1.8Al2O3:13.4(TMA)2O:700H2O 22 40-80 [35]

LTA 0.22Na2O:5.0SiO2:Al2O3:8.0(TMA)2O:400H2O 63 130 [64]

LTL 10-12.5K2O:1.0Al2O3:16-40SiO2:250-450H2O 140-190 30-70 [48]

LTL 10K2O:1Al2O3:20SiO2:400H2O 175 50-60 [65]

LTL 8.0K2O:Al2O3:20SiO2:200H2O 72.5,82.5,

92.5 30-75 [66]

LTL 0.005BaO:0.25K2O:0.08Al2O3:1.0SiO2:15H2O 170 140 [67]

MEL 0.55Na2O:1.26(TBA)2O:10SiO2:150H2O 67.5,90 50-200 [68]

MEL SiO2:0.3TBAOH:4.0EtOH:18H2O 22+100 90 [69]

MEL 0.35TBAOH:1.0TEOS:12H2O 60-90 114 [69]

17

MFI 5-9TPAOH:0-0.3Na2O:25SiO2:0-0.25Al2O3:480-1500H2O 98 130-230 [70]

MFI 1Al2O3:60SiO2:21.4TPAOH:650H2O 170 10-100 [71]

MFI 9TPAOH:0.16Na2O:1Al2O3:50Si:300-495H2O:0/100EtOH 165 15-60 [72]

MFI Al2O3:60SiO2:11TPAOH:900H2O 70,90 10-20 [73]

MFI 0/0.53Na2O:0.62-1.52(TPA)2O:10SiO2:60/143H2O 50,60, 80 25-80 [44]

MFI 9TPAOH:0/0.1Na2O:25SiO2:480/1500H2O:100EtOH 98 95-180 [74]

MFI 0.01-0.443TPAOH:20-80H2O:TEOS 115 >90 [75, 76]

MFI 9TPAOH:25SiO2:480H2O:100EtOH

60,80,60+

100,

80+100

60-80 [77]

MFI 3-13TPAOH:25SiO2:480H2O:100EtOH 95 100 [78]

MFI 3.0/4.5/9.0TPAOH:16NaOH:50SiO2:495/2000H2O:100EtOH 60-98,165 20-1000 [79]

MFI 0.27Na2O:5TPAOH:25SiO2:420H2O 22+230 130-260 [80]

MFI 9TPAOH:25SiO2:0.13Na2O:595H2O:100EtOH 60,100

60-

170,100-

300

[81]

MFI 9TPAOH:25SiO2:0.13Na2O:595H2O:100EtOH 22+60,22

+100

55-

160,70-

230

[82]

MFI 9TPAOH:0.25TiO2:25SiO2:404H2O:100EtOH 100 85 [83]

MFI 0.36TPAOH:0.06TiO2:1.00SiO2:16.2H2O:4EtOH:0.24BuOH 22+175(m

w) <100 [84]

MFI 25 SiO2:3-9TPAOH:480-1500H2O:100EtOH. 100 100 – 200 [85-87]

MFI 20SiO2 : 9TPAOH : 9500H2O 40 100 [88]

MFI 20SiO2 : 7.2TPAOH : 360H2O 92 20 - 50 [89]

MFI 25SiO2 : 9TPAOH : 400H2O 95 100 [90]

MOR 6Na2O:2Al2O3:30SiO2:780H2O+seeds 150 63 [91]

OFF 2.78(TMA)2O:0.47-0.98K2O:0-

0.5Na2O:Al2O3:9.90H2O:91H2O 85 45-60 [44, 92]

OFF 10SiO2:1.0Al2O3:110-220H2O:0.12Na2O:0.5K2O:3-

4.5(TMA)2O 100 30-250 [92, 93]

SOD 14(TMA)2O:0.85Na2O:1.0Al2O3:40SiO2:805H2O 100 37 [43]

ZSM-

22 1.52(TMA)2O:0.53Li2O:0-0.08Na2O:3.4SiO2:315H2O 100 49-108 [61]

Parameters affecting the crystal size

From the perspective of crystallization theory, the crystal size is a function of the

ratio between rate of nucleation and rate of growth.[23] Thus, to obtain nanozeolites, one

should optimize these following conditions: (i) attaining very high nucleation rates and (ii)

providing stabilization of nuclearsized entities. The first condition is controlled by many

parameters such as temperature, alkalinity, aging… whereas the second one depends

chiefly on the role of SDAs. Here is a brief review of these parameters.

- Temperature: a low crystallization temperature (80 – 100oC) is often applied. This

is because temperature raises growth rate more than nucleation rates. However, it should be

18

noted that a too low temperature usually results in poor crystallinity, low efficiency, and

longer crystallization time.

- Aging: the aging of the synthesis mixture at room temperature has significant

influence on the nucleation rates. This is due to the fact that nucleation rate is favored at

room temperature, but the growth rate is negligible, and thus the nuclei prevail until the

temperature is raised.

- Alkalinity: the concentration of OH- ions strongly increases the solubility of

silicate species. In general, smaller zeolite crystals tend to formed at higher alkalinity.[60]

- The concentration of the clear solution strongly affects the degree of saturation of

the system. At lower supersaturation, growth is favored at the expense of nucleation.

Dilution of the solution can cause large crystal to form. Hence a high concentration is a

desired parameter.

- The solubility of the silica source plays an important role in the synthesis of

nanozeolites. Smaller crystals are formed from monomeric silicate solutions than by

dissolution of colloidal silica.

- Metal cations usually facilitate the crystal growth. Hence, they should be present

in the synthesis solution at low concentration. For example, it has been found that sodium is

the growth-limiting nutrient in the formation of Y-type zeolite.[43, 59] The crystallizations

of Y-type and A-type zeolites are very sensitive to sodium content. In some cases, a small

variation in this factor results in the formation of a different crystal phase, for the synthesis

of zeolites with the clear synthesis gel having molar composition of 2.46 (TMA)2O: x

Na2O: 1.0 Al2O3 : 3.40 SiO2 : 370 H2O:13.6 EtOH (0.03 < x < 0.43), the sodium

concentration of the batch is crucial for controlling which zeolite phase crystallizes. Greater

Na2O/Al2O3 ratio (0.43) in the batch favors the formation of zeolite A with a higher yield of

56.5% after a shorter crystallization time. Lower Na2O/Al2O3 ratio (0.03) in the batch

produces smaller zeolite Y crystals with a lower yield of 8.1% after a longer crystallization

time.

19

- Structure directing agent (SDA): SDAs are often quaternary of the type [R4N]+OH

-

(where R is an alkyl group, typically CH3, C2H5, C3H7 or C4H9). The presence of SDA in

the synthesis solution helps assist the formation of desired zeolite structure. Furthermore,

SDA is responsible to the stabilization of silicate subcolloidal particles[94] as well as the

nanozeolites.[95] In the synthesis of nanosilicalite-1 using tetrapropylammonium hydroxide

(TPAOH) as SDA, it was revealed[95] that, there were two different environments for the

TPA+ cation: (i) TPA

+ occluded in the channel intersections and (ii) TPA

+ adsorbed on to

the external surface of the particles. Bulky quaternary ammonium cations adsorbed on to

particle surfaces provide steric stabilization, preventing aggregation upon collision. In an

aqueous medium, zeolite particles will acquire a negative surface charge due to dissociation

of the surface silanol groups. Such a surface charge will cause organic cations in the

surrounding solution to align along the particles‟ surface, creating an electric double layer.

This stabilizing barrier of bulky organic cations restricts the close approach of similar

particles, such that the attractive potential between them is insufficient to cause aggregation

or flocculation.

1.2.2.3. Synthesis using growth inhibitor

The synthesis of nanozeolites using growth inhibitor is in fact derived from the

synthesis from clear solution. In this method, an organic additive other than SDA is

introduced to inhibit the growth process, thus resulting in small zeolite crystals. The

reactivity of the additive and its content in the synthesis mixture are two significant factors.

The addictive should be able to adsorb on to or react with the surface of the silicate

particles, thus, protects them from further aggregation. If the concentration of the additive

is too high, zeolite might not be obtained since there would be not sufficiently free

aluminosilicate species for the formation of zeolite structure. In contrast, if the

concentration is too low, the inhibition effect might be inadequate.

Hosokawa et al[96] revealed that, a nonionic surfactant (polyoxyethylene lauryl

ether, C12E6) and polyethylene glycol (PEG 600) were able to act as the growth inhibitors

in the synthesis of nanosized A-type zeolites. The inhibitor was added with the synthesis

mixture prior to crystallization. TEM investigation showed that the resulting zeolite A was

20

in the form of aggregated 30 – 40 nm particles which was nearly similar to the size of the

aluminosilicate precursor species available in the synthesis gel prior to crystallization. This

observation indicated that the growth of the zeolite crystals should be inhibited.

The introduction of a growth inhibitor can be postponed until viable zeolite

precursors become prevalent. Thus the undesired interference of the inhibitor in the

formation of precursors is avoided. This approach was developed by Naik et al[97] for

preparation of nanosilicalite-1. The procedure involves: (i) prepare a clear solution that is

known to produce colloidal TPA-silicalite upon extended hydrothermal reaction; (ii)

subject the solution to hydrothermal condition but stop before the appearance of colloidal

silicalite; (iii) protect the TPA-silicalite precursor nanoparticles with cationic surfactant

(CTABr) and collect them as flocculated mass; (iv) convert the precursor/surfactant hybrid

into nanocrystals via high temperature steaming. A steaming temperature of 150 oC was

found enough to convert the collected precursor into nanocrystals. The obtained

nanocrystals were smaller than 30 nm. However these particles are hard aggregated and can

not redispersed in water.

Serrano et al[98] reported the use of organosilane as the growth inhibitor.

Organosilane is a good silylating agent which has been widely used for the

functionalization of zeolites. According to the author, MFI and beta zeolites were

successfully synthesized, using phenylaminopropyl-trimethoxysilane (PHAPTMS). The

synthesis is based on reducing the growth of zeolite crystals by silanization of the zeolitic

seeds to hinder their further aggregation. Typically, the synthesis of MFI zeolite is as

follows: a clear solution of TPA-aluminosilicate was produced. The precursor solution was

precrystallized under reflux with stirring (100 rpm) at 90 ºC for 20 h. Then, the zeolite

seeds obtained were functionalized by reaction with PHAPTMS at 90 ºC for 6 h. Finally,

the resulting solution was subjected to hydrothermal treatment at 170 oC for 5 days.

However, as investigated by TEM analysis, the MFI sample obtained consisted of particles

of about 300 – 400 nm which were formed by aggregation of ultra small primary units of

10 nm. Having that large size, the sample was hardly considered as true nanozeolite.

21

1.2.2.4. Confined space synthesis

In the field of zeolite science, the term “confined space” was first used by Jabobsen

et al. [99-102] in 1999 to describe a novel method for zeolites synthesis which allows

preparation of nanosized zeolites crystals with a controlled crystal size distribution. The

principle of confined-space synthesis is to synthesize the zeolite inside the mesoporous of

an inert matrix. The maximum crystal size is limited by the diameter of the mesopores as

shown in Figure 1.11

The zeolite gel is introduced into the mesopores of the matrix by sequential

incipient wetness impregnations the matrix with the gel precursor solution. For the

synthesis of ZSM-5, the carbon black was impregnated to incipient wetness with a clear

solution of TPAOH, H2O, NaOH and ethanol. After aging for 3 h at room temperature, the

carbon black was subjected to hydrothermal treatment at 180 oC for 48 h in an autoclave.

The product was then recovered by calcinations to remove the matrix.[99]

Figure 1.11. Schematic illustration of confined space synthesis.[101]

This method has also been applied to the synthesis of beta, X and A zeolites.[101,

102] Two carbon black matrixes were used with a pore diameter of 31.6 and 45.6 nm,

respectively. Generally, the crystal size distributions of the zeolites obtained were governed

22

by the pore size of the carbon black matrix and were typically in the range 30-45 nm.

Crucial factors in the synthesis are (i) restriction of the crystallization of the zeolite gel

within the pore system of the matrix, which was achieved by the incipient wetness

impregnation method employed to load the mesopores with a synthesis gel, and (ii)

prevention of diffusion of the zeolite gel species from the mesopores, which was ensured

by avoiding direct contact between the impregnated carbon black matrix and the water at

the bottom of the autoclave.[9] Jacobsen et al. later reported that the failure in controlling

these two factors could lead to the formation of mesoporous zeolites rather than

nanozeolites.[103] Although the confined space synthesis by Jacobsen et al. eliminates the

problem of recovering nanocrystals form solution, it has its own drawback. For example the

carbon matrix must have a uniform distribution of mesopores to ensure the size distribution

of the product. Careful procedures were needed to impregnate the synthesis solution just

inside not outside. And finally, the zeolite particles are aggregated and not discrete.

Another type of confined space synthesis has been developed by Wang et al.[104]

The technique works on the thermoreversibility of gelling polymers. These polymer gels

are hydrogels with three-dimensional networks of polymer chains that are cross-linked via

either physical or chemical bonds, and they can entrap a large volume of water. The

interesting gelation behavior of thermoreversible polymer hydrogels is that it is reversibly

responsive to temperature. In particular, the polymers gel at elevated temperatures and turn

back to solution at room temperature. This feature is attractive because the temperature

profile of their solution-gel transition can very nicely fit that of hydrothermal synthesis of

zeolite. The three-dimensional pores of polymer hydrogels can potentially serve as

microreactors or nanoreactors for controlling zeolite growth. The general synthesis

procedure is illustrated in Figure 1.12.[104] A (20 – 180 nm) and X (10 – 100 nm) zeolites

were prepared using this technique, however, the size distributions were broad. In addition,

this method encounters the following difficulties: (i) Gelation of methyl cellulose is

complicated. The thermoreversibility depends strongly on the heating and cooling rates.

Any variation in these rates can lead to different gel properties.[105] (ii) It is difficult for

the gelation of methyl cellulose to occur in the presence of zeolite synthesis gel since the

high basic content of the gel facilitates the solubility of methyl cellulose.[106] (iii) The

authors applied slight hydrothermal conditions to prepare A and X zeolites, i.e. at 80oC for

23

2 – 3 h. However, for the synthesis of other zeolite structure, the conditions generally

involve the temperature higher or equal to 100oC and the crystallization time of several

days. It is likely that the gel structure of methyl cellulose would not be able to maintain

under these hard conditions.

Figure 1.12. Schematic representation of synthesis of template-free zeolite nanocrystals by

using in situ thermoreversible polymer hydrogels.[104]

24

Figure 1.13. Schematic representation of revered microemulsion

Finally, it would be worthy of mentioning the synthesis of nanozeolites using

microemulsion. Microemulsions are colloidal „nano-dispersions‟ of water in oil (or oil in

water) stabilized by surfactants.[107] These thermodynamically stable dispersions can be

considered as truly nanoreactors which can be used to carry out chemical reactions and, in

particular, to synthesize nanomaterials. The main idea behind this technique is that by

appropriate control of the synthesis parameters one can use these nanoreactors to produce

tailor-made products down to a nanoscale level with new and special properties (Figure

1.13). Reversed microemulsions have been wildly used to prepare inorganic

nanoparticles.[107, 108] Thus, the application of this technique in the synthesis of

nanozeolites is of interest. The first attempt of using microemulsion to prepare zeolites can

be credited to Dutta et al.[109, 110] The authors reported the synthesis of 1 – 2 m A-type

zeolite and 0.6 m zeolite-type zicophosphate. Nevertheless, since the preparation of

Surfactant

Water droplets

Increase

[water]/[surfactant]

25

zeolites is somewhat different from that of inorganic particles. The technique has the

following difficulties:

(i) The crystallization of zeolites often involves heating at high temperature. Under

these conditions the microemulsion becomes unstable, the nanoreactor effect is not

attained. Manna et al.[111] reported the synthesis of silicalite-1 in a microemulsion of

tetraethylenepentamine–sodium bis(2-ethylhexly) sulfosuccinate (AOT)–water system

containing fluoride ions. At the crystallization temperature of 170 oC, the microemulsion

system was destroyed and turned into bicontinuous emulsion. Hence the resulting silicalite-

1 crystals were twinned and had large diameter of 4 m.

(ii) The interaction between surfactant and aluminosilicate species is complicated

and can affect the stability of microemulsion and the morphology of zeolites. Shantz et

al.[112-114] found that, initially, the microemulsion acts as a confined space, effectively

inhibiting zeolite growth in the early stages of synthesis as compared to bulk syntheses.

However, once the particles reach a critical size, approximately 100 nm, the effect of

surfactant adsorption at the aluminosilicate surface becomes so important that the small

particles formed in the microemulsions aggregate to form large particles.

Chen et al.[115] reported the successful synthesis of 40-80 nm A-type nanozeolite.

To overcome these above problems, the microemulsion system containing the syntheis gel

was crystallized at a low temperature of 75 oC in very short duration less than 60 min. To

facilitate the crystallization, microwave heating was applied instead of conventional

heating. In spite of that, the size distribution is still broad (Figure 1.14).

26

Figure 1.14. Schematic representation of microemulsion-microwave synthetic method[115]

1.2.3. Recent advances in application of nanozeolites

The expected application of nanozeolite is its catalytic activity due to three main

factors. First is the accessibility. The external surface of nanozeolite is higher, hence

exposing catalytically active to large molecule. Second is higher activity. Also increasing

with the external surface is the number of active sites.[116] As the low-coordinated corner

and edge sites are more active, the catalytic activity should increase with decrease in size of

the nanozeolite. And third is the improvement in the diffusion into the internal active sites

of nanozeolites thanks to their smaller size.

Beside direct applications in catalysts, in the last five years there has been a great

interest in the use of nanozeolite to build advanced catalysts. The search for mesoporous

materials with zeolitic walls has received a lot of attention and effort from the microporous

and mesoporous materials community.[117, 118] These hierarchically nanoporous zeolites

27

would combine the hydrothermal stability, catalytic activity of zeolite with the accessibility

of mesopores, thus it is possible to design efficient catalyst for reactions involving large

and bulky molecules. However, the synthesis of these materials is not easy. The straight

forward methods which introduce both zeolite template and mesoporous template do not

work as the mesoprous template and zeolite cancel each other out. In 2006, a report by

Ryoo et al suggested a novel method to synthesize hierarchically nanoporous zeolites.[119,

120] Nanozeolite ZSM-5 was synthesized in such a long duration that affords sufficient

time for the freshly formed nanoparticles to gather into large aggregates. The aggregates

feature the intraparticle mesopores as well as the zeolitic micropore. NH3-TPD and

spectroscopy analyses of the hierarchically nanoporous zeolite showed its acidity as high as

conventional zeolites.[121] Thus as the catalyst for reactions of small reactants and

products such as methanol to olefin/gasoline conversion, it exhibits a catalytic activity on

par with bulk zeolite. However the superior activity of this hierarchically nanoporous

zeolite is revealed when larger molecules that cannot diffuse easily inside bulk zeolites are

introduced. The mesoporous MFI zeolite exhibits much higher catalytic activity and

selectivity in the jasminaldehyde (α-n-amylcinnamaldehyde) synthesis reaction than bulk

zeolite. It also displayed an outstanding catalytic activity in the synthesis of vesidryl

(2‟,4,4‟-trimethoxychalcone).[119] Not only activity but the stability of nanozeolite-based

materials is also improved. Hierarchical structure exhibited remarkably high resistance to

deactivation in catalytic activity of various reactions such as isomerization of 1,2,4-

trimethylbenzene, cumene cracking, and esterification of benzylalcohol with hexanoic acid,

as compared with conventional MFI and mesoporous aluminosilicate MCM-41.[122] Ryoo

et al also showed that nanozeolites are also better support than bulk zeolite or mesoporous

silica.[123] Palladium acetate was immobilized on the mesopore wall of hierarchical

MFI nanozeolite, and tested as a catalyst for Suzuki coupling reaction in water. The

catalyst exhibited very high activity in the coupling of various aryl bromides with

arylboronic acids. Moreover, the catalyst could be recycled without a significant loss

of catalytic activity.[124] There have been reports on the synthesis using MFI, SOD, BEA,

LTA zeolites.[123]

The year 2009 saw another breakthrough in nanozeolite applications.[125, 126] A

special type of nanozeolite, the zeolite nanosheet of ZSM-5 which is composed of only two

28

layers of a microporous channel in the perpendicular direction was successfully

prepared.[125, 126] The author found that, in the case of the methanol-to-hydrocarbon

conversion, the MFI zeolite nanosheet exhibited much longer catalytic lifetime than the

bulk zeolite. This result illustrates the facile diffusion of coke precursors in the internal

acidic catalytic sites out of nanozeolite.

29

1.3. Metal-organic frameworks

1.3.1. Background

The advent of zeolite has inspired researches for zeolite analogs or zeotype. From

the basis of zeolite structure which is built on the oxygen connected tetrahedral TO4 (T =

Si, Al) it is a rational step to propose possible structure which is built on other polyhedra of

other atom other than Si and Al such as octahedral TO6 pentacoordinated TO5, or pyramidal

TO4, TO3 units. Several zeotype structures have been found. These include a large number

of main block phosphates, such as those of gallium, indium, and tin, as well as several

transition metal phosphates, including systems based upon vanadium, molybdenum, cobalt,

and iron.[127, 128] Attempts to preplace the oxygen atom with sulfur, chloride, and

nitroten were also reported. [127]

Figure 1.15. Rational design of zeotype and MOF from the basis of zeolite

But in retrospective, the most radical design of porous materials which can be traced

back to zeolite is metal-organic framework (MOF). By definition, MOF is

a crystalline compound consisting of metal ions or clusters coordinated to often

Replace SiO4 with PO4

Zeotype AlPO4

Replace Al, Si with M

Replace O with L

MOF

Zeolite

30

rigid organic molecules to form one-, two-, or three-dimensional structures.[129, 130] In

the MOF framework, the metal or metal cluster is called the node and the organic molecule

linking these nodes is called linker.[131] Hence, MOF is a completely new redesign of

zeotype structure: replacing both the TO4 and the oxygen atom with the node and linker,

respectively. The design principle of MOF is illustrated in Figure 1.15.

From the point of coordination chemistry, MOF is often called coordination

polymer. In most cases, both of these terms, MOF and coordination polymer can be

interchangeably used. However, their use and scope are still subject to debate.[132] MOF is

often used to imply a 3D crystalline structure while coordination polymer refers to 1D and

2D infinitive array. An IUPAC project was initiated in 2009 to address the terminology

issues in this area. A progress report has been published.[133]

There are some similarities in history of MOF discovery and that of zeolite. Both

were found by accident without deliberate purpose to obtain porous structures. While

zeolite was discovered about 200 years before its full potentials were realized, the time it

took chemists to appreciate the amazing world of MOF is even astonishingly longer. The

first successful synthesis of MOF, the Prussian Blue dates back to 1705 by Diesbach but it

would be more than 270 years later, the structure of Prussian Blue, Fe4[Fe(CN)6]3.xH2O

was resolved.[134] In the intervene and sometime after, some preparations of MOF such as

M(imidazole)2 (M = Ni, Cu, Zn and Ag)[135] and iron(III) dicarboxylate[136] were

reported with suggestion of their polymeric nature. However, the general consensus was

that they are just a kind of coordination compounds with expected and typical properties

such as magnetism.[137] The framework aspect of these compounds which would be later

considered as MOFs received little attention. It was only after 1989 with the landmark

paper of Robson and Hoskin and their subsequent paper in 1990[138, 139] that the real

exciting framework of MOFs was realized.

The breakthrough in the discovery of Robson and Hoskin is that it is possible to

design coordination polymers or MOFs.[140, 141] Applying the idea of Wells[142] which

describes the crystal in terms of network with nodes and linkers, they showed that it is

possible to produce MOF of predetermined topology and pore size and functionality by

rational selection of node and linker and appropriate experiment conditions. Thus the

31

porosity and the catalytic activity of MOF can be engineered. The idea was soon picked up

and developed by several prominent researchers[143-149] making important contributions.

However, it would take another decade since the first report of Robson and Hoskin on

MOFs before the interest in MOFs took off and soon exploded thanks to the two separate

discoveries of HKUST-1 (Cu3(btc)2.3H2O) (btc: benzenetricarboxylic acid) and MOF-5

(ZnO4(bdc)3) (bdc: benzenedicarboxylic acid) reported in Science[150] and Nature[151],

respectively, in 1999.

Figure 1.16. Crystal structure of HKUST-1 (left) and MOF-5 (right)

Before the discoveries of HKUST-1 and MOF-5, despite early promising reports of

new MOF structures[149, 152], the stability and the porosity of MOFs were met with

skepticism because of the weak coordination bonds in MOFs.[132] But the coming of

HKUST-1 and MOF-5 changed everything. MOF-5 exhibits spectacular and unprecedented

specific area of 4400 m2/g[153] which would be easily five times the average specific area

of zeolite and put MOF-5 in the class of ultraporous materials. HKUST-1 displays a much

lower specific area of 690 m2/g which is on par with zeolites but exhibits thermostability up

to 240 oC.[150] In addition catalytic potentials and its flexibility in ligand exchange were

clearly demonstrated. But HKUST-1 and MOF-5 are not only striking in their properties,

what is also startling is their rational design. Both the nodes of HKUST-1 and MOF-5, the

copper cluster and the zinc cluster, have been known in copper acetate and basic zinc

32

acetate, respectively for decades.[154] The beauty of the design of HKUST-1 and MOF-5

is that it was simple and logical: replacing acetic acid with ditopic linker terephtalic (bdc)

or tritopic linker trimesic acid (btc) to obtaine MOF-5 or HKUST-1. In a retrospective view

10 years later, even one of the co-inventors of MOF-5, Michael O‟Keeffe still wondered

why no one had come up with that idea before.[132]

Figure 1.17. Number of metal–organic framework (MOF) structures reported in the

Cambridge Structural Database (CSD) from 1978 through 2006. The bar graph illustrates

the recent dramatic increase in the number of reports, while the inset shows the natural log

of the number of structures as a function of time, indicating the extraordinarily short

doubling time for MOF structures compared to the total number of structures archived in

the database.[129]

Fast forwards to the present, the research of MOF is now at an explosive mode with

an exponential growth in the number of research papers and reviews appearing in literature

(Figure 1.17).[129] Almost every issue of any chemistry related journal features at least one

research paper or review on MOF. Exciting properties of MOF in every aspect seem to be

reported daily and non-stop: catalysis,[155-158] adsorption,[159-165] separations,[166-

33

168] optics,[169] drug delivery,[170] magnetism[171] and luminescence[172]… MOFs

also begin to see their industrial application.[173] Still the chemistry of MOF is not

matured yet, there are lot of wonders left to be found.[174] And even new discoveries of

MOF sometimes could catch their veteran researchers by surprise, raising questions

whether there is a limit to MOF.[175]

1.3.2. Design principles of MOFs

The crystal structure is a complicated result of the connectivity, the geometry, and

the shape and position of its elements. Among these parameters, the connectivity between

the linkers and the nodes is the most important one. This parameter can be described by

using the network concept.

A network (net) is a topology classification of MOFs that can aid the description

and the understanding of MOFs structure. A network is a polymeric collection of

interlinked nodes; each link connects two nodes and each node is linked to three or more

other nodes. A node cannot be connected to only two nodes; in this case it then becomes a

link. Similarly, a link can only connect two nodes; if it connects more than two it is a node.

The network must also have a repeating pattern and thus a finite number of unique nodes

and links.

A MOF structure can be described by a corresponding net which provides with

information on the number of different nodes and there connectivities. Since the net does

not involve the geometry parameters of the structure, it is often much simpler than the

crystal structure it represents while still keeping vital information of how the structure can

be constructed. For example two hypothetical structures are showed in Figure 1.18, which

are clearly different, but in fact they share the same net, a 3-connecting net.[131]

34

Figure 1.18. Two geometrically different but topologically identical nets.

The naming of a net name as a three letter code often refers to the most well-known

structure that it represents. For example the dia net represent the four-connecting

framework of diamond. The sod is that of sodalite zeolite.[176]

In addition to the net concept, to facilitate the design of MOF, Yaghi et al suggested

the use of the concept secondary building unit (SBU).[176, 177] The SBU of a MOF

structure is the abstraction of the repeating metal cluster in the MOF structure, the metal

cluster is regarded merely as a node in the net with its number of connectivities.

For example, the MOF-5 structure is formed by the Zn4O clusters are linked by bdc

linkers. The Zn4O as an SBU can be represented as an octahedron with its 6 vertices being

the 6 connectors. Hence the structure of MOF-5 is the linking of these octohedra (Figure

1.19)

35

Figure 1.19. MOF-5 structure from linking octahedral SBUs

To obtain a desired structure, it is necessary to apply two transformations to its net:

decoration and expansion. Decoration is to replace a vertex by a group of vertices. A

special case of decoration is augmentation; this is a replacement of the vertices of an N-

connected net by a group of N-vertices. Expansion is to increase the spacing between

vertices in a net by using longer links, which in principle means that a bond is replaced by a

sequence of bonds. Both decoration and expansion are the main principles of the reticular

synthesis which have produced series of MOF having the same topology.[178] To

illustrate, the design of MOF-5 from the simple structure of NaCl is explained. NaCl

structure is a 6 connecting framework, by augmentation, the node of NaCl is replaced with

the Zn4O cluster octahedra with the same 6 connectors. Then in the expansion step the

links between these octahedra are increased by replacing the link with bdc, thus the MOF-5

structure is obtained.

36

Figure 1.20. Design of MOF-5 from simple fcc structure

1.3.3. Synthesis

The synthesis of MOF bears some similarities to that of zeolites. Both employ

solvothermal technique. While water is almost the exclusive solvent in the synthesis of

zeolites, the use of other solvents are often encountered in the synthesis of MOF. The

equipment for the synthesis of zeolites thus can be used for the synthesis of MOF without

any modification. In a typical synthesis, a solution containing metal salts and the linkers are

prepared at room temperature, the solution is then transferred into a capped autoclave

which is afterwards heated at determined temperature for certain duration. Solid product is

recovered by filtration or centrifugation. Just like the synthesis of zeolites, the parameters:

solvent, temperature, pH, concentration, synthesis time and even anion types are important

factors affecting the formation of MOF. Eventually, the right synthesis parameters should

favor the formation of the SBUs while maintaining the integrity of the organic linkers.

Solvent: water would be the first to be considered for the synthesis due to its

convenience: high availability, ease of handling and non-toxicity. However, as the organic

linkers often exhibit low solubility in water, other organic solvents such as

dimethylformamide (DMF), diethylformamide (DEF) and alcohol are often used.

Sometimes a cosolvent is necessary to tune the solubility. Solvents have a great impact on

the structure of MOFs. For example, in the synthesis of MOF-5, the SBU Zn4O is very

sensitive to solvents, and it was found that alkyl formamide such as DMF and DEF favor its

formation.[151, 178] That‟s why DMF is often used in the synthesis involving Zn4O.[178,

37

179] In contrast, water is detrimental to Zn4O, even with a trace amount of water in DMF

could render the synthesis of MOF-5 fruitless.[153]

Temperature of the synthesis should be high enough to increase the solubility of the

reactants and accelerate the reaction, however, the temperature should not be too high to

avoid the thermal decomposition of the organic linker. The synthesis temperature can be

elevated to 210 oC as in the synthesis of Cr-MIL-101 of which the formula is

Cr3O(bdc)3F.3H2O.[180] One of the reasons for such a high temperature is the low

solubility of bdc in water which is used as solvent at low temperature. Sometimes from the

same synthesis composition change in temperature can result in different MOF structures.

This is the case when there is a competition between the kinetically favored structure and

the thermodynamically favored one. The two MOFs: MIL-88B and MIL-101 have the same

formula Fe3O(bdc)3Cl.xDMF but different structures.[180, 181] In addition, MIL-88B is

the kinetically favored structure while MIL-101 is thermodynamically favored one. From

the same mixture of FeCl3 and bdc in DMF, one can obtain MIL-88B at temperature less

than 100 oC and MIL-101 at temperature higher than 170

oC.[182]

pH is another important factor to consider. When acidic organic linkers such as

polycarboxylic are used, a basic environment is needed to facilitate the deprotonation of the

linkers, so that they can be assembled into the MOF frameworks. The syntheses of MIL-

88B, MIL-88A structure would demand a basic strength of NaOH to initiate.[183, 184] It

should be noted that the agent used to change the pH does not interfere with the SBU by

directly bonding to it, that‟s why trimethyl amine is a common choice to increase pH due to

its high basicity and low ability in forming a coordination bond to metal.[149, 177] In most

of the syntheses not only the protonation of linkers but the desired SBU only coexist in a

very short pH range, hence control of pH is a vital factor. A classic example of the careful

control of pH is the synthesis of MOF-5 in DMF using Zn(NO3)2.6H2O and bdc without

any additional pH agent.[151] The linker bdc needs to be deprotonated, hence a basic pH is

required, but a strong basic environment does not favor the SBU Zn4O. The use of DMF as

solvent perfectly overcomes this problem. At high temperature, DMF slowly and partly

decomposes to release dimethyl amine, which in return deprotonate bdc.[185, 186] Due to

its slow release, the amine is almost used up by the deprotonation and thus the overall

38

increase in pH is very small and does not have any detrimental impact on the SBU. In this

case the combined factors of both solvent and temperature allow a precise control over pH.

Sometimes the need of deprotonation is outweighed by the need to maintain the integrity of

the synthesis mixture. In the synthesis of MIL-101 as the temperature is as high as 210 oC,

thus the chance is that Cr(III) ion would undergo hydrolysis yielding oxide, to prevent this,

HF is added, reducing the pH, hence preventing the hydrolysis.[180, 184]

Concentration in general affects the speed of the synthesis. Traditionally, the

ultimate goal of the synthesis of MOF is to obtain high quality single crystals for

subsequent structural analysis and characterization. Thus the synthesis is carried at high

dilution so that the crystal growth be controlled. But when high yield is desired over the

crystal quality, one can increase concentration.[130] In some instance increase in

concentration would promote the formation of kinetically favored structure over the

thermodynamically favored one. Fe3O-MIL-101 can be prepared in a diluted solution,

increase in the concentration results in the exclusive formation of MIL-88B.[182, 187]

For thermodynamically stable products, longer crystallization would increase the

crystal quality, however for kinetically favored products, care should be taken to choose the

right crystallization time, to obtain the maximum yield in product with high quality in

crystallinity. Jhung et al observed that the synthesis of MIL-101 should be in one-day

course, upon prolonging to two days, MIL-101 dissolve, giving room for another more

stable MOF: MIL-53 to come.[188]

1.4. Some applications of MOFs

1.4.1. Adsorption

Specific surface areas of MOFs often come at thousands m2/g. In fact, the material

having the highest specific surface area per a mass unit is MOF, up to 6000 - 7000

m2/g.[175, 179, 189] So adsorption is the obvious application of MOFs. Earlier attempts

involved adsorption of hydrogen[190] due to interest in the hydrogen-powered technology

for mobile vehicles.[191] Now adsorption analyses of MOFs have embraced almost every

common gases: N2, CO, CO2, alkanes, alkenes, aromatics, H2O and H2S. [159-165, 192]

39

Adsorption of hydrogen: as an alternative to mobile transportation based on

combustion engine, hydrogen technology is very promising. The engine built on the

hydrogen fuel cell would be completely “green” since it only takes in hydrogen and oxygen

and exhausts water. Oxygen is freely available in air, so the issue for the fuel cell is the

storage of hydrogen. The storage of hydrogen is in fact the biggest problem. Other factors

in hydrogen technology such as the price of the fuel cell and cost and safety requirement of

hydrogen production plants are also important but they are merely cost driven factors and

would be relieved when the fuel cell automobiles are produced at mass scale. But the

storage of hydrogen is a scientific and technological problem. An on-board storage of

hydrogen should allow a running course of 500 km or greater for one charge of hydrogen

while meeting all performance metrics. Unfortunately, this is impossible for current storage

technology, regardless of cost.[191] The U.S. Department of Energy (DOE) target of

hydrogen storage by the year 2017 is 5.5 wt % in gravimetric capacity, 40 g/L of

volumetric capacity at an operating temperature of 40 - 60 oC under a maximum delivery

pressure of 100 atm. The targets are for a complete system, including tank, material, valves,

regulators, piping, mounting brackets, insulation, added cooling capacity, and/or other

balance-of-plant components. So the actual adsorption capacity of the materials should be

greater.[163]

Due to its high porosity, studies of MOFs as hydrogen storage materials have been

extensively carried out. The highest excess hydrogen storage capacity reported so far for

MOFs is 99.5 mg/g at 56 bar and 77 K in NU-100 (NU = North-western University, Figure

1.21), which has a total capacity of 164 mg/g at 77 K and 70 bar.[193] The highest total

hydrogen storage capacity reported is 176 mg/g (excess 86 mg/g) in MOF-210 at 77 K and

80 bar (Figure 1.22).[189]

40

Figure 1.21. Structure of NU-100 and its linker

Figure 1.22. The two different linkers used in MOF-210 (left) and the crystal structure of

MOF-210 displaying its unique topology with two different types of pores (right).

41

Figure 1.23. Excess hydrogen uptake at 77 K versus BET specific surface area (BET ssa)

for various high-porosity MOFs. The symbols denote measurements conducted by different

research groups (circles: at 20 bar by Hirscher et al.[193] triangles: at 60 bar by Kaskel et

al.[194]; squares: saturation values by Yaghi et al.).[195]

A general trend is that there is a linear correlation between the hydrogen uptake at

77 K and the specific area (Figure 1.23).[194, 195] Hence the strategy to improve hydrogen

storage of MOFs is to increase its surface area and pore volume. This can be done by

elongation of the linker as in the case of NU-100 or use of mixed linker as in the case of

MOF-210. In fact at 77 K NU-100, MOF-210 and other MOFs readily exhibits hydrogen

storage capacity greater than the DOE target, however, at room temperature the storage

capacity drops drastically and does not meet the DOE target. This is due to the nature of

physical adsorption of hydrogen on MOFs. The heat of adsorption is a useful parameter

representing the stability of the adsorption of hydrogen on MOFs. The heat of hydrogen

adsorption on MOFs is small and in the range of 5 – 9 kJ/mol.[196] Optimal heat of

adsorption should be about 15 kJ/mol.[197] One could overcome this problem by

introducing chemically active site into MOFs to strengthen the bond between MOF and

hydrogen. The active sites can be the metal site or the function group in the organic linkers.

42

Recent theoretical studies showed that impregnation MOFs with light metal that can form

hydride with H2 can improve greatly the heat of adsorption.[198, 199]

Figure 1.24. MOF-117 structure and comparison of the volumetric CO2 capacity of

crystalline MOF-177 relative to zeolite 13X pellets, MAXSORB carbon powder, and

pressurized CO2.[200]

43

Figure 1.25. ZIF-69 structure

Another application in adsorption is CO2 capture. There is growing concern that

anthropogenic CO2 emissions are contributing to global climate change. Therefore, it is

critical to develop technologies to mitigate this problem. One very promising approach to

reducing CO2 emissions is CO2 capture at a power plant, transport to an injection site, and

sequestration for long-term storage in any of a variety of suitable geologic formations.[201]

Compared with common solid porous materials MOFs exhibit superior CO2 adsorption

capacity. Zeolite 13X and activated carbon MAXSORB are those conventional materials

that exhibit the highest reported gravimetric CO2 capacity of 7.4 mmol/g and 25 mmol/g,

respectively and at 32 – 35 bar.[202, 203] But the gravimetric CO2 capacity of MOF-177 of

33.5 mmol/g readily exceeds these standard materials, having 150% of their capacity.[200]

In terms of volume, one liter of MOF-177 could hold 9 liters of CO2 at ambient

temperature. This was a record for MOF back in 2005, however it was soon surpassed by

an exceptional MOF: the ZIF-69 structure (Figure 1.25) which is composed

44

of tetrahedrally-coordinated Zn ion connected by organic imidazole derivative linkers. One

liter of ZIF-69 can hold 83 liters of CO2.[204]

The selective CO2 adsorption of MOFs can be improved by using metal ion with

high affinity to CO2.[205] Among the family of MOF-74 built on different metal ion: Mg,

Zn, Ni, Co, the Mg-MOF-74 exhibits the highest selectivity in adsorption of CO2 at room

temperature and very low pressure of 0.1 atm. One unit cell of Mg-MOF-74 can hold 12

CO2 molecules, twice as much as the other MOF-74.[206]

Figure 1.26. A portion of the structure of the sodalite-type framework of Cu-BTTri (1)

showing surface functionalization of a coordinatively unsaturated Cu(II) site with

ethylenediamine, followed by attack of an amino group on CO2 .

Functionalization of MOFs with amine group can lead to strong binding of CO2. As

illustrated in Figure 1.26, alkylamine incorporation onto the open metal sites of Cu-BTTri

(BTTri: 1,3,5-benzenetristriazolate) was found to be an effective method for

postsynthetically modifying this MOF to enhance the CO2 binding.[207] A 3.5-fold

increase in gravimetric capacity at 0.15 bar (to ∼ 9.5 wt %) compared with the

45

nonfunctionalized Cu-BTTri framework was achieved. Significantly, the functionalized

framework exhibits a higher uptake of CO2 at very low pressures compared with the

nongrafted material and displays a record isosteric heat of adsorption of 90 kJ/mol.

1.4.2. MOF as catalysts

Catalysis is potentially one of the most important applications of MOF. As is the

case of zeolites, the catalytic applications of MOFs come from the following factors:

catalytic activity of the metal ions, catalytic activity of the organic linkers, and catalyst

support. Moreover some distinctive catalytic properties of zeolites such as shape selectivity,

confinement effect are expected to be also available for MOFs [208].

1.4.2.1 Metal ions as catalytic sites

It is well known that transition metal complexes can be used as active homogeneous

catalysts in selective synthetic routs under mild operating conditions for valuable chemicals

from basic organic precursors. Some of major reactions catalyzed by transition metal

complexes are hydrogenation, oxidation, hydroformylation, carbonylation, carbon-carbon

bond formation reactions. Some of the important commercial applications of homogeneous

catalysis are: hydroformylation of olefins to aldehydes/alcohols, carbonylation of methanol

to acetic acid, synthesis of L-dopa by asymmetric hydrogenation, oxidation of p-xylene to

terephthalic acid, hydrocyanation of butadiene to adiponitrile, ethylene oligomerization etc.

Though these homogeneous catalysts play an extremely important role in highly efficient

processes, yet there are some serious drawbacks, mainly in terms of catalyst-product

separation from the reaction mixture and the reusability of the catalyst.

The use of MOF as alternative catalysts can overcome these shortcoming. In this

approach, the active transitional metals are introduced into the MOF structure, playing the

role of connectors, being coordinated by imidazolate linkers. Thus a heterogeneous catalyst

which emulates the catalytic properties of homogeneous complex is obtained. Compared

with the counterpart homogeneous catalysts, these MOF heterogeneous catalysts provide

obvious advantages: the ease of recovery and the ability of being used repeatedly.

It should be kept in mind that, the MOF structure must be activated in an

appropriate way to make sure that the coordination spheres of the metal ion are not

46

completely blocked by the organic linkers or solvent molecules, so that they are accessible

for the reactant molecules [209]. The use metal complexes in MOF structures as active sites

for catalysis has been reported recently by the group of A. Corma [209, 210].

Figure 1.27. Detail of Pd-MOF, showing the 4-membered and the two 6-membered rings.

3D arrangement of the sodalite cages in sodalite-type frameworks.

They found that, [210] Pd-MOF is a very active catalyst for alcohol oxidation,

Suzuki C-C coupling, and olefin hydrogenation for which several homogeneous Pd

complexes are known to perform well. The Pd-MOF is composed of palladium and 2-

hydroxypyrimidinolate as organic linker, [Pd(2-pymo)2]·3H2O(2-pymo = 2-hydroxy-

pyrimidinolate). This material is topologically related to a 3D cubic sodalite-type

framework, with accessible Pd(II) ions in a square planar coordination [211] as illustrated

in Figure 1.27. Moreover, the regular pore system of the material (accessible through two

different hexagonal windows with free openings of 4.8 and 8.8 Å) conferred potential

shape-selective properties to the material that was demonstrated for the hydrogenation of

olefins with different steric hindrance. The structure of Pd-MOF was found to remain intact

under the reaction conditions tested, allowing recovery and reuse for successive catalytic

cycles.

47

The Corma‟s group also probed the catalytic activity of Co-MOF and Cu-MOF

[209]. Two metal–organic frameworks, [Cu(2-pymo)2] and [Co(PhIM)2] (2-pymo = 2-

hydroxypyrimidinolate; PhIM = phenylimidazolate) as shown in (Figure 1.28), containing

respectively Cu2+

and Co2+

ions and anionic diazaheterocyclic ligands (pyrimidinolate and

phenylimidazolate) as organic linkers, were successfully used for the aerobic oxidation of

tetralin, yielding α-tetralone (T=O) as the main product. Both materials are stable and

recyclable under the reaction conditions. Kinetic studies revealed significant differences

between the two MOFs, as a consequence of the different catalytic behavior of their central

metal ions. [Cu(2-pymo)2] is highly active for the activation of tetralin to produce

tetralinhydroperoxide (T–OOH), and less efficient in reacting the peroxide. Meanwhile, the

use of the cobalt catalyst involves a long induction period for the reaction. However, once

T–OOH is formed, Co2+

rapidly and efficiently transforms this into T=O, with high

tetralone-to-tetralol ratio (T=O/T–OH of ca. 7). As a result of this, a compromise between

the two effects has been reached with a mixture containing 10 wt% Cu-MOF and 90 wt%

Co-MOF, that yields the best performance in terms of activity, selectivity and low level of

hydroperoxides [209].

Figure 1.28. (a) Basic building block of Cu(2-pymo)2 and (b) Diagram of the asymmetric

unit of the Co(PhIM)2 framework

1.4.2.2 Ligands as catalytic sites

Just like their counterpart metal connectors, ligand linkers can play the role of

catalytic sites. The functionalization of the zeolitic imidazolate frameworks (ZIF) a special

(a) (b)

48

type of MOFs illustrates the importance of the organic linkers in catalysis. Basically, ZIFs

are composed of tetrahedrally-coordinated Zn ion connected by organic imidazole

derivative linkers.[204, 212] The interesting in the structure of ZIFs is that it is similar to

zeolitic structures: (i) Im bridges make the M–Im–M angle, close to 145°, which is

coincident with the Si–O–Si angle which is preferred and commonly found in many

zeolites; (ii) the metal ion should have a tetrahedral configuration so that the bond angle

Im-M-Im is of 109.5, equal to that of O-Si-O. Hence there is a perfect analogy between

zeolites and ZIFs, metal ion plays the role of T ion whereas Im is the O bridges. For each

imidazole used as a linker, beside two N atoms forming the bonds, three C-H bonds are left

available for tailoring with functional groups, hence providing the linkers with desired

catalytic properties.

Interestingly, the introduction of active sites into linkers can be done easily by

choosing the imidazole derivatives that already possess the functional groups. For example,

to increase the hydrophilicity of ZIFs, one may want to substitute the H atom at the C-H

bond with an aldehyde group. To achieve that, the use of imidazolecarboxylicaldehyde in

the synthesis of ZIFs is an obvious choice. Furthermore, there are many options for this

imidazolecarboxylicaldehyde: (i) the aldehyde can be attached at one of three positions in

the heterocylic of imidazole, each of these positions has different contribution to the

hydrophilicity of the resulting ZIF (ii) the number of attached aldehyde group can be as

many as three, as the number of aldehyde group attached increases, it is expected that the

hydrophilicity increases. Hence, in this case, the properties of the ZIF material have been

predetermined by picking up the right imidazole derivative, or its properties have been

designed.

Other small functional groups such as -Cl, -Br, -NO2 can be attached with the

linkers using this design method, provided that their attachments with imidazole ring do not

affect the formation of ZIF structures.

However, this method may become difficult to carry out when large functional

groups are desired. This is due to the fact that large functional groups may cause steric

hindrance and change the polarity of the bond N-H in the imidazole ring. Hence, the ZIF

49

structure could not be obtained. In addition, the functional groups may become vulnerable

to chemical conversion under the synthesis conditions.

This drawback can be overcome by using post synthesis treatment methods. The

principal of this method is to introduce functional groups bearing active components onto

the organic linkers in the already-formed ZIF structure. The organic linkers now become

anchoring sites to immobilize the desired functional group. The anchoring can occur via

covalent or noncovalent bonding. When the covalent bond obtained, it can be robust

enough to withstand the harsh conditions of catalytic reactions.

Figure 1.29. Transformation of ZIF-90 (A) by Reduction with NaBH4, and reaction with

ethanolamine to give ZIF-91 (B) and ZIF-92 (C) [213]

An interesting example which illustrates the capacity of immobilization of

functional groups onto ZIF structure has been recently reported [213]. The ZIF-90 structure

which is consisted of tetrahedral Zn(II) joined by imidazolate-2-carboxilyaldehye linkers

was employed to modify with different functional group (Figure 1.29). The anchoring site

of interest is the aldehyde group which is already available in the imidazole linkers. Two

kinds of reactions were applied for the functionalization, which in return, give two different

modified ZIFs from the same starting ZIF-90. When the modification involves a strong

50

reducing agent e.g. NaBH4 the aldehyde group is reduced to the corresponding alcohol

derivative, resulting in a new ZIF-91. If other reactant e.g. ethanolamine is used, the

aldehyde group converts to an imine, giving to a ZIF-92. Remarkably, XRD and BET

analysis of these two functionalized ZIF showed that, the zeolitic structure and the high

specific surface area of ZIF-90 are maintained during the transformation.

1.4.3.3. Support and encapsulator for catalysts

Thanks to their well-defined pore structures and high thermal stability, it would be

logical to use MOFs as catalyst supports. Given that zeolite supports have been widely

employed as an ideal medium for dispersion and stabilization of active metal and metal

oxide nanoparticles [214-216], one may want to try these nanoparticles on MOF supports.

Although having a uniform pore system structure similar to zeolites, the difference of

components in MOF structures (metal ions and organic linkers) would have different effect

on the supported nanoparticles. Because of that, new catalytic properties can be obtained.

An advantage of MOF structures is that their uniform pore systems can selectively

immobilize an individual molecule inside their cages, in this case the molecule is said to be

encapsulated. The encapsulation is possible thanks to the shape selectivity nature and the

confinement effect of zeolitic structure [208]. Hence, if a homogeneous catalyst molecule

can be encapsulated in a MOF structure, the potential applications would be overwhelming.

The encapsulation of a molecule can be carried out in a post synthesis treatment.

The molecule is inserted via diffusion provided that it is small enough to travel through the

pore system of the MOF.

Another approach is to introduce the desired guest molecule during the synthesis of

MOF. Under proper conditions, the obtained MOF can have the guest molecules

encapsulated inside its pore structure.

51

Figure 1.30. (a) Eight-coordinate molecular building block that could be represented as a

tetrahedral building unit, (b) [H2TMPyP]4+

porphyrin, (c) crystal structure of rho-ZMOF

(left), hydrogen atoms omitted for clarity, and schematic presentation of [H2TMPyP]4+

porphyrin ring enclosed in rho-ZMOF R-cage (right, drawn to scale) [217]

Using this approach, Alkordi et al[217] have successfully encapsulated

metalloporphyrins inside a MOF material (Figure 1.30). Metalloporphyrins are very active

homogeneous catalysts which have been used in many reactions such as hydroxylation

epoxidation of hydrocarbons. Many attempts to immobilize metalloporphyrins on

conventional host materials such as metal oxide, mesoporous silica and zeolites have been

made [218-225]. However the catalytic activity of the obtained systems is limited by low

loading, leaching problem as well as the dispersion of metalloporphyrins. To overcome

these limitations the authors propose the use of recently synthesized (indium-imidazoledi-

52

carboxylate)-based rho-MOF (topologically analogous to zeolite RHO) as the host matrix.

Indeed, the large voids inside its periodic R-cages and their anionic nature can afford the

ability for encapsulation of cationic porphyrins. In addition, the synthesis of this porphyrin-

encapsulated MOF is facile, porphyrin, 5,10,15,20-tetrakis(1-methyl-4-pyridinio)porphyrin

tetra(p-toluenesulfonate) ([H2TMPyP] [p-tosyl]4), was mixed together with starting

materials In(NO3)3.xH2O and 4,5-imidazoledicarboxylic acid in organic solvent to produce

the desired product. The obtained product showed high catalytic activity in the oxidation of

cyclohexane without any trace of leaching and outperformed the catalytic activity of rho-

MOF impregnated with Mn-metallated porphyrin. These facts clearly suggest the

outstanding property of the porphyrin-encapsulated MOF.

53

References

1 M. E. Davis, R. F. Lobo, Chem. Mater. 1992, 4, 756.

2 D. W. Breck, Zeolite molecular sieves: structure, chemistry, and use, Wiley,

London 1973.

3 C. Baerlocher, W. M. Meier, D. H. Olson, Atlas of Zeolite Framework Types,

Elsevier, New York 2001.

4 P. A. Jacobs, Carboniogenic Activity of Zeolites, Elsevier New York 1977.

5 Y. Ono, T. Baba, Catal. Today 1997, 38, 321.

6 M. Huang, A. Adnot, S. Kaliaguine, J. Catal. 1992, 137, 322.

7 A. Corma, J. Catal. 2003, 216, 298.

8 H. International, Zeolites: industry trends and worldwide markets in 2010, Wiley,

New York 2000.

9 L. Tosheva, V. P. Valtchev, Chem. Mater. 2005, 17, 2494.

10 K. Rajagopalan, A. W. Peters, G. C. Edwards, Applied Catalysis 1986, 23, 69.

11 T. Sano, H. Ikeya, T. Kasuno, Z. B. Wang, Y. Kawakami, K. Soga, Zeolites 1997,

19, 80.

12 M. A. Camblor, A. Corma, A. Martínez, F. A. Mocholí, J. P. Pariente, Applied

Catalysis 1989, 55, 65.

13 K. J. Klabunde, J. Stark, O. Koper, C. Mohs, D. G. Park, S. Decker, Y. Jiang, I.

Lagadic, D. Zhang, The Journal of Physical Chemistry 1996, 100, 12142.

14 M. E. Davis, Nature 2002, 417, 813.

15 P. Yang, T. Deng, D. Zhao, P. Feng, D. Pine, B. F. Chmelka, G. M. Whitesides, G.

D. Stucky, Science 1998, 282, 2244.

16 B. T. Holland, C. F. Blanford, A. Stein, Science 1998, 281, 538.

17 O. D. Velev, P. M. Tessier, A. M. Lenhoff, E. W. Kaler, Nature 1999, 401, 548.

18 C. S. Cundy, P. A. Cox, Microporous Mesoporous Mater. 2005, 82, 1.

19 C. S. Cundy, P. A. Cox, Chem. Rev. 2003, 103, 663.

20 R. A. van Santen, Nature 2006, 444, 46.

21 A. Aerts, C. E. A. Kirschhock, J. A. Martens, Chem. Soc. Rev. 2010, 39, 4626.

22 S. Bordiga, F. Bonino, K. P. Lillerud, C. Lamberti, Chem. Soc. Rev. 2010, 39, 4885.

23 J. W. Mullin, Crystallization, Butterworth-Heinemann, Oxford 1993.

24 E. M. Flanigen, D. W. Breck, presented at 137th Meeting of the ACS, Cleveland,

OH, 1960.

25 D. W. Breck, J. Chem. Ed. 1964, 41.

26 G. T. Kerr, J. Phys. Chem. 1966, 70.

27 G. T. Kerr, Zeolites 1989, 9.

28 P.-P. E. A. de Moor, T. P. M. Beelen, B. U. Komanschek, L. W. Beck, P. Wagner,

M. E. Davis, R. A. van Santen, Chemistry – A European Journal 1999, 5, 2083.

29 R. Ravishankar, C. Kirschhock, B. J. Schoeman, P. Vanoppen, P. J. Grobet, S.

Storck, W. F. Maier, J. A. Martens, F. C. De Schryver, P. A. Jacobs, The Journal of

Physical Chemistry B 1998, 102, 2633.

30 R. Ravishankar, C. E. A. Kirschhock, P.-P. Knops-Gerrits, E. J. P. Feijen, P. J.

Grobet, P. Vanoppen, F. C. De Schryver, G. Miehe, H. Fuess, B. J. Schoeman, P. A.

Jacobs, J. A. Martens, The Journal of Physical Chemistry B 1999, 103, 4960.

31 C. E. A. Kirschhock, R. Ravishankar, F. Verspeurt, P. J. Grobet, P. A. Jacobs, J. A.

Martens, The Journal of Physical Chemistry B 1999, 103, 4965.

54

32 C. E. A. Kirschhock, R. Ravishankar, L. V. Looveren, P. A. Jacobs, J. A. Martens,

The Journal of Physical Chemistry B 1999, 103, 4972.

33 C. E. A. Kirschhock, R. Ravishankar, P. A. Jacobs, J. A. Martens, The Journal of

Physical Chemistry B 1999, 103, 11021.

34 S. Mintova, N. H. Olson, T. Bein, Angew. Chem., Int. Ed. 1999, 38, 3201.

35 S. Mintova, N. H. Olson, V. Valtchev, T. Bein, Science 1999, 283, 958.

36 T. M. Davis, T. O. Drews, H. Ramanan, C. He, J. Dong, H. Schnablegger, M. A.

Katsoulakis, E. Kokkoli, A. V. McCormick, R. L. Penn, M. Tsapatsis, Nat Mater

2006, 5, 400.

37 A. Rivas-Cardona, M. Chovanetz, D. F. Shantz, Microporous Mesoporous Mater.

2012, 155, 56.

38 G. Xomeritakis, S. Nair, M. Tsapatsis, Microporous Mesoporous Mater. 2000, 38,

61.

39 T. O. Drews, M. Tsapatsis, Microporous Mesoporous Mater. 2007, 101, 97.

40 S. Bosnar, J. Bronic, T. Antonic Jelic, B. Subotic, CrystEngComm 2012, 14, 3069.

41 A. Palcic, J. Bronic, D. Brlek, B. Subotic, CrystEngComm 2011, 13, 1215.

42 J. Bronić, A. Mužic, T. Antonić Jelić, J. Kontrec, B. Subotić, J. Cryst. Growth 2008,

310, 4656.

43 B. J. Schoeman, J. Sterte, J. E. Otterstedt, Zeolites 1994, 14, 110.

44 J. P. Verduijn, WO9308125,, 1993.

45 M. Vilaseca, S. Mintova, K. Karaghiosoff, T. H. Metzger, T. Bein, Appl. Surf. Sci.

2004, 226, 1.

46 G. Zhu, S. Qiu, F. Gao, G. Wu, R. Wang, B. Li, Q. Fang, Y. Li, B. Gao, X. Xu, O.

Terasaki, Microporous Mesoporous Mater. 2001, 50, 129.

47 C. S. Cundy, J. O. Forrest, J. Microporous Mesoporous Mater. 2004, 72, 67.

48 H. Du, M. Fang, W. Xu, X. Meng, W. Pang, J. Mater. Chem. 1997, 7, 551.

49 S. Mintova, S. Mo, T. Bein, Chem. Mater. 1998, 10, 4030.

50 M. A. Camblor, A. Corma, A. Mifsud, J. Pérez-Pariente, S. Valencia, Stud. Surf.

Sci. Catal. 1997, 105, 341.

51 M. A. Camblor, A. Corma, S. Valencia, Microporous Mesoporous Mater. 1998, 25,

59.

52 B. J. Schoeman, E. Babouchkina, S. Mintova, V. P. Valtchev, J. Sterte, J. Porous

Mater. 2001, 8, 13.

53 S. Mintova, M. Reinelt, T. H. Metzger, J. Senker, T. Bein, Chem. Commun. 2003,

326.

54 B. Z. Zhan, M. A. White, M. Lumsden, J. Mueller-Neuhaus, K. N. Robertson, T. S.

Cameron, M. Gharghouri, Chem. Mater. 2002, 14, 3636.

55 S. Mintova, V. Valtchev, Stud. Surf. Sci. Catal. 1999, 125, 141.

56 V. P. Valtchev, K. N. Bozhilov, J. Phys. Chem. B 2004, 108, 15587.

57 B. A. Holmberg, H. Wang, J. M. Norbeck, Y. Yan, Microporous Mesoporous

Mater. 2003, 59, 13.

58 B. A. Holmberg, H. Wang, Y. Yan, Microporous Mesoporous Mater. 2004, 74, 189.

59 Q. Li, D. Creaser, J. Sterte, Chem. Mater. 2002, 14, 1319.

60 T. F. Chaves, H. O. Pastore, D. Cardoso, Microporous Mesoporous Mater. 2012,

161, 67.

61 J. Kecht, B. Mihailova, K. Karaghiosoff, S. Mintova, T. Bein, Langmuir 2004, 20,

5271.

55

62 G. Zhu, S. Qiu, J. Yu, Y. Sakamoto, F. Xiao, R. Xu, O. Terasaki, Chem. Mater.

1998, 10, 1483.

63 Q. Li, D. Creaser, J. Sterte, Stud. Surf. Sci. Catal. 2001, 135, 02.

64 J. Hedlund, B. Schoeman, J. Sterte, Chem. Commun. 1997, 1193.

65 M. Tsapatsis, M. Lovallo, T. Okubo, M. E. Davis, M. Sadakata, Chem. Mater. 1995,

7, 1734.

66 M. H. Anthonis, M. Mertens, J. P. Verduijn, WO9703021, 1997.

67 O. Larlus, V. P. Valtchev, Chem. Mater. 2004, 16, 3381.

68 S. Mintova, N. Petkov, K. Karaghiosoff, T. Bein, Microporous Mesoporous Mater.

2001, 50, 121.

69 J. P. Dong, J. Zou, Y. C. Long, Microporous Mesoporous Mater. 2003, 57, 9.

70 A. E. Persson, B. J. Schoeman, J. Sterte, J. E. Otterstedt, Zeolites 1995, 15, 611.

71 R. Van Grieken, J. L. Sotelo, J. M. Menéndez, J. A. Melero, Microporous

Mesoporous Mater. 2000, 39, 135.

72 W. Song, R. E. Justice, C. A. Jones, V. H. Grassian, S. C. Larsen, Langmuir 2004,

20, 8301.

73 J. Aguado, D. P. Serrano, J. M. Escola, J. M. Rodríguez, Microporous Mesoporous

Mater. 2004, 75, 41.

74 A. E. Persson, B. J. Schoeman, J. Sterte, J. E. Otterstedt, Zeolites 1994, 14, 557.

75 B. J. Schoeman, J. Sterte, KONA 1997, 15, 150.

76 C. S. Tsay, A. S. T. Chiang, Microporous Mesoporous Mater. 1998, 26, 89.

77 Q. Li, D. Creaser, J. Sterte, Microporous Mesoporous Mater. 1999, 31, 141.

78 S. Yang, A. Navrotsky, D. J. Wesolowski, J. A. Pople, Chem. Mater. 2004, 16, 210.

79 W. Song, R. E. Justice, C. A. Jones, V. H. Grassian, S. C. Larsen, Langmuir 2004,

20, 4696.

80 V. P. Valtchev, A. C. Faust, J. Lézervant, Microporous Mesoporous Mater. 2004,

68, 91.

81 Q. Li, B. Mihailova, D. Creaser, J. Sterte, Microporous Mesoporous Mater. 2000,

40, 53.

82 Q. Li, B. Mihailova, D. Creaser, J. Sterte, Microporous Mesoporous Mater. 2001,

43, 51.

83 G. Zhang, J. Sterte, B. Schoeman, J. Chem. Mater. 1997, 9, 210.

84 C. S. Cundy, J. O. Forrest, R. Plaisted, J. Microporous Mesoporous Mater. 2003,

66, 143.

85 B. Tokay, M. Somer, A. Erdem-Şenatalar, F. Schüth, R. W. Thompson,

Microporous Mesoporous Mater. 2009, 118, 143.

86 B. Tokay, O. Karvan, A. Erdem-Şenatalar, Microporous Mesoporous Mater. 2010,

131, 230.

87 B. Tokay, A. Erdem-Şenatalar, Microporous Mesoporous Mater. 2012, 148, 43.

88 S. Kumar, R. L. Penn, M. Tsapatsis, Microporous Mesoporous Mater. 2011, 144,

74.

89 S. A. Pelster, R. Kalamajka, W. Schrader, F. Schüth, Angew. Chem. 2007, 119,

2349.

90 A. Aerts, M. Haouas, T. P. Caremans, L. R. A. Follens, T. S. van Erp, F. Taulelle, J.

Vermant, J. A. Martens, C. E. A. Kirschhock, Chemistry – A European Journal

2010, 16, 2764.

56

91 B. O. Hincapie, L. J. Garces, Q. Zhang, A. Sacco, S. L. Suib, Microporous

Mesoporous Mater. 2004, 67, 19.

92 J. P. Verduijn, WO9703020, 1997.

93 J. Hedlund, E. Kurpan, Stud. Surf. Sci. Catal. 2001, 135, 224.

94 R. K. Iler, The Chemistry of Silica, Wiley, New York 1979.

95 R. Ravishankar, C. Kirschhock, B. J. Schoeman, P. Vanoppen, P. J. Grober, S.

Storck, W. F. Maier, J. A. Martens, F. C. De Schryver, P. A. A. Jacobs, J. Phys.

Chem. B 1998, 102, 2633.

96 H. Hosokawa, K. Oki, Chem. Lett. 2003, 32, 586.

97 S. P. Naik, J. C. Chen, A. S. T. Chiang, Microporous Mesoporous Mater. 2002, 54,

293.

98 D. P. Serrano, J. Aguado, J. M. Escola, J. M. Rodríguez, Á. Peral, Chem. Mater.

2006, 18, 2462.

99 I. Schmidt, A. Krogh, K. Wienberg, A. Carlsson, M. Brorson, C. J. H. Jacobsen,

Chem. Commun. 2000, 2157.

100 C. Madsen, C. J. H. Jacobsen, Chem. Commun. 1999, 673.

101 I. Schmidt, A. Boisen, E. Gustavsson, K. Ståhl, S. Pehrson, S. Dahl, A. Carlsson, C.

J. H. Jacobsen, Chem. Mater. 2001, 13, 4416.

102 I. Schmidt, C. Madsen, C. J. H. Jacobsen, Inorg. Chem. 2000, 39, 2279.

103 C. J. H. Jacobsen, C. Madsen, J. Houzvicka, I. Schmidt, A. Carlsson, J. Am. Chem.

Soc. 2000, 122, 7116.

104 H. Wang, B. A. Holmberg, Y. Yan, J. Am. Chem. Soc. 2003, 125, 9928.

105 L. Li, P. M. Thangamathesvaran, C. Y. Yue, K. C. Tam, X. Hu, Y. C. Lam,

Langmuir 2001, 17, 8062.

106 D. Klemm, B. Philipp, T. Heinze, U. Heinze, W. Wagenknecht, Comprehensive

Cellulose Chemistry, Vol. 1, Wiley-VCH, New York 2004.

107 M. A. López-Quintela, Current Opinion in Colloid & Interface Science 2003,

8, 137.

108 B. L. Cushing, V. L. Kolesnichenko, C. J. O'Connor, Chem. Rev. 2004, 104, 3893.

109 P. K. Dutta, D. Robins, Langmuir 1991, 7, 1048.

110 P. K. Dutta, M. Jakupca, K. S. N. Reddy, L. Salvati, Nature 1995, 374, 44.

111 A. Manna, B. D. Kulkarni, R. K. Ahedi, A. Bhaumik, A. N. Kotasthane, J. Colloid

Interface Sci. 1999, 213, 405.

112 C. S. Carr, S. Kaskel, D. F. Shantz, Chem. Mater. 2004, 16, 3139.

113 S. Lee, D. F. Shantz, Chem. Commun. 2004, 680.

114 C. S. Carr, D. F. Shantz, Microporous Mesoporous Mater. 2005, 85, 284.

115 Z. Chen, S. Li, Yan, Chem. Mater. 2005, 17, 2262.

116 J. M. Thomas, W. J. Thomas, J. Anderson, M. Boudart, Principles and practice of

heterogeneous catalysis, Vol. 638, VCH New York, 1997.

117 D. T. On, S. Kaliaguine, Angew. Chem., Int. Ed. 2001, 40, 3248.


119 M. Choi, H. S. Cho, R. Srivastava, C. Venkatesan, D.-H. Choi, R. Ryoo, Nat Mater

2006, 5, 718.

120 M. Choi, R. Srivastava, R. Ryoo, Chem. Commun. 2006, 4380.

121 K. Suzuki, Y. Aoyagi, N. Katada, M. Choi, R. Ryoo, M. Niwa, Catal. Today 2008,

132, 38.

122 R. Srivastava, M. Choi, R. Ryoo, Chem. Commun. 2006, 4489.

57

123 D.-H. Lee, M. Choi, B.-W. Yu, R. Ryoo, Chem. Commun. 2009, 74.

124 M.-J. Jin, A. Taher, H.-J. Kang, M. Choi, R. Ryoo, Green Chemistry 2009, 11, 309.

125 M. Choi, K. Na, J. Kim, Y. Sakamoto, O. Terasaki, R. Ryoo, Nature 2009, 461,

246.

126 A. Corma, Nature 2009, 461, 182.

127 A. K. Cheetham, G. Férey, T. Loiseau, Angew. Chem. Int. Ed. 1999, 38, 3268.

128 Y. Ma, W. Tong, H. Zhou, S. L. Suib, Microporous Mesoporous Mater. 2000, 37,

243.

129 J. R. Long, O. M. Yaghi, Chem. Soc. Rev. 2009, 38, 1213.

130 J. L. C. Rowsell, O. M. Yaghi, Microporous Mesoporous Mater. 2004, 73, 3.

131 S. R. Batten, S. M. Neville, D. R. Turner, Coordination polymers : design, analysis

and application, Royal Society of Chemistry, 2009.

132 M. O'Keeffe, Chem. Soc. Rev. 2009, 38, 1215.

133 S. R. Batten, N. R. Champness, X.-M. Chen, J. Garcia-Martinez, S. Kitagawa, L.

Ohrstrom, M. O'Keeffe, M. P. Suh, J. Reedijk, CrystEngComm 2012, 14, 3001.

134 H. J. Buser, D. Schwarzenbach, W. Petter, A. Ludi, Inorg. Chem. 1977, 16, 2704.

135 J. E. Bauman, J. C. Wang, Inorg. Chem. 1964, 3, 368.

136 C. T. Dziobkowski, J. T. Wrobleski, D. B. Brown, Inorg. Chem. 1981, 20, 671.

137 D. M. L. Goodgame, D. J. Williams, R. E. P. Winpenny, Angewandte Chemie

International Edition in English 1987, 26, 1044.

138 B. F. Hoskins, R. Robson, J. Am. Chem. Soc. 1989, 111, 5962.

139 B. F. Hoskins, R. Robson, J. Am. Chem. Soc. 1990, 112, 1546.

140 S. R. Batten, R. Robson, Angew. Chem. Int. Ed. 1998, 37, 1460.

141 R. Robson, J. Chem. Soc., Dalton Trans. 2000, 3735.

142 A. F. Wells, A. Wells, Three-dimensional nets and polyhedra, Wiley New York,

1977.

143 S. Kitagawa, S. Matsuyama, M. Munakata, T. Emori, J. Chem. Soc., Dalton Trans.

1991, 2869.

144 M. Fujita, Y. J. Kwon, S. Washizu, K. Ogura, J. Am. Chem. Soc. 1994, 116, 1151.

145 L. R. MacGillivray, S. Subramanian, M. J. Zaworotko, J. Chem. Soc., Chem.

Commun. 1994, 1325.

146 L. Carlucci, G. Ciani, D. M. Proserpio, A. Sironi, J. Chem. Soc., Chem. Commun.

1994, 2755.

147 O. M. Yaghi, G. Li, Angewandte Chemie International Edition in English 1995, 34,

207.

148 O. M. Yaghi, G. Li, H. Li, Nature 1995, 378, 703.

149 O. M. Yaghi, H. Li, C. Davis, D. Richardson, T. L. Groy, Acc. Chem. Res. 1998, 31,

474.

150 S. S.-Y. Chui, S. M.-F. Lo, J. P. H. Charmant, A. G. Orpen, I. D. Williams, Science

1999, 283, 1148.

151 H. Li, M. Eddaoudi, M. O'Keeffe, O. M. Yaghi, Nature 1999, 402, 276.

152 H. Li, M. Eddaoudi, T. L. Groy, O. Yaghi, J. Am. Chem. Soc. 1998, 120, 8571.

153 S. S. Kaye, A. Dailly, O. M. Yaghi, J. R. Long, J. Am. Chem. Soc. 2007, 129,

14176.

154 F. A. Cotton, G. Wilkinson, C. A. Murillo, M. Bochmann, Advanced inorganic

chemistry, Vol. 5, Wiley New York, 1988.

155 L. Ma, C. Abney, W. Lin, Chem. Soc. Rev. 2009, 38, 1248.

58

156 A. Corma, H. Garc a, F. X. Llabrés i Xamena, Chem. Rev. 2010, 110, 4606.

157 M. Yoon, R. Srirambalaji, K. Kim, Chem. Rev. 2012, 112, 1196.

158 J. Lee, O. K. Farha, J. Roberts, K. A. Scheidt, S. T. Nguyen, J. T. Hupp, Chem. Soc.

Rev. 2009, 38, 1450.

159 J. Graetz, Chem. Soc. Rev. 2009, 38, 73.

160 S. S. Han, J. L. Mendoza-Cortes, W. A. G. Iii, Chem. Soc. Rev. 2009, 38, 1460.

161 J.-R. Li, R. J. Kuppler, H.-C. Zhou, Chem. Soc. Rev. 2009, 38, 1477.

162 L. J. Murray, M. Dinca, J. R. Long, Chem. Soc. Rev. 2009, 38, 1294.

163 M. P. Suh, H. J. Park, T. K. Prasad, D.-W. Lim, Chem. Rev. 2012, 112, 782.

164 K. Sumida, D. L. Rogow, J. A. Mason, T. M. McDonald, E. D. Bloch, Z. R. Herm,

T.-H. Bae, J. R. Long, Chem. Rev. 2012, 112, 724.

165 H. Wu, Q. Gong, D. H. Olson, J. Li, Chem. Rev. 2012, 112, 836.

166 J.-R. Li, J. Sculley, H.-C. Zhou, Chem. Rev. 2012, 112, 869.

167 D. Zacher, O. Shekhah, C. Woll, R. A. Fischer, Chem. Soc. Rev. 2009, 38, 1418.

168 O. Shekhah, J. Liu, R. A. Fischer, C. Woll, Chem. Soc. Rev. 2011, 40, 1081.

169 C. Wang, T. Zhang, W. Lin, Chem. Rev. 2012, 112, 1084.

170 P. Horcajada, R. Gref, T. Baati, P. K. Allan, G. Maurin, P. Couvreur, G. Férey, R.

E. Morris, C. Serre, Chem. Rev. 2012, 112, 1232.

171 W. Zhang, R.-G. Xiong, Chem. Rev. 2012, 112, 1163.

172 Y. Cui, Y. Yue, G. Qian, B. Chen, Chem. Rev. 2012, 112, 1126.

173 A. U. Czaja, N. Trukhan, U. Muller, Chem. Soc. Rev. 2009, 38, 1284.

174 New J. Chem. 2010, 34, 2355.

175 O. K. Farha, I. Eryazici, N. C. Jeong, B. G. Hauser, C. E. Wilmer, A. A. Sarjeant, R.

Q. Snurr, S. T. Nguyen, A. Ö. Yazaydın, J. T. Hupp, J. Am. Chem. Soc. 2012, 134,

15016.

176 M. O‟Keeffe, M. A. Peskov, S. J. Ramsden, O. M. Yaghi, Acc. Chem. Res. 2008,

41, 1782.

177 M. Eddaoudi, D. B. Moler, H. Li, B. Chen, T. M. Reineke, M. O'Keeffe, O. M.

Yaghi, Acc. Chem. Res. 2001, 34, 319.

178 O. M. Yaghi, M. O'Keeffe, N. W. Ockwig, H. K. Chae, M. Eddaoudi, J. Kim,

Nature 2003, 423, 705.

179 K. Koh, A. G. Wong-Foy, A. J. Matzger, J. Am. Chem. Soc. 2009, 131, 4184.

180 G. Férey, C. Mellot-Draznieks, C. Serre, F. Millange, J. Dutour, S. Surblé, I.

Margiolaki, Science 2005, 309, 2040.

181 S. Surble, C. Serre, C. Mellot-Draznieks, F. Millange, G. Ferey, Chem. Commun.

2006, 284.

182 S. Bauer, C. Serre, T. Devic, P. Horcajada, J. r. m. Marrot, G. r. Férey, N. Stock,

Inorg. Chem. 2008, 47, 7568.

183 T. Chalati, P. Horcajada, R. Gref, P. Couvreur, C. Serre, J. Mater. Chem. 2011, 21,

2220.

184 P. F. G. S. C. G. R. C. P. Horcajada-cortes, US Patent 20100226991, 2010.

185 S. Hausdorf, F. Baitalow, J. Seidel, F. O. R. L. Mertens, The Journal of Physical

Chemistry A 2007, 111, 4259.

186 S. Hausdorf, J. r. Wagler, R. Moβig, F. O. R. L. Mertens, The Journal of Physical

Chemistry A 2008, 112, 7567.

59

187 P. Horcajada, F. Salles, S. Wuttke, T. Devic, D. Heurtaux, G. Maurin, A. Vimont,

M. Daturi, O. David, E. Magnier, N. Stock, Y. Filinchuk, D. Popov, C. Riekel, G.

Férey, C. Serre, J. Am. Chem. Soc. 2011, 133, 17839.

188 N. A. Khan, S. H. Jhung, Crystal Growth & Design 2010, 10, 1860.

189 H. Furukawa, N. Ko, Y. B. Go, N. Aratani, S. B. Choi, E. Choi, A. O. Yazaydin, R.

Q. Snurr, M. O'Keeffe, J. Kim, O. M. Yaghi, Science 2010, 329, 424.

190 Dinca, Mircea, J. R. Long, Angew. Chem. Int. Ed. 2008, 47, 6766.

191 S. Satyapal, J. Petrovic, C. Read, G. Thomas, G. Ordaz, Catal. Today 2007, 120,

246.

192 J. L. C. Rowsell, E. C. Spencer, J. Eckert, J. A. K. Howard, O. M. Yaghi, Science

2005, 309, 1350.

193 O. K. Farha, A. Özgür Yazaydın, I. Eryazici, C. D. Malliakas, B. G. Hauser, M. G.

Kanatzidis, S. T. Nguyen, R. Q. Snurr, J. T. Hupp, Nat Chem 2010, 2, 944.

194 A. G. Wong-Foy, A. J. Matzger, O. M. Yaghi, J. Am. Chem. Soc. 2006, 128, 3494.

195 B. Panella, M. Hirscher, H. Pütter, U. Müller, Adv. Funct. Mater. 2006, 16, 520.

196 Y. E. Cheon, M. P. Suh, Chem. Commun. 2009, 2296.

197 S. K. Bhatia, A. L. Myers, Langmuir 2006, 22, 1688.

198 S. S. Han, W. A. Goddard, J. Am. Chem. Soc. 2007, 129, 8422.

199 S. S. Han, W. A. Goddard, The Journal of Physical Chemistry C 2008, 112, 13431.

200 A. R. Millward, O. M. Yaghi, J. Am. Chem. Soc. 2005, 127, 17998.

201 J. D. Figueroa, T. Fout, S. Plasynski, H. McIlvried, R. D. Srivastava, Int. J.

Greenhouse Gas Ctr. 2008, 2, 9.

202 S. Cavenati, C. A. Grande, A. E. Rodrigues, J. Chem. Eng. Data 2004, 49, 1095.

203 S. Himeno, T. Komatsu, S. Fujita, J. Chem. Eng. Data 2005, 50, 369.

204 R. Banerjee, A. Phan, B. Wang, C. Knobler, H. Furukawa, M. O'Keeffe, O. M.

Yaghi, Science 2008, 319, 939.

205 J. An, N. L. Rosi, J. Am. Chem. Soc. 2010, 132, 5578.

206 S. R. Caskey, A. G. Wong-Foy, A. J. Matzger, J. Am. Chem. Soc. 2008, 130, 10870.

207 A. Demessence, D. M. D‟Alessandro, M. L. Foo, J. R. Long, J. Am. Chem. Soc.

2009, 131, 8784.

208 A. Corma, Chem. Rev. 1995, 95, 559.

209 F. X. Llabres i Xamena, O. Casanova, R. Galiasso Tailleur, H. Garcia, A. Corma, J.

Catal. 2008, 255, 220.

210 F. X. Llabres i Xamena, A. Abad, A. Corma, H. Garcia, J. Catal. 2007, 250, 294.

211 J. A. R. Navarro, E. Barea, J. M. Salas, N. Masciocchi, S. Galli, A. Sironi, C. O.

Ania, J. B. Parra, Inorg. Chem. 2006, 45, 2397.

212 A. Phan, C. J. Doonan, F. J. Uribe-Romo, C. B. Knobler, M. O‟Keeffe, O. M.

Yaghi, Acc. Chem. Res. 2010, 43, 58.

213 W. Morris, C. J. Doonan, H. Furukawa, R. Banerjee, O. M. Yaghi, J. Am. Chem.

Soc. 2008, 130, 12626.

214 Y. Xu, L. Lin, Applied Catalysis A: General 1999, 188, 53.

215 E. P. Reddy, L. Davydov, P. Smirniotis, Applied Catalysis B: Environmental 2003,

42, 1.

216 V. Meille, Applied Catalysis A: General 2006, 315, 1.

217 M. H. Alkordi, Y. Liu, R. W. Larsen, J. F. Eubank, M. Eddaoudi, J. Am. Chem. Soc.

2008, 130, 12639.

60

218 W. S. Cardoso, M. S. P. Francisco, R. Landers, Y. Gushikem, Electrochimica Acta

2005, 50, 4378.

219 T. A. Khan, J. A. Hriljac, Inorganica Chimica Acta 1999, 294, 179.

220 Z. Li, C.-G. Xia, X.-M. Zhang, Journal of Molecular Catalysis A: Chemical 2002,

185, 47.

221 H. Nur, H. Hamid, S. Endud, H. Hamdan, Z. Ramli, Materials Chemistry and

Physics 2006, 96, 337.

222 I. L. V. Rosa, C. M. C. P. Manso, O. A. Serra, Y. Iamamoto, Journal of Molecular

Catalysis A: Chemical 2000, 160, 199.

223 F. C. Skrobot, I. L. V. Rosa, A. P. A. Marques, P. R. Martins, J. Rocha, A. A.

Valente, Y. Iamamoto, Journal of Molecular Catalysis A: Chemical 2005, 237, 86.

224 F. S. Vinhado, C. M. C. Prado-Manso, H. C. Sacco, Y. Iamamoto, Journal of

Molecular Catalysis A: Chemical 2001, 174, 279.

225 W. Xu, H. Guo, D. L. Akins, J. Phys. Chem. B 2001, 105, 1543.

61

Chapter 2. Experimental

2.1. Synthesis

A number of methods were studied in this thesis. The preparation of clear gel

solution was the first step of all these methods. The obtained solutions were then processed

in different procedures depending on the method applied. The use of the same starting

solution among these methods can provide the ease of comparison between them (Figure

2.1).

Figure 2.1. Methods studied for the synthesis of nanozeolites

2.1.1. Preparation of clear zeolite gel solution

Silicalite-1

In a typical recipe according to Trong-On Do et al,[1] 14 g of tetrapropylammonium

hydroxide (TPAOH) 20% in water was added to 7.8 g of tetraorthosilicate Si(OC2H5)4. The

mixture was stirred vigorously for 24 h at room temperature. The molar gel composition

was: 2.68 SiO2 : 1 TPAOH : 168 H2O.

FAU zeolites

Clear gel solution

Synthesis in

aqueous media Single phase

synthesis

Two phase

synthesis

62

The preparation of clear synthesis solution for FAU nanozeolites followed Mintova

et al.[2] First a 38.4 g solution of NaOH 0.05 N (Fischer) was diluted with 121.6 g H2O.

Then 52.3 g tetramethylammonium hydroxide solution (Aldrich, 25% in water) and

aluminium isopropoxide, Al(iPr)3, (Aldrich, 98%) were added in that order, and stirred

vigorously until the solution became clear. To this solution 21.66 g of tetraorthosilicate was

added. This clear solution was aged for 3 days under vigorous stirring at room temperature.

The final molar composition was: 2.46 (TMA)2O : 0.032 Na2O : 1 Al2O3 : 3.4 SiO2 : 370

H2O.

2.1.2. Synthesis of nanozeolites using clear gel solution in aqueous

medium (conventional method)

In this method, a Teflon-lined stainless steel autoclave (Figure 2.2) was employed

to perform the crystallization under hydrothermal conditions. Before use, the Teflon beaker

was cleaned by immersing in hydrofluoric acid (HF, 20%) for 24 h, and then washed with

distilled water. This is to remove the crystalline residue on the surface of the Teflon beaker

from previous synthesis. After being filled with the prepared clear solution, the autoclave

was completely sealed and heated in a convection oven at desired temperature.

For the synthesis of silicalite-1, the temperature was 100oC and the crystallization

time was 24 h, whereas, for that of FAU zeolites, the temperature and crystallization time

were 100oC and 6 days, respectively.

The solid product was recovered by centrifugation at the speed of 20.000 rpm for 1

h, washed several times with distilled water, dried over night at 80 oC, and calcined in air at

550 oC for 8 h.

In this thesis the zeolites produced using this method were referred to as reference

samples or nanozeolites synthesized using conventional method.

63

Figure 2.2. Scheme of the autoclave: (1) a cylindrical stainless steel vessel, (2) a Teflon

cylindrical beaker, (3) a flat Teflon cover for closing the Teflon beaker, (4) a flat stainless

steel cover which was tightened up to part (1) by six screws.[3]

2.1.3. Synthesis of nanozeolites in organic medium

2.1.3.1. Synthesis of nanozeolites using single-phase method

The procedure is consisted of the following steps:

(i) Preparing a clear zeolite gel solution. This clear gel was heated at 80ºC for 12 hours

to produce zeolite seeds.

(ii) To this clear zeolite gel solution, an organic solvent containing organosilanes was

added and stirred for several hours at 40-100ºC.

64

(iii) The organic phase containing organosilane-functionalized zeolite seeds was

extracted. This organic phase was transferred into an autoclave and then heated at desired

temperature for crystallization.

(iv) After the crystallization, the resulting nanozeolite product was recovered by

centrifugation and then washed with ethanol and water for several times. The product was

then dried at 80ºC for 24 hours and calcined at 550ºC for 5 hours.

Typically, 10 g of zeolite gel solution was added to 500 ml of a solution of toluene

containing n-butanol (30% wt) and a certain amount of hexadecyltrimethoxysilane. The

organosilane was in a proportion of less than 10% mol in regards to the silica content in the

gel. After vigorous stirring at 60 oC for 48 h a mixture clear to naked eyes was obtained.

This mixture was then transferred to an autoclave for further hydrothermal treatment at

150oC for 3 days and 170

oC for 1 day for faujasite and silicalite-1, respectively.

2.1.3.2. Synthesis of nanozeolites using two-phase method

The procedure is consisted of the following steps

i) Preparing a clear zeolite gel solution. This clear solution was heated at 80 ºC for

12 hours to produce zeolite seeds.

ii) To this clear solution, an organic solvent (such as n-octane, toluene, etc.)

containing oganosilanes (for example, hexadecyltrimethoxysilane) in an organic solvent

was added. The organosilane was in proportion of less than 10% mol with respect to the

silica content in the gel. Since the solvent is insoluble in water, a two-phase system was

obtained and was stirred for 12 hours at 80 ºC.

iii) The two-phase mixture was transferred into an autoclave and then heated at

elevated temperature (100-200 ºC) for crystallization.

iv) After the crystallization, the organic phase containing organosilane-

functionalized nanozeolite was extracted. In the organic phase, the resulting nanozeolite

product was recovered by centrifugation and then washed with ethanol and water for

several times. The product was then dried at 80 ºC for 24 hours and calcined at 550 ºC for 5

hours.

65

For the synthesis of nanosilicalite-1, typically, 21.8 g of clear zeolite gel solution

was heated at 80oC for 12 h, the obtained solution was added with 93 g of a solution of

toluene containing 1.22% wt hexadecyltrimethoxysilane, resulting in a two-phase mixture.

After stirring for 12 hours at 60oC, the organic phase mixture was extracted and transferred

into an autoclave; and heated at 170oC for 24 hours.

2.1.4. Preparation of silica containing nanozeolites

Two types of silica containing zeolites which had the zeolite content of 20 %wt and

50% wt respectively were prepared. These two samples were referred to as nano-faujasite-

20 and nano-faujasite-50, respectively. Typically, for the synthesis of silica containing

20%wt zeolite, 30 g of distilled water was added with 100 g of ethanol (Aldrich 98%)

under stirring. Next, 14.16 g of TEOS (Aldrich, 98%) was added and stirred until the

solution became clear. Subsequently, 1 g of FAU nanozeolite, which was prepared using

conventional method, was added. The obtained mixture was then heated at 80 oC with

reflux under vigorous stirring overnight. The product was separated by filtration and

washed with distilled water for several times, dried overnight at 120 oC, then calcined at

550 oC for 8 h.

2.1.5. Synthesis of MIL-88B metal-organic framework

In a typical synthesis, 0.67 mmol of FeCl3.6H2O 99%, 0.33 mmol of corresponding

Ni(NO3)2.6H2O 97% and 1 mmol of bdc 98% were dissolved in 10 ml of DMF. To this

clear solution, 0.4 mmol of NaOH was added under stirring for 15 min. The mixture was

then transferred into a Teflon-lined autoclave and heated at 100 oC for 15 h. Solid product

was then recovered by filtration and washed several times with DMF. The sample was

treated with water, pyridine (Py), pyrazine (Pz) and 4-,4‟-bipyridine (Bp) to obtain Fe2Ni-

MIL-88B.H2O, Fe2Ni-MIL-88B.Py, Fe2Ni-MIL-88B.Pz and Fe2Ni-MIL-88B.Bp,

respectively (see Supporting Information). For comparison, the single metal Fe3-MIL-

88B.DMF was also prepared using the procedure of Férey et al

For the mechanism investigation of MIL-88B the syntheses have the same molar

ratios with respect to the metal cluster (Fe3O or Fe2NiO). In addition, the molar ratio of bdc

(benezendicarboxylic) over metal cluster is kept stoichiometric (and is of 3). Single metal

66

synthesis: vials of 10 ml DMF solution of 10 mmol of FeCl3.6H2O 99% or Fe(NO3)3.9H2O

98% were added with 10 mmol bdc under stirring. Next 4 ml of NaOH 2M was rapidly

injected with continuous stirring. The vials were then capped and heated at 100 oC for

different times: 0 h (5 min after the addition of NaOH at room temperature), 1h, 2h, 3h, and

12 h. Mixed metal synthesis: vials of 10 ml DMF solution of 3,33 mmol of Ni(NO3)2.6H2O

and 6.67 mmol of FeCl3.6H2O 99% or Fe(NO3)3.9H2O 98% were added with 10 mmol bdc

under stirring. Next 4 ml of NaOH 2M was rapidly injected with continuous stirring. The

vials were then capped and heated at 100 oC for different times: 0 h (5 min after the

addition of NaOH at room temperature), 1h, 2h, 3h, and 12 h. Solids products were

recovered by centrifugation at 5000 rpm for 5 min. The solids were then dried in vacuum

for 24 h at 50 oC. In general, the samples prepared with FeCl3.6H2O yield firm solids.

However, those prepared with Fe(NO3)3.9H2O become thick gel during the heat treatment

and thus, their corresponding solid products are not as firm as the solids from the Cl- based

samples.

2.2. Characterization

2.2.1. FTIR Spectroscopy

Fourier transform infrared spectroscopy (FTIR) is the subset of spectroscopy that

deals with the infrared part of the electromagnetic spectrum; it can be used to identify a

compound and to investigate the composition of a sample. Typically, when a molecule is

exposed to infra-red (IR) radiation, it absorbs specific frequencies of radiation. The

frequencies which are absorbed are dependent upon the functional groups within the

molecule and the symmetry of the molecule. IR radiation can only be absorbed by bonds

within a molecule, if the radiation has exactly the right energy to induce a vibration of the

bond. This is the reason why only specific frequencies are absorbed.

Infrared spectroscopy focuses on the frequency range 400 - 4000 cm-1

, where cm-1

is known as wavenumber (1/wavelength), which is a unit of measure for the frequency. To

generate the infrared spectrum, radiation containing all frequencies in the IR region is

passed through the sample. Those frequencies which are absorbed appear as a decrease in

the detected signal. This information is displayed as a spectrum of percentage transmitted

67

radiation plotted against wavenumber. When determining structure types of zeolites, the

region of 400-1400 cm-1

is interesting. This region contains the framework vibrations of

zeolite structure, and stretching and bending modes of the silica-alumina TO4 tetrahedra.

The framework vibrations of zeolite-type materials can be divided into structure insensitive

and structure sensitive vibrations, as shown in Table 2.1. For the MIL-88B, the region of

400 – 800 cm-1

is important since it includes the vibrations of the metal clusters (Table 2.2).

Table 2.1. Structure insensitive and sensitive framework vibrations of zeolites [4]

Structure insensitive vibrations Wavenumber (cm-1

)

Asymmetric stretching vibrations 1200 – 1000

Symmetric stretching vibrations 850 – 700

Bending vibrations 600 – 400

Structure sensitive vibrations Wavenumber (cm-1

)

Asymmetric stretching vibrations 1050 – 1150

Symmetric stretching vibrations 750 – 820

Double ring vibrations 500 – 650

Pore opening vibrations 300 – 420


Band (cm-1

) Assignment

750 C-H[5, 6]

720 Fe2NiO[7, 8]

690 C-C[5, 6]

660 OCO[5, 6]

624 Fe3O[9, 10]

550 Fe-O, Ni-O[11]

To measure a sample, 1 mg of the sample was well-mixed with 99 mg of KBr

powder (1 % sample in KBr) by grinding using an agate mortar and pestle. The obtained

68

powder was then crushed in a mechanical die press to form a translucent wafer. Finally, the

wafer was placed in a FTIR spectrometer for measurement. A pure KBr wafer was also

made for the background corrections.

2.2.2. Raman spectroscopy

The Raman scattering technique is a vibrational molecular spectroscopy which

is derived from an inelastic light scattering process. With Raman spectroscopy, a laser

photon is scattered by a sample molecule and loses (or gains) energy during the process.

The amount of energy lost is seen as a change in energy (wavelength) of the

irradiating photon. This energy loss is characteristic for a particular bond in the molecule.

Raman can provide a precise spectral fingerprint, unique to a molecule or an individual

molecular structure. In this respect it is similar to the more commonly found FT-IR

spectroscopy. Raman analysis was carried out with a Horiba U100 Raman spectrometer

using excitation wavelength of 514 nm.

2.2.3. UV-Vis spectroscopy

While IR and Raman spectroscopy techniques allow studying the vibrations of the

atoms and the molecules, the UV-Vis (ultraviolet and visible, with the wavelength ranging

from 200 to 800 nm) spectroscopy provides information about the electron transition inside

them. When continuous radiation strike a material, a portion of the radiation may be

absorbed. If that occurs, the residual or reluctant radiation, when it is passed through a

prism, yields a spectrum with gaps in it, called an absorption spectrum. As a result of

energy absorption, atoms or molecules pass from a state of low energy (the initial, or

ground state) to a state of higher energy (the excited state). Figure 2.3 depicts this

excitation process, which is quantized.

69

Figure 2.3. The excitation process

The electromagnetic radiation that is absorbed has energy exactly equal to the

energy difference between the excited and ground states. In the case of ultraviolet and

visible spectroscopy, the transitions that result in the absorption of electromagnetic

radiation in this region of the spectrum are transitions between electronic energy levels. As

a molecule absorbs energy, an electron is promoted from an occupied orbital to an

unoccupied orbital of greater potential energy. Generally, the most probable transition is

from the highest occupied molecular orbital (HOMO) to the lowest unoccupied molecular

orbital (LUMO). The energy differences between electronic levels in most molecules vary

from 125 to 650 kJ/mole (kilojoules per mole).

Figure 2.4. Electronic energy levels and transitions.

70

For most molecules, the lowest-energy occupied molecular orbitals are the

orbitals, which correspond to s bonds. The orbitals lie at somewhat higher energy levels,

and orbitals that holdunshared pairs, the nonbonding (n) orbitals, lie at even higher

energies. The unoccupied, or antibonding orbitals ( * and *), are the orbitals of highest

energy. Figure 2.4a shows a typical progression of electronic energy levels. In all

compounds other than alkanes, the electrons may undergo several possible transitions of

different energies. Some of the most important transitions are

Figure 2.4b illustrates these transitions. Clearly, the energy required to bring about

transitions from the highest occupied energy level (HOMO) in the ground state to the

lowest unoccupied energy level (LUMO) is less than the energy required to bring about a

transition from a lower occupied energy level. Thus, in Figure 2.4b an n * transition

would have a lower energy than a * transition. For many purposes, the transition of

lowest energy is the most important. For transition metal complex the UV-Vis spectra can

yield information about the d – d transfer and the charge transfer band. In our study, UV-

Vis analysis was carried out in a Cary 300 instrument, using MgO wafer.

71

2.2.4. Energy-dispersive X-ray spectroscopy

Figure 2.5. Principle of EDX spectroscopy

Energy-dispersive X-ray spectroscopy (EDX) is an analytical technique used for the

elemental analysis or chemical characterization of a sample. When a high-energy beam of

charged particles such as electrons or protons, or a beam of X-rays, is focused into the

sample being studied, the incident beam may excite an electron in an inner shell of the

atoms, ejecting it from the shell while creating an electron hole where the electron was. An

electron from an outer, higher-energy shell then fills the hole, and the energy released from

this may be released in the form of an X-ray. The number and energy of the X-rays emitted

from a specimen can be measured by an energy-dispersive spectrometer. As the energy of

the X-rays is characteristic of the difference in energy between the two shells, and of the

atomic structure of the element from which they were emitted, this allows the elemental

composition of the specimen to be measured. The EDX analysis was carried out using the

Tecnai G2 F20 200 kV scanning transmission electron microscope. The probe diameter is

about 3 nm and the acquiring time was set to 200 s.

2.2.5. X-ray Diffraction (XRD)

X-ray Diffraction (XRD) is a technique in crystallography in which the pattern

produced by the diffraction of X-rays through the closely spaced lattice of atoms in a

72

crystal is recorded and then analyzed to reveal the nature of that lattice. XRD is a powerful

technique for determining zeolite structure. Moreover, sample preparation is relatively

easy, and the test itself is often rapid and non-destructive.

Figure 2.6. Diffraction of X-ray beams on a crystal lattice

The mechanism of XRD was well explained by Bragg in 1913.[12] A crystal can be

considered to be composed of discrete parallel planes or layers of atoms. As the wave

enters the crystal, it will be partially reflected by the first layer of atoms, while the rest of it

will continue through to the second layer, where the process continues. The separately

reflected waves will remain in phase if the difference in the path length of each wave

(2dsin) is equal to an integer multiple of the wavelength (n) (Figure 2.6).

nd sin2 (2.1)

This equation is known as Bragg‟s law, where:

- n is an integer,

- λ is the wavelength of x-rays, and moving electrons, protons and neutrons,

- d is the spacing between the planes in the atomic lattice, and

- θ is the angle between the incident ray and the scattering planes.

Waves that satisfy this condition interfere constructively and result in a reflected

wave of significant intensity, causing a peak in the pattern diffraction.

73

The pattern of powder diffraction peaks can be used to quickly identify materials

(thanks to the zeolite pattern database [13]), and changes in peak width or position can be

used to determine crystal size, crystallinity of zeolites.[12]

The crystallinity of zeolite can be derived from XRD as follow:

(hkl)

(hkl)

Intensity of peak sampleCrystallinity (%) 100

Intensity of peak reference

(2.2)

The crystal size, a is calculated according to Scherrer‟s equation:

cos

91.0a

(2.3)

Where, is the full width at half maximum of the corresponding peaks. However,

as noticed by Jacobsen et al,[14] the large crystals contribute more to the average size

determined by XRD than the small crystals. Hence, crystal sizes determined from XRD of

very small zeolite crystals are not very accurate and should be taken as first approximations

to the true crystal size.

In a typical test, zeolite sample was characterized by powder wide-angle XRD,

recorded on a Siemens D5000 X-ray diffractometer using CuKα radiation (λ = 1.54184 Å).

The samples were scanned over a range of 2 values from 5 to 50o with a scan step size of

0.02o and a scan step time of 1 s.

2.2.6. 29

Si Magic Angle Spinning Nuclear Magnetic Resonance

Spectroscopy (MAS NMR)

The 29

Si MAS NMR spectroscopy now belongs among the most powerful

techniques for the characterization of molecular sieves such as zeolites and related

materials. The basis of this success was the invention of effective line narrowing techniques

and two-dimensional experiments that make the detection of highly resolved solid-state

NMR spectra and the separation of different spectral parameters possible. The technique

74

allows the direct investigation of the framework of zeolites and related materials, of extra-

framework cations and of the different types of hydroxyl groups.[15]

Lippmaa et al were among the first to show that the chemical shifts of the 29

Si in

solid silicates were approximately equal to the chemical shifts of species in solution.

Therefore it is easy to differentiate between Si(OSi)4 (Q4), (HO)Si(OSi)3 (Q

3),

(HO)2Si(OSi)2 (Q2) silicate species because the chemical shifts lie approximately 10 ppm

apart (Q4 at -110 ppm, Q

3 at -100 ppm, Q

2 at -90 ppm etc).[16] The

29Si MAS NMR

technique is also useful when investigating the incorporation of organosilane species.

Exchange of one Si-O bond against a Si-C bond (the transformation of a Q-silicon into a T-

silicon) causes a shift of about 45 ppm and again there is a separation of approximately 10

ppm between the silicon T3 RCSi(OSi)3, T

2 (RC)2(SiOSi)2 and T

1 (RC)3SiOSi. Thus T

3 can

be found close to -110 + 45 = -65 ppm T2 = -100 + 45 = -55 ppm and T

1 at -90 + 45 = 45

ppm.[17]

For measurement, 29

Si MAS NMR spectra were recorded at a frequency of 59.60

MHz using 30 o pulses of 3 μs duration, 2600 scans and 30 s recycle delays at room

temperature on a Bruker ASX 300 spectrometer.

2.2.7. Scanning electron microscope (SEM)

Scanning electron microscope (SEM) is a microscope that uses electrons rather than

light to form an image. As the primary electron beam “scans” across the sample, the

electrons on the surface of the sample are excited. This excitation leads to the emission of

the secondary electron beam from the surface which produces the image. The SEM can

produces images of high resolution, which means that closely spaced features can be

examined at a high magnification. Images obtained from this technique can provide

information about the surface and particle size of the samples. Preparation of the samples is

relatively easy since most SEMs only require the sample to be conductive. In this study, a

JEOL JSM-840 scanning electron microscope operated at 15 kV was used.

75

2.2.8. Transmission Electron Microscope (TEM)

Unlike the SEM technique, where electrons are detected by beam emission,

transmission electron microscopy (TEM) make use of the electron beam that has been

partially transmitted through the very thin (and so semitransparent for electrons) specimen.

This beam is detected to provide the image of the sample. TEM allows investigating the

information about the inner structure of the specimen as well as their particle size.

Generally, the TEM resolution is about an order of magnitude better than the SEM

resolution. All the TEM images reported in this thesis were obtained on a JEOL 2011

transmission electron microscope.

2.2.9. Nitrogen Adsorption/Desorption Isotherms

The isotherms of nitrogen adsorption and desorption at 77K is the technique which

has been often used for surface characterization of zeolites. It can provide information

about: specific surface, external surface, internal surface as well as the diameter of the

mesopores (if available). According to IUPAC, the isotherms are classified into six groups

as shown in Figure 2.7. The isotherms of zeolites fall into the type I group. This group are

distinguished by a plateau which is nearly or quite horizontal, and which may cut the P/Po =

1 axis sharply or may show a “tail” as saturated pressure is approached.[18]

Figure 2.7. Types of sorption isotherms.[18]

76

The specific surface area of the specimen is derived from the isotherm using BET

equation:[19]

00

0 *11

1 p

p

CV

C

CVppV

pp

mm

(2.4)

Where,

- V is the volume adsorbed,

- Vm is the volume adsorbed on a monolayer

- P is the pressure of the gas

- Po is the saturation pressure of adsorbates at the temperature of adsorption

- c is the BET constant, which is expressed by:

RT

EEc L1exp . E1 is the heat of

adsorption for the first layer, and EL is that for the second and higher layers and is equal to

the heat of liquefaction.

Equation (2.4) can be plotted as a straight line with 1/v[(Po/P) − 1] on the y-axis and

P/Po on the x-axis according to experimental results. This plot is called a BET plot. The

linear relationship of this equation is maintained only in the range of 0.05 < P/Po < 0.30.

The value of the slope and the y-intercept of the line are used to calculate the monolayer

adsorbed gas quantity Vm (reduced in STP) and the BET constant c. Thus the specific

surface area S is given by:

201022414

LaV

S mm [m

2/g]

(2.5)

where, am is the average area occupied by a molecule of adsorbate in the completed

monolayer for nitrogen am = 16.2 Ǻ2 and L is the Avogadro constant.

77

The external surface area which is strongly dependent on the particle size of zeolites

is calculated from the isotherm using t-plot method. This method was developed by de Boer

et al.[20, 21] It assumes that during the adsorption when the micropores are already filled-

up, the adsorption proceeds to occur on the external area. This stage of adsorption can be

regarded as an adsorption on a flat surface. Then the adsorption within this pressure region

may be described by a simple linear dependence:

o

extmicro

o P

PtSka

P

Pa .max,

(2.6)

where:

-

oP

Pa : volume adsorbed, reduced to STP,

- amicro,max - adsorption in saturated micropores, corresponding to total volume of

micropores,

- Sext - "external" surface area; here it is the surface area of pores larger than

micropores,

-

oP

Pt - estimated statistical thickness of adsorbed layer, according to Harkins et

al.[22] :

2/1

log034.0

99.13

o

o

P

PP

Pt

- k - coefficient which depends on units used for the values of adsorption a, layer

thickness t and surface area Sext, e.g. for t [nm], S [m2/g] and a [cm

3/g STP] we obtain: k =

0.6489.

The t-plot is illustrated in Figure 2.8.

78

In this study, the nitrogen adsorption/desorption measurements were carried out

using an Omnisorp-100 automatic analyser at -196 oC after degassing about 30 mg of

calcined sample at 200 oC for at least 4 h under vacuum (10

-4 -10

-5 torr).

Figure 2.8. t-plot method.

2.2.10. Cracking reaction

Cracking experiments were performed in an automated fixed-bed microactivity test

(MAT) unit (Zeton Automat IV), which was a modified version of ASTM D 5154. The unit

was equipped with collection systems for gas and liquid products. The distribution of

gaseous products was analyzed by gas chromatographies. The boiling point (bp) range of

the liquid products was determined by simulated distillation gas chromatography.

The catalysts were tested in the MAT unit at 510 °C with a weight hourly space

velocity (WHSV) of 8 h-1

. MAT results reported include conversion, yields of dry gas (H2,

H2S, C1 and C2), liquefied petroleum gas (LPG, i.e., C3-C4), gasoline (> C5, bp up to 215

°C), LCO (bp 215-345 °C), heavy cycle oil (HCO, bp above 345 °C) and coke. Conversion

was determined from the difference between the amount of feed and the amount of

amico,max

a

t

a = k.S.t

a = amicro,max + k.Sext.t

79

unconverted material defined as liquid product boiling above 215 °C (i.e., LCO + HCO).

The same vacuum gaz oil (VGO) was used to all MAT runs

References

1 D. Trong On, S. M. J. Zaidi, S. Kaliaguine, Microporous Mesoporous Mater. 1998,

22, 211.

2 S. Mintova, N. H. Olson, T. Bein, Angew. Chem., Int. Ed. 1999, 38, 3201.

3 V.-T. Hoang, Vol. PhD, Laval University, Quebec 2005.

4 A. Jentys, J. A. Lercher, Elsevier, Amsterdam 2001.

5 J. F. Arenas, J. I. Marcos, Spectrochim. Acta, Part A 1980, 36, 1075.


7 L. Meesuk, U. A. Jayasooriya, R. D. Cannon, Spectrochim. Acta, Part A 1987, 43,

687.

8 R. Wu, U. A. Jayasooriya, R. D. Cannon, Spectrochim. Acta, Part A 2000, 56, 575.

9 M. K. Johnson, D. B. Powell, R. D. Cannon, Spectrochim. Acta, Part A 1981, 37,

995.

10 L. Montri, R. D. Cannon, Spectrochim. Acta, Part A 1985, 41, 643.

11 K. Nakamoto, Infrared spectra of inorganic and coordination compounds, Wiley-

Interscience New York 1986.

12 H. Koningsveld, J. M. Bennett, in Structures and Structure Determination, Vol. 2

(Eds: C. Baerlocher, J. M. Bennett, W. Depmeier, A. N. Fitch, H. Jobic, H.

Koningsveld, W. M. Meier, A. Pfenninger, O. Terasaki), Springer Berlin

Heidelberg, 1999, 1.

13 C. Baerlocher, W. M. Meier, D. H. Olson, Atlas of Zeolite Framework Types,

Elsevier, New York 2001.

14 I. Schmidt, C. Madsen, C. J. H. Jacobsen, Inorg. Chem. 2000, 39, 2279.

15 M. Hunger, E. Brunner, in Characterization I, Vol. 4 (Eds: H. Karge, J. Weitkamp),

Springer Berlin Heidelberg, 2004, 201.


17 A. van Blaaderen, A. Vrij, J. Colloid Interface Sci. 1993, 156, 1.

18 S. Gregg, K. S. W. Sing, 1983.

19 S. Brunauer, P. H. Emmett, E. Teller, J. Am. Chem. Soc. 1938, 60, 309.

20 B. C. Lippens, J. H. de Boer, J. Catal. 1965, 4, 319.

21 B. C. Lippens, B. G. Linsen, J. H. d. Boer, J. Catal. 1964, 3, 32.

22 W. D. Harkins, G. Jura, J. Am. Chem. Soc. 1944, 66, 1366.

81

Chapter 3. A New Route for the Synthesis of Uniform

Nanozeolites with Hydrophobic External Surface in

Organic Solvent Medium

Gia-Thanh Vuong and Trong-On Do *

Department of Chemical Engineering, Laval University, Quebec G1K 7P4, Canada

Published in Journal of the American Chemical Society 2007, 129 (13), 3810-3811

DOI: 10.1021/ja069058p

Résumé

Des nanozéolithes silylées de tailles uniformes ont été synthétisés dans un milieu

organique en présence d‟hexadecytrimethylsilane. Les nanozéolithes silylées présentent en

plus des petits cristaux uniformes, avec une surface externe hydrophobe. Ces nanozéolithes

présentent des applications potentielles en catalyse dans les réactions impliquant de grosses

molécules. Ces applications sont directement liées à leur grande surface externe, à la

réduction de diffusion et à l‟exposition des sites actifs.

Abstract

Uniform silylated nanozeolites were synthesized in organic solvent medium in the

presence of hexadecytrimethylsilane. The resulting zeolites have not only small/uniform

crystal sizes, but also a hydrophobic external surface. These zeolites are considered of

potential for application as catalysts in reactions involving bulky molecules, owing to their

high external surface area, reduced diffusion pathways, and exposed active sites.

https://vpn-externe1.ulaval.ca/+CSCO+ch756767633A2F2F63686F662E6E70662E626574++/action/doSearch?action=search&author=Vuong%2C+Gia%5C-Thanh&qsSearchArea=author

https://vpn-externe1.ulaval.ca/+CSCO+ch756767633A2F2F63686F662E6E70662E626574++/action/doSearch?action=search&author=Do%2C+Trong%5C-On&qsSearchArea=author

https://vpn-externe1.ulaval.ca/+CSCO+dh756767633A2F2F63686F662E6E70662E626574++/doi/full/10.1021/ja069058p#ja069058pAF1

82

Nanosized zeolite crystals (nanozeolites) with narrow particle-size distributions and

sizes less than 100 nm have received much attention because of their great potential

applications in catalysis and adsorption. The decrease in the crystal sizes results in high

external surface areas, reduced diffusion path lengths, and more exposed active sites. For

example, small Y-type zeolite crystals have been reported to increase catalytic activity and

improve the selectivity of intermediate cracked products such as gasoline and light gas oil

in catalytic cracking of heavy gas oil. This catalyst exhibited lower deactivation rates since

the coke formation was suppressed.[1] Moreover, nanozeolites can be used as “building

units” for constructing hierarchical materials. Zeolite nanoclusters have been employed for

assembling mesoporous alumosilicates.[2] Materials with semicrystalline zeolitic mesopore

walls[3] and nanozeolite coated mesoporous aluminosilicates[4] were reported by our

group. Recently, zeolites with intracrystal mesopores and strong acidity were also

synthesized.[5] The resulting materials are considered of potential for application in

catalysis and separation, owing to easier transport of guest molecules through the

mesopores and shorter diffusion pathways in the zeolitic walls.

Syntheses of nanozeolites are often carried out in the aqueous phase. During the

crystallization, once the nanozeolite precursors are formed, the aqueous phase acts as an

effective environment for the incorporation of soluble aluminosilicate species and the

aggregation of zeolite crystals. This could lead to the formation of large crystals and

aggregates.[6] Direct synthesis using a clear gel solution of aluminosilicates can also

produce nanozeolites by careful control of the gel composition and crystallizing

conditions.[7] Another method, which is called confined space synthesis, has been

developed for the preparation of nanosized zeolite crystals. The synthesis is conducted

within an inert matrix such as porous carbon matrices,[8] thermoreversible polymer

hydrogels, or microemulsions[9] which provides a steric hindered space for zeolite crystal

growth.

Several synthetic routes have been reported for the preparation of nanocrystalline

zeolites. However, none of these attempts has produced an easy means of controlling the

small size. Furthermore, the external surface of nanocrystalline zeolites is hydrophilic and

83

thereby has mostly silanol groups that limit catalytic reactivity to the internal pore

surface.[6]

Serrano et al.[10] have recently reported the use of organosilane as growth inhibitor.

In this study, MFI and β zeolites were synthesized in the aqueous medium, using

phenylaminopropyl−trimethoxysilane (PHAPTMS). The synthesis is based on reducing the

growth of zeolite crystals by silanization of the zeolitic seeds to hinder their further

aggregation. However, as investigated by TEM analysis, the obtained MFI sample

consisted of particles of about 300−400 nm which were formed by the aggregation of

ultrasmall units of 10 nm. Having that large size, the sample was hardly considered as true

nanozeolite.

Herein, we demonstrate a new route for the synthesis of controlled uniform size

nanozeolites with the hydrophobic external surface. An organic solvent is used as the

medium for crystallization instead of water. The zeolite precursors are functionalized with

organic silane groups. They thus become hydrophobic and highly dispersed in the organic

solvent. Because the crystallization occurs in the organic phase and the zeolite precursors

are protected by functional groups, catastrophic aggregation can be prevented hence,

resulting in small and uniform nanozeolites with hydrophobic external surface.

The MFI and faujasite zeolites were selected to illustrate our approach, because they

are widely recognized for their unique properties as catalysts. As seen in Scheme 3.1, this

approach is simple and was found to be reproducible when using

hexadecyltrimethoxysilane as organosilane agent and the mixture of toluene and n-butanol

as organic medium.

84

Scheme 3.1. Schematic Representation of the Single-Phase Synthesis Method

The XRD patterns of the as-made nanozeolite samples prepared from silylated seeds

are shown in Figure 3.1 Samples prepared by the conventional method in aqueous medium

from the same clear zeolite gel solution without organosilane were used as references. The

XRD pattern of the nanosilicalite-1 sample is identical to that of the reference, indicating

the MFI structure of this nanosilicalite-1 sample. However, there is a clear broadening of

the reflections, which is attributed to small crystals. In addition, no significant peak at 2θ =

20−30° which is characteristic of amorphous phase was observed, indicating a relatively

high crystallinity of this sample. A similar trend was also observed for the nanofaujasite

sample (Figure 3.1B). Furthermore, the FTIR spectra of both as-made nanosilicalite-1 and

nanofaujasite match well with the typical FTIR peaks assigned to silicalite-1 and zeolite Y,

respectively (not shown).[11]

85

Figure 3.1. XRD patterns of the as-made silylated zeolite and zeolite samples prepared

from the same zeolite gel in solvent medium in the presence of organosilane and in aqueous

medium in the absence of organosilane, respectively: (A) silicalite-1; (B) faujasite.

Representative TEM micrographs of the as-made silylated zeolite samples are

shown in Figure 3.2 and exhibit very uniform nanocrystal sizes mostly with spherical and

cubic shaped particles for silicalite-1 and zeolite Y, respectively. The particle size is about

21 nm for silicalite-1 and 27 nm for faujasite. The standard deviation established from the

analysis of more than 1500 particles in representative TEM pictures of each sample showed

a standard deviation of less than 10% (Supporting Information (SI) Table 1). Interestingly,

the samples are composed of discrete particles rather than aggregates. Owing to the

organosilane being grafted on the zeolite precursor, the nanoparticles remain highly

dispersed in the organic medium and protected against drastic aggregation during

crystallization.

86

Figure 3.2. TEM images of the as-made samples: (A) silylated nanosilicalite-1, (B)

silylated nanofaujasite.

The 29

Si MAS NMR spectra of the as-made samples, silicalite-1 and silylated

silicalite-1, prepared from the same zeolite gel solution were investigated (SI Figure 1A).

The NMR spectrum of the as-made silicalite-1 sample shows a main resonance at ca. −110

ppm and a weak resonance peak at ca. −100 ppm which are attributed to Si(OSi)4, Q4 and

Si(OSi)3OH, Q3 species, respectively. For the as-made silylated silicalite sample, only one

resonance peak at ca. −110 ppm attributable to Q4 species was observed; however, an

additional peak at −65 ppm assigned to R-C-Si-(OSi)3 species is present.[11] This

additional peak is the result of the reaction between the silicon in the organosilane and the

silanol groups of zeolite nuclei during the synthesis. This also suggests the silanization on

the external surface of nanosilicalite-1, which acts to heal defect sites (e.g., silanol groups)

in the zeolite surface. Furthermore this calcined sample also shows essentially a single

resonance peak Q4 at ca. −110 ppm (SI Figure 1A). Thus, it can be concluded that the

presence of only one resonance Q4 even after calcination of the silylated silicalite-1 sample

suggests its hydrophobic surface character. Similar results were also obtained for the

silylated faujasite sample (SI Figure 1B). For the silylated faujasite sample, besides the

resonance peaks at −88, −95, −100, and −103 ppm corresponding to Si(3Al), Si(2Al),

Si(1Al), and Si(0Al), respectively,[11] the peak attributed to R-C-Si-(OSi)3 species at −65

ppm was also observed. This peak at −65 ppm is absent in the faujasite sample prepared in

aqueous medium in the absence of organosilane. For the silylated faujasite sample,

Q4 signals became much broader with higher intensity as compared to those of the faujasite

87

one. This means that the silanization led to the transformation of Q3 to Q

4 silicon species

during the synthesis.

The physicochemical properties and crystal size of the nanozeolite samples are

presented in SI Table 1 and SI Figure 2. The crystal sizes determined using the Scherrer

equation correspond reasonably well to those estimated from the TEM pictures. The BET

surface area is 570 and 545 m2/g for the nanosicalite-1 and nanofaujasite samples,

respectively. The external surface area based on t-plot calculation is 150 and 96 m2/g, in

that order. This high external surface value also indicates the small crystal size of the

sample. Detailed studies of the sorption behavior and catalytic properties are underway.

In conclusion, a new approach has been developed for the synthesis of uniform

zeolite nanocrystals with hydrophobic external surface. We anticipate that with this

approach the syntheses of other types of nanozeolites should also be possible. Furthermore,

a large variety of silylating agents allows also the tailoring of nanozeolite properties such as

crystal size and chemical nature of the surface.

Acknowledgment

We thank Prof. S. Kaliaguine for stimulating discussions and comments and Dr. B.

Nohair for assistance in connection with MAS NMR spectra. We thank the Natural

Sciences and Engineering Research Council of Canada (NSERC) for financial support of

this research and the Vietnam ministry of education and training for the scholarship

(G.T.V.)

Supporting Information Available

Experimental procedure, SI Figures 1−3, SI Table 1.

https://vpn-externe1.ulaval.ca/+CSCO+dh756767633A2F2F63686F662E6E70662E626574++/doi/suppl/10.1021%2Fja069058p

88

References

1. (a) Sano, T.; Ikeya, H.; Kasuno, T. Zeolites 1997, 19, 80. (b) Camblor, M. A.; Corma,

A.; Martinez, A.; Mocholi, F. A.; Pariente, J. P. Appl. Catal. 1989, 55, 65. (c) Tonetto,

G.; Atias, J.; de Lasa, H. Appl. Catal. 2004, 270, 9. (d) Do, T. O.; Kaliaguine, S.

In Nanoporous Materials Science and Engineering; Imperial College Press: London,

2004; Vol. 4, p 47.

2. (a) Liu, Y.; Zhang, W.; Pinnavaia, T. J. Angew. Chem., Int. Ed. 2001, 40, 1255. (b)

Zhang, Z.; Han, Y.; Zhu, L.; Wang, R.; Yu, Y.; Qui, S.; Zhao, D.; Xiao, F.-S. Angew.

Chem., Int. Ed. 2001, 40, 1258.

3. (a) Do, T. O.; Kaliaguine, S. Angew. Chem. Int. Ed. 2001, 40. (b) Do, T. O.; Kaliaguine,

S. Angew. Chem., Int. Ed. 2002, 41, 1036.

4. (a) Do, T. O.; Kaliaguine, S. J. Am. Chem. Soc. 2003, 125, 618. (b) Do, T. O.; Nossov,

A.; Springuel, M. A.; Schneider, C.; Bretherton, J. L.; Fyfe, C. A.; Kaliaguine, S. J. Am.

Chem. Soc. 2004, 126, 14324.

5. (a) Choi, M.; Cho, H. S.; Srivastava, R.; Venkatesan, C.; Choi, D. H.; Ryoo, R. Nature

Mater. 2006, 5, 718. (b) Wang, H.; Pinnavaia, T. J. Angew.Chem., Int. Ed. 2006, 45,

7603.

6. Tosheva, L.; Valchev, V. P. Chem. Mater. 2005, 17, 2494.(b) Cundy, C. S.; Cox, P.

A. Microporous Mesoporous Mater. 2005, 82, 1.

7. (a) Schoeman, B. J.; Sterte, J.; Otterstedt, J. Zeolites 1994, 14, 110. (b) Mintova, S.;

Olson, N. H.; Valtchev, V.; Bein, T. Science, 1999, 283, 958. (c) Davis, T. M.; Drews,

T. O.; Ramanan, H.; He, C.; Dong, J.; Schnablegger, H.; Katsoulakis, M.; Kokkoli, E.;

Mccormick, A. V.; Penn, R. L.; Tsapatsis, M. Nat. Mater. 2006, 5, 400.

8. (a) Schmidt, I.; Krogh, A.; Wienberg, K.; Carlsson, A.; Brorson, M.; Jacobsen, C. J. H.

Chem. Commun. 2000, 2157. (b) Jacobsen, C. J. H.; Madsen, C.; Houzvicka, J.;

Schmidt, I.; Carlson, A. J. Am. Chem. Soc. 2000, 122, 7116.

9. Wang, H.; Holmberg, A.; Yan, Y. J. Am. Chem. Soc. 2003, 125, 9928. (b) Chen, Z.; Li,

S.; Yan, Y. Chem. Mater. 2005, 17, 2262.

10. Serrano, D. P.; Aguado, J.; Escola, J. M.; Rodriguez, J. M.; Peral, A. Chem.

Mater. 2006,18, 2462.

11. (a) Holmberg, B. A.; Wang, H.; Norbeck, J. M.; Yan. Y. Microporous Mesoporous

Mater. 2003, 59, 13. (b) Ravishankar, R.; Kirschhock, C.; Schoeman, B. J.; Vanoppen,

P.; Grobet, P. J.; Stork, S.; Maier, W. F.; Martens, J. A.; Schryver, F. C.; Jacobs, P.

A. J. Phys. Chem. B 1998, 102, 2633.

89

Supporting information

Synthesis: The synthesis of nanozeolites involves two steps: (i) In the fist step,

clear zeolite gel solutions were prepared with molar composition of 2.5 TPAOH: 10 SiO2 :

250 H2O and 0.07 Na2O : 2.4 (TMA)2O : Al2O3 : 4 SiO2 : 264 H2O for silicalite-1 and

faujasite, respectively. In a typical silicalite-1 gel synthesis, 35g of 20% aqueous solution

of TPAOH was added to 19.5 g of TEOS in 25 g of water. The resulting clear solution was

stirred for 24 h at room temperature. These gel solutions were then heated at 80oC for 24 h

to speed up the formation of protozeolitic species (known as zeolite seeds). (ii) In the

second step, 10 g of gel solution was added to 500 ml of a solution of toluene containing n-

butanol (30% wt) and a proper amount of hexadecyltrimethylsilane. The organosilane was

in a proportion of less than 10 mol% in regards to the silica content in the gel. Tolulene is a

suitable medium for the modification of zeolite with organosilane and n-butanol acts as a

surfactant. The functionalization reaction was carried out batchwise in a glass reactor under

stirring at 60oC for 12 h and reflux. After 12 h of stirring, a mixture of only one clear liquid

phase was observed. This mixture was then transferred to an autoclave for further

hydrothermal treatment at 150oC for 3 days, and 180

oC for 5 days for faujasite and

silicalite-1 respectively. After the crystallization, the crude solution of nanozeolite product

was precipitated with ethanol and further isolated by centrifugation and then washed with

ethanol for several times. The product was then dried at 100ºC for 24 hours and calcined at

550ºC for 5 hours.

The conventional synthesis of the zeolites in aqueous medium was carried out

according to the procedure described in the literature.7 For the synthesis of silicalite-1, the

temperature was 150oC and the crystallization time was 3 days, whereas, for that of FAU

zeolites, the temperature and crystallization time were 100oC and 5 days.

Characterization: The FTIR spectra were recorded using a Biorad FTS-60

spectrometer on sample wafers. Powder XRD patterns of the materials were recorded on a

Philips X-ray diffractometer (PW 1010 generator and PW 1050 computer assisted

goniometer) using nickel-filtered CuKa ( = 1.5406 Ǻ) radiation, 0.0258 step size and a 1 s

step time. The crystal size of the zeolites was estimated from the broadening of the XRD

90

peaks using the Scherrer equation: d = 0.9λ /(w-w1) cos θ ; where d is the crystal diameter,

w and w1 are the half-intensity width of the relevant diffraction peak and the instrumental

broadening, respectively, λ is the X-ray wavelength and θ is the angle of diffraction. The

following reflections were used for the crystal size determination: for silicalite-1, [501] and

[151] plans corresponding to the 2θ peaks at 23.10 and 23.75º; and for faujasite, [440] and

[733] planes corresponding to those at 20.30 and 29.55º.

Nitrogen adsorption/desorption isotherms at -196oC were established using an

Omnisorp-100 apparatus. The specific surface area (SBET) was determined from the linear

part of the BET equation (P/Po = 0.05 - 0.15). High-resolution TEM images were obtained

on a JEOL 200 CX transmission electron microscope operated at 120 kV. The samples for

TEM were prepared by dispersing the fine powders of the products in slurry in ethanol onto

honeycomb carbon copper grids. Solid-state 29

Si MAS NMR spectra were recorded at room

temperature on a Bruker ASX 300 spectrometer.

91

S-Table 1. Physicochemical properties of the calcined silylated nanozeolite and zeolite

samples prepared from the same zeolite gel, in solvent medium in the presence of

organosilane, and in aqueous medium in absence of organosilane, respectively.

Sample

SBET

(m2/g)

SEXT

(m2/g)

SMIC

(m2/g)

VMIC

(cm3/g)

VMESO

(cm3/g)

RPORE

(nm)

Crystal size

(nm)-TEM

Crystal size*

(nm)-XRD

Nanosilicalite-1

570

150

420

0.154

0.230

3.7

~20

~23

Silicalite-1

512

15

479

0.162

-

-

~300

-

Nanofaujasite

545

96

449

0.149

0.195

6.5

~30

~24

Faujasite

479

19

460

0.173

-

-

~400

-

*The crystal sizes were calculated by applying the Scherrer’s equation to the XRD

reflections of [501] [151] planes for nanosilicalite-1 and those of [440], [733] planes for

nanofaujasite.

92

S-Figure 1. 29

Si MAS NMR spectra of the as-made silylated samples prepared from the

same zeolite gel, in solvent medium in the presence of organosilane, and in aqueous

medium in absence of organosilane, respectively.

(A) silicalite-1: a) as-made silicalite-1, b) as-made nanosilicalite-1, c) calcined

nanosilicalite-1.

(B) faujasite: a) as-made faujasite, b) as-made nanofaujasite.

93

S-Figure 2. Nitrogen adsorption/desorption isotherms and BJH pore radius distribution

(inset) of the calcined nanosilicalite-1 and nanofaujasite samples

94

S-Figure 3. TEM image of the calcined silylated nano faujasite sample

95

Chapter 4. Synthesis of Silylated Nanozeolites in the

Presence of Organic Phase: Two-Phase and Single Phase

Methods

Gia-Thanh Vuong and Trong-On Do *


Published in Microporous and Mesoporous Materials 2009, 120 (3), 310-316

DOI: 10.1016/j.micromeso.2008.11.029

Résumé

Deux méthodes de synthèse, l‟une mono- l‟autre biphasique ont été développées

pour la préparation des nanozéolithes silylées. Dans ces méthodes, un solvant organique a

été utilisé comme milieu de cristallisation de ces nanozéolithes. Pour illustrer ces deux

méthodes, la zéolithe avec la structure MFI telle que silicalite-1 a été utilisée dans cette

étude. Différentes techniques de caractérisation incluant XRD, MET, MEB, BET et RMN

ont été utilisés pour caractériser les propriétés physico-chimiques de ces matériaux. Les

résultats montrent que la méthode à phase unique produit (des nanocristaux uniformes) de

silicalite-1, tandis que la méthode bi-phasique donne deux produits : nanozéolithe et

microzéolithe dans la phase organique et la phase aqueuse respectivement. Cependant, la

RMN indique que le caractère hydrophobe de ces silicalites-1 est conservé après

calcination.

Abstract

Two synthesis methods: two-phase method and single-phase one, for the

preparation of silylated nanozeolites were reported. In these methods an organic solvent

was used as an effective medium for the crystallization of zeolites. To illustrate these two

methods, silicalite-1 was selected in this study. Various techniques including XRD, TEM,

SEM, BET and NMR techniques were used to monitor the physico-chemical properties of

these synthesized materials. The results revealed that the single-phase method allows

producing uniform/small nanosilicalite-1, whereas the two-phase one can bring two

separate products: nanosized and microsized zeolite crystals in organic phase and in

aqueous phase, respectively. Furthermore, the NMR results indicate that the hydrophobic

surface character of these silicalite-1 samples can be obtained even after calcination.

Keywords: nanozelites, nanosilicalite-1, synthesis, organic medium, surface modification.




96

4.1. Introduction

Zeolites are crystalline aluminosilicate molecular sieves with uniform pores of

molecular dimensions. They have been widely used as heterogeneous catalysts and

adsorbents in the field of oil refining and in the petrochemical industry. Conventional

zeolites with crystal sizes of several micrometers are synthesized under hydrothermal

conditions. However, due to the pore size constraints (<15 Å), the unique catalytic

properties of zeolites are limited to reactant molecules having kinetic diameters below

10 Å [1-3]. To overcome this limitation, there is an intensive research to either minimize

the size of the zeolite crystals or to generate hierarchical zeolites containing a bimodal

porosity, both micro-/mesopores [4-11] or micro-/macropores [12, 13]. The former strategy

has led to the discovery of various types of zeolitic porous materials such as composite

MFI/MCM-41 materials, UL-zeolites and MSU-S [4-11].

Nanozeolite with crystal sizes of less than 100 nm, has been receiving much of

interest due to the fact that the reduction of zeolite crystal size from the micrometer to the

nanometer scale leads to increase the external surface area and reduce diffusion path

lengths [14, 15]. Furthermore, the reduction of crystal size to the nanometer scale results in

substantial changes in the properties of materials, which have an impact on the performance

of zeolites in traditional application areas such as catalysis and separation[16, 17].

Therefore, the synthesis of zeolite nanocrystals with small uniform diameter is highly

desirable.

A number of synthesis methods has been reported, which allows the syntheses of

nanozeolites with different structures, such as FAU, MFI, LTA and MOR [15, 18-20].

Synthesis methods are often carried out in the aqueous phase, and the key to the success

was the use of so-called “clear gel solution” [15]. Another method, which is called confined

space synthesis, has been developed for the preparation of nanosized zeolite crystals. The

synthesis is conducted within an inert matrix such as porous carbon matrices [21-23],

thermo-reversible polymer hydrogels, or microemulsions [24-29] which provides a steric

hindered space for zeolite crystal growth. However, the external surface of the nanozeolites

is generally hydrophilic and has high concentration of silanol groups that limit the catalytic

97

activity to the internal pore surface. Therefore, the synthesis of nanozeolites with

hydrophobic external surface is highly interesting [30-32]. To achieve that, the nanozeolites

have to undergo the post-synthesis treatment which often involves the silanization of the

nanozeolite surfaces with organosilane agents [33-35].

Serrano et al. [36] have reported the use of organosilane as growth inhibitor. MFI

and beta zeolites were synthesized in the aqueous medium, using phenylaminopropyl-

trimethoxysilane (PHAPTMS). The synthesis is based on reducing the growth of zeolite

crystals by silanization of the zeolitic seeds to hinder their further aggregation. However, as

investigated by TEM analysis, the silylated MFI samples were prepared with a broad

crystal size distribution in the range of 200–400 nm, which were formed by aggregation of

ultra small units of 10 nm. Recently, we reported a new method, namely a single organic

phase for the synthesis of controlled uniform size nanozeolites with the hydrophobic

external surface. An organic solvent is used as the medium for crystallization instead of

water. The zeolite precursors are functionalized with organic silane groups. They thus

become hydrophobic and well dispersed in the organic solvent. Because the crystallization

occurs in the organic phase and the zeolite precursors are protected by functional groups,

catastrophic aggregation can be prevented hence resulting in small and uniform silylated

nanozeolites with hydrophobic external surface [37]. Clear zeolite gel precursor was

prepared using the method described in the literature [38, 39]. The as-made silylated

nanozeolites are readily hydrophobic, hence, there is no need for the post-synthesis

treatment. Therefore it can be suggested that organic solvent is an effective medium for the

synthesis of nanozeolites.

The two-phase organic solvent/water synthesis is an appealing method which has

been used in the synthesis of transition metal nanoparticles such as Pt, TiO2, ZrO2,

CdS [40-47]. For the synthesis of TiO2 nanocrystals, for example, a toluene solution

containing titanium n-propoxide and oleic acid (OA) as a capping agent and an aqueous

solution of tert-butylamine are mixed and heated. The formation of nanoparticles takes

place at the toluene-water interface. The freshly formed nanoparticles capped with OA

become hydrophobic and could be dispersed in toluene. They therefore stop growing and

finally, monodisperse nanoparticles can be achieved [41-43]. The two phase synthesis can

98

be carried out with an appropriate choice of capping agents. These agents contain both

hydrophilic groups at one end that can be bound to the nanoparticle surface and

hydrophobic alkyl-chains at the other end [48].

The prospect of applying this strategy for the synthesis of nanozeolites is attractive,

it would be an effective approach for spatial isolation of zeolitic seeds from further growth

and the size could be tuned via changing the capping agent content. However, the

crystallization of zeolites is somewhat different and difficult as compared to the synthesis

of transition metal nanoparticles. Care should be taken in choosing the appropriate capping

agent and solvent for the synthesis of nanozeolites. One of the problems is that common

capping agents such as oxalic acid, stearic acid are incapable of “capping” zeolitic seeds. In

the synthesis of transition metal nanoparticles, these acids that contain donor group can

attach to metal atoms by coordinate bonding. Thus, a transition metal of which the atom has

an incomplete d shell is more likely to form the complex with the acids, compared to the

main group Si and Al ions (e.g., no transition metal ions) in the structure of zeolites.

To overcome this problem, the synthesis strategy is based on hindering the growth

of zeolite crystals by silanization of the zeolitic seeds using an organosilane. An aqueous

clear solution containing zeolitic seeds was added in an organic solvent containing

organosilane agent. As the organic solvent is insoluble in water, the organic phase stays on

top of the aqueous phase. Since the zeolitic seeds are rich in silanol groups, they can react

with the organosilane agents at the interface between the two phases. Because of the

protecting organic groups, the functionalized zeolitic precursors become more hydrophobic

and can diffuse into the organic phase.

In this paper, we report a two-phase method approach in which silylated

nanosilicalite-1 was prepared in the presence of both organic and aqueous phases.

Hexadecyltrimethoxysilane and toluene were used as silylating agent and organic solvent,

respectively. These materials were characterized using XRD, TEM, BET and 29

Si MAS

NMR techniques, and compared to those obtained by the single organic solvent phase

method.

99

4.2. Experimental

4.2.1. Synthesis of silylated silicalite-1 using the two phase and single-

phase methods

The synthesis of nanozeolites involves two steps: (i) In the fist step, clear zeolite gel

solution was prepared according to the method described in the literature [38, 39]. (ii) The

obtained clear gel solutions were then processed in different procedures depending on the

method applied. The use of the same starting zeolite gel solution can provide the ease of

comparison between these synthesis methods.

4.2.2. Two-phase method

The procedure is consisting of the following two steps (see Figure 4.1): (i) A clear

zeolite gel solution was prepared with a molar gel composition of 2.68 SiO2:1 TPAOH:168

H2O. In a typical recipe, 14 g of tetrapropylammonium hydroxide (TPAOH) 20% in water

was added to 7.8 g of tetraethoxysilane, Si(OC2H5)4. The mixture was stirred vigorously for

24 h at room temperature. This clear solution was then heated at 80 °C for 12 h to speed up

the formation of protozeolitic species (known as zeolite seeds). (ii) Typically, 21.8 g of

clear zeolite gel solution was added with 93 g of a solution of toluene containing 1.22 wt%

hexadecyltrimethoxysilane (note that the organosilane was in proportion of less than 10%

mol with respect to the silica content in the gel). Since the solvent is insoluble in water, a

two-phase system was obtained. The functionalization reaction was carried out batchwise

in a glass reactor under stirring at 60 °C for 12 h and reflux. This two-phase mixture was

transferred into a Teflon-lined stainless steel autoclave and then heated at 180 °C for 5

days. After crystallization, the two silicalite-1 products in organic phase (toluene) and

aqueous phase were recovered separately by centrifugation and then washed with ethanol

and water for several times. The products were then dried at 80 °C for 24 h and were

denoted as OP (organic phase) and AP (aqueous phase) samples.

100

Figure 4.1. Schematic representation of the two-phase synthesis method.

4.2.3. Single-phase method

The synthesis of silylated nanosilicalite-1 was carried out according to the

procedure described in reference [37]. 10 g of clear gel solution containing zeolitic seeds

was added to 500 ml of a solution of toluene containing n-butanol (30 wt%) and a proper

amount of hexadecyltrimethoxysilane. An amount of n-butanol is introduced in the organic

phase, which is expected to increase the miscibility of the organic phase toward the

aqueous one, since n-butanol is miscible in hydrocarbon solvent, however it is moderately

soluble in water. The mixture was heated in a glass reactor under stirring at 60 °C for 12 h

and reflux. After 12 h of stirring, a mixture of only one clear liquid phase was observed.

101

This mixture was then transferred to a Teflon-lined stainless steel autoclave for further

hydrothermal treatment at 180 °C for 5 days. After the crystallization, the crude solution of

nanosilicalite-1 product was precipitated with ethanol and further isolated by centrifugation

and then washed with ethanol for several times. The product was then dried at 80 °C for

24 h and is denoted as SOP sample.

4.2.4. Conventional method (synthesis of nanozeolites in aqueous

medium)

The conventional synthesis of the zeolites in aqueous medium was carried out

according to the procedure described in the literature [19, 38, 39]. After being filled with

the same starting silicalite-1 gel solution for the single-phase and two-phase methods, the

Teflon-lined stainless steel autoclave was completely sealed and heated in a convection

oven at 150 °C for 3 days. The solid product was recovered by centrifugation, washed

several times with distilled water, dried over night at 80 °C. The products synthesized using

this conventional method, were denoted as reference or conventional silicalite-1 samples.

4.2.5. Characterization

The FTIR spectra were recorded using a Biorad FTS-60 spectrometer on sample

wafers. Powder XRD patterns of the materials were recorded on a Philips X-ray

diffractometer (PW 1010 generator and PW 1050 computer assisted goniometer) using

nickel-filtered Cu Kα (λ = 1.5406 Ǻ) radiation, 0.0258° step size and a 1 s step time.

The nitrogen adsorption/desorption measurements were carried out using an

Omnisorp-100 automatic analyzer at −196 °C after degassing about 30 mg of calcined

sample at 200 °C for at least 4 h under vacuum (10−4

to 10−5

torr). The specific surface area

(SBET) was determined from the linear part of the BET equation (P/Po = 0.05 − 0.15).

High-resolution TEM images were obtained on a JEOL 200 CX transmission electron

microscope operated at 120 kV. The samples for TEM were prepared by dispersing the fine

powders of the products in slurry in ethanol onto honeycomb carbon copper grids.

However, for scanning electron microscope (SEM), JEOL JSM-840 scanning electron

microscope operated at 15 kV was used. Solid-state29Si MAS NMR spectra were recorded

at room temperature on a Bruker ASX 300 spectrometer.

102

4.3. Results and discussion

The synthesis of nanosilicalite-1 was selected to demonstrate the two phase method

approach (Figure 4.1). Hexadecyltrimethoxysilane and toluene were used as silylating

agent and organic solvent, respectively[29, 36]. After crystallization, the products were

recovered and denoted as OP in the organic phase and AP in the aqueous phase. The mass

ratio of the OP and AP products is about 6/4 in this work. For the single-phase method [37],

the water/oil ratio was reduced significantly (see Figure 4.2). To obtain a mixture of only

one clear liquid phase after stirring at 60 °C for 12 h, an appropriate amount of clear

aqueous zeolite gel solution is added to a solution of toluene containing both n-butanol and

hexadecyltrimethoxysilane (see Section 5.2 for details). The product is denoted as the SOP

sample.

The physico-chemical properties of the calcined silylated nanozeolite and zeolite

samples prepared from the same clear zeolite gel, using different methods: the two-phase,

single-phase and conventional methods are summarized in Table 4.1.

Table 4.1. Physico-chemical properties of the calcined silylated nanozeolite and zeolite

samples prepared from the same clear zeolite gel, using different methods: the two-phase,

single-phase and conventional methods.

*Product recovered in aqueous phase

The specific surface area, SBET determined from the linear part of the BET equation (P/P0 = 0.05–0.15). The mesopore size distribution

calculated using the desorption branch of the N2 adsorption/desorption isotherms and the Barrett–Joyner–Halenda (BJH) formula. The mesopore

surface area (SEXT) and mesopore volume (VBJH) obtained from the pore size distribution curves. The micropore surface area (SMIC) obtained

as SBET–SEXT and the average mesopore radius, (RPORE) calculated as 4VBJH/SBJH[49].

103

Figure 4.2. Schematic representation of the single-phase synthesis method.

104

2

5 10 15 20 25 30

Intensity

(d) Conventional method

(c) AP

(b) OP

(a) SOP

Figure 4.3. XRD patterns of the as-made silicalite-1 samples, (a) sample prepared using the

conventional method in aqueous medium, (b) AP silicalite-1, (c) OP silicalite-1 using the

two-phase method and (d) SOP silicalite-1 using single-phase method.

Figure 4.3 shows the X-ray powder pattern of the as-made silylated SOP sample

using the single-phase method, and as-made silylated OP and AP samples using the two

105

phase method. Sample prepared by the conventional method in aqueous medium from the

same clear zeolite gel solution without organosilane was used as a reference (d). The XRD

patterns of these samples are identical to that of the reference sample, indicating the MFI

structure of the samples. This also suggests that the presence of organosilane in the

synthesis mixture did not affect the formation of the desired structure. However, for the

SOP sample prepared using the single-phase method and the OP sample, there is a clear

broadening of the reflections, which is attributed to small crystals. In addition, no

significant peak at 2θ = 20–30° which is characteristic of amorphous phase was observed

suggesting a relatively high crystallinity of the silylated samples.

The crystal size of these samples was also investigated by scanning and

transmission electron microscope (SEM and TEM) techniques. The SEM micrographs of

the as-made silylated OP and AP silicalite-1 samples (e.g., the products in organic phase

and aqueous phase, respectively) are shown in Figure 4.4. The crystal size of the OP

silicalite-1 sample was uniform, ranging from 30 to 50 nm. In contrast, large crystals of

5μm were observed for the AP silicalite-1 sample. A difference in crystal size between the

two samples suggests the effect of media for crystallization on crystal size of the product.

Further, for the as-made silylated SOP silicalite-1 sample, a representative TEM

micrograph is also shown in Figure 4.5 and exhibits very uniform nanocrystal sizes with

mostly spherical and cubic shaped particles. The particle size is about 21 nm. Owing to the

organosilane being grafted on the zeolite precursors, the nanoparticles remain well

dispersed in the organic medium and are protected from drastic aggregation during

crystallization.

106

Figure 4.4. SEM micrographs of the as-made samples, (A) silylated OP silicalite-1 and (B)

silylated AP silicalite-1.

107

Figure 4.5. TEM micrograph of the as-made SOP nanosilicalite-1 sample prepared using

the single-phase method.

It is well known that water phase which is used as a crystallization medium, plays a

role of transporting monomers to the surface of zeolite crystals and large crystal sizes were

obtained. However, using an organic solvent instead of water, the zeolite precursors are

functionalized with organic silane groups. They thus become hydrophobic and highly

dispersed in the organic phase. Because of the crystallization occurs in the organic phase

with a limited water amount, catastrophic aggregation can be prevented. As a result, small

and uniform nanozeolites were observed for the OP silicalite-1 and SOP silicalite-1

samples. Figure 4.6 also shows the FTIR spectra of the samples prepared using the

conventional and single-phase methods (the calcined SOP sample). The FTIR peak

positions are identical to those of the MFI structure. In particular, the band at 550 cm−1

is

assigned to the asymmetric stretching mode of the five-membered ring present in ZSM-5

which is an indication of the MFI structure. Splitting of this lattice-sensitive band into a

doublet has been observed in nanophase silicalite-1 [39]. The FTIR spectrum of the SOP

silicalite-1 sample prepared using the single-phase method in Figure 4.6 shows the doublet

band at 561/547 cm−1

which indicates the formation of nanocrystals.

108

Wavenumber (cm-1

)

400600800100012001400

Tra

nsm

itta

nce

544

555 450

795

1080

1220

(A)

(B)

Figure 4.6. FTIR spectra of the silicalite-1 samples prepared using the single-phase method

(A) and the conventional method (B).

109

Figure 4.7 29

Si MAS NMR spectra of the silicalite-1 samples prepared from the same

zeolite gel solution using (a) the conventional method in aqueous medium without

organosilane, (b) as-made SOP nanosilicalite-1 using single-phase method in organic

solvent and (c) calcined SOP nanosilicalite-1.

110

Figure 4.8 29

Si MAS NMR spectra of the as-made silicalite-1 samples prepared from the

same zeolite gel solution using the two-phase method: (a) as-made AP nanosilicalite-1 and

(b) as-made OP nanosilicalite-1.

The 29

Si MAS NMR spectra of the as-made silicalite-1 samples prepared using

different methods from the same zeolite gel solution were investigated (Figure 4.7

and Figure 4.8). The NMR spectrum of the as-made silicalite-1 sample prepared using the

conventional method, in the absent of organosilane, shows a main resonance at

approximately −110 ppm and a weak resonance peak at approximately −100 ppm which are

attributed to Si(OSi)4, Q4 and Si(OSi)3OH, Q

3 species, respectively (Figure 4.7a). However,

for the as-made SOP silicalite-1 sample, only one resonance peak at approximately

−110 ppm attributable to Q4 species was observed, an additional intense peak at −65 ppm

assigned to R-C-Si-(OSi)3 species is present [33]. This additional peak is the result from the

reaction between the silicon in the organosilane and the silanol groups of zeolite nuclei

111

during the crystallization. This NMR broad peak at 50–70 ppm could be contributed

to T2 and T

3 which correspond to two and three OH groups consumed by one organosilane

molecule. The calcined sample also shows essentially a single resonance peak Q4 at

approximately −110 ppm (Figure 4.7b and c). It clearly indicates that the silanol groups

which disappear in the chemical interaction with the organosilane do not reappear upon

calcination. This could be due to two or three OH groups on the nanoparticle surface

consumed by one organosilane molecule. As a result, only one resonance Q4 of this sample,

even after calcination, suggests its hydrophobic surface character.

Furthermore, the as-made OP and AP silicalite-1 samples exhibit also only one

NMR peak at approximately −110 ppm characteristic of Q4 along with a NMR broad peak

at 50–70 ppm attributed to T2 and T

3 for the OP sample and a very weak peak for the AP

sample (Figure 4.8). This also suggests the silanization on the external surface of silicalite-

1, which acts to heal defect sites (e.g., silanol groups) on the zeolite surface. Thus, it can be

concluded that the presence of only one resonance Q4 even after calcination of the silylated

silicalite-1 samples suggests their hydrophobic surface character.

112

Figure 4.9. Nitrogen adsorption/desorption isotherms of the calcined samples: (A) SOP

silicalite-1, (B) OP silicalite-1 and (C) AP silicalite-1 (inset: t-plot curve).

Figure 4.9 shows the N2 adsorption/desorption isotherms and the BJH pore radius

distribution of the calcined SOP nanosilicalite-1 and OP nanosilicalite-1 samples. At low

relative P/Po pressure, a steep rise in uptake, followed by a flat curve, corresponds to filling

Micropore volume

Micropore volume

Micropore volume

113

of micropores with nitrogen. An inflection at higher pressures (e.g., in P/Po range from 0.7

to 0.9) and a hysteresis loop for the OP nanosilicalite-1 sample are characteristic of

capillary condensation and are related to the range of mesopores due to the interparticles.

For the AP silicalite-1 sample, mesopores were not observed (see Figure 4.9B) owing to its

large particle size. However, for the SOP nanosilicalite-1 sample no hysteresis loop is

present; even its particle size is smaller than that of the OP sample. This can be explained

by uncompacted nanoparticles of this sample [49]. The specific surface areas are 570 and

520 m2/g, and the external surface areas based on t-plot calculation are 150 and 106 m

2/g

for the SOP nanosilicalite-1 and OP nanosilicalite-1 samples, respectively. However, for

the AP silicalite-1 sample with large crystal size, the external surface area is very low, only

10 m2/g (Table 4.1). The high external surface values indicate the small crystal size of the

sample.

Indeed, silanol groups on silicalite-1 are mainly located on the external surface and

therefore the quantity of silanol sites is related to the zeolite crystal size. Due to the bulky

size of hexadecyltrimethoxysilane, only the silanol groups on the external surface of zeolite

crystals are accessible for the silanization. Hence the high external surface of nanozeolites

provides silanol sites which are available for the chemical functionalization. The order of

the external surface areas and of the amount of silanol sites reacted with organic silane

groups based on the NMR peak intensity at −65 ppm was found to be: SOP silicalite-

1 > OP silicalite-1 > AP silicalite-1. In contrast, the order of crystal size is: AP silicalite-

1 > OP silicalite-1 > SOP silicalite-1. Furthermore, as seen the NMR results, the silylated

silicalite-1 sample synthesized by these methods exhibit a hydrophobic character, even after

calcination.

4.4. Conclusion

In conclusion, we have demonstrated two methods in the presence of organic

medium for the synthesis of nanozeolites with hydrophobic external surface. Depending on

the method of synthesis, different crystal sizes can be synthesized: Uniform and small

nanozeolites using the single-phase method, however, using the two phase method, large

nanosized and microsized zeolite crystals in organic phase and in aqueous phase,

respectively. We believe that these methodologies can be applied to the synthesis of other

114

types of nanozeolites. Detailed studies of the sorption behavior and catalytic properties of

this class of nanozeolites are underway.

Acknowledgments

We thank Prof. S. Kaliaguine and Prof. F. Kleitz and Dr. V.T. Hoang for

stimulating discussions and comments. This study was supported by the Natural Sciences

and Engineering Research Council of Canada (NSERC).

115

References

1. D. Breck Zeolite Molecular Sieves: Structure, Chemistry and Use Wiley

Interscience, New York (1974)

2. C. Baerlocher, W.M. Meier, D. Olson Atlas of Zeolite Framework Types Elsevier

(2001)

3. L.B. McCusker, C. Baerlocher H.V. Bekkum, P.A. Jacobs, E.M. Flanigen, J.C.

Jansen (Eds.), Introduction to Zeolite Science and Practice (second ed.), Elsevier,

Amsterdam (2001), p. 37

4. Y. Liu, W. Zhang, T.J. Pinnavaia Angew. Chem. Int. Ed., 40 (2001), p. 1255

5. Z. Zhang, Y. Han, L. Zhu, R. Wang, Y. Yu, S. Qiu, D. Zhao, F.-S. Xiao Angew.

Chem. Int. Ed., 40 (2001), p. 1258

6. A. Karlsson, M. Stöcker, R. Schmidt Micropor. Mesopor. Mater., 27 (1999), p. 181

7. S. Kaliaguine, T.O. Do, Mesoporous Zeolitic Materials with Microporous Crystalline

Mesopore Walls: UL-zeolites, US 6,669,924 B1.

8. S. Kaliaguine, T.O. Do, Mesoporous Zeolitic Materials with Microporous Crystalline

Mesopore Walls: UL-zeolites, EP 1237818 B1.

9. T.O. Do, S. Kaliaguine Angew. Chem. Int. Ed., 40 (2001), p. 3248

10. T.O. Do, S. Kaliaguine Angew. Chem. Int. Ed., 41 (2002), p. 1036

11. T.O. Do, A. Nossov, M.A. Springuel-Huet, C. Schneider, J.L. Bretherton, C.A. Fyfe,

S. Kaliaguine J. Am. Chem. Soc., 126 (2004), p. 14324

12. S. Vaudreuil, B. Echchahed, P. Reinert, M. Bousmina, S. Kaliaguine, L. Bonneviot J.

Porous Mater., 14 (2007), p. 173

13. B.T. Holland, L. Abrams, A. Stein J. Am. Chem. Soc., 121 (1999), p. 4308

14. T.O. Do, S. Kaliaguine G.Q. Lu, X.S. Zhao (Eds.), Nanoporous Materials Science

and Engineering, Imperial College Press Publisher (2004), p. 47

15. L. Tosheva, V.P. Valtchev, Chem. Mater., 17 (2005), p. 2494

16. S.S. Kim, J. Shah, T.J. Pinnavaia, Chem. Mater., 15 (2003), p. 1664

17. E.F.S. Aguiar, M.L.M. Valle, M.P. Silva, D.F. Silva, Zeolites, 15 (1995), p. 620

18. C.S. Cundy, P.A. Cox, Chem. Rev., 103 (2003), p. 663

19. C.S. Cundy, J.O. Forrest, R.J. Plaisted, Micropor. Mesopor. Mater., 66 (2003), p. 143

20. T.M. Davis, T.O. Drews, H. Ramanan, C. He, J. Dong, H. Schnablegger, M.A.

Katsoulakis, E. Kokkoli, A.V. McCormick, R.L. Penn, M. Tsapatsis, Nat. Mater., 5

(2006), p. 400

21. C. Madsen, C. Madsen, C.J.H. Jacobsen, Chem. Commun. (1999), p. 673

22. C.J.H. Jacobsen, C. Madsen, T.V.W. Janssens, H.J. Jakobsen, J. Skibsted, Micropor.

Mesopor. Mater., 39 (2000), p. 393

23. I. Schmidt, C. Madsen, C.J.H. Jacobsen, Inorg. Chem., 39 (2000), p. 2279

24. H. Wang, B.A. Holmberg, Y. Yan, J. Am. Chem. Soc., 125 (2003), p. 9928

25. C.S. Carr, D.F. Shantz, Micropor. Mesopor. Mater., 85 (2005), p. 284

26. S.P. Naik, J.C. Chen, A.S.T. Chiang, Micropor. Mesopor. Mater., 54 (2002), p. 293

27. S. Lee, D.F. Shantz, Chem. Commun. (2004), p. 680

28. S. Lee, D.F. Shantz, Chem. Mater., 17 (2005), p. 409

29. Z. Chen, S. Li, Y.S. Yan, Chem. Mater., 17 (2005), p. 2262

30. R.S. Bowman, Micropor. Mesopor. Mater., 61 (2003), p. 43

31. P. Huttenloch, K.E. Roehl, K. Czurda, Environ. Sci. Technol., 35 (2001), p. 4260

116

32. W. Song, G. Li, V.H. Grassian, S.C. Larsen, Environ. Sci. Technol., 39 (2005), p.

1214

33. B.Z. Zhan, M.A. White, M. Lumsden, Langmuir, 19 (2003), p. 4205

34. W. Song, J.F. Woodworth, V.H. Grassian, S.C. Larsen, Langmuir, 21 (2005), p. 7009

35. B.Z. Zhan, M.A. White, P. Fancy, C.A. Kennedy, M. Lumsden, Macromolecules, 37

(2004), p. 2748

36. D.P. Serrano, J. Aguado, J.M. Escola, J.M. Rodriguez, A. Peral, Chem. Mater., 18

(2006), p. 2462

37. G.T. Vuong, T.O. Do, J. Am. Chem. Soc., 129 (2007), p. 3810

38. B.J. Schoeman, J. Sterte, J.E. Otterstedt, Zeolites, 14 (1994), p. 110

39. R. Ravishankar, C. Kirschhock, B.J. Schoeman, P. Vanoppen, P.J. Grobet, S. Storck,

W.F. Maier, J.A. Martens, F.C. De Schryver, P.A. Jacobs, J. Phys. Chem. B, 102

(1998), p. 2633

40. M. Niederberger, G. Garnweitner, F. Krumeich, R. Nesper, H. Colfen, M. Antonietti,

Chem. Mater., 16 (2004), p. 1202

41. D. Pan, S. Jiang, L. An, B. Jiang, Adv. Mater., 16 (2004), p. 982

42. D. Pan, Q. Wang, S. Jiang, X.J.L. An, Adv. Mater., 17 (2005), p. 176

43. D. Pan, N. Zhao, Q. Wang, S. Jiang, X. Ji, L. An, Adv. Mater., 17 (2005), p. 1991

44. Q. Wang, D. Pan, S. Jiang, X. Ji, L. An, B. Jiang, Chem. Eur. J., 11 (2005), pp. 3843–

3848

45. S.L. Horswell, C.J. Kiely, I.A. O‟Neil, D.J. Schiffrin, J. Am. Chem. Soc., 121 (1999),

p. 5573

46. M. Niederberger, M.H. Bartl, G.D. Stucky, J. Am. Chem. Soc., 124 (2002), p. 13642

47. M. Niederberger, M.H. Bartl, G.D. Stucky, Chem. Mater., 14 (2002), p. 4364

48. B.L. Cushing, V.L. Kolesnichenko, C.J. O‟Connor, Chem. Rev., 104 (2004), p. 3893

49. S.J. Gregg, K.S.W. Sing, Adsorption, Surface Area and PorosityAcademic Press,

London (1982)

117

Chapter 5. Synthesis of Nanozeolite-Based FCC

Catalysts and their Catalytic Activity in Gasoil Cracking

Reaction

Gia-Thanh Vuong, Vinh-Thang Hoang, Dinh-Tuyen Nguyen and Trong-On Do *


Institute of Chemistry, Vietnamese Academy of Science and Technology, Viet Nam

Published in Applied Catalysis A: General 2010, 382 (2), 231-239

DOI: 10.1016/j.apcata.2010.04.049

Résumé

Une nouvelle méthode de synthèse de nanozéolithe en milieu organique utilisant le

formamide et le toluène comme milieu de cristallisation a été développé. Ces deux solvants

ont été utilisé à la place de l‟eau, en présence d‟agent de silylation. L‟effet des solvants

organiques est majeur dans la synthèse des nanozéolithes. Le formamide qui a des

propriétés similaires à l‟eau est un bon candidat de solvant pour la synthèse de

nanozéolithes. Cette méthode de synthèse nous facilite le contrôle de la taille des

nanoparticules des zéolithes. Dans cette étude, nanoparticules de zéolithe avec différentes

tailles; 25, 40 et 100 nm ont été préparées dans le toluène et le formamide agissant comme

solvant. À fin d‟étudier l‟effet de la taille des nanoparticules zéolithiques sur l‟activité

catalytique, une série de nanozéolithes a été utilisée pour le craquage catalytique. Les

résultats montrent une très bonne corrélation entre les tailles des nanozéolithes et l‟activité

catalytique. Les nanozéolithe de taille petite ont montré une activité catalytique supérieure

à celle des nanozéolithe de taille supérieure.




118

Abstract

A new method for the synthesis of nanosized zeolites in organic solvents, such as

formamide and toluene as crystallization medium instead of water, in the presence of

organosilane has been developed. Organic solvents have a great impact on the synthesis of

nanozeolites. Formamide, which has similar properties to water, is a good candidate as the

solvent for the synthesis of nanosized zeolites. This synthetic method allows easy

manipulation with the control of crystal sizes. In this study, different crystal sizes such as

25, 40 and 100 nm were prepared in toluene and formamide solvents. To study the effect of

crystal nanosizes on the catalytic performance of nanosized zeolites, nanozeolite-based

FCC catalysts were also prepared using different nanozeolite sizes as active component and

silica as inactive matrix. The activity of these catalysts was evaluated with FCC feedstock.

The results revealed a good correlation between the crystal size of zeolites and the activity:

smaller nanozeolite-based FCC catalyst exhibits higher catalytic activity.

Keywords: Nanozeolites; Formamide; Non-aqueous synthesis; FCC catalysts; FCC

cracking

119

5.1. Introduction

Nanozeolites with the size of less than 200 nm have received much of interest

recently, because of their great potential applications not only in catalysis and adsorption,

but also in a variety of new applications including chemical sensing, medicine,

optoelectronics etc. [1, 2]. The decrease in the crystal sizes results in higher external

surface areas, reduced diffusion path lengths, and more exposed active sites, which have an

impact on the performance of the nanosized zeolites as compared to that of conventional

zeolites of which the size is often of microns [1, 3]. Besides the well known applications of

such zeolites in catalysis and adsorption, nanozeolites can also find their applications as

seeds and as building blocks for the preparation of mesoporous zeolitic materials [1, 4-9].

Crystalline structure of zeolites with tridimensional network of well-defined micropores

(pore diameter less than 15 Å) brings both (i) advantage and (ii) disadvantage. (i) This

feature provides zeolite with a consistent adsorption behavior toward guest molecules. Only

molecules of size less than or equal to pore size aperture can have access to the vast internal

surface area of zeolites. Thus, when the catalytic reaction occurs inside the zeolite pores,

zeolites can exhibit high selectivity toward small guest molecules [2, 10, 11]. (ii) However,

the unique catalytic properties of zeolites are limited to reactant molecules having kinetic

diameters below 15 Å, due to the pore size constraints. Reactions involving large molecules

on zeolites hence must resort to only the external surface of zeolite [12].

The use of nanosized zeolites could overcome this limitation, the ratio of external to

internal number of atoms increases rapidly as the particle size decreases, and zeolite

nanoparticles have large external surface areas and high surface activity. The external

surface acidity is of importance, when the zeolite is used as catalyst in reactions involving

bulky molecule. The nanosized zeolites could bring better performance due to a high

accessibility of active phase and high external surface area. For example, in catalytic

cracking of gas oil, most of the hydrocarbon molecules are barred from zeolite pores and

thus only the external surface of zeolite contributes to the gas oil conversion. Most of

cracking of these molecules is realized on the interface of zeolite–matrix component of the

FCC catalysts [13, 14]. Rajagopalan et al. have shown that in cracking gas oil, when the

crystallite size of zeolite decreases, both conversion and selectivity clearly increase [15].

120

On this aspect, the use of nanozeolites is a workaround and an improvement for FCC

catalysts. Since the external surface of nanozeolites is expectedly higher and this type of

surface is accessible, cracking of large hydrocarbon molecules on nanozeolites with high

efficiency is possible. Hence a study of a nanozeolite-based FCC catalyst is of great

interest.

Synthesis of nanozeolites has been studied extensively [1]. A common approach is

to modify the general method of synthesis of zeolites, which is carried out in an aqueous

phase [18-20]. Careful adjustment of the parameters such as gel composition, temperature,

crystallization time, aging time etc. can allow nanozeolites to form. The principle of the

synthesis is derived from the classic nucleation and crystallization theory: facilitating the

nucleation, which produces nuclei as much as possible; and controlling a subsequent slow

growth of crystal particles. Ideally, the nucleation and growth processes should be

completely separate from each other.

There are two possible mechanisms of nucleation in the synthesis of zeolites [21]:

homogeneous nucleation and heterogeneous nucleation. Homogeneous nucleation occurs

from the mother liquid while heterogeneous nucleation happens within the gel.

Heterogeneous nucleation and growth are hardly separate process. Hence, regarding the

synthesis of nanozeolites, it is very important to obtain the starting synthesis gel in the state

of a “clear solution” or a clear gel solution in the hope that the homogeneous nucleation

would take place instead of the heterogeneous one. Other factors such as aging, pH,

crystallization time, gel composition are also subject to change to control the nucleation

and growth process.

In this paper, we report a new route for the synthesis of nanozeolites of FAU by the

soft controlling method using different types of solvents as crystallization medium instead

of water. The crystal size of the nanozeolites can be manipulated to some extent by

changing the solvent type. To evaluate the potential application, a series of FCC catalysts

based on these nanozeolites with various particle sizes are also prepared. The obtained

catalysts were tested against commercial catalysts in a standard test of gas oil cracking.

121

5.2. Materials and methods

5.2.1. Synthesis of nanofaujasite

Three kinds of samples were prepared. The synthesis followed what we have

reported [16]. In a typical procedure, Al(iPr)3 (19.5 g) was added into 78.36 g of TMAOH

25% under stirring for 3 h. Then 40.68 g of TEOS 98% was added. The stirring was

continued overnight to make sure TEOS was completely hydrolyzed. Then, 64 mL of

NaOH 0.1 M was added and stirred for another 3 h. The resulting clear solution was then

aged at 90 °C for 2 (or 4) days to speed up the formation of protozeolitic species known as

zeolite seeds. Subsequently, 10 g of the aged gel was added into 100 mL of

hexadecyltrimethoxysilane (HDMT, 10%) containing toluene (or formamide). The clear

homogeneous mixture was then transferred into an autoclave and heated for 5 days at

160 °C temperature. The silylated nanozeolite product was then recovered by centrifuge

and washed with ethanol three times before drying at 100 °C for 24 h. The samples,

prepared using toluene, were designated as FAU–TOLxD, while the ones using formamide

were designated as FAU–FORxD, where x is the aging time in day; the yield of synthesis

was 41% and 47%, respectively. Zeolite Y reference was used from Strem Chemical.

5.2.2. Synthesis of nanofaujasite-based FCC catalysts

35 g of TEOS was dissolved in 100 mL of ethanol. To this mixture 10 g of as-made

nanofaujasite was added. The mixture was stirred overnight and then evacuated under

reduced pressure. The collected solid was dried at 100 °C for 24 h then calcinated at 600 °C

for 6 h. The FCC catalyst samples were designated as FCC–FAU–TOLxD and FCC–FAU–

FORxD, where x is the aging time of zeolite gel in day.

5.2.3. Characterization

The FT-IR spectra were recorded using a Biorad FTS-60 spectrometer on sample

wafers. Powder XRD patterns of the materials were recorded on a Philips X-ray

diffractometer using nickel-filtered CuKα(λ = 1.5406 Ǻ) radiation.

The nitrogen adsorption/desorption measurements were carried out using an

Omnisorp-100 automatic analyzer at −196 °C after degassing about 30 mg of calcined

sample at 200 °C for at least 4 h under vacuum (10−4

–10−5

Torr). The specific surface area

https://vpn-externe1.ulaval.ca/+CSCO+00756767633A2F2F6A6A6A2E667076726170727176657270672E70627A++/science/article/pii/S0926860X10003224#bib16

122

(SBET) was determined from the linear part of the BET equation (P/Po = 0.05–0.15). TEM

images were obtained on a JEOL 200 CX transmission electron microscope operated at

120 kV. The samples for TEM were prepared by dispersing the fine powders of the

products in slurry in ethanol onto honeycomb carbon copper grids. For scanning electron

microscope (SEM), JEOL JSM-840 scanning electron microscope operated at 15 kV was

used. Solid-state 29Si MAS NMR spectra were recorded at room temperature on a Bruker

ASX 300 spectrometer.

5.2.4. MAT cracking evaluation

Cracking experiments were performed in an automated fixed-bed microactivity test

(MAT) unit (Zeton Automat IV), which was a modified version of ASTM D 5154. A

simplified drawing of the MAT unit is shown in Scheme 5.1. The unit was equipped with

collection systems for gas and liquid products. The distribution of gaseous products was

analyzed by gas chromatographies. The boiling point (bp) range of the liquid products was

determined by simulated distillation gas chromatography.

The catalysts were tested in the MAT unit at 510 °C with a weight hourly space

velocity (WHSV) of 8 h−1

. All samples were steamed with 20% water vapor in N2 at

550 °C for 24 h before the catalytic tests. MAT results reported include conversion, yields

of dry gas (H2, H2S, C1 and C2), liquefied petroleum gas (LPG, i.e., C3–C4), gasoline (>C5,

bp up to 215 °C), LCO (bp 215–345 °C), heavy cycle oil (HCO, bp above 345 °C) and

coke. Conversion was determined from the difference between the amount of feed and the

amount of unconverted material defined as liquid product boiling above 215 °C (i.e.,

LCO + HCO). The same vacuum gas oil (VGO) was used to all MAT runs [17].

https://vpn-externe1.ulaval.ca/+CSCO+00756767633A2F2F6A6A6A2E667076726170727176657270672E70627A++/science/article/pii/S0926860X10003224#bib17

123

Scheme 5.1. Simplified diagram of the microactivity test MAT unit for cracking

experiments.

5.3. Results and discussion

5.3.1. Synthesis of nanozeolites

Crystallization of zeolites is complicated and sensitive to synthesis conditions. Its

mechanism is still under debate. And a small change in the synthesis parameters could

result in fruitless products. Hence it is very often that the products of the syntheses of

124

nanozeolites using clear gel method are poorly crystalline and sometimes desired structures

cannot be obtained [22, 23].

An alternative approach is to apply a physical restriction into the synthesis

environment [24-27]. The physical restriction provides a nanospace for the crystallization

of zeolites inside it but prevents them from growing larger than the size of the nanospace.

Porous carbon matrices, micro emulsion and methyl cellulose have been found being a

good physical restrictor. Nevertheless, there are some difficulties that needed to be

overcome: (i) the uniformity in the nanospace size of the restrictor of carbon matrix and

methyl cellulose is not perfect, (ii) full introduction of synthesis gel into the restricting

environment is almost impossible and (iii) the stability of the restrictor under the synthesis

conditions are not acceptable.

Recently, we and other authors [16, 28-30] have proposed a novel approach for the

synthesis of nanozeolites. The idea is to apply a “soft” restriction on the crystal growth

process. This is done using an organosilane to silanize the freshly formed nanozeolites

during the crystallization, the resulting functionalized nanozeolites thus become stable

toward the subsequent growth process. In our method, an organic solvent is introduced

which can disperse these functionalized nanozeolites and completely protect them from the

growth process. Hence, fine nanoparticles can be obtained. The introduction of organic

solvent is an attractive option; the dispersion of the synthesis gel into the organic solvent

depends largely on the affinity of the solvent toward water. A study of the influence of the

solvent on the preparation of nanozeolites would be necessary and worthwhile. When a

hydrophobic solvent is used, a large amount of the solvent is needed to obtain a complete

dispersion of the synthesis gel. But for a hydrophilic solvent, the expectation is that gel

dispersion would be easier. And thanks to the higher affinity toward the gel, higher impact

on the crystal size of the final product is anticipated.

In our previous study, we used toluene as the solvent, which is hydrophobic [16, 29,

30]; hence it was difficult to obtain a homogeneous mixture of the aqueous synthesis gel in

toluene. Thus, to adjust the affinity of this solvent to water, an addition of butanol as an

additive was necessary. However, as the content of butanol increases the crystal size

becomes larger; this is due to the fact that alcoholic systems tend to favor formation of

125

large crystals [31]. So there is a compromise of butanol content; it should be sufficient for a

complete dispersion of the synthesis gel but not too high so as the effect on crystal size is

not significant. According to Qiu et al. [31], alcohol with dielectric constant lower than that

of water would slow down the polymerization and thus the crystallization rate; hence large

crystals are favored. So a good alternative solvent for the synthesis of functionalized

nanozeolites should meet the following requirements: (i) high polarity and (ii) high

solvating capacity. In short, the solvent must resemble water in terms of physicochemical

properties as much as possible while maintaining dissolution capacity of organosilane

agent.

Bearing that in mind it is clear that formamide would be a perfect solvent. The

ability of formamide as a water replacement has been well established [32-34]. It should be

noted that as formamide is an aprotic solvent, it contributes no protons to the synthesis gel.

Hence, it is expected that the role of formamide would be neutral during the synthesis

process.

To demonstrate the advantage of using formamide, we show here three

representative samples of FAU nanozeolite, the first sample FAU–TOL prepared using

toluene as the main solvent and the last two samples FAU–FOR prepared using formamide.

The obtained FT-IR spectra in the region of framework vibrations are shown in Figure 5.1.

The band at 460 cm−1

is assigned to the internal vibration of TO4 (T = Si or Al) tetrahedra.

This vibration is always observable on aluminosilicate species [10]. The band at

565 cm−1

is attributed to the vibration of the double-ring D6R units [35]. This band can be

regarded as a confirmation of the presence of a zeolitic structure. The bands at 685 and

775 cm−1

are assigned to external linkage symmetrical stretching and internal tetrahedral

symmetrical stretching, respectively. Furthermore, the bands at 1010 and 1080 cm−1

are

assigned to internal tetrahedral asymmetrical stretching and external linkage asymmetrical

stretching, respectively [20]. Overall, the FT-IR spectra of these samples match well with

the typical FT-IR absorption peaks of zeolite Y (Figure 5.1).

126

Wavenumber (cm-1)

400600800100012001400

Tra

nsm

itta

nce

(A)

(B)

(C)

Figure 5.1. FT-IR spectra of the prepared nanofaujasite samples: (A) FAU–TOL2D

prepared using toluene and pre-heated zeolite gel for 2 days at 90 °C, (B) FAU–FOR2D

prepared using formamide and pre-heated zeolite gel for 2 days at 90 °C, (C) FAU–FOR4D

prepared using formamide and pre-heated zeolite gel for 4 days at 90 °C, and (D) zeolite Y

reference.

The XRD patterns of the samples (Figure 5.2) are identical to that of the FAU

structure. There is a clear broadening of the reflections from the sample, which is attributed

to small crystals. Furthermore, no evident peak at around 2θ = 20–30° which is

https://vpn-externe1.ulaval.ca/+CSCO+00756767633A2F2F6A6A6A2E667076726170727176657270672E70627A++/science/article/pii/S0926860X10003224#fig2


127

characteristic of amorphous phase, was observed indicating that the samples are highly

crystalline.

2 theta

5 10 15 20 25 30 35

(A)

(B)

(C)

Figure 5.2. XRD patterns of nanofaujasite samples prepared: (A) FAU–TOL2D in toluene,

(B) FAU–FORM2D in formamide from the zeolite gel pre-heated at 90 °C for 2 days, (C)

FAU–FORM4D in formamide from the zeolite gel pre-heated at 90 °C for 4 days, and (D)

zeolite Y standard.

128

Figure 5.3. TEM images of (A) the sample FAU–TOL2D prepared in toluene from the

zeolite gel pre-heated at 90 °C for 2 days, (B) sample FAU–FOR2D prepared in formamide

from the zeolite gel pre-heated at 90 °C for 2 days, and (C) the sample FAU–FOR4D

prepared in formamide from the zeolite gel pre-heated at 90 °C for 4 days.

Representative micrographs of the as-made nanofaujasite samples are shown

in Figure 5.3. The crystals appear very uniform. This is expected since the nanozeolite

particles were protected from aggregation during the crystallization. The crystal size values



129

of these samples FAU–TOL2D, FAU–FOR2D and FAU–FOR4D are 40, 25 and 100 nm,

respectively. For the samples prepared in the presence of formamide, for example, the

sample FAU–FOR4D which was prepared from the clear gel that was pre-heated for 4 days

at 90 °C has larger crystal size than that of the sample FAU–FOR2D which was prepared

from the gel pre-heated for 2 days at 90 °C. It is interesting to note that, while the FAU–

FOR2D sample exhibits typical cubic single nanocrystals, the FAU–FOR4D sample shows

spherical particles. The formation of these spherical particles is attributed to the Ostwald

ripening effect, which aggregates the nanocrystals into larger one.

Figure 5.4 shows the 29

Si MAS NMR spectra of the as-made faujasite prepared in

aqueous medium in the absence of organosilane (conventional method) and silylated

nanofaujasite samples prepared in solvent medium in the presence of organosilane. For the

as-made silylated nanozeolite samples, besides the resonance peaks at −88, −95, −100 and

−103 ppm corresponding to Si(3Al), Si(2Al), Si(1Al) and Si(0Al), respectively, the peak at

−65 ppm attributed to R–C–Si–(OSi)3 species. This peak results in the reaction between the

silicon in the organosilane and the silanol groups of zeolite nuclei during the crystallization.

The NMR broad peak at 50–70 ppm could be contributed to T2 and T

3 which correspond to

two and three OH groups consumed by one organosilane molecule. This peak at −65 ppm is

absent in the faujasite sample prepared in aqueous medium in the absence of

organosilane [36,37]. As seen in Figure 5.4 for the silylated nanofaujasite samples,

Q4 signals became much broader with higher intensity as compared to those of the faujasite

one. This means that the silanization led to the transformation of Q3 to Q

4 silicon species

during the crystallization. Thus, it can be concluded that the 3 samples of functionalized

nanozeolites were obtained.



130

Figure 5.4. 29

Si MAS NMR spectra of the as-made faujasite prepared in aqueous medium

in the absence of organosilane (conventional method) and silylated faujasite samples: (A)

FAU–TOL4D using toluene pre-heated for 4 days, (B) FAU–FOR2D using formamide pre-

heated for 2 days, (C) FAU–TOL2D using toluene pre-heated for 2 days, and (D) FAU-

Standard using conventional method.

The pre-heating treatment of gel at 90 °C was an attempt to populate the

protozeolitic species which were functionalized with organosilane agent for the next

process of crystallization. The duration of the pre-heating process of zeolite gel is a

significant parameter. It should be long to make sure that the population of protozeolitic

species becomes sufficient. As the pre-heating treatment of zeolite gel was done, for the

process of crystallization in the organic solvent, larger nanoparticles obviously grow at the

expense of smaller ones. As a result, these large species even functionalized with

organosilane agent would be precipated. In this case, they settle down on the bottom of the

teflon-line, and these species aggregate into larger ones.

131

However, the preparation using formamide allows production of nanozeolites with

controlled crystal sizes. This fact is of important interest since it opens up a new method to

synthesize nanozeolite crystals with predetermined crystal size. As discussed above, it is

expected that protozeolitic species in synthesis gel pre-heated at 90 °C for 4 days would be

larger in size than those in synthesis gel pre-heated for 2 days. Hence the dispersion of the

gel pre-heated for 4 days in an organic solvent such as toluene would be more difficult

since large protozeolitic species tend to aggregate at higher extent. Nevertheless, using

formamide allows a tolerance toward these zeolite gels; hence it is well dispersed into the

solvent. This is due to the fact that formamide has physicochemical properties similar to

water, while still retaining great dissolution power toward the organosilane agents.

However, the drawback could be the increase in crystal size. Figure 5.5 shows the

N2 adsorption/desorption isotherms of different silylated nanofaujasite samples after

calcination: FAU–TOL2D, FAU–FOR2D and FAU–FOR4D. The isotherms represent a

steep rise in uptake at low relative P/Po pressure and a flat curve following, which is typical

for microporous materials. However, for FAU–TOL2D and FAU–FOR4D (Figure 5.5A

and C), an inflection at P/Po of 0.7–0.9 and a hysteresis loop are characteristic of capillary

condensation and are related to the range of mesopores owing to the interparticles, while

for FAU–FOR2D, a hysteresis loop was essentially not observed (Figure 5.5B). This could

be due to its smaller particle size (25 nm), as compared to the 40 and 100 nm size of the

other ones. The specific surface areas are 505, 515 and 570 m2/g, and the external surface

areas based on t-plot calculation are 80, 115 and 65 m2/g for FAU–TOL2D, FAU–FOR2D

and FAU–FOR4D, respectively. In addition, the external surface areas of the samples are in

agreement with the TEM analysis. The sample with a smaller size as indicated by TEM

images shows higher external surface area. Some physicochemical properties of the

faujasite samples are tabulated in Table 5.1.

132

Figure 5.5. N2 adsorption desorption isotherms of (A) FAU–TOL2D prepared in toluene

from the zeolite gel pre-heated at 90 °C for 2 days, (B) FAU–FOR2D prepared in

formamide from the zeolite gel pre-heated at 90 °C for 2 days, and (C) FAU-FOR4D

prepared in formamide from the zeolite gel pre-heated at 90 °C for 4 days.

133

Table 5.1. Physicochemical properties of nanofaujasite samples.

Sample Particle size (nm) SBET [m2/g] Sexternal[m

2/g] Pore volume [cm

3/g]

FAU–TOL2D 40 505 80 0.43

FAU–FOR2D 25 520 130 0.60

FAU–FOR4D 100 570 65 0.45

5.3.2. Synthesis of FCC

Figure 5.6. XRD patterns of the nanozeolite-based FCC catalyst samples prepared from the

corresponding 40, 24 and 100 nm nanozeolites: (A) FCC–TOL2D, (B) FCC–FOR2D and

(C) FCC–FOR4D.

XRD patterns of the nanozeolite-based FFC catalyst samples with different

nanozeolite sizes are shown in Figure 5.6. The presence of the FAU structure is observed;

however, a broad peak at 2θ = 20–30° is available, implying the presence of amorphous

matrix. The SEM images of these samples show that the FCC catalyst samples are

aggregated into micro-size particles, which are composed of uniform spheres of ~200 nm.

These spheres are merely silica, and nanozeolites are well dispersed and incorporated along



134

the silica spheres (Figure 5.7). For these resulting FCC catalysts, the silica matrix could

stabilize nanozeolites and increase the resistance of zeolite to steam deactivation and

therefore increase the FCC catalyst real-life. The N2 adsorption/desorption isotherms of

these samples are shown in Figure 5.8. The specific surface area values are 360, 355 and

315 m2/g and the pore volumes are 0.90, 0.60 and 1.10 cm

3/g for FCC–FAU–TOL2D,

FCC–FAU–FOR2D and FCC–FAU–FOR4D, respectively (Table 5.2) which are also

namely FCC-40, FCC-25 and FCC-100.

Table 5.2. BET analysis of nanozeolite-based FFC catalyst samples.

Sample SBET [m2/g] Sexternal[m

2/g] Pore volume [cm

3/g]

FCC–FAU–TOL2D 360 312 0.90

FCC–FAU–FOR2D 355 254 0.60

FCC–FAU–FOR4D 315 312 1.10

Figure 5.7. SEM image of (A) FCC–FAU–TOL2D, (B) FCC–FAU–FOR2D and (C) FCC–

FAU-FOR4D.





https://vpn-externe1.ulaval.ca/+CSCO+00756767633A2F2F6A6A6A2E667076726170727176657270672E70627A++/science/article/pii/S0926860X10003224#tbl2

https://vpn-externe1.ulaval.ca/+CSCO+00756767633A2F2F6A6A6A2E667076726170727176657270672E70627A++/science/article/pii/S0926860X10003224#tbl2

135

Figure 5.8. N2 adsorption desorption isotherms of (A) FCC–FAU–TOL2D, (B) FCC–

FAU–FOR2D and (C) FCC–FAU–FOR4D.

136

5.3.3. Catalytic test

Before discussing the catalytic test results we should mention here two points: (i)

nanozeolite particles are the main active components of these FCC catalysts and their

activity in the cracking reaction is of our interest. The cracking of single hydrocarbon over

nanozeolite has been reported by several authors [12, 38, 39]. This kind of reaction can

provide a general suggestion on the potential of nanozeolite. However, it is necessary to

evaluate the activity of nanozeolites in real-life application; hence the cracking of a typical

feed for FCC cracking over nanozeolite-based catalysts was carried out. (ii) the matrix

component of our nanozeolite-based catalysts was deliberately made almost neutral

(amorphous silica) to the cracking reaction so that the impact of silica matrix as inactive

matrix on the overall activity of the catalyst is negligent. Consequently, the activity of the

catalysts can be supposed to stem from only the zeolite component. In a typical FCC

catalyst, matrix component also plays an active role in the cracking of large hydrocarbon,

contributing to the conversion as a whole [17]. Thus the activity of our nanozeolite-based

catalysts regarding conversion is expected to be lower than that of the commercial ones.

The relation between conversion and catalyst-to-oil ratio is shown in Figure 5.9. A

general trend can be observed. The conversion increased with the catalyst-to-oil ratio and

eventually it reached a plateau. This trend is explainable: as the catalyst-to-oil ratio rises the

number of active sites available for the cracking reaction becomes higher resulting in

higher conversion. As the ratio reaches a critical value, this effect is less pronounced; the

conversion approaches a steady state. In agreement with our anticipation, the conversion

over these catalysts is not very high. The highest value was observed on the sample FCC-

25 (nanozeolite size ~25 nm), which is about 50%.

137

Figure 5.9. Relationship between conversion and catalyst-to-oil ratio of different prepared

FCC-samples.

The most appealing conclusion drawn from the changed course of conversion is that

it clearly demonstrates the ability of the impact of nanoparticles on catalytic activity.

Reaction activity over the catalysts rises with the decrease in zeolite particle size. At the

same catalyst-to-oil ratio the catalyst that bears the smallest nanoparticle size has the

highest value among the three samples. Since the matrix components are identical and

neutral among the catalysts, the change in activity is attributed to the larger external surface

area, hence giving higher accessibility for large hydrocarbon molecules.

The correlation of dry gas with conversion is plotted in Figure 5.10. The dry gas is

the lightest fraction of the cracking reaction. It contains C1–C2 hydrocarbons and other light

gaseous molecules H2, H2S, CO and CO2 etc. Dry gas is undesired since it has low value

and hence its amount should be kept as low as possible. In most cases, dry gas is the

product of thermal cracking and the overcracking of gasoline. Thus, it is reasonable that an

increase in dry gas is observed at higher conversion. All the three catalyst samples

exhibited a very low production of dry gas. As the conversion reaches the maximum value,

10

20

30

40

50

60

2 4 6 8 10 12 14 16 18

C/O RATIO, g/g

MA

T C

ON

VE

RS

ION

, w

t %

FCC-100

FCC-40

FCC-25



138

the highest value of dry gas obtained over these catalysts is about only 2.5% wt. However,

at the same given conversion, the yield of dry gas is in the following order: FCC-

100 > FCC-40 > FCC-25. This order suggests that on nanozeolite, the secondary cracking

and thermal cracking reactions are subdued.

Figure 5.10. Correlation of dry gas yield with conversion of different prepared FCC-

samples.

LPG fraction is one of the products of cracking reaction that is valuable. The as-

produced LPG contains C3 and C4 hydrocarbons which can be used in the commercial LPG

and as a feedstock for further chemical upgrade to other chemicals of great value such as

the octane boosters: MTBE and ETBE etc. As shown in Figure 5.11, the LPG content

increased with the rise of the conversion. Although the profiles of the yield curve of LPG

over FCC-100 and FCC-40 are similar, in general, at a given conversion, the FCC-100 gave

the highest LPG yield, followed by FCC-40 and FCC-25.

1.0

1.2

1.4

1.6

1.8

2.0

2.2

2.4

2.6

2.8

20 25 30 35 40 45 50 55

MAT CONVERSION, wt %

DR

Y G

AS

, w

t %

FCC-100

FCC-40

FCC-25



139

Figure 5.11. Correlation of LPG yield with conversion of different prepared FCC-samples.

Gasoline is the objective of the FCC process. The relation between conversion and

gasoline yield is shown in Figure 5.12. The relation profiles clearly demonstrate the

advantage of nanozeolites. The catalyst containing smaller zeolite particles give higher

gasoline yield, and the yield gap among these catalysts rises with the increase of

conversion. Furthermore, an important parameter to evaluate the efficiency of an FCC

catalyst is the gasoline selectivity, which is defined as the ratio of gasoline to conversion.

The relationship between gasoline selectivity and conversion is shown in Figure 5.13.

Generally, the selectivity of gasoline decreased as the conversion increased. However, the

catalyst with smaller zeolite particles retained its higher selectivity.

2

4

6

8

10

12

14

16

20 25 30 35 40 45 50 55

LP

G, w

t %


FCC-100

FCC-40

FCC-25





140

Figure 5.12. Correlation between gasoline yield and conversion of different prepared FCC-

samples.

LCO (Light Cycle Oil) is the product of which the value changes seasonally. LCO

is used as the feedstock to be upgraded to diesel and/or fuel oils. In an ideal cracking

process, LCO is the intermediate product of a chain cracking reaction:

HCO => LCO => gasoline. LCO is both the product of the cracking of HCO and the

reactant for the cracking to gasoline. The content of LCO produced can be taken as a

parameter reflecting the competition between these two reactions. Cracking of large

molecule cannot be done inside zeolite pores due to the small opening of its pores. In

addition, the matrix component of the catalyst is essentially inactive. Hence, cracking of

LCO and HCO must be realized on the external surface of nanozeolites. Thus, the catalyst

offering more of external surface area would give higher efficiency in cracking of large

molecules. Taking into account the chain cracking scheme above, it is expected that, at the

low conversion, the content of large molecules that are likely to be cracked is high; hence,

the LCO produced rises as the conversion rises. However, with an increase of the

conversion, the source of these molecules depleted; hence, after a maximum value of

0

10

20

30

40

20 25 30 35 40 45 50 55

GA

SO

LIN

E,

wt %


FCC-100

FCC-40

FCC-25

141

conversion, the rate of the cracking of large molecule is exceeded by the rate of cracking

LCO; thus the content of LCO is decreased.

Figure 5.13. Relationship between gasoline selectivity and conversion of different prepared

FCC-samples.

Figure 5.14 shows the relationship between the LCO yield and the conversion. The

convex curves of the LCO profiles over the FCC-100 and FCC-25 are noticed. For the

FCC-40, the trend is different; the LCO yield exhibits a continuous decrease with the

increase of conversion. But there is a consistent order among these three catalysts: at a

given conversion, the yield of LCO is as follows: FCC-25 > FCC-40 > FCC-100.

40

45

50

55

60

65

70

20 25 30 35 40 45 50 55

GA

SO

LIN

E S

ELE

CT

IVIT

Y, w

t %


FCC-100

FCC-40

FCC-25

142

Figure 5.14. Relationship between LCO yield and conversion.

HCO fraction is the undesired product of the FCC process. It contains the aromatic

hydrocarbons that are difficult to crack and sulfur. Hence, HCO yield should be diminished

to minimum. The relationships between HCO and conversion are in agreement with our

expectation (Figure 5.15): HCO yield is reduced using nanosized zeolites; the smaller the

zeolite particles the lower the HCO yield.

Coke is an inevitable product and the only product that cannot be recovered. Being

the catalyst poison and apparently giving no value in commercial applications, the coke

formation is undesired and its amount should be as low as possible. The relationship

between coke yield and conversion is shown in Figure 5.16. The FCC-25 showed the least

coke selectivity among the three catalysts.

15

17

19

21

23

25

20 25 30 35 40 45 50 55

LC

O, w

t %


FCC-100

FCC-40

FCC-25





143

Figure 5.15. Relation between HCO yield and conversion of different prepared FCC-

samples.

In conclusion of the evaluation of FCC cracking, a clear trend has been noticed: the

activity increases with the decrease in crystal size of the nanozeolites. This is due to the fact

that cracking of FCC feed is heavily realized on external surface, which is higher on

nanozeolite. The activity of the catalyst as a whole (the conversion) was not very high; it is

deliberate since the matrix component was made neutral, and it is likely that the acidity of

nanozeolites is not sufficient. However, the addition of nanozeolite in FCC catalyst as main

component or additive is an interesting option.

20

30

40

50

60

20 25 30 35 40 45 50 55

HC

O, w

t %


FCC-100

FCC-40

FCC-25

144

Figure 5.16. Relationship between coke yield and conversion of different prepared FCC-

samples.

5.4. Conclusion

In this study, we have reported new methods of preparing nanozeolites using

toluene and formamide solvents as crystallization medium instead of water. Different

crystal sizes, e.g., 25, 40 and 100 nm, were prepared in toluene and formamide solvents. It

was demonstrated that the solvents play an important role in giving the zeolite crystal with

desired size. The key parameter that is important to choose the suitable solvent is its

solvating power and its hydrophobicity. Nanozeolite-based FCC catalysts were prepared

using silica as inactive matrix in order to study the effect of crystal size on the performance

of nanozeolites. These FCC catalysts were evaluated with FCC feedstock. The relationship

between gasoline selectivity and conversion is a function of nanozeolite size. In general, the

performance of these catalyst is in the following order FCC-25 > FCC-40 > FCC-100.

Acknowledgments

1

2

3

4

5

6

7

8

20 25 30 35 40 45 50 55

CO

KE

, w

t %


FCC-100

FCC-40

FCC-25

145

This work was supported by the Natural Sciences and Engineering Research

Council of Canada (NSERC) and the Research Application Development Society

(SOVAR-Quebec). G.T.V. thanks the foundation of Laval University (FUL) for the

scholarship. The authors thank Prof. S. Kaliaguine for stimulating discussions and

comments.

146

References

1. L. Tosheva, V.P. Valtchev, Chem. Mater., 17 (2005), pp. 2494–2513

2. A. Corma, Chem. Rev., 97 (1997), pp. 2373–2420

3. A. Corma, E. Sastre, J. Catal., 129 (1991), pp. 177–185

4. T.-O. Do, A. Nossov, M.-A. Springuel-Huet, C. Schneider, J.L. Bretherton, C.A. Fyfe,

S. Kaliaguine, J. Am. Chem. Soc., 126 (2004), pp. 14324–14325

5. T.-O. Do, S. Kaliaguine, Angew. Chem. Int. Ed., 40 (2001), pp. 3248–3251

6. T.-O. Do, S. Kaliaguine, Angew. Chem. Int. Ed., 41 (2002), pp. 1036–1040

7. A. Karlsson, M. Stöcker, R. Schmidt, Micropor. Mesopor. Mater., 27 (1999), pp. 181–

192

8. L. Yu, Z. Wenzhong, J.P. Thomas, Angew. Chem. Int. Ed., 40 (2001), pp. 1255–1258

9. Z. Zongtao, H. Yu, Z. Lei, W. Runwei, Y. Yi, Q. Shilun, Z. Dongyuan, X. Feng-Shou,

Angew. Chem. Int. Ed., 40 (2001), pp. 1258–1262

10. D. Breck, Zeolite Molecular Sieves: Structure, Chemistry, and Use, Wiley Interscience,

New York (1974)

11. W.M.M.Ch. Baerlocher, D.H. Olson, Atlas of Zeolite Framework Types, Elsevier

(2001)

12. B.W. Wojciechowski, A. Corma, Catalytic Cracking: Catalysts, Chemistry, and

Kinetics (1986)

13. E.F.S. Aguiar, M.L.M. Valle, M.P. Silva, D.F. Silva, Zeolites, 15 (1995), pp. 620–623

14. M.A. Camblor, A. Corma, A. Martínez, F.A. Mocholí, J.P. Pariente, Appl. Catal., 55

(1989), pp. 65–74

15. K. Rajagopalan, A.W. Peters, G.C. Edwards, Appl. Catal., 23 (1986), pp. 69–80

16. G.-T. Vuong, T.-O. Do, Micropor. Mesopor. Mater., 120 (2009), pp. 310–316

17. S. Al-Khattaf, Energy Fuels, 17 (2003), p. 62

18. R. Ravishankar, C. Kirschhock, B.J. Schoeman, P. Vanoppen, P.J. Grobet, S. Storck,

W.F. Maier, J.A. Martens, F.C. De Schryver, P.A. Jacobs, J. Phys. Chem. B, 102

(1998), pp. 2633–2639

19. P. Morales-Pacheco, F. Alvarez-Ramirez, P. Del Angel, L. Bucio, J.M. Dominguez, J.

Phys. Chem. C, 111 (2007), pp. 2368–2378

20. P. Morales-Pacheco, F. Alvarez, L. Bucio, J.M. DominguezJ. Phys. Chem. C, 113

(2009), pp. 2247–2255

21. T.M. Davis, T.O. Drews, H. Ramanan, C. He, J. Dong, H. Schnablegger, M.A.

Katsoulakis, E. Kokkoli, A.V. McCormick, R.L. Penn, M. Tsapatsis, Nat. Mater., 5

(2006), pp. 400–408

22. C.S. Cundy, P.A. Cox, Chem. Rev., 103 (2003), pp. 663–702

23. C.S. Cundy, J.O. Forrest, R.J. Plaisted, Micropor. Mesopor. Mater., 66 (2003), pp.

143–156

24. C.J.H. Jacobsen, C. Madsen, T.V.W. Janssens, H.J. Jakobsen, J. Skibsted, Micropor.

Mesopor. Mater., 39 (2000), pp. 393–401

25. I. Schmidt, C. Madsen, C.J.H. Jacobsen, Inorg. Chem., 39 (2000), pp. 2279–2283

26. H. Wang, B.A. Holmberg, Y. Yan, J. Am. Chem. Soc., 125 (2003), pp. 9928–9929

27. Z. Chen, S. Li, Y.S. Yan, Chem. Mater., 17 (2005), pp. 2262–2266

28. D.P. Serrano, J. Aguado, J.M. Escola, J.M. Rodriguez, A. Peral, Chem. Mater., 18

(2006), pp. 2462–2464

147

29. G. T. Vuong, T.-O. Do, in: US (Ed.), US, 2008.

30. G.T. Vuong, T.-O. Do, J. Am. Chem. Soc., 129 (2007), pp. 3810–3811

31. S. Qiu, J. Yu, G. Zhu, O. Terasaki, Y. Nozue, W. Pang, R. Xu, Micropor. Mesopor.

Mater., 21 (1998), pp. 245–251

32. A. Lattes, E. Perez, I. Rico-Lattes, C. R. Chim., 12 (2009), pp. 45–53

33. M. Almgren, S. Swarup, J.E. Loefroth, J. Phys. Chem., 89 (1985), pp. 4621–4626

34. I. Rico, A. Lattes, J. Phys. Chem., 90 (1986), pp. 5870–5872

35. G. Coudurier, C. Naccache, J.C. Vedrine, J. Chem. Soc., Chem. Commun. (1982), pp.

1413–1415

36. B.Z. Zhan, M.A. White, M. Lumsden, Langmuir, 19 (2003), pp. 4205–4210

37. B.Z. Zhan, M.A. White, P. Fancy, C.A. Kennedy, M. Lumsden, Macromolecules, 37

(2004), pp. 2748–2753

38. P.B. Venuto, E.T. Habib Jr. Fluid Catalytic Cracking with Zeolite Catalysts, CRC

Press (1979)

39. B.C. Gates, J.R. Katzer, G.C.A. Schuit Chemistry of Catalytic Processes, McGraw-Hill

College (1979)

149

Chapter 6. Synthesis and Engineering Porosity of Mixed

Metal Fe2Ni-MIL-88B Metal-Organic Framework

Gia-Thanh Vuong, Minh-Hao Pham and Trong-On Do *


Published in Dalton Transactions 2013, 42 (2), 550-557

DOI: 10.1039/C2DT32073H

Résumé

Une nouvelle approche a été développé pour la synthèse de Fe2Ni MIL-88B en

utilisant des clusters neutres de métaux mixtes, Fe2Ni(µ3-O). Ces clusters occupent les

nœuds du réseau MIL-88B, au lieu du mono-métal, Fe3 (µ3-O) avec un anion compensateur

présent dans le matériau Fe3MIL-88B non-poreux qui est obtenu par la méthode

conventionnelle. De ce fait, en absence des anions compensateurs dans la structure, Fe2Ni

MIL-88B devient un matériau poreux. De plus, la combinaison de la flexibilité de MIL-

88B et des métaux mixtes comme nœuds dans le réseau, la porosité peut être contrôlée par

échange des ligands terminaux du réseau. Ceci nous a permis de moduler d‟une manière

réversible la porosité ainsi que la surface spécifique du Fe2Ni MIL-88B à différents

niveaux, dépendamment de la taille des ligands échangés.

Abstract

A new rational approach has been developed for the synthesis of mixed metal MIL-

88B metal organic framework based on neutral mixed metal cluster, such as Fe2Ni(µ3-O)

cluster. Unlike the conventional negative charged single metal cluster, the use of neutral

mixed metal cluster as nodes in the framework avoids the need of compensating anion

inside porous MIL-88B system; thus mixed metal MIL-88B becomes porous. The

flexibility of the mixed metal MIL-88B can be controlled by terminal ligands with different

steric hindrance. This allows us to reversibly customize the porosity of MIL-88B structure

at three levels of specific surface area as well as the pore volume




150

6.1. Introduction

MOFs (Metal-organic frameworks) are ordered structures of metal clusters

connected by organic linkers.1 This combination of an inorganic entity and an organic one

can result in highly porous MOF crystals which show promising applications in adsorption,

catalysis and drug delivery.2-6

A classic example is the combination of trimeric M3(µ3-O)

clusters (M = Fe, Cr) with benzenedicarboxylate which gives rise to MIL-101, MIL-88B

and MOF-235 structures of which only MIL-101 shows porosity.7-13

As the metal is

trivalent, a guest anion is needed to balance the charge of the cluster14

. With regard to

porosity, this anion is not desired. It blocks the pores, rendering the MOF structure non-

porous as in the case of MIL-88B and MOF-235.9,11

Only MIL-101 which has sufficiently

large pore size can afford the high porosity.10

In return, the dense MIL-88B structure is not

rigid but flexible, the network can expanse upon adsorption of particular solvents. Such a

great increase of 125% of unit cell volume can be obtained.10,15

This distinctive feature

coined breathing effect by Férey et al. is of great interest since it provides a facile control

over the porosity by changing solvents.10,15

Unfortunately, the flexibility of MIL-88B

structure is not permanent without solvent molecules. Solvent molecules must remain in its

pores to sustain the expansion hence despite there is a gain in pore size and pore volume,

the pores are filled with solvent molecules and thus become inaccessible for adsorbates.

Upon removal of solvents, the structure shrinks, and MIL-88B returns to its dense state. A

workaround to sustain the porosity of MIL-88B structure has been recently reported.16

Instead of the straight and simple linker benzenedicarboxylate, functional groups were

introduced to the phenyl ring of the linker thus giving a steric hindrance effect to the linker.

The obtained series of functionalized MIL-88B can resist the shrinkage, and one of them,

MIL-88B(4CH3), can yield specific area up to 1200 m2/g.

16 However the reversibility of the

breathing effect was not reported by the authors. And it is likely that the breathing

magnitude in functionalized MIL-88B could become less due to the hindrance effect of the

functional group, the structure is more likely “fixed”.

On the other hand, most of MOFs are based on single metals.17,18

Many properties

of MOFs are dependent on the metal component such as the stability, magnetic behavior,

density etc. Therefore, a successful preparation of mixed metal MOFs would provide an

151

effective way to fine tune the properties of MOFs. Direct syntheses of mixed metal MOFs

are difficult, due to the high complexity upon addition of a second metal. Each metal can

form separate clusters irrelevant and unconnected to the other metal clusters, resulting in a

mixture of discrete MOFs. Rare-earth metals with high and flexible coordination capacity

are able to join together with transition metals in some MOF structures.19-21

However, it is

not the case for main group and transition metals. The syntheses of mixed transition metal

MOFs often employ linkers bearing free reactive functional groups that can coordinate with

other metal in a pre- or post-synthesis treatment.22-28

In this fashion, the second metal

contributes nothing to the construction of the framework structure, but resides within it as

an attachment. Few attempts which involve the selective introduction of the second metal

are reported to include both metals in the nodes of the framework. In these mixed metal

MOFs, the transition metal and the other one are distributed periodically among the nodes

of the framework.18

Although mixed metal clusters such as trimeric metal carboxylate have been

reported since 1911,14,29

however, to the best of our knowledge, there has been no report on

MOF structures built on them. In fact, synthesis and characterization of trimeric mixed

metal acetates of Fe(III) and first row divalent transition metal such as Co, Mn, Ni and Zn

have been well established.14

Thanks to the presence of the divalent metal, the mixed metal

cluster is balanced in charge without a compensating anion. Judging from the similarity

between the mixed metal cluster and the single metal cluster, we believe that it is possible

to obtain analogs of those MOF structures such as MIL-101, MIL-88 from these mixed

metal trimeric clusters. Because no compensating anion is needed in the framework, the

pore blockage by anions is avoided, and the mixed metal MOF could become porous

(Scheme 6.1).

152

Scheme 6.1. Mixed metal building unit forming porous MOF. Black: C, yellow: divalent

metal (Ni), green: Fe, olive: balancing anion (Cl, Br etc.), Red: O

153

Scheme 6.2. Reversible ligand exchange and porosity control of Fe2Ni-MIL-88B, for

clarity, only terminal ligands bonding to Ni are showed.

In this paper, we report a new approach for the synthesis of mixed metal MIL-88B

using mixed metal cluster (Scheme 6.1) and the control of the breathing of the obtained

mixed metal MOF via the steric hindrance of the terminal ligands (Scheme 6.2). To

illustrate our approach, we have selected the synthesis of the MIL-88B structure based on

Fe2Ni(µ3-O) cluster (denoted as Fe2Ni-MIL-88B). Our Fe2Ni-MIL-88B product shows

drastic change over the original single metal Fe3-MIL-88B in regard to porosity, and

154

considerately high N2 and CO2 adsorption. But the most interesting feature of Fe2Ni-MIL-

88B is its switchable porosity and specific surface area (30 m2/g to 1120 m

2/g) and pore

volume (10 x 10-3

cm3/g to 448 x 10

-3 cm

3/g). Unlike the conventional single metal Fe3-

MIL-88B, Fe2Ni-MIL-88B can switch reversibly and permanently from dense state to

porous one upon ligand exchange with certain terminal ligands.

6.2. Experiments

Sample preparation: In a typical synthesis, 0.67 mmol of FeCl3.6H2O 99%, 0.33

mmol of corresponding Ni(NO3)2.6H2O 97% and 1 mmol of bdc 98% were dissolved in 10

ml of DMF. To this clear solution, 0.4 mmol of NaOH was added under stirring for 15 min.

The mixture was then transferred into a Teflon-lined autoclave and heated at 100 oC for 15

h. Solid product was then recovered by filtration and washed several times with DMF.

Elemental analysis of the synthesized sample showed Fe 12.5 wt%, Ni 6.2 wt%, N 5.3

wt%. Thus, the suggested formula Fe2NiO(OOC-C6H4-COO)3.3DMF is designated as

Fe2Ni-MIL-88B.DMF. The sample was treated with water, pyridine (Py), pyrazine (Pz) and

4-,4‟-bipyridine (Bp) to obtain Fe2Ni-MIL-88B.H2O, Fe2Ni-MIL-88B.Py, Fe2Ni-MIL-

88B.Pz and Fe2Ni-MIL-88B.Bp, respectively (see Supporting Information). For

comparison, the single metal Fe3-MIL-88B.DMF was also prepared using the procedure of

Férey et al.30

Characterization: N2 and CO2 adsorption tests were carried out in an Autosorb 1

instrument, before analysis the samples were outgassed in vacuum for 3 hours at 150 oC.

Specific surface area was calculated with the BET model in the linear range of P/Po =

0 – 0.15. FTIR was carried in a FT-BIORAD 450s system using KBr disc. UV-VIS was

carried in a Cary 300 instrument using MgO disc. Powder X-ray diffraction (XRD) patterns

were collected on a Bruker SMART APEX II X-ray diffractometer with Cu Kα radiation (λ

= 1.5406 Å) in the 2θ range of 1 – 50° at a scan rate of 1.0° min–1

. For XRD measurement

of samples in Figure 1 and for crystal lattice calculation, the samples were dried in vacuum

overnight at 100 oC, then the analysis was taken immediately. Scanning electron

microscopy (SEM) images were taken on a JEOL 6360 instrument at an accelerating

voltage of 3 kV.

155

6.3. Results

6.3.1. Synthesis of Mixed Metal Fe2Ni-MIL-88B with Different Terminal

Ligands

Details of the synthesis and ligand exchanges are described in the experimental

section and in Supporting Information. Elemental analysis of the synthesized sample

showed Fe 12.5 wt%, Ni 6.2 wt%, N 5.3 wt%. Thus, the suggested formula Fe2NiO(OOC-

C6H4-COO)3.3DMF is designated as Fe2Ni-MIL-88B.DMF. The sample was treated with

water, pyridine (Py), pyrazine (Pz) and 4-,4‟-bipyridine (Bp) to obtain Fe2Ni-MIL-

88B.H2O, Fe2Ni-MIL-88B.Py, Fe2Ni-MIL-88B.Pz and Fe2Ni-MIL-88B.Bp, respectively.

For comparison, the pristine Fe3-MIL-88B.DMF was also prepared using the procedure of

Férey et al.30

2

5 10 15 20 25 30 35 40 45 50

(a)

(b)

Figure 6.1. XRD patterns of Fe2Ni-MIL-88B.H2O (a) and XRD simulation of the Fe3-MIL-

88B (b).

156

The XRD patterns of all of Fe2Ni-MIL-88B samples were collected, however, due

to the flexibility of the MIL-88B structures, which will be explained in the next session, the

best way to determine the structure of the samples is to compare the completely dense

phase Fe2Ni-MIL-88B.H2O with the XRD simulation of the standard Fe3-MIL-88B. As

showed in Figure 6.1, the XRD of the Fe2Ni-MIL-88B.H2O is identical to the simulation

one, in addition no guest phase is found, implying the MIL-88B structure with the high

purity of the sample.

Table 6.1. IR analysis of the MIL-88B samples

Sample

Wavenumber (cm-1)

asymC sym(OCO) 31 DMF 32,33 Py 34-36 Pz 34 Bp 35 Fe2NiO/Fe3O 37,38

Fe2Ni-MIL-88B.Py 1606 1382 224 1486,

1447,

703 633

718

Fe2Ni-MIL-88B. H2O 1590 1381 209 717

Fe2Ni-MIL-88B.DMF 1609 1385 224 1660 718

Fe2Ni-MIL-88B.Pz 1603 1383 220 1416 717

Fe2Ni -MIL-88B.Bp 1609 1385 224 1431,

1408, 635

718

Fe3-MIL-88B 1601 1393 208 1666 624

The FTIR analysis of the samples is in agreement with the suggested MIL-88B

formula (for details, see Table 6.1 and Figure S1-S6, Supporting Information). There are

four remarks: (i) no free acid (no band at 1700 cm-1

) is found, (ii) the value = asym(OCO)

–sym(OCO) corresponds well to the bridge coordination mode of metal carboxylate,31

(iii)

the bands characteristic of H2O, DMF, Py, Pz and Bp are present on the samples Fe2Ni-

MIL-88B.H2O, Fe2Ni-MIL-88B.DMF, Fe2Ni-MIL-88B.Py, Fe2Ni-MIL-88B.Pz and Fe2Ni-

MIL-88B.Bp, respectively; and (iv) the vibration of cluster Fe2Ni(µ3-O) (~718 cm-1

) is

observed, while the vibration of Fe3(µ3-O) (620cm-1

) is only noticed in the single metal

Fe3-MIL-88B sample.37,38

157

Figure 6.2. UV-Vis spectra of Fe2Ni-MIL-88B.Bp (a), Fe2Ni-MIL-88B.Pz (b), Fe2Ni-

MIL-88B.Py (c), Fe2Ni-MIL-88B.DMF (d) Fe2Ni-MIL-88B.H2O (e) and Fe3-MIL-88B (f)

158

Table 6.2. Crystal parameters of the Fe2Ni-MIL-88B samples

Sample a [Å] c [Å] Unit cell volume

[Å3] V/V (%)[a]

Fe2Ni-MIL-88B.H2O 11.2 19.1 2075 ~

Fe2Ni-MIL-88B.DMF

- Open phase 13.7 17.8 2893 39.4

- Dense phase 11.3 19.1 2112 1.8

Fe2Ni-MIL-88B.Py

- Open phase 14.2 17.4 3039 46.4

- Dense phase 11.3 18.5 2046 -1.4

Fe2Ni-MIL-88B.Pz

- Open phase 14.4 17.8 3197 54.1

- Dense phase 11.0 19.0 1991 -4.0

Fe2Ni-MIL-88B.Bp 14.1 17.4 2996 44.4

[a] The increase in unit cell volume compared with the Fe2Ni-MIL-88B.H2O sample

The UV-Vis spectra of the Fe2Ni-MIL-88B samples with different terminal ligands

are shown in Figure 6.2. The UV-Vis analysis also confirms the presence of Ni as well as

its octahedral coordination mode in the MIL-88B structures. The band at 244 nm is

assigned to the ligand-to-metal charge transfer, implying the bonding of carboxylate

oxygen to metal. Most of the transition bands of Ni2+

are obscured by those of Fe3+

.

However, the presence of the band at 760 nm, which is characteristic of the transition [3A2g

=> 1Eg(D)] of Ni in the tri-nuclear complex is observed. The transition [

6A1g =>

4A1g +

4Eg(G)] in Fe

3+ is also found at 350 - 500 nm.

39,40 Especially, the [

6A1g =>

4T2g] transitions

at 550 - 650 nm of Fe3+

, which are sensitive to the ligand field energy, clearly reveal the

bonding of the terminal ligand to the metal and the effect of the divalent metal Ni in the

complex. These transitions of the samples Fe2Ni-MIL-88B.H2O and Fe2Ni-MIL-88B.DMF

which involve the weak field terminal ligand H2O and DMF, are observed at 575 nm.

Accordingly, for the samples involving the strong field ligands Py, Pz and Bp, these bands

are shifted to lower energy (higher wavelength at 625 nm). This behavior is in agreement

with the higher ligand field energy of Py, Pz, and Bp than that of H2O and DMF.40

In the

Fe2Ni complex under the effect of Ni, the ligand field in Fe reduces,39

thus the [6A1g =>

4T2g] transition in Fe3-MIL-88B.DMF is at 525 nm while it is observed at 575 nm in the

Fe2Ni-MIL-88B.DMF. As seen in Figure S7, Supporting Information, the change in color

159

of the samples depends on the nature of terminal ligand (a-d) as well as on the presence of

Ni in the structure (d, f).

Figure 6.3. XRD patterns of Fe2Ni-MIL-88B samples, the planes of open phase are in

black, the planes of dense phase are in red and placed in boxes. Fe2Ni-MIL-88B.Bp (a),

Fe2Ni-MIL-88B.Pz (b), Fe2Ni-MIL-88B.Py (c), Fe2Ni-MIL-88B.DMF (d) and Fe2Ni-MIL-

88B.H2O (e)

6.3.2. Reversible Breathing Control Using Terminal Ligand

The breathing behavior of the MIL-88B has been well documented by Férey et al.8

and can be studied in details by investigation of the XRD patterns in the 2 range from 7 to

12o. Figure 6.3 shows the XRD patterns of the mixed metal samples with different terminal

ligands, which also confirms the MIL-88B structure. For comparison the XRD pattern of

160

Fe3-MIL-88B is also displayed. As Férey et al. have reported,8 the swelling up of MIL-88B

structure causes the splitting and shifting to low 2 of the plane (100) and (101), while the

plane (002) is shifted to higher 2. Thus these planes can be used as the indicators of the

swollen (open) phase as well as the dense phase in the samples. In Figure 6.3, the planes

assigned to the dense phase are in red and in box, while the ones assigned to the open phase

are in black and without box. For the Fe2Ni-MIL-88B.H2O sample, only the dense phase is

observed. The Fe2Ni-MIL-88B.DMF, Fe2Ni-MIL-88B.Py and Fe2Ni-MIL-88B.Pz samples

feature both open phase and dense phase. The Fe2Ni-MIL-88B.Bp sample exhibits an open

structure without any trace of the dense phase. Assuming the Fe2Ni-MIL-88B samples

taking the same hexagonal lattice structure as the original Fe3-MIL-88B, calculations of the

unit cell parameters using the assigned planes in Figure 6.3 were carried out. The results

listed in Table 6.2 are in agreement with the plane assignment. Upon swelling, the lattice

constant a increases from 11.0 Å to 14.4 Å while the c constant decreases from 19.1 Å to

17.5 Å. Consequently, these changes are reflected in the unit cell volume. The unit cell

volume of the open phase is about 40 - 50 % higher than that of the dense phase.

Obviously, the XRD patterns have demonstrated the effect of the terminal ligand on the

swelling degree of the samples. In terms of steric hindrance, the terminal ligands can be

classified into three groups: low hindrance of H2O, intermediate of DMF, Py and Pz and

high hindrance of Bp. The small ligand water with low steric hindrance hence cannot swell

up the structure. In fact, the Fe2Ni-MIL-88B.H2O sample exhibits a dense phase. The

introduction of larger ligands DMF, Py and Pz with higher steric hindrance brings about the

openness of the structure of the obtained Fe2Ni-MIL-88B.DMF, Fe2Ni-MIL-88B.Py and

Fe2Ni-MIL-88B.Pz samples, but the openness is not full yet as some of the dense phase still

remains. The samples show both open and dense phases. Eventually, the ligand Bp, which

has the largest size and the steric hindrance among the terminal ligands used in this study,

yields the fully opened Fe2Ni-MIL-88B.Bp.

6.3.3. Adsorption Analysis

Figure 6.4A shows the nitrogen adsorption isotherms of the Fe2Ni-MIL-88B

samples with different terminal ligands, which are also in agreement with the XRD results.

All the isotherms of the Fe2Ni-MIL-88B samples show the characteristic of microporous

161

materials with the isotherm reaching a plateau at very low partial pressure. As the steric

hindrance of the terminal ligand increases from the small size of water to the middle size of

DMF, Py, Pz and finally to the large size of Bp, the BET specific surface area of the

samples also exhibits three levels: (i) dense (non-porous to N2), 30 m2/g for Fe2Ni-MIL-

88B.H2O; (ii) porous, 355 - 550 m2/g for Fe2Ni-MIL-88B.DMF, Fe2Ni-MIL-88B.Py,

Fe2Ni-MIL-88B.Pz; and (iii) highly porous, 1120 m2/g for Fe2Ni-MIL-88B.Bp. The same

trend for their micropore volumes is also observed, showing three levels of pore volume. If

the micropore volume is taken as a factor measuring the breathing, then the pore volume of

the Fe2Ni-MIL-88B.Bp sample is about 44 times higher than that of the Fe2Ni-MIL-

88B.H2O sample. Interestingly, the micropore diameters calculated by the HK model41

are

also consistent with the steric hindrance of the corresponding terminal ligands (Figure

6.4B). The Fe2Ni-MIL-88B.Bp sample shows an average pore size of 6.3 Å, the highest

value among the samples. The Fe2Ni-MIL-88B.DMF, Fe2Ni-MIL-88B.Py and Fe2Ni-MIL-

88B.Pz samples exhibit the value of 5.2 Å, while the pore volume of the dense structure

Fe2Ni-MIL-88B.H2O is negligible. Details of the porosity and the specific surface area of

the samples are shown in Table 6.3.

In addition to N2 adsorption, the CO2 adsorption measurements at low pressure (up

to 1 atm) at 273 K were also carried out (Figure 6.5). Again, the CO2 adsorption capacity of

the samples increases with the increase in porosity of the samples. The Fe2Ni-MIL-88B.Bp

sample shows the highest capacity, 101 cc/g STP of CO2, corresponding to a capacity of 20

wt %. This is one of the highest values (at 273 K at 1 atm) among MOF materials to date.42

The high CO2 adsorption capacity of the Fe2Ni-MIL-88B.Bp could be also due to the

presence of free pyridyl group of Bp in Fe2Ni-MIL-88B.Bp, which has a high affinity with

CO2.

162

P/Po

0.0 0.2 0.4 0.6 0.8 1.0

Vo

lum

e a

ds

orb

ed

[cc/g

]

0

100

200

300

400

(a)

(b)

(c)

(d)

(e)

Pore diameter [Å]

4 5 6 7 8

dD

(w)

[a.u

.]

(a)

(b)

(c)

(d)

(e)

(A) (B)

Figure 6.4. N2 adsorption isotherms at 77 K (A) and pore size distributions (B) of Fe2Ni-

MIL-88B.Bp (a), Fe2Ni-MIL-88B.Pz (b), Fe2Ni-MIL-88B.Py (c), Fe2Ni-MIL-88B.DMF (d)

and Fe2Ni-MIL-88B.H2O (e).

163

P [torr]

0 76 152 228 304 380 456 532 608 684 760

Volu

me a

dsorb

ed [

cc/g

]

0

20

40

60

80

100(a)

(b)

(c)

(d)

(e)

Figure 6.5. CO2 adsorption isotherms at 273 K of Fe2Ni-MIL-88B.Bp (a), Fe2Ni-MIL-

88B.Pz (b), Fe2Ni-MIL-88B.Py (c), Fe2Ni-MIL-88B.DMF (d) and Fe2Ni-MIL-88B.H2O (e)

Table 6.3. Porosity of Fe2Ni-MIL-88B

Samples Specific surface area

[m2/g]

Micropore volume

[10-3

cm3/g]

Pore width [Å]

Fe2Ni-MIL-88B.H2O 30 10 ~

Fe2Ni-MIL-88B.DMF 355 140 5.2

Fe2Ni-MIL-88B.Py 549 216 5.2

Fe2Ni-MIL-88B.Pz 465 186 5.2

Fe2Ni-MIL-88B.Bp 1120 448 6.3

6.4. Discussion

It is clear from the XRD analysis that our synthesized samples have the MIL-88B

structure with high purity. Both FTIR and UV-Vis analyses and elemental analysis indicate

the presence of Ni in the mixed metal MIL-88B samples. Interestingly, the spectroscopy

164

techniques provide us with unambiguous evidence of the presence of Fe2NiO cluster which

is the building unit of the mixed metal MIL-88B. The presence of Fe3O cluster which

implies Fe3-MIL-88B was not detected by FTIR and UV-Vis analyses of the mixed metal

samples. Hence this suggests the mixed metal Fe2Ni-MIL-88B structure.

Mixed metal Fe2Ni-MIL-88B exhibits stable, controllable and permanent porosity,

much better than single metal Fe3-MIL-88B. The spectroscopy results revealed that the

substituting ligands Py, Pz, and Bp are bound to the framework via a chemical bond to the

metal atom, not physically packing. In fact, DMF, H2O, Py, Pz and Bp are the terminal

ligands in the Fe2NiO cluster. The XRD analysis and adsorption isotherm measurements of

the samples Fe2Ni-MIL-88B.H2O, Fe2Ni-MIL-88B.DMF, Fe2Ni-MIL-88B.Py, Fe2Ni-MIL-

88B.Pz and Fe2Ni-MIL-88B.Bp demonstrate that the porosity of the mixed metal Fe2Ni-

MIL-88B can be controlled by the size of the terminal ligands. Since the terminal ligand is

chemically bonded to the metal rather than a weak physical packing, the impact is

permanent. The samples retained their porosity after various types of treatment: drying in

vacuum at 150 °C for 24h, and/or exposure to air at room temperature for days. In contrast,

this behavior was not observed on single metal Fe3-MIL-88B treated with Py or Bp under

the same conditions; no improvement in porosity was found on Fe3-MIL-88B after the Py

exchange. This drastic difference in the breathing behavior between the Fe2Ni-MIL-88B

and original Fe3-MIL-88B stems from their different breathing mechanisms. While

breathing effect in Fe3-MIL-88B originates from the packing of solvent molecules, this

effect in Fe2Ni-MIL-88B comes from the bonding of neutral ligands to the trinuclear

clusters. The “swelling up” in Fe2Ni-88B would be dependent on the size of the ligand and

its orientation in bonding with the ligand. The larger the ligand is the higher is the swelling

up. And since bonding to each metal atom in MIL-88B structure is a terminal ligand (H2O

or DMF), which is ready to be replaced, it is possible to introduce larger terminal ligand

into the MIL-88B structure which helps retain the breathing effect. Common terminal

ligands for trimeric clusters are pyridine and its derivatives such as pyrazine and 4,4‟-

bipyridine,14,43

which have different steric hindrances and thus can be used as the breathing

agent of the MIL-88B structure, as illustrated in Scheme 7.2.

165

It should be noted that for the original single metal Fe3-MIL-88B, the activation

conditions affect strongly the specific surface area. For example badly activated Fe3-MIL-

88B(CH3) solid that still contains some DMF, exhibit a significant BET surface area while

the fully water exchange sample is non porous when dried.16

This behaviour is attributed to

the dependency of Fe3-MIL-88B on the solvent to retain its porosity, as long as the solvent

is present, the pores are opened. Thus solvent molecule is the pore opening agent for Fe3-

MIL-88B. In case of Fe2Ni-MIL-88B it is the terminal ligand that keeps the pores open and

since the terminal ligand is chemically bonded to the framework, it is expected that the

porosity of Fe2Ni-MIL-88B would be more stable. However, a more quantitative work such

as PXRD simulation will be necessary to confirm the mechanism of the pore opening in

mixed metal MIL-88B.

For rigid structure, addition of guest molecule into the pores always results in

decrease in surface area; this is due to the fact that the guest molecule will partially block

the pores, thus reducing the accessible surface. For example the attachment of chiral

organic ligand to Cr3-MIL-101 reduces the surface area by 70%.44

However this trend is

reversed in mixed metal Fe2Ni-MIL-88B, the larger the ligand is, the higher the surface

area is obtained. It is the flexibility of the mixed metal Fe2Ni-MIL-88B that gives this

interesting behavior. As the XRD patterns shows (Figure 7.3), the large ligand can trigger

the mixed metal Fe2Ni-MIL-88B framework to expand further, opening the pores, thus

gaining surface area.

The secondary divalent metal (Ni) in the mixed metal Fe2Ni-MIL-88B is also very

important to the obtained structure. As discussed earlier, the introduction of the divalent

metal helps avoid the need of the blocking anions, in addition the bond strength of the

cluster to the terminal ligand can be improved by selection of the second divalent metal

which has higher affinity to the ligand. For the trimeric mixed metal Fe2III

MII(µ3-O)(µ2-

O2CCH3)6L3 with L being pyridine derivative terminal ligand and M divalent metal,

Novitchi et al.43

found that the stability of the bond Ni-L is 44 times higher than Fe-L; and

Co-L is 6 times higher than Fe-L. Moreover, in a classic paper by Irving and Williams the

stability constants of the pyridine-based [MII(Py)(H2O)5]

2+ follow the order: Mn

2+(0.14) <

Fe2+

(0.6) < Co2+

(1.14) < Ni2+

(1.78). 45

In fact, the improvement in stability of Ni-based

166

complexes has been well established, thanks to the maximal crystal field stabilization

energy of Ni2+

.45

Hence, it is suggested that the mixed metal MOF based on Fe2Ni(µ3-O)

cluster would exhibit stronger binding affinity to pyridine-like terminal ligands. Moreover,

taking into account the anion-free state of the mixed metal MOF, the ligand can orient with

less restriction inside the pores to attain a proper bond to the metal.

Beside the exceptional advantage of permanent porosity, the use of terminal ligand

as a swelling agent has another fascinating feature: reversibility. The terminal ligand is

exchangeable without affecting the linkers and the nodes of the framework. As illustrated

in Scheme 2, the samples of Fe2Ni-MIL-88B.DMF, Fe2Ni-MIL-88B.H2O, Fe2Ni-MIL-

88B.Py, and Fe2Ni-MIL-88B.Pz can be mutually converted with the corresponding ligand

exchange. The XRD and N2 adsorption isotherm results show that the samples retain their

crystallinity as well as porosity after the various conversion cycles. The Fe2Ni-MIL-88B.Bp

sample can be converted to Fe2Ni-MIL-88B.DMF and then from this state it can be

converted to bear any other ligand of Py, Pz, DMF and H2O. It means that the porosity of

Fe2Ni-MIL-88B can be switched in-situ to yield the necessary porosity, one of the desired

features of smart porous materials.

Unlike zeolites and mesoporous inorganic materials regarded as rigid and fixed-

pore structures, the reversible change in pore size of Fe2Ni-MIL-88B is achieved by the

breathing of the whole structure with the use of the terminal ligands to sustain it.

Consequently, Fe2Ni-MIL-88B can also provide at least three different states of porosity

(Scheme 6.2). In fact, breathing can bring about 300 % change in volume in some MOFs.

In addition, the terminal ligands are available in many sizes as well as the bond strength.

Hence we believe the pore size control in MOFs is much easier and higher in magnitude

and importantly not exclusive to Fe2Ni-MIL-88B, but available to other MOFs capable of

breathing.

Another general advantage of trimeric mixed metal is that the ability to tether

functional groups to the MOF structure. In this study, with the use of Pz and Bp, the free

pyridyl group becomes available in MIL-88B structure. It is very likely that other

functional groups such as carboxyl, aldehyde can be introduced in the same manner using

the corresponding ligands such as nicotinic acid and pyridinecarboxaldehyde. The key

167

point about trimeric mixed metal MOFs is that the functional group can be prepared

separately, and then attaches to the MOF structure as a removable module, the MOF hence

becomes a flexible docking station for various types of functional modules.

6.5. Conclusion

In the field of adsorption and membrane technologies, a smart material which can

switch itself from high to low porosity or vice versa could be extremely beneficial. Imagine

an adsorbent that can take in desired gas molecules with high capacity, but it can also

become inaccessible for them upon terminal ligand exchange; or a membrane which can let

certain gas molecules pass through can be shut off entirely to them if necessary. Many

applications could benefit from this kind of “on and off” materials.

We regard the Fe2Ni-MIL-88B material as one step toward the creation of a truly

smart porous material. Its versatility lies in its switchable and reversible three-level

porosity. Different levels could be attained depending on the nature of stimulant terminal

ligand. Their obvious applications are of course adsorption and separation in which the

pore size of the material can be reversibly controlled to be wide opened, half opened or be

completely closed at will.

In conclusion, we have succeeded synthesizing MIL-88B structure based on mixed

metal cluster of Fe(III) and Ni(II). This mixed metal cluster helps bring in the porosity to

the MOF product and an exact control over the porosity and surface area by using simple

stimulant terminal ligands. We believe that this rationale approach is not restricted to the

MIL-88B structure but it can cover all other flexible MOF structures which are based on

trinuclear metal carboxylates MIL-101, MIL-100, MIL-88 and MOF-235 etc. With the rich

choices of the mixed metal clusters from a large collection of trimeric mixed metal cluster

M2III

MII (M

III: Fe, Cr, Mn, Rh and M

II: Ca, Ba, Mg, Ni, Mn, Co) and the great selections of

terminal ligands among N-based, S-based O-based ligands new fascinating properties of

trimeric mixed metal MOFs are more to come

† Electronic Supplementary Information (ESI) available: Details of synthesis, XRD

patterns, FTIR spectra SEM images and reversible ligand exchange. See

DOI: 10.1039/b000000x/

168

Refrences

1. O. M. Yaghi, M. O'Keeffe, N. W. Ockwig, H. K. Chae, M. Eddaoudi and J. Kim,

Nature, 2003, 423, 705-714.

2. A. Corma, H. Garc a and F. X. Llabrés i Xamena, Chem. Rev., 2010, 110, 4606-4655.

3. J. Lee, O. K. Farha, J. Roberts, K. A. Scheidt, S. T. Nguyen and J. T. Hupp, Chem. Soc.

Rev., 2009, 38, 1450-1459.

4. J.-R. Li, R. J. Kuppler and H.-C. Zhou, Chem. Soc. Rev., 2009, 38, 1477-1504.

5. L. J. Murray, M. Dinca and J. R. Long, Chem. Soc. Rev., 2009, 38, 1294-1314.

6. P. Horcajada, T. Chalati, C. Serre, B. Gillet, C. Sebrie, T. Baati, J. F. Eubank, D.

Heurtaux, P. Clayette, C. Kreuz, J.-S. Chang, Y. K. Hwang, V. Marsaud, P.-N. Bories,

L. Cynober, S. Gil, G. Ferey, P. Couvreur and R. Gref, Nat Mater, 2010, 9, 172-178.

7. M.-H. Pham, G.-T. Vuong, A.-T. Vu and T.-O. Do, Langmuir, 2011, 27, 15261-15267.

8. C. Serre, C. Mellot-Draznieks, S. Surble, N. Audebrand, Y. Filinchuk and G. Ferey,

Science, 2007, 315, 1828-1831.

9. S. Surble, C. Serre, C. Mellot-Draznieks, F. Millange and G. Ferey, Chem. Commun.,

2006, 284-286.

10. G. Ferey, C. Mellot-Draznieks, C. Serre, F. Millange, J. Dutour, S. Surble and I.

Margiolaki, Science, 2005, 309, 2040-2042.

11. A. C. Sudik, A. P. Côté and O. M. Yaghi, Inorg. Chem., 2005, 44, 2998-3000.

12. J. H. Yoon, S. B. Choi, Y. J. Oh, M. J. Seo, Y. H. Jhon, T.-B. Lee, D. Kim, S. H. Choi

and J. Kim, Catal. Today, 2007, 120, 324-329.

13. M.-H. Pham, G.-T. Vuong, F.-G. Fontaine and T.-O. Do, Crystal Growth & Design,

2011, 12, 1008-1013.

14. R. D. Cannon and R. P. White, in Prog. Inorg. Chem., John Wiley & Sons, Inc., 1988,

vol. 36, pp. 195-298.

15. G. Ferey and C. Serre, Chem. Soc. Rev., 2009, 38, 1380-1399.

16. P. Horcajada, F. Salles, S. Wuttke, T. Devic, D. Heurtaux, G. Maurin, A. Vimont, M.

Daturi, O. David, E. Magnier, N. Stock, Y. Filinchuk, D. Popov, C. Riekel, G. Férey

and C. Serre, J. Am. Chem. Soc., 2011, 133, 17839-17847.

17. D. J. Tranchemontagne, J. L. Mendoza-Cortes, M. O'Keeffe and O. M. Yaghi, Chem.

Soc. Rev., 2009, 38, 1257-1283.

18. S. R. Caskey and A. J. Matzger, Inorg. Chem., 2008, 47, 7942-7944.

19. B. Zhao, X.-Y. Chen, P. Cheng, D.-Z. Liao, S.-P. Yan and Z.-H. Jiang, J. Am. Chem.

Soc., 2004, 126, 15394-15395.

20. B. Zhao, P. Cheng, Y. Dai, C. Cheng, D.-Z. Liao, S.-P. Yan, Z.-H. Jiang and G.-L.

Wang, Angew. Chem. Int. Ed., 2003, 42, 934-936.

21. J.-W. Cheng, J. Zhang, S.-T. Zheng, M.-B. Zhang and G.-Y. Yang, Angew. Chem. Int.

Ed., 2006, 45, 73-77.

22. E. D. Bloch, D. Britt, C. Lee, C. J. Doonan, F. J. Uribe-Romo, H. Furukawa, J. R. Long

and O. M. Yaghi, J. Am. Chem. Soc., 2010, 132, 14382-14384.

23. S.-H. Cho, B. Ma, S. T. Nguyen, J. T. Hupp and T. E. Albrecht-Schmitt, Chem.

Commun., 2006, 2563-2565.

169

24. C. J. Doonan, W. Morris, H. Furukawa and O. M. Yaghi, J. Am. Chem. Soc., 2009, 131,

9492-9493.

25. S. J. Garibay, Z. Wang and S. M. Cohen, Inorg. Chem., 2010, 49, 8086-8091.

26. F. Song, C. Wang, J. M. Falkowski, L. Ma and W. Lin, J. Am. Chem. Soc., 2010, 132,

15390-15398.

27. K. C. Szeto, K. P. Lillerud, M. Tilset, M. Bjørgen, C. Prestipino, A. Zecchina, C.

Lamberti and S. Bordiga, The Journal of Physical Chemistry B, 2006, 110, 21509-

21520.

28. K. C. Szeto, C. Prestipino, C. Lamberti, A. Zecchina, S. Bordiga, M. Bjørgen, M. Tilset

and K. P. Lillerud, Chem. Mater., 2007, 19, 211-220.

29. R. Weinland and H. Holtmeier, Z. Anorg. Allg. Chem., 1928, 173, 49-62.

30. S. Bauer, C. Serre, T. Devic, P. Horcajada, J. r. m. Marrot, G. r. Férey and N. Stock,

Inorg. Chem., 2008, 47, 7568-7576.

31. U. Kumar, J. Thomas and N. Thirupathi, Inorg. Chem., 2009, 49, 62-72.

32. A. Fateeva, P. Horcajada, T. Devic, C. Serre, J. Marrot, J.-M. Grenèche, M. Morcrette,

J.-M. Tarascon, G. Maurin and G. Férey, Eur. J. Inorg. Chem., 2010, 2010, 3789-3794.

33. Y. J. Kim and C. R. Park, Inorg. Chem., 2002, 41, 6211-6216.

34. M. I. Zaki, M. A. Hasan, F. A. Al-Sagheer and L. Pasupulety, Colloids Surf., A, 2001,

190, 261-274.

35. K. Nakamoto, in Infrared and Raman Spectra of Inorganic and Coordination

Compounds, John Wiley & Sons, Inc., 2008, pp. 1-273.

36. K. N. Wong and S. D. Colson, J. Mol. Spectrosc., 1984, 104, 129-151.

37. M. Yazdanbakhsh, H. Tavakkoli, M. Taherzadeh and R. Boese, J. Mol. Struct., 2010,

982, 176-180.

38. L. Meesuk, U. A. Jayasooriya and R. D. Cannon, Spectrochimica Acta Part A:

Molecular Spectroscopy, 1987, 43, 687-692.

39. A. B. Blake and A. Yavari, J. Chem. Soc., Chem. Commun., 1982, 1247-1249.

40. A. B. Blake, A. Yavari, W. E. Hatfield and C. N. Sethulekshmi, J. Chem. Soc., Dalton

Trans., 1985, 2509-2520.

41. G. Horváth and K. Kawazoe, J. Chem. Eng. Jpn., 1983, 16, 470-475.

42. K. Sumida, D. L. Rogow, J. A. Mason, T. M. McDonald, E. D. Bloch, Z. R. Herm, T.-

H. Bae and J. R. Long, Chem. Rev., 2011, 112, 724-781.

43. G. Novitchi, F. Riblet, R. Scopelliti, L. Helm, A. Gulea and A. E. Merbach, Inorg.

Chem., 2008, 47, 10587-10599.

44. M. Banerjee, S. Das, M. Yoon, H. J. Choi, M. H. Hyun, S. M. Park, G. Seo and K. Kim,

J. Am. Chem. Soc., 2009, 131, 7524-7525.

45. H. Irving and R. J. P. Williams, J. Chem. Soc., 1953, 3192-3210.

170

Supporting information

Preparation

In a typical synthesis, 0.67 mmol of FeCl3.6H2O 99%, 0.33 mmol of corresponding

Ni(NO3)2.6H2O 97% and 1 mmol of bdc 98% were dissolved in 10 ml of DMF. To this

clear solution, 0.4 mmol of NaOH was added under stirring for 15 min. The mixture was

then transferred into an Teflon-lined autoclave and heated at 100 oC for 15 h. Solid product

was then recovered by filtration and washed several times with DMF.

Characterization

N2 and CO2 adsorption tests were carried out in an Autosorb 1 instrument, before analysis

the samples were outgassed in vacuum for 3 hours at 150 oC. Specific surface area was

calculated with the BET model in the linear range of P/Po = 0 – 0.15. KBr solid state

FTIR was carried in a FT-BIORAD 450s system. MgO solid state UV-VIS was carried in a

Cary 300 instrument. Powder X-ray diffraction (XRD) patterns were collected on a Bruker

SMART APEX II X-ray diffractometer with Cu Kα radiation (λ = 1.5406 Å) in the 2θ

range of 4 – 20° at a scan rate of 1.0° min–1

. For XRD measurement of samples in Figure 3

and for crystal lattice calculation, the samples were dried in vacuum overnight at 100 oC,

then the analysis was taken immediately. Peak fitting was carried out using Jade software

package (http://www.materialsdata.com/).Simulation of Fe3-MIL-88B XRD pattern was

done on the crystalography data reported by Férey et al 1 using Mercury software package

(https://www.ccdc.cam.ac.uk/products/mercury/) Scanning electron microscopy (SEM)

images were taken on a JEOL 6360 instrument at accelerating voltage of 3 kV

FTIR

http://www.materialsdata.com/

https://www.ccdc.cam.ac.uk/products/mercury/

171

Figure S1. FTIR spectra of Fe2Ni-MIL-88B.DMF

Figure S2. FTIR spectra of Fe2Ni-MIL-88B.Py

400600800100012001400160018002000

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

1.1

Wavenumbers [1/cm]

Tra

nsm

itta

nce

400600800100012001400160018002000

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

1.1

Wavenumbers [1/cm]

Tra

nsm

itta

nce

asym

C

Py

sym

(OCO)

asym

C

Fe2NiO

sym

(OCO)

DMF

Fe2NiO Py

172

Figure S3. FTIR spectra of Fe2Ni-MIL-88B.Pz

Figure S4. FTIR spectra of Fe2Ni-MIL-88B.Bp

400600800100012001400160018002000

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

1.1

Wavenumbers [1/cm]

Tra

nsm

itta

nce

400600800100012001400160018002000

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

1.1

Wavenumbers [1/cm]

Tra

nsm

itta

nce

asym

C

asym

C

sym

(OCO)

sym

(OCO)

Fe2NiO

Fe2NiO

Pz

Bp

Bp

173

Figure S5. FTIR spectra of Fe2Ni-MIL-88B.H2O

Figure S6. FTIR spectra of Fe3-MIL-88B.DMF

400600800100012001400160018002000

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

1.1

Wavenumbers [1/cm]

Tra

nsm

itta

nce

400600800100012001400160018002000

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

1.1

Wavenumbers [1/cm]

Tra

nsm

itta

nce

asym

C

asym

C

sym

(OCO)

sym

(OCO)

Fe2NiO

Fe2NiO

Bp

174

Figure S7. Images of Fe2Ni-MIL-88B.Bp (a), Fe2Ni-MIL-88B.Pz (b), Fe2Ni-MIL-88B.Py

(c), Fe2Ni-MIL-88B.DMF (d) Fe2Ni-MIL-88B.H2O (e) and Fe-MIL-88B.DMF (f).

d

b

d c

f e

a

175

Figure S8. SEM image of as-synthesized Fe2Ni-MIL-88B.DMF

Ligand exchange reactions:

+ DMF => H2O: 0.25 g of Fe2Ni-MIL-88B.DMF was added with 10 ml of water. The

obtained mixture was stirred at room temperature for 3 hours, and then the solid was

recovered by filtration and dried at 100 ⁰C overnight. XRD pattern showed in Figure 3e

+ H2O=>DMF: 0.25 g of Fe2Ni-MIL-88B.H2O was added with 10 ml of DMF. After

stirring for 30 min, the mixture was transferred into an autoclave and it was placed in oven

176

at 110 ⁰C for 3 days. The product was recovered by filtration and washed several times

with DMF (Figure S9), BET specific surface area: 320 m2/g.

Figure S9: XRD pattern of Fe2Ni-MIL-88B.DMF obtained from Fe2Ni-MIL-

88B.H2O

+ DMF=>Py: 0.25 g of Fe2Ni-MIL-88B.DMF was added with 10 g of pyridine. Very

quickly the solid changed its color from brown to olive green. The mixture was stirred for 3

hours. The product was filtrated and dried in vacuum at 100 oC overnight. XRD pattern

showed in Figure 3c.

177

+ Py = > DMF. 0.25 g of Fe2Ni-MIL-88B.Py was added with 10 ml of DMF. The mixture

was stirred at 100 oC for 3 days while the color gradually changed from olive green to

yellow. The product was filtrated, washed with DMF and dried in vacuum at 100 oC

overnight (Figure S10). BET specific surface area: 340 m2/g.

Figure S10: XRD pattern of Fe2Ni-MIL-88B.DMF obtained from Fe2Ni-MIL-88B. Py

+ DMF = > Pz: 0.25 g of Fe2Ni-MIL-88B.DMF was added with 10 g of pyrazine. The

mixture was heated to 70 oC as pyrazine melted, stirring was applied for 3 hours. The olive

green product was recovered by hot filtration and dried in vacuum at 100 oC overnight.

XRD pattern showed in Figure 3b.

178

+ Pz => DMF: 0.25 g Fe2Ni-MIL-88B.Pz was added with 10 ml of DMF, the mixture was

transferred into an autoclave and heated at 100 oC for 3 days. Brown solid product was

filtrated and washed with DMF before drying in vacuum at 100 o

C overnight (Figure S11).

BET specific surface area: 325 m2/g.

Figure S11: XRD pattern of Fe2Ni-MIL-88B.DMF obtained from Fe2Ni-MIL-88B.

Pz

179

+ Pz = > H2O: 0.25 of Fe2Ni-MIL-88B.Pz in a vial was added with 15 ml of water. The vial

was then sealed and stirred at 95 oC. After 3 hours the brown product was filtrated and

dried in vacuum at 100 oC overnight (Figure S12). BET specific surface area: 15 m

2/g.

Figure S12: XRD pattern of Fe2Ni-MIL-88B.H2O obtained from Fe2Ni-MIL-88B. Pz

+ H2O => Pz: 0.25 g of Fe2Ni-MIL-88B.H2O was added with 10 g of pyrazine. The mixture

was heated at 100 oC for 3 days. The olive green product was recovered by filtration and

dried in vacuum at 100 oC overnight (Figure S13). BET specific surface area: 420 m

2/g.

180

Figure S13: XRD pattern of Fe2Ni-MIL-88B.Pz obtained from Fe2Ni-MIL-88B. H2O

+ Py => H2O: 0.25 of Fe2Ni-MIL-88B.Py in a vial was added with 15 ml of water. The vial

was then sealed and stirred at 95 oC. After 3 hours the brown product was filtrated and

dried in vacuum at 100 oC overnight (Figure S14), BET specific surface area: 10 m

2/g.

181

Figure S14: XRD pattern of Fe2Ni-MIL-88B.H2O obtained from Fe2Ni-MIL-88B. Py

+ H2O => Py: 0.25 g of Fe2Ni-MIL-88B.H2O was added with 10 ml of pyridine. The

mixture was sealed in a vial and stirred at 100 oC for 4 days. Olive green product was

filtered and dried in vacuum at 100 oC overnight (Figure S14). BET specific surface area:

530 m2/g

182

Figure S15: XRD pattern of Fe2Ni-MIL-88B.Py obtained from Fe2Ni-MIL-88B. H2O

+ DMF => Bp: 0.16 g of bipyridine was introduced to 2 ml of DMF, to this solution 0.12 g

of Fe2Ni-MIL-88B.DMF was added. The mixture was then stirred at 100 oC for 4 days.

Olive green product was filtered and dried in vacuum at 100 oC. XRD pattern showed in

Figure 3a

183

Figure S16: XRD pattern of Fe2Ni-MIL-88B.DMF obtained from Fe2Ni-MIL-88B.Bp

+ Bp => DMF: 0.12 g of Fe2Ni-MIL-88B.Bp was added with 15 ml of DMF, the mixture as

transferred into an autoclave and it was placed in an oven at 100 oC for 6 days. Brown

product was filtered and washed with DMF before drying in vacuum overnight at 100 oC

(Figure S16). BET specific surface area: 330 m2/g.

References

(1) S. Bauer, C. Serre, T. Devic, P. Horcajada, J. r. m. Marrot, G. r. Férey, N. Stock,

Inorg. Chem. 2008, 47, 7568

185

Chapter 7. Direct Synthesis and Mechanism for the

Formation of Mixed Metal Fe2Ni-MIL-88B

Gia-Thanh Vuong, Minh-Hao Pham and Trong-On Do *


Submitted to CrystEngComm 2013.

Résumé

Le mécanisme de synthèse a été étudié pour la synthèse Fe3-MIL88B et Fe2Ni-

MIL88B. Ces matériaux ont été caractérisés par différentes techniques, telles que la

spectroscopie UV-Vis, IR et Raman et la diffraction des RX. Les résultats montrent que

pour la synthèse de Fe3-MIL88B, le mono-métal Fe3-MOF-235 se forme en première étape

de synthèse et joue le rôle de précurseur (germe) pour la formation de Fe3-MIL-88B: les

germes MOF-235 sont formés, ensuit se transforment en Fe3-MIL-88B. Dans le cas

d‟utilisation du cluster de métaux mixtes Fe2Ni(µ3-O), les mono-métaux Fe3-MOF-235

formés en premier étape jouent le rôle comme germes pour la croissance de matériau

Fe2Ni-MIL88B. L‟anion FeCl4- est très important pour le succès de la formation de MOF-

235. Un mécanisme d‟anion médiateur dans la formation de MOF-235 a été suggéré.

Abstract

The direct synthesis of Fe3-MIL-88B and Fe2Ni-MIL-88B was analyzed using

different characterization techniques including UV-Vis, IR, Raman spectroscopies and

XRD. It was found that single metal Fe3-MOF-235 seeds which were formed from the first

stage of synthesis are as precursors for the formation of MIL-88B. Fe3-MOF-235 seeds

formed in the first stage of synthesis were then transformed to Fe3-MIL-88B in the case of

single metal, and to mixed Fe2Ni-MIL88B in the case of mixed metal synthesis. In the both

cases of Fe3-MIL-88B and Fe2Ni-MIL-88B, FeCl4- anion is a key feature to the formation

of MOF-235. An anion mediated mechanism for the formation of MOF-235 structure is

also suggested.




186

7.1. Introduction

Metal-organic frameworks (MOFs) are one of the fastest growing fields of

chemistry.[1] The structure of MOFs is formed by a polymeric connection of a metal

cluster in coordination bond with an organic linker, which results in a vast collection of

MOFs.[2] As a typical MOF structure, MIL-88B is of special interest due to its potentials in

adsorption, catalysis, biomedicine.[3-7] Its structure is built on the connection of 1,4-

bezenedicarboxylate (bdc) with trinuclear oxo-centered metal cluster (Me3O, Me = Fe, Cr,

Sc). Several trinuclear metal clusters have been reported in successful synthesis of MIL-

88B yielding Fe3-MIL-88B, Cr3-MIL-88B and Sc3-MIL88B.[6, 8, 9] In our recent

publication,[10] a novel route to prepare a new type of mixed metal MIL-88B structure was

reported. Unlike the conventional negative charged single metal cluster, the use of neutral

mixed metal cluster as nodes in the framework avoids the need of compensating anion

inside porous MIL-88B system. As a result, this mixed metal MIL-88B becomes porous.

Furthermore, the flexibility of the mixed metal MIL-88B can be controlled by terminal

ligands with different steric hindrance. This allows us to reversibly customize the porosity

of MIL-88B structure at three levels of specific surface area as well as the pore volume.[10]

In the synthesis of trinuclear-based MOFs such as MIL-88B and MIL-101, when

iron and 1,4-bezenedicarboxylic acid (bdc) are used in the presence of DMF, the kinetically

and thermodynamically stable phases are MIL-88B and MIL-101 which form at low

temperatures (≤100 °C), and thermodynamically stable MIL-53, which forms at higher.[11,

12] When a second metal is introduced in the reaction medium to obtain the mixed metal

MOF, the situation becomes more complicated. There are competing reactions to yield

single metal and mixed metal MOFs. It is also necessary to determine which factors

promote the single metal MOF and which factors promote the mixed metal MOFs and why.

A detailed study of the synthesis of both the single metal and mixed metal of trinuclear

based MOF could contribute to the understanding of the kinetics and mechanism of the

MOF formation

In this study, as the continuing part of the previous publication,[10] we report the

effect of several factors on the formation of both single metal Fe3-MIL-88B and mixed

metal Fe2Ni-MIL-88B. It occurred to us that even though the synthesis of mixed metal

187

MOFs is more complicated than that of the single metal one, the presence of the second

metal was found to be helpful in determination of the important initial solid Fe3-MOF-235,

revealing a vital template effect from a surprising source: halogen anion. The impacts of

pH, concentration, second metal were considered and illustrated. We also found similarities

in the principles of synthesis between zeolites and MOFs, which can be used as a guideline

for the synthesis of porous materials.

7.2. Experimental Section

Chemicals: FeCl3.6H2O (99%) and Fe(NO3)3.9H2O (98%), 1,4-bezenedicarboxylic

acid (bdc, 98%), NaOH (99%), N,N-dimethyl-formamide (DMF) were used as purchased.

Synthesis: Syntheses of Fe3-MIL-88B and Fe2Ni-MIL-88B were carried out

following our previous work.[10] Two series of samples with different times of

crystallization were prepared using two different iron sources, FeCl3.6H2O and

Fe(NO3)3.9H2O: single metal and mixed metal based MIL-88B. (i) For single metal based

MIL-88B synthesis: 5 vials of 10 ml DMF solution containing 10 mmol of FeCl3.6H2O

99% (or Fe(NO3)3.9H2O 98%) was added with 10 mmol 1,4-bezenedicarboxylic acid (bdc)

under stirring at room temperature. Subsequently, 4 ml of NaOH 2M was rapidly injected

under continuous stirring. The vials were then capped and heated at 100 oC for different

times: 0 h (e.g., 5 min after the addition of NaOH at room temperature), 1h, 2h, 3h, and 12

h. (ii) For mixed metal based MIL-88B synthesis: the same procedure was also applied for

the synthesis of mixed metal MIL-88B, except that 10 ml DMF solution containing 3.33

mmol of Ni(NO3)2.6H2O and 6.67 mmol of FeCl3.6H2O 99% (or Fe(NO3)3.9H2O 98%)

were used. Solids products were recovered by centrifugation at 5000 rpm for 5 min. The

solids were then dried in vacuum for 24 h at 50 oC. In general, the samples prepared with

FeCl3.6H2O yield firm solids. However, those prepared from Fe(NO3)3.9H2O become thick

gel during the heat treatment and thus, their corresponding solid products are not as firm as

the solids prepared from FeCl3.6H2O. The samples are designated as [Metal cluster]-[Anion

type]-[Synthesis time]. For example, Fe2Ni-Cl-5h is the mixed metal MIL-88B sample

prepared at 100 oC for 5 h using FeCl3

.6H2O.

188

Characterization Methods: FTIR was carried in a FT-BIORAD 450s instrument

using KBr disc. FTIR spectra were normalized by setting the transmittance value of the

band at 750 cm-1

which represents the strong vibration of the C-H bond to 0.05. UV-Vis

analysis was carried out in a Cary 300 instrument using MgO disc as the reference sample.

Normalization of the spectra was done by setting the value of the strongest absorbance

band at 350 nm to 1. Raman analysis was carried out with a Horiba U100 Raman

spectrometer using excitation wavelength of 514 nm. The spectra were normalized by

setting the value of the strong absorbance band of the benzene ring at 1615 cm-1

to 1.

Powder X-ray diffraction (XRD) patterns were collected on a Bruker SMART APEX II X-

ray diffractometer with Cu Kα radiation (λ = 1.5406 Å) in the 2θ range of 1 - 20° at a scan

rate of 1.0° min-1

. HRTEM analysis was carried out with a Hitachi HF-2000 Field Emission

TEM. EDS analysis was done with a fine electron probe of 3 nm, the acquiring time was set

to 200 sec. Samples were dispersed on a copper grid.

7.3. Results

UV-Vis spectra: UV-Vis spectra of the two series of single metal and mixed metal

based samples are shown in Figure 7.1 and Figure 7.2. For the single metal synthesis, the

UV-Vis spectra of the samples do not change much over the synthesis times; it is likely that

the Fe3+

species (octahedral configuration) remains the same throughout the synthesis. For

the mixed metal samples, most of the transition bands of Ni2+

are obscured or overlapped

by those of Fe3+

, however, the presence of octahedral Fe3+

is verified. The transition [6A1g

=> 4A1g +

4Eg(G)] in Fe

3+ is found at 350 - 500 nm and the [

6A1g =>

4T2g] transition at 550 -

650 nm is also attributed to Fe3+

.[13, 14] In contrast to the single metal, the UV-Vis

spectra of the mixed metal samples change considerably with the synthesis time. In the

Fe2Ni complex under the effect of Ni,[13] the ligand field in Fe reduces, thus the [6A1g =>

4T2g] transition in single metal samples (Fe3O) is at 525 nm while it is observed at 575 nm

in the Fe2Ni. In the spectra of mixed metal Fe2Ni-MIL-88B, the band at 760 nm is

observed. This band is characteristic of the transition [3A2g =>

1Eg(D)] of Ni in the tri-

nuclear cluster [13] and thus it can be used as an indicator of the formation of the mixed

metal cluster Fe2NiO. The evolution of the UV-Vis spectra over synthesis time of both

syntheses using Fe(NO3)3.9H2O and FeCl3.6H2O exhibits a similar trend. At 0 h, the spectra

189

are similar to the single metal samples; no band at 760 nm is observed. However after 3 h,

the band 760 nm is found in two series of mixed metal samples prepared using

Fe(NO3)3.9H2O and FeCl3.6H2O, the intensity of this band increases with synthesis time,

up to 12 h (Fig. 2). This suggests that at the early stage of synthesis, only Fe3+

is present in

the solid, and after that the mixed metal cluster Fe2Ni is formed. In other words, the

formation of Fe3O cluster takes place first, followed by that of the mixed metal cluster

Fe2NiO. The difference between the mixed metal syntheses using Fe(NO3)3.9H2O and

FeCl3.6H2O is the rate of the formation of Fe2NiO cluster. For the sample prepared using

Fe(NO3)3.9H2O, the band at 760 nm is very weak even after 12 h of synthesis; however,

for the sample prepared using FeCl3.6H2O, this band is readily observed after only 2 h of

synthesis and it still remains its strong intensity after 12 h (Figure 7.2).

FT-IR spectra: The FTIR band at 1660 cm-1

which is characteristic of DMF was

observed on all of the samples (Supporting information), as also reported in ref [15]. The

presence of free bdc linker (FTIR band at 1700 cm-1

) [16] was observed on the samples

prepared from Fe(NO3)3.9H2O but not on those prepared from FeCl3.6H2O. This absence of

free acid bdc in the Cl- based samples is interesting given the fact that no effort was made

to wash the obtained solid off free bdc acid. This observation implies that under these

investigated conditions; H2BDC is mostly deprotonated when Cl- is used. The free bdc

observed in the NO3- based samples also relates to its gel-like behavior of the products

while the Cl- based synthesis produces firmly solid products.

Detailed analysis was focused on the wavelength range from 400 – 800 cm-1

which

includes the well-documented framework vibration of the trinuclear cluster.[17, 18] The

FTIR band assignments in this range are showed in Table 7.1. Beside the presence of the

vibrations of the organic ligand bdc, the vibration of the central oxygen in single Fe3O and

mixed Fe2NiO clusters at 720 and 620 cm-1

was observed, respectively. Since the central

oxygen is available only in the trinuclear clusters, these FTIR bands are considered as their

indicators. And the presence of Fe3O and Fe2NiO can be distinguished.

For the single metal samples at different synthesis times, only the Fe3O vibration

(600 – 625 cm-1

) which is of interest is shown in Figure 7.3. The vibration of Fe3O is

observed in all the samples. The band is well defined at synthesis time 0 h and continues to

190

remain strong at the synthesis time 12 h. This implies that the cluster Fe3O is formed fast

and remains stable under the synthesis conditions. For the samples that employ Cl-, the

FTIR band characterisric of Fe3O is more intense and well-defined than those using NO3-.

The FT-IR spectra of the mixed metal samples are shown in Figure 7.4. The spectra

exhibit a gradual development of Fe2NiO clusters, and concomitantly with a decrease of

Fe3O ones. For the samples using Cl-, at 0 h, (Figure 7.4B) after the addition of NaOH at

room temperature, the FTIR band of Fe3O was visible, while the band of Fe2NiO was not

found. However, after 1h, the FTIR band of Fe2NiO was observed and the one of Fe3O

decreased. At 2 h and 3 h, the FTIR band of Fe2NiO is prominent while that of Fe3O

becomes very weak. Finally after 12 h, only the well-defined band Fe2NiO was observed

while the band of Fe3O almost disappeared. For the samples using NO3- (Figure 7.4A), the

same trend was also observed. The FTIR band of Fe3O appears first right at 0 h and

decreases with the synthesis time. In contrast, the band of Fe2NiO was observed only after

3 h. Thus, using Fe(NO3)3.9H2O, the formation of Fe2NiO clusters is much slower than that

using FeCl3.6H2O. After 12 h, both FTIR bands of Fe3O and Fe2NiO were still observed.

In short, FTIR analysis of the mixed metal synthesis using NO3- (e.g.,

Fe(NO3)3.9H2O) or Cl- (e.g., FeCl3.6H2O) shows that both Fe3O and Fe2NiO clusters were

produced, first Fe3O then followed by Fe2NiO. For the mixed metal synthesis using Cl-,

with increasing synthesis time, the Fe2NiO cluster becomes the main one, while the Fe3O

cluster diminishes. In contrast, for the mixed metal synthesis using NO3-, both types of the

clusters are observed at the end of the synthesis.

Raman spectra: The similar results were also observed for the Raman spectra. The

bdc linker was found on all of the samples. The band at 631 cm-1

is assigned to the in-plane

bending of the carboxylate group OCO.[19-21] The weak band at 1125 cm-1

is attributed to

the vibration of C-COO.[22] The benzene ring in bdc gives rise to the vibrations at 860,

1146 and 11616 cm-1

. [19-21] The Raman spectra also reveal some key inorganic

components in the samples. The vibrations of the trinuclear cluster are found at two weak

bands 175 and 267 cm-1

.[23] The medium band at 430 cm-1

is assigned to the metal oxygen

bond.[17] The bands at around 567cm-1

found only on the mixed metal samples are

tentatively assigned to the asymmetric stretching mode of Fe2NiO cluster.[19-21]

191

However, the most interesting results in the Raman analysis are the determination of

the compensating anions for the Fe3O cluster, e.g. NO3- and FeCl4

-. As discussed earlier, for

the single metal Fe3O-based material, each Fe3O carboxylate cluster in the framework

needs an anion to balance the charge. A presence of Fe3O clusters in the framework implies

the necessary accompany of these anions. For synthesis using Fe(NO3)3.9H2O, nitrate is

likely the only anion available.

For the mixed metal syntheses using FeCl3.6H2O, beside the available nitrate which

comes from Ni(NO3)2, the source of anions can also include Cl- and FeCl4

-. Although it is

impossible to determine Cl- in Raman spectra, the detection of FeCl4

- and NO3

- is

feasible.[17] The stretching vibration of FeCl4- leads to a well-defined band at 330 cm

-

1.[17, 24, 25] For NO3

-, the symmetric N-O stretching vibrations give rise to the strong

band at 1044 cm-1

. [17, 26, 27] Details of the band assignments are listed in Table 2. For

the single and mixed metal syntheses using Fe(NO3)3.9H2O thus having only NO3- as the

anion source, the presence of nitrate is found on all the samples, as observed by the Raman

band at 1044 cm-1

. Nitrate anion content correlates well with Fe3O clusters in the

framework by the intensity of this Raman band at1044 cm-1

(Figure 7.6A). In the single

metal synthesis, the band 1044 cm-1

of nitrate remains stable, in correlation with the readily

formed Fe3O clusters (Figure 7.5A). In the mixed metal synthesis using Fe(NO3)3.9H2O,

the band 1044 cm-1

grows in strength and shape over the synthesis time, implying an

increase of the single metal cluster Fe3O (Figure 7.6A). Taking into account of the FTIR

spectra results, it suggests that although there is a competition from the formation of

Fe2NiO, the formation of the Fe3O cluster is still favored when Fe(NO3)3.9H2O is used. In

the mixed metal syntheses using FeCl3.6H2O, Raman spectra (Figure 7.6B) shows the

presence of anion FeCl4- by the Raman band at 330 cm

-1, which diminished during the

synthesis; however, there is no nitrate found in the sample, this fact implies the high

selectivity of FeCl4- and Cl

- anions over NO3

- of the clusters. In the single metal synthesis,

FeCl4- is found right at 0 h, however the Raman band at 330 cm

-1 of FeCl4

- decreases

sharply during the synthesis, after 3 h, this band of FeCl4- is vanished (Figure 7.5B). It is

likely that FeCl4- decomposes, providing additional iron source for the growing Fe3O

clusters and leaving Cl- as the balancing anion in the cluster. In the mixed metal synthesis

using FeCl3.6H2O, this Raman band of FeCl4- at 330 cm

-1 is much smaller. Also the

192

decomposition of FeCl4- is much faster, after 2 h, no significant band of FeCl4

- was

observed, implying the formation of mixed metal cluster readily dominates and no

compensating anion is needed as compared to that of the single metal one.

Thus, the spectroscopy data have allowed us to determine and distinguish Fe3O and

Fe2NiO clusters as well as the balancing anions that accompany the formation of Fe3O

clusters. The results from the spectroscopy analysis suggest that: (i) in the single metal

synthesis, Fe3O clusters as nodes in the MOF structure were formed, regardless of anion

used; (ii) in the mixed metal syntheses, when Cl- was used in the synthesis mixture, single

Fe3O clusters were formed at the first stage, followed by the formation of mixed metal

Fe2NiO in the framework, which subsequently becomes the main clusters with increasing

the synthesis time. At the initial stage, FeCl4- is the main balancing anion for the Fe3O

clusters in the framework. It is then decomposed. Finally, the Cl- becomes the balancing

anion for the Fe3O cluster. When NO3- was used, both Fe2NiO and Fe3O clusters were

produced, however, free bdc was found. Trinuclear Fe3O cluster exhibits high preference of

Cl- based anion (FeCl4

- or Cl

-) as balancing anion over NO3

-. It is however noted that the

spectroscopy data cannot reveal how the nodes and linkers arrange in space, in other words,

they cannot confirm the crystalline or amorphous structure of the solids. To determine the

structure and phase, the XRD analysis is needed.

XRD analysis: XRD patterns of the Fe3- NO3-x (A) and Fe3-Cl-x (B) samples as a

function of synthesis time are shown in Figure 7.7. The results revealed that, for single

metal syntheses at 100 o

C, the use of NO3- as anion in the synthesis mixture yields no

definitive structure even after 12 h of synthesis, while using Cl-, the Fe3-Cl-x sample

exhibits a gradual structure change from MOF-235 to MIL-88B as a function of synthesis

time. As seen in Figure 7.7B, the solid obtained right after the addition of NaOH (at 0 h)

exhibits readily the XRD pattern of the MOF-325 structure. And then in the next three

hours at 100 oC, its XRD pattern is much more intense implying the increase of its

crystallinity; and only MOF-235 phase was observed. However after 12 hours at 100 oC,

the XRD pattern shows a mixture of both MIL-88B and MOF-235.

The similar trend is also observed for the synthesis of mixed metal MIL-88B

(Figure 8). Again, Fe2Ni-NO3-x samples show essentially noncrystalline phase regardless

193

of the synthesis time (Figure 8A). In contrast, the sample Fe2Ni-Cl-x clearly exhibits faster

phase transition than the single metal synthesis. At 0 h after the addition of NaOH, although

the MOF-235 phase was dominant, the MIL-88B was observed with the presence of weak

peaks of the plane (101) and (002) (Figure 8B). After 1 h, the MIL-88B structure was

established as the prominent and characteristic planes (100) (101) (002) appeared.

However, the MOF-235 phase is still pronounced as its (101) plane is still intense. After 3

h, the MIL-88B phase became the major phase over the MOF-235 phase. The peaks

characteristic of MOF-235 phase were very weak or disappeared. Eventually after 12 h,

only MIL-88B in open form is present. Hence, the transformation from MOF-235 to MIL-

88B in the mixed metal synthesis is faster than that of the single metal synthesis.

High-resolution transmission electron microscopy (HRTEM) and energy-

dispersive X-ray spectroscopy (EDS) analysis: Other important information is whether

the Fe/Ni ratio varies along its crystal. For this purpose, HRTEM and EDS techniques were

employed; different Fe2Ni-Cl-12h crystals were observed. A representative HRTEM image

of a crystal of Fe2Ni-Cl-12h sample is shown in Figure 7.9. The crystal is an elongated

hexagonal bipyramid, which is 500 nm long and 80 nm wide. This crystal shape is

frequently encountered for MIL-88B.[28, 29] EDS spectra were acquired on a large number

of positions in the crystal. For example, as shown in Figure 7.9, a selection of 5 different

positions are shown, they are two positions (1 and 4) near the external part of the crystal

and three positions (2, 3 and 5) approaching the crystal center. The atomic ratios of Fe and

Ni from the EDS spectra are also displayed in

Table 7.3. The results exhibit that the Fe/Ni ratio is not identical but does vary

throughout the crystal. In terms of Fe/Ni ratio, the crystal is rich in Fe in the center but Ni

content increases as one move outward. At the outer part of the crystal, the Fe/Ni reaches

the value of 2, in agreement with the stoichiometric ratio of Fe2NiO cluster. This behavior

implies that the Fe2Ni-MIL-88B crystal indeed includes both Fe3O and Fe2NiO clusters in

the framework.

194

7.4. Discussion

General remarks: The phase selection and transition in the synthesis of MIL

materials have been investigated by several groups.[30-33] The first report on this matter

by Férey et al.[32] dealt with the synthesis of Fe based MIL-53 using dimeric metal cluster

as its framework nodes. The authors found that, MOF-235 is formed at the first stage of

synthesis, followed by the formation of MIL-53 product. As the SBUs of MIL-53 and

MOF-235 are different, it would be impossible for a solid phase transition from MOF-235

to MIL-53. It suggests that during the MIL-53 synthesis, MOF-235 dissolves releasing free

monomers to yield subsequent MIL-53. For the synthesis Cr-based MIL-53, Jhung et

al.[30, 31] reported that MIL-101, but not MOF-235 is the transient phase and is

subsequently converted into MIL-53 under the studied synthetic conditions. Recently,

Stavitski et al. [33] studied on the synthesis of MOFs based on tri-nuclear cluster of Al, the

authors‟ findings are in agreement with the previous reports by Férey and Jhung groups. In

fact, they also suggest that MOF-235 is the first to appear and then forms the MIL-101, and

finally they dissolve to yield MIL-53. Thus regardless of the metal cluster, a general rule is

observed: MOF-235, which is kinetically favored, will be appeared first and play as the

precursor to form MIL-101. Finally the most thermodynamically stable MIL-53 is formed

at the expense of both MIL-101 and MOF-235. And as MIL-101 and MOF-235 are built on

the same trinuclear cluster while MIL-53 employs the dimeric oxo-cluster, it is likely that

trinuclear cluster is kinetically favored over dimeric one, however, tri-nuclear cluster is

thermodynamically less stable than the dimeric one. To the best of our knowledge, no

information about the phase transformation and the formation MIL-88B during the

synthesis has been reported. Table 7.4 summarizes the crystal parameters of MIL-88B and

MOF-235. For single metal Fe, both MOF-235 and MIL-88B structures are cationic

framework of the same formula Fe3O(bdc)3. The difference in terms of composition is the

compensating anions which are Cl- in the MIL-88B and FeCl4

- in MOF-235. In regard to

topology, MOF-235 and MIL-88B are identical; they are both built on the acs net,[34]

having the same crystal and space group. In addition, thanks to the breathing capacity,[35]

the a lattice constant can increase from 11 Å to 14 Å, while the c constant decreases from

19 Å to 14 Å, accordingly. This flexible range of lattice constant of MIL-88B structure

comprehensively encompasses the lattice parameters of MOF-235. Hence topologically

195

speaking, structures of MIL-88B and MOF-235 are likely to be inter-convertible; one could

be converted to other by distortion without breaking the linker bdc and the tri-nuclear

cluster. And with this high similarity in structure of MIL-88B and MOF-235, it is likely

that MIL-88B would form fast at initial stages, but does it come before or after MOF-235?

In our study of the MIL-88B synthesis, a secondary metal (Ni) which can form stable

mixed metal complex with Fe, is introduced, the situation would become more

complicated, there would be the additional competition between mixed metal MIL and

single metal MIL. However, as it will be explained later, the secondary metal Ni turns out

to be useful, playing a role of spectroscopically labeled atom in the detection and

distinction of single and mixed trinuclear clusters and thus it is possible to distinguish

MOF-235 and MIL-88B structures.

Determination of phase transformation during the synthesis: As seen above in

the FTIR, Raman and UV-Vis spectroscopy analyses, all the syntheses at first produce

single metal solid (Fe) even in the case with the presence of secondary metal Ni. FTIR

spectra of the solids also revealed the tri-nuclear Fe3O cluster which is the building unit of

MOF-235 and MIL-88B. However, based on the XRD results, these initial solids exhibit

that only those prepared using the FeCl3.6H2O source are structural and were identified as

MOF-235, while no structural products were observed using Fe(NO3)3.9H2O as the Fe

source during the synthesis. Furthermore, for the Cl- based samples, the solid structure

gradually changes from MOF-235 to MIL-88B with increasing the synthesis time. Along

with this phase transformation, for the mixed metal synthesis, the mixed metal tri-nuclear

cluster Fe2NiO begins to appear. As seen in the FTIR spectra (Figure 7.4), the vibration of

mixed metal cluster becomes pronounced over the synthesis time, while the vibration of

single metal cluster diminishes. This accompanies with the decrease of FeCl4- anions as

shown in Raman spectra (Figure 7.6). Hence, the trend observed for the samples using

FeCl3.6H2O is as follows: MOF-235 comes first, and then MIL-88B which is more stable

comes later and gradually takes over the MOF-235. The MIL-88B develops thanks to the

transformation from the MOF-235 structure. The phase transformation from MOF-235 to

MIL-88B is also observed and is found to be fast. In addition, there are competing reactions

forming Fe3-MIL-88B and Fe2Ni-MIL-88B. The formation of single metal cluster Fe3O

dominates at the initial stages with the formation of MIL-235, and then Fe3-MIL-88B.

196

Subsequently, it is surpassed by the formation of mixed metal cluster Fe2NiO for the Fe2Ni-

MIL-88B synthesis. However, this behavior is not observed for the mixed metal synthesis

using Fe(NO3)3.9H2O. No MOF-235 was produced during the synthesis; especially,

amorphous solid product was yielded. Strictly speaking, Cl- is required in composition of

MOF-235. This also implies that MOF-235 would play an important role for the phase

transformation from MOF-235 to MIL-88B.

The HRTEM and EDS analyses show that in the mixed metal synthesis using Cl-,

the Fe3 and Fe2NiO clusters are not distributed separately in two kinds of crystals Fe3-MIL-

88B and Fe2Ni-MIL-88B, respectively. In fact these clusters reside in the same crystal in

which the Fe3O cluster prefers the center while the Fe2Ni ones take the outward place.

Taking into account the findings of the spectroscopy data and XRD analysis, it is suggested

that, the formation of Fe2Ni-MIL-88B at first starts with single Fe3O cluster in form of

MOF-235 and then Fe2NiO cluster comes in as the crystal grows. The kinetically favored

MOF-235 could be seeds or in other word precursors to subsequent growing of

thermodynamically stable MIL-88B structure. Without MOF-235, the formation of the

coordination framework is much more difficult as in the case of nitrate based synthesis.

Although there are readily Fe3 and/or Fe2NiO cluster in the synthesis mixture, the lack of

MOF-235 seeds as precursor results in an amorphous gel in the final product of the NO3-

based synthesis. The possible whole transformation is illustrated in Scheme 8.1.

Anion effect: When only nitrate is used, the solid products are amorphous on all

runs. However, when chlorate is introduced in the form of FeCl3.6H2O, MOF structures

were obtained under certain conditions, and high selectivity of Cl- is also observed, even in

the case of mixed metal synthesis, there are both NO3- and Cl

-, only Cl

- is present in the

final product. In our syntheses involving FeCl3.6H2O, FeCl4- as balancing anion is found at

the beginning of the syntheses then it gradually disappears. In fact, the necessity of halogen

for the synthesis of trinuclear MOF has been reported in the synthesis of the trinuclear

based MIL by several authors. [30-33],46-48

The first report on the synthesis of Cr3-MIL-101

mentioned the use of Cr(NO3)3.9H2O without chlorate, however, other halogen anion in the

form of HF is required.[36] Later, Jhung et al. found out that when CrCl3 is used, there is

no need of HF to obtain MIL-101(Cr).[30, 31] Reports on the synthesis of Al3-MIL-101

197

also emphasized the successful use of AlCl3 and the infertile use of Al(NO3)3.[33, 37]

Successful preparations of Fe3-MIL-101 also involve the use of FeCl3.6H2O.[12, 38, 39]

But from the experimental point of view, the difference in products obtained from different

anions deserves a rational explanation. The role of ion type in the synthesis of MOF has

been emphasized,[11] but so far, to the best of our knowledge, there is few comprehensive

investigation published. Férey et al. suggested a vague mineral role of F- and a possible

template effect of Cl-, but the detail is unknown.[11, 12, 40] The original formula of MOF-

235 is Fe3O(bdc)3.FeCl4 which requires Cl-, thus strictly speaking, nitrate cannot give

original MOF-235 structure. But the question is why a similar structure based one nitrate

Fe3O(bdc)3.NO3 was not observed, instead a gel is form. We believe that the difference lies

in the template effect of the anion. Let‟s consider the case when FeCl4- and NO3

- residing in

the pore structure. FeCl4- is a tetrahedron of which four vertexes are Cl anion and the center

is the Fe cation. The -1 charge of FeCl4- is distributed evenly in the four Cl anions. The

NO3- features a triangle structure with nitrogen in the center and three oxygens at the

corners. The -1 charge of NO3- is distributed evenly in the three O anions, each carrying -

2/3 charge. For every 3 Fe atoms in the cluster, one anion is needed to balance the charge.

It would be safe to assume that the charge is distributed evenly among the Fe atoms in the

framework. Hence a tetrahedral structure of FeCl4- would be better at balancing the charge

of its MOF surrounding than a flat structure of nitrate. However, to confirm this suggestion,

comprehensive theoretical calculation and simulation would be necessary.

A similar role of compensating anion can be found in zeolite science, the role of

counter ions is not only to balance the framework but also to initiate the ordering structure

in the nucleation.[41-43] As zeolite framework is positive charged, the cations will assume

the template role, organizing around themselves negative charged oligomers in an

energetically favored fashion, thus forming certain favored geometry. This idea has been

suggested since the early days of the zeolite science and has been consolidated and

developed ever since, and become widely accepted.[41-44] In a similar fashion, the

formation of MOF-235 could be started with the assembly of the positive charged metal

carboxylate clusters around the negative charged FeCl4- in a geometry that favors the

formation of the acs net of MIL-235. The bulkier and more spacious of FeCl4- is a template

guiding the formation of ordered structure, while the nitrate is less effective resulting in

198

disordered structure. The possible mechanism is illustrated in Scheme 8.2. Beside the

anions, the template effect can be drawn from the use of bulky and appropriate terminal

ligand such as CH3CN as reported by Choi et al.[28] Other possible ligands could be

pyridine, THF. Bulkier linkers could also provide a template effect.[45] The use of

functional bdc (NH2-bdc) could improve the selection of MOF-235 as well as MIL-88B

without the need of chlorate.[29] The steric hindrance of the function group could stabilize

and increasing the preference of MOF-235 structure over the amorphous solid.

7.5. Conclusion

The synthesis of Fe2Ni-MIL-88B provides us an opportunity to have a detailed

investigation of the synthesis of MOF. We found in it many aspects of crystal nucleation

and growth: phase transformation, phase selectivity, precursor and template. The key to the

explanation of all these phenomena is to understand the kinetic and thermodynamic

difference between Fe3O and Fe2NiO. Fe3O is kinetically favored while Fe2NiO is

thermodynamically favored. FeCl4- anion is suggested to be template for the formation of

MOF-235 as well as the MIL-88B structure. Although our suggestions on the mechanisms

and the anion effect are in agreement with the experiment results, a theoretical calculation

is actually needed.

We also notice resemblances between zeolite science and MOF science. In fact,

concepts and ideas that have been well developed in the synthesis of zeolite such as:

template, SBU (secondary unit building), seeding, aging etc. can be used and applied to

MOF synthesis. In return, it is hoped that advances in MOF science could also inspire new

discovery in zeolites.

† Electronic Supplementary Information (ESI) available: Details of FTIR spectra HRTEM

and EDS spectra. See DOI: 10.1039/b000000x/

199

References

1 H.-C. Zhou, J. R. Long, O. M. Yaghi, Chem. Rev. 2012, 112, 673.

2 H. Li, M. Eddaoudi, M. O'Keeffe, O. M. Yaghi, Nature 1999, 402, 276.

3 orma ar a X la r s i Xamena Chem. Rev. 2010, 110, 4606.

4 P. Horcajada, R. Gref, T. Baati, P. K. Allan, G. Maurin, P. Couvreur, G. Férey, R. E.

Morris, C. Serre, Chem. Rev. 2011, 112, 1232.

5 P. Horcajada, T. Chalati, C. Serre, B. Gillet, C. Sebrie, T. Baati, J. F. Eubank, D.

Heurtaux, P. Clayette, C. Kreuz, J.-S. Chang, Y. K. Hwang, V. Marsaud, P.-N. Bories,

L. Cynober, S. Gil, G. Ferey, P. Couvreur, R. Gref, Nat Mater 2010, 9, 172.

6 L. Mitchell, B. Gonzalez-Santiago, J. P. S. Mowat, M. E. Gunn, P. Williamson, N.

Acerbi, M. L. Clarke, P. A. Wright, Catal. Sci. Tech. 2013, 3, 606.

7 A. Dhakshinamoorthy, M. Alvaro, H. Chevreau, P. Horcajada, T. Devic, C. Serre, H.

Garcia, Catal. Sci. Technol. 2012, 2, 324.

8 P. Horcajada, F. Salles, S. Wuttke, T. Devic, D. Heurtaux, G. Maurin, A. Vimont, M.

Daturi, O. David, E. Magnier, N. Stock, Y. Filinchuk, D. Popov, C. Riekel, G. Férey,

C. Serre, J. Am. Chem. Soc. 2011, 133, 17839.

9 S. Surble, C. Serre, C. Mellot-Draznieks, F. Millange, G. Ferey, Chem. Commun. 2006,

284.

10 G.-T. Vuong, M.-H. Pham, T.-O. Do, Dalton Trans. 2013, 42, 550.

11 G. Ferey, Chem. Soc. Rev. 2008, 37, 191.

12 auer erre Devi or aja a r m arrot r re to k Inorg.

Chem. 2008, 47, 7568.

13 A. B. Blake, A. Yavari, J. Chem. Soc., Chem. Commun. 1982, 1247.

14 A. B. Blake, A. Yavari, W. E. Hatfield, C. N. Sethulekshmi, J. Chem. Soc., Dalton

Trans. 1985, 2509.

15 Y. J. Kim, C. R. Park, Inorg. Chem. 2002, 41, 6211.

16 A. Fateeva, P. Horcajada, T. Devic, C. Serre, J. Marrot, J.-M. Grenèche, M. Morcrette,

J.-M. Tarascon, G. Maurin, G. Férey, Eur. J. Inorg. Chem. 2010, 2010, 3789.

17 K. Nakamoto, Infrared spectra of inorganic and coordination compounds, Wiley-

Interscience New York 1986.

18 R. D. Cannon, R. P. White, in Prog. Inorg. Chem., Vol. 36, John Wiley & Sons, Inc.,

1988, 195.

19 G. Xue, Prog. Polym. Sci. 1994, 19, 317.



22 P. Larkin, Infrared and Raman Spectroscopy; Principles and Spectral Interpretation,

Elsevier, Amsterdam 2011.

23 M. K. Johnson, D. B. Powell, R. D. Cannon, Spectrochim. Acta, Part A 1981, 37, 995.

24 H. Hellman, R. S. Laitinen, L. Kaila, J. Jalonen, V. Hietapelto, J. Jokela, A. Sarpola, J.

Rämö, J. Mass Spectrom. 2006, 41, 1421.

25 M. S. Sitze, E. R. Schreiter, E. V. Patterson, R. G. Freeman, Inorg. Chem. 2001, 40,

2298.

26 J. Sanders-Loehr, W. D. Wheeler, A. K. Shiemke, B. A. Averill, T. M. Loehr, J. Am.

Chem. Soc. 1989, 111, 8084.

27 A. Ianoul, T. Coleman, S. A. Asher, Anal. Chem. 2002, 74, 1458.

28 W. Cho, S. Park, M. Oh, Chem. Commun. 2011, 47, 4138.

200

29 M.-H. Pham, G.-T. Vuong, A.-T. Vu, T.-O. Do, Langmuir 2011, 27, 15261.

30 N. A. Khan, S. H. Jhung, Cryst. Growth Des 2010, 10, 1860.

31 N. A. Khan, J. W. Jun, S. H. Jhung, Eur. J. Inorg. Chem. 2010, 2010, 1043.

32 F. Millange, M. I. Medina, N. Guillou, G. Férey, K. M. Golden, R. I. Walton, Angew.

Chem. Int. Ed. 2010, 49, 763.

33 E. Stavitski, M. Goesten, J. Juan-Alcañiz, A. Martinez-Joaristi, P. Serra-Crespo, A. V.

Petukhov, J. Gascon, F. Kapteijn, Angew. Chem. Int. Ed. 2011, 50, 9624.

34 N. W. Ockwig, O. Delgado-Friedrichs, M. O'Keeffe, O. M. Yaghi, Acc. Chem. Res.

2005, 38, 176.

35 C. Serre, C. Mellot-Draznieks, S. Surble, N. Audebrand, Y. Filinchuk, G. Ferey,

Science 2007, 315, 1828.

36 G. Férey, C. Mellot-Draznieks, C. Serre, F. Millange, J. Dutour, S. Surblé, I.

Margiolaki, Science 2005, 309, 2040.

37 P. Serra-Crespo, E. V. Ramos-Fernandez, J. Gascon, F. Kapteijn, Chem. Mater. 2011,

23, 2565.

38 P. F. G. S. C. G. R. C. P. Horcajada-cortes, US Patent 20100226991, 2010.

39 K. M. L. Taylor-Pashow, J. D. Rocca, Z. Xie, S. Tran, W. Lin, J. Am. Chem. Soc.

2009, 131, 14261.

40 D.-Y. Hong, Y. K. Hwang, C. Serre, G. Férey, J.-S. Chang, Adv. Funct. Mater. 2009,

19, 1537.

41 D. W. Breck, Zeolite molecular sieves: structure, chemistry, and use, Wiley, London

1973.

42 C. S. Cundy, P. A. Cox, Chem. Rev. 2003, 103, 663.

43 C. S. Cundy, P. A. Cox, Microporous Mesoporous Mater. 2005, 82, 1.

44 G.-T. Vuong, T.-O. Do, J. Am. Chem. Soc. 2007, 129, 3810.

45 P. Horcajada, F. Salles, S. Wuttke, T. Devic, D. Heurtaux, G. Maurin, A. Vimont, M.

Daturi, O. David, E. Magnier, N. Stock, Y. Filinchuk, D. Popov, C. Riekel, G. Férey,

C. Serre, J. Am. Chem. Soc. 2011, 133, 17839.

46 L. Meesuk, U. A. Jayasooriya, R. D. Cannon, Spectrochim. Acta, Part A 1987, 43, 687.

47 R. Wu, U. A. Jayasooriya, R. D. Cannon, Spectrochim. Acta, Part A 2000, 56, 575.

48 L. Montri, R. D. Cannon, Spectrochim. Acta, Part A 1985, 41, 643.

49 A. C. Sudik, A. P. Côté, O. M. Yaghi, Inorg. Chem. 2005, 44, 2998.

201

FIGURE CAPTIONS, SCHEMES AND TABLES

0 h

Wavelength (nm)

200 300 400 500 600 700 800

1h

2 h

3 h

Adsorp

tion (

a.u

.)

12 h

(A)

202

0 h

Wavelength (nm)

200 300 400 500 600 700 800

1h

2 h

3 h

12 h

(B)

Figure 7.1. UV-Vis spectra of the samples Fe3-NO3-x (A) and Fe3-Cl-x (B) prepared using

Fe(NO3)3.9H2O and FeCl3.6H2O, respectively, at different synthesis times.

203

0 h

Wavelength (nm)

200 300 400 500 600 700 800

3 h

2 h

1 h

Adsorp

tion (

a.u

.)

12 h

(A)

204

0 h

Wavelength (nm)

200 300 400 500 600 700 800

1h

2 h

3 h

Adsorp

tion (

a.u

.)

12 h

(B)

Figure 7.2. UV-Vis spectra of Fe2Ni-NO3-x (A) and Fe2Ni-Cl-x (B) samples prepared

using Fe(NO3)3.9H2O and FeCl3.6H2O, respectively at different synthesis times

205

Wavenumber cm-1

600610620630640650

Fe3O Fe3O

12 h12 h

3 h

3 h

2 h

2 h

1h 1 h

0 h 0h

(A)(B)

Wavenumber cm-1

600610620630640650

Figure 7.3. Transmittance FTIR spectra of the samples of Fe3-NO3-x (A) and Fe3-Cl-x (B)

at different synthesis times

206

Fe2NiO Fe3O

3 h

2 h

1 h

0 h

12 h (A)

Wavenumber [1/cm]

600625650675700725750

207

Fe2NiO Fe3O

3 h

2 h

1 h

0 h

12 h

Wavenumber [1/cm]

600625650675700725750

(B)

Figure 7.4. Transmittance FTIR spectra of Fe2Ni-NO3-x (A) and Fe2Ni-Cl-x (B) at

different synthesis times

208

0 h

Wavelength (1/cm)

10030050070090011001300150017001900

1 h

2 h

3 h

12 h

(A)

209

0 h

Wavelength (1/cm)

10030050070090011001300150017001900

1 h

2 h

3 h

12 h

(B)

Figure 7.5. Raman spectra of the samples Fe3-NO3-x (A) and Fe3-Cl-x (B) at different

synthesis times

210

0 h

Wavelength (1/cm)

10030050070090011001300150017001900

1 h

2 h

3 h

12 h

(A)

211

0 h

Wavelength (1/cm)

10030050070090011001300150017001900

1 h

2 h

3 h

12 h

(B)

Figure 7.6. Raman spectra of Fe2Ni-NO3-x (A) and Fe2Ni-Cl-x (B) at different synthesis

times.

212

0 h

3 h

12 h

(A)

2

5 8 11 14 17 20

2

5 8 11 14 17 20

12 h

3 h

2 h

1 h

0 h

#

*

#

#

*

#*

#

#

*

(B)

Figure 7.7. XRD patterns of Fe3-NO3-x (A) and Fe3-Cl-x (B) at different synthesis times.

(*) MOF-235 phase, (#): MIL-88B phase.

213

2

5 8 11 14 17 20

#

#

#

##

#

#

#

#

#

*

#

#

##

*

*#

# #

*

*

*

#

#

#

**

*

*

*

*

#

#

#

#

#

#

#

#

*

*

0 h

1 h

2 h

3 h

12 h

(B)

2

5 8 11 14 17 20

0 h

3 h

12 h

(A)

Figure 7.8. XRD patterns of Fe2Ni-NO3 (A) and Fe2Ni-Cl (B) at different synthesis time.

(*) MOF-235 phase, (#): MIL-88B phase

214

Figure 7.9. Representative HRTEM and EDS acquiring positions of Fe2Ni-Cl-12h crystal

1

2

3

5 4

215

SCHEMES

Scheme 8.1. Possible mechanism of the formation of MIL-88B samples using FeCl3.6H2O

216

Schem 8.2. Possible proposition of anion mediated formation of MIL-88B

217


Band (cm-1

) Assignment

750 C-H [20, 21]

720 Fe2NiO [46, 47]

690 C-C [20, 21]

660 OCO [20, 21]

624 Fe3O [23, 48]

550 Fe-O, Ni-O [17]

Table 7.2. Raman band assignments

Band (cm-1

) Assignment

175 nsym(M3O)

267 δasym(M3O)

330 FeCl4-

440 M-O

567 Fe2NiO

631 OCO

860 CC

1050 NO3-

1125 CX

1146 CH

1431 CH3 (bending mode in DMF)

1454 CH3 (bending mode in DMF)

1616 CC

Table 7.3. Fe and Ni atomic percentages calculated from EDS spectra

218

Position Fe (atomic %) Ni (atomic %) Fe/Ni

1 75 36 2.1

2 67 35 2.2

3 81 29 2.8

4 78 37 2.1

5 89 19 4.7

Table 7.4. Comparison of the crystal parameters of MIL-88B and MOF-235

Parameter MIL-88B [9, 12] MOF-235 [49]

Chemical formula Fe3O(bdc)3Cl.3DMF Fe3O(bdc)3FeCl4.3DMF

Molar weight (g.mol-1

) 905.8 1092.8

Net acs acs

Crystal system Hexagonal Hexagonal

Space group P -6 2 c P -6 2 c

a (Å) 11.1075 12.531

b (Å) 11.1075 12.531

c (Å) 19.0925 18.476

90 90

90 90

120 120

Volume (Å3) 2040.8 2512.6

219

Supporting Information

1. FTIR spectra of samples with wavelength 400 – 4000 cm-1

1.1.Single metal synthesis using Fe(NO3)3.9H2O

0 h

Wavenumber [cm-1

]

5001000150020002500300035004000

Tra

nsm

itta

nce

1 h

Wavenumber [cm-1

]

5001000150020002500300035004000

Tra

nsm

itta

nce

220

2 h

Wavenumber [cm-1

]

5001000150020002500300035004000

Tra

nsm

itta

nce

3 h

Wavenumber [cm-1

]

5001000150020002500300035004000

Tra

nsm

itta

nce

221

12 h

Wavenumber [cm-1

]

5001000150020002500300035004000

Tra

nsm

itta

nce

1.2.Single metal synthesis using FeCl3.6H2O

0 h

Wavenumber [cm-1

]

5001000150020002500300035004000

Tra

nsm

itta

nce

222

1 h

Wavenumber [cm-1

]

5001000150020002500300035004000

Tra

nsm

itta

nce

2 h

Wavenumber [cm-1

]

5001000150020002500300035004000

Tra

nsm

itta

nce

223

3 h

Wavenumber [cm-1

]

5001000150020002500300035004000

Tra

nsm

itta

nce

12 h

Wavenumber [cm-1

]

5001000150020002500300035004000

Tra

nsm

itta

nce

1.3.Mixed metal synthesis using Fe(NO3)3.9H2O

224

0 h

Wavenumber [cm-1

]

5001000150020002500300035004000

Tra

nsm

itta

nce

1 h

Wavenumber [cm-1

]

5001000150020002500300035004000

Tra

nsm

itta

nce

225

2 h

Wavenumber [cm-1

]

5001000150020002500300035004000

Tra

nsm

itta

nce

3 h

Wavenumber [cm-1

]

5001000150020002500300035004000

Tra

nsm

itta

nce

226

12 h

Wavenumber [cm-1

]

5001000150020002500300035004000

Tra

nsm

itta

nce

1.4.Mixed metal synthesis using FeCl3.6H2O

0 h

Wavenumber [cm-1

]

5001000150020002500300035004000

Tra

nsm

itta

nce

227

1 h

Wavenumber [cm-1

]

5001000150020002500300035004000

Tra

nsm

itta

nce

2 h

Wavenumber [cm-1

]

5001000150020002500300035004000

Tra

nsm

itta

nce

228

3 h

Wavenumber [cm-1

]

5001000150020002500300035004000

Tra

nsm

itta

nce

12 h

Wavenumber [cm-1

]

5001000150020002500300035004000

Tra

nsm

itta

nce

229

2. HRTEM and EDS analysis

EDS spectra of 5 positions taken on the same Fe2Ni-MIL-88B crystal (see Figure 9 in the

article).

Since the samples were dispersed on a copper grid for analysis. Signal of copper was

observed in the EDS spectra.

EDS spectroscopy of position 1



230



3. XPS analysis of Fe2Ni-MIL-88B-12h, dried

The X-ray photoelectron spectra (XPS) were taken on a photoelectron spectrometer

(KRATOS AXIS-ULTRA) with a monochromatic X-ray source of Al K . The operating

conditions for recording high-resolution spectra were as follows: 1486.6 eV and 225 W;

pass energy of 160 eV with anoperating pressure of 10-9

Torr.

CasaXPS software was use to analys the collected XPS spectra. 1 All spectra were

calibrated using the adventitious C 1s peak with a fixed value of 284.8 eV. Shirley

background was then applied and subtracted.

As the Fe cation in trinuclear cluster is at high spin state,2,3

envelope of Fe 2p3/2 spectrum

was fit with peaks corresponding to the GS multiplets, surface structures and shake-up-

related satellites.4-6

All the four GS multiplets have the same full width at half-maximum

(FWHM) of 1.6 eV and their peak areas are in similar to those of multiplets. The rest of the

envelop was filled with one surface structure peak and one satellites peak.

Fitting result is in agreement with Fe3+

GS multiplets , suggesting the presence of only Fe3+

not Fe2+

in the sample.

Gupta and Sen (GS) multiplet peak parameters used to fit the high-spin Fe3+

231

Sample

Peak 1 Peak 2 Peak 3 Peak 4

eV %

area E eV

%

area E

(eV) eV

%

area E

(eV) eV

%

area E

(eV)

Fe2Ni-MIL-88B-12h 713.2 7 1.3 711.9 21 1.1 710.8 35 1.5 709.3 37

GS Fe3+ multiplets 10 0.6 20 1.3 30 1.6 40

References

1. Fairley N. CasaXPS Version 2.2.19, copyright 1999–2003

2. Boudalis A. K. et al Polyhedron 2005, 24, 1540

3. Psycharis, V et al Eur. J. Inorg. Chem 2006, 2006, 3710-

4. Gupta RP, Sen SK. Phys. Rev. B. 1975; 12: 15

5. Grosvenor A.P et al. Surf. Interface Anal. 2004, 36, 1564

6. Mullet M. et al. Surf. Interface Anal. 2008, 40, 323

233

Chapter 8. Conclusion

For the study of nanozeolites, two novel methods, single-phase and two-phase

synthesis, have been developed for the synthesis of nanozeolites. Both of them employed

the use of organosilane agent to restrain the growth of zeolite during the crystallization

process thus, resulting in uniform nanozeolites. This growth control works on the principle

that organosilane can react with silanol groups which are abundant on the external surface

of nanozeolites. Upon reacting with organosilane, zeolite nanocrystals are functionalized

with the organic group of the organosilane. These functionalized crystals therfore cannot

incorporate more aluminosilicate species which would lead to the larger crystal.

In the two-phase synthesis, in addition to the aqueous medium, an organic solvent

was introduced. As the nanocrystals are functionalized with organic group and thus become

hydrophobic, it can disperse into the organic phase, and thus the growth process is

completely stopped. It is clear that the diffusion is strongly dependent on the hydophobicity

of the functionalized crystals. This parameter is governed by two opposite factors, the

crystal size (the hydrophilic factor) and the degree of functionalization (the hydrophobic

factor) of the functionalized crystals. Generally, increase in crystal size results in higher

value of the hydrophibicity factor, given the same degree of functionalization. In contrast,

the hydrophobic factor is favored by the increase in the external surface. As the results in

the two-phase synthesis, functionalized nanozeolites are more likely to be found in the

organic solvent than in the aqueous phase.

The two-phase synthesis provides a useful way to produce nanozeolites dispersed in

the organic phase. In contrast to the two-phase synthesis, the aim of the single-phase

method is to well disperse the zeolites synthesis solution in an organic solvent. This is

carried out with the aid of organosilane and n-butanol. Organosilane is used with the same

purpose as in the two-phase synthesis, i.e. to functionalize and “hydophobilize” the

produced nanozeolites. The introduction of n-butanol is intended to increase the dispersion

of the aqueous medium in the organic phase. In our study, careful selecting the organic

phase/inorganic phase ratio and mixing at proper temperature could lead to a single-phase

234

solution for further crystallization. Due to the absence of the separate aqueous phase as in

the case of the two-phase synthesis, only nanozeolites are produced, there is no trace of

large crystal in the obtained product. Using this method we have successfully synthesized

silicalite-1 and FAU nanozeolites.

The application potentials of our synthesized zeolites were also evaluated. The

catalyst was prepared by hydrolysis of TEOS in the presence of FAU nanozeolites to

incorporate the zeolites in the amorphous silica. The cracking reaction of FCC feed stock

was chosen to determine the catalytic activity of nanozeolites. The results showed that our

catalysts gave high activity and it is recommended that it can be used to improve the

activity of FCC catalysts. Direct application of nanozeolites as catalyst is not the only way

to utilize it. Another approach to maximize the potential of nanozeolite is to use them as

nano-unit to build superstructure materials. Some pioneering studies by prominent

researchers in the field have emphasized the significance of this approach: nanozoelite to

form zeolitic mesoporous-nanoporous materials, nanozeolite to form zeolitic nanosheets.

For our study of MOFs, we have showed that there are many opportunities to

customize these materials. As MOFs are composed of two components: the metal cluster

and the organic linker, the general strategy would be focused on these two. The common

one is to play with the organic linker while keeping the metal cluster. Due to its organic

nature, the linker can be modified in many aspects without much change in its reactivity in

regards to the connectivity to the metal clusters. Hence connectivities between metal cluster

and linker remain intact, thus the synthesis is likely to be successful. Almost all of

approaches to modify a linker have been used: functionalization, prolonging, isomerization

etc. yielding explosive number of new MOFs. The second strategy to modify the metal

clusters, are much less popular and less successful. That is not because the modification of

clusters is any less interesting but in fact it is much more difficult than modification of

linkers. Having second metal into MOF structure, hence mixed metal MOF is great but, the

complexity of the synthesis would rise abruptly. Different reactivities of metals do not

allow the incorporation of multi metals into one single MOF structure.

We believe that to increase the chance of success in the synthesis of mixed metal

MOF, it would be better to start with an original single metal MOF structure and apply the

235

modification of the metal cluster with the introduction of the second metal such that the

mixed metal cluster will retain the original shape and connectivity, hence, the obtained

mixed metal MOF will not suffer any disruption in connection due to the second metal. We

focused on MIL-88B which had been regarded as a dense structure to illustrate this

approach. The single metal cluster Fe3O in the original MIL-88B was replaced with mixed

metal Fe2NiO successfully to obtaine mixed metal Fe2Ni-MIL-88B. The real benefit of this

modification is clear as mixed MIL-88B becomes porous because the need of

accompanying and blocking anion is avoided by using mixed metal cluster. But another

significant change in the mixed metal MIL-88B is the improvement in bond strength

between the mixed metal clusters and the terminal ligands. This boost in bond strength has

allowed us to take advantage of the breathing effect of MIL-88B to control its porosity.

Strongly attached terminal ligands in mixed metal MIL-88B now become pillars to sustain

its structure. The larger the ligands are the greater the porosity is. We can control the

porosity of MIL-88B at three different levels. One may find this discovery contradicts

common sense that large molecules would block the pores instead of enlarging them. But

this fact just perfectly embodies how powerful and exciting the MOF materials are:

interesting and surprising properties are often found in MOFs.

Of course, this approach of using mixed metal can be applied to other MOFs. MIL-

88A, MIL-88C and MIL-88D are the obvious candidates. In fact it is possible to use mixed

metal cluster Fe2NiO to replace any single metal trimeric cluster of the type M3O. Other

mixed metal clusters of Fe(III), Cr(III) with divalent metal such as: Ca, Mg, Ba, Sr, Mn, Co

and Zn should be able for the synthesis of mixed metal MOFs. The breathing effect

combined with the controllable porosity of mixed metal MIL-88B would be of great

interests for membrane technology. It could allow creation of smart membrane that can

change its pore size “on the fly” increasing the selectivity. The terminal ligands attached to

mixed metal clusters are also a useful agent to functionalize MOFs. Function groups could

be introduced via terminal ligands and in return could be removable by ligand exchange.

Thus we can envisage mixed metal MIL-88B like a chemical “main board” which carries

various function group modules.

236

The study on the synthesis of mixed metal MOFs also led us to the discovery of the

underlying synthesis mechanism. The importance of anion sources has been recognized, but

never been well explained. In the synthesis of MOFs based trimeric cluster such as Fe3O as

in the case of MIL-88, MIL-101 families, the necessary of halogen anion was remarked but

how it contributes to the synthesis was not mentioned in literature. With the synthesis of

MIL-88B we found that a template should be available for the structure to form. The

template for MIL-88B is FeX4- (X is halogen) and that is why halogen ion is required.

FeX4- should be available in the synthesis so that the cationic frameworks of MIL-88B, the

precursor MOF-235 can be built around it. It would be better if we could do some

theoretical calculations to prove the preference in energy of our MIL-88B structure bearing

FeX4 over other anions such as nitrate, sulphate… But we would like to emphasize the

similarity between the well known cation template mechanism suggested by Breck and our

proposal of anion template mechanism. It turns out that knowledge of zeolites could be

very helpful in dealing with MOFs.

Another similar feature to zeolites in the synthesis of MOFs is that, the synthesis

involves phase transformations in which the more stable phase replace the less stable one,

the less stable phase is thus the precursor to the stable phase. For the synthesis of MIL-88B,

MOF-235 is the precursor phase, one should make sure the synthesis conditions favor the

formation of MOF-235 otherwise MIL-88B will not come out. In the synthesis of zeolite a

reactive gel phase may come and go in the middle of the process, acting as intermediate

phase before the formation of zeolite phase. This is possibly due to the gain in energy from

gel phase to zeolite phase is great enough. For MOFs, the weaker coordination bond does

not afford enough energy necessary for a transformation from a gel phase to a MOF phase.

This is why the intermediate phase, the precursor phase of MIL-88B should be another

MOF phase, as we can obtaine MOF-235 in a matter of several minutes of synthesis. Any

formation of gel phase at the initial stage of the synthesis would signal its failure since the

chance the gel phase can transform to MOF phase is unlikely.

In conclusion, we have showed in this study that both nanozeolites and MOFs are

potential nanoporous materials. There are plenty of ways to exploit them. Together they

form the foundation of the science of nanoporous materials. But is it possible to obtain

237

materials composed of both zeolites and MOFs? We think yes. As our nanozeolites are

readily functionalized with organic functions, it would possible to attach them to a MOF

“main board”. The result would be an ultimate super nanoporous materials and that is what

we are dreaming about now.

238

239

List of Pulications

1. G.T. Vuong and T.O. Do, A new route for the synthesis of uniform nanozeolites with

hydrophobic external surface in organic solvent medium. Journal of the American

Chemical Society (JACS), 2007. 129(13): p. 3810-3811.

2. G.T. Vuong and T.O. Do, Nanozeolites and process for preparation thereof, 2008,

WO Patent WO/2008/058,398.

3. G.T. Vuong., S. Kaliaguine, and T.O. Do, A strategy towards macroporous sponge-like

networks of metal oxide-surfactant mesophases and bulk metal oxides. Journal of

Porous Materials, 2008. 15(6): p. 679-683.

4. G.T. Vuong and T.O. Do, Synthesis of silylated nanozeolites in the presence of organic

phase: Two-phase and single-phase methods. Microporous and Mesoporous Materials,

2009. 120(3): p. 310-316.

5. G.T. Vuong., V.T. Hoang, D.T. Nguyen, and T.O. Do, Synthesis of nanozeolites and

nanozeolite-based FCC catalysts, and their catalytic activity in gas oil cracking

reaction. Applied Catalysis A: General, 2010. 382(2): p. 231-239.

6. Pham, M.H., G.T. Vuong, F.G. Fontaine, and T.O. Do, A Route to Bimodal Micro-

Mesoporous Metal–Organic Frameworks Nanocrystals. Crystal Growth & Design,

2011. 12(2): p. 1008-1013.

7. Pham, M.H., G.T. Vuong, A.T. Vu, and T.O. Do, Novel Route to Size-Controlled Fe-

MIL-88B–NH2 Metal–Organic Framework Nanocrystals. Langmuir, 2011. 27(24): p.

15261-15267.

8. Pham, M.H., G.T. Vuong, F.G. Fontaine, and T.O. Do, Rational Synthesis of Metal-

Organic Framework Nanocubes and Nanosheets Using Selective Modulators and Their

Morphology-Dependent Gas-Sorption Properties. Crystal Growth & Design, 2012.

12(6): p. 3091-3095.

9. G.T. Vuong, M.H. Pham, and T.O. Do, Synthesis and Engineering Porosity of mixed

metal Fe2Ni- MIL-88B Metal-Organic Framework. Dalton Transactions, 2013, 42, 550-

557.

10. G.T. Vuong, M.H. Pham, and T.O. Do, Direct Synthesis and Mechanism for the

Formation of Mixed Metal Fe2Ni-MIL-88B. CrystatEngComm, submitted, 2013.

Documents

Synthesis and characterization of nanoporous materials